The Challenge of Digital Transformation in The Automotive Industry-2020

The challenge of digital
transformation in the
automotive industry
Jobs, upgrading and the prospects
for development
—
Edited by
Jan Drahokoupil
The challenge of digital transformation in the automotive industry.
Jobs, upgrading and the prospects for development
The challenge of digital
transformation in the
automotive industry
Jobs, upgrading and the prospects
for development
—
Edited by
Jan Drahokoupil
ETUI publications are published to elicit comment and to encourage debate. The views expressed are those
of the authors alone and do not necessarily represent the views of the ETUI nor those of the members of its
general assembly.
Brussels, 2020
© Publisher: ETUI aisbl, Brussels
All rights reserved
Print: ETUI printshop, Brussels
D/2020/10.574/19
ISBN : 978-2-87452-569-8 (print version)
ISBN : 978-2-87452-570-4 (electronic version)
The ETUI is financially supported by the European Union. The European Union is not
responsible for any use made of the information contained in this publication.
Contents
Jan Drahokoupil
Chapter 1
Introduction: Digitalisation and automotive production networks in Europe ................................... 7
The headquarters perspective .............................................................................................................................. 23
Pamela Meil
Chapter 2
Inside looking out: Digital transformation in the German automobile sector and its
effects on the value chain ........................................................................................................................................ 25
Digital transformations in factory economies ............................................................................................. 45
Andrea Szalavetz
Chapter 3
Digital transformation and local manufacturing subsidiaries in central and
eastern Europe: Changing prospects for upgrading? ................................................................................... 47
Ricardo Aláez-Aller, Carlos Gil-Canaleta, Juan Carlos Longás-García and Miren Ullibarri-Arce
Chapter 4
Digitalisation and the role of MNC subsidiaries in the Spanish automotive industry ................. 65
Krzysztof Gwosdz, Grzegorz Micek, Arkadiusz Kocaj, Agnieszka Sobala-Gwosdz and

Agnieszka Świgost-Kapocsi
Chapter 5
Industry 4.0 and the prospects for domestic automotive suppliers in Poland ................................ 89
Andrea Szalavetz
Chapter 6
Digital entrepreneurs in factory economies: Evidence from the automotive industry
in Hungary .................................................................................................................................................................. 107
The challenge of digital transformation in the automotive industry 5

Impact on workers .................................................................................................................................................. 125
Matteo Gaddi
Chapter 7
Technological and organisational innovation under Industry 4.0 – Impact on working
conditions in the Italian automotive supply sector .................................................................................. 127
Monika Martišková
Chapter 8
The transformation of jobs and working conditions: Towards a policy response ....................... 153
List of contributors ..................................................................................................................................................... 177
6 The challenge of digital transformation in the automotive industry

Chapter 1
Introduction: Digitalisation and automotive production
networks in Europe
Jan Drahokoupil
This edited volume analyses the challenges of digital transformation, typically discussed
under the heading of Industry 4.0, from the perspective of production locations. Digital
transformation represents a dual challenge from such a perspective.
First, production locations face a fundamental challenge as far as their role in the
production network is concerned. Increased automation may undermine competitive
advantage in terms of lower labour costs and more flexible labour market arrangements.
New technology may also replace the more knowledge-intensive tasks conducted by
engineers in production locations, leading to the effective downgrading of the position
of manufacturing units in the value chain. At the same time, digital technologies may
support the decentralisation of advanced activities across the production network,
allowing production sites to be upgraded through advanced manufacturing technologies.
Second, there is the jobs challenge. It has been particularly noticed that countries
specialising in production are particularly vulnerable to the job displacement effects
of new technologies as many of the tasks in which they specialise can be automated.
However, new technology also changes the nature of work in production. This entails
changes in the nature of the skills required from workers and the autonomy they have
in carrying out tasks, and also in work intensity.
The digital transformation of production is associated with so-called Industry 4.0

technologies. These combine data analytics, the Internet of Things and production
machinery into cyber-physical systems. The list of technologies associated with Industry
4.0 includes industrial sensors, robots and collaborative robots (cobots), predictive
analytics, machine learning, autonomous in-plant logistics, simulation, augmented
and virtual reality, wearables and 3D printing. However, the distinctive feature is
the networking between humans and both physical and digital industrial production
processes throughout the value chain. The integration of physical and digital production,
together with continuous and real-time analysis, should improve the optimisation of
production, resulting in a more flexible and efficient production process (WEF 2014).
Improved profitability can be achieved by value creation through an improved product
offer (flexibility), asset utilisation (optimisation) and reduced labour costs (automation).
This book addresses the twin challenges of digital transformation by analysing the
impact of new technologies at company level. It focuses in particular on the automotive
industry which has been at the forefront of introducing new technologies such as
industrial robots. We analyse their impact on working conditions and employment as
well as on the role of production sites in the value chain. The book also addresses the

Jan Drahokoupil
extent to which digital transformation represents an opportunity, or a challenge, for

the countries that specialise in manufacturing production as far as their development
prospects and competitiveness are concerned.
The automotive industry in Europe is characterised by a division of labour – organised

by a few multinational corporations (MNCs) – between headquarters locations where
production is collocated with business and technology development and other intangible
activities; and peripheral sites that specialise in production activities. The tradition of
industrial peripheries in the automotive sector encompasses Spain, where automotive
production continues to play an important role. In Italy, while still characterised by the
presence of headquarters functions, product specialisation has also moved towards that
of a peripheral location. More recently, in the context of EU enlargement, a large part
of automotive manufacturing has been relocated to central and eastern Europe (CEE),
which now constitutes the European industrial heartland as far as production activities
are concerned.
These challenges in the automotive industry are addressed through case studies of old
and new peripheries in the European automotive industry. The case studies presented
in individual chapters were conducted within an ETUI research project on digitalization
in production. The project primarily examined MNC affiliates, including both original
equipment manufacturers (OEMs) and their suppliers. We focused, in particular, on
leaders in the implementation of Industry 4.0 technologies. However, the attention on
MNC affiliates is complemented by two studies covering domestic companies. These
include both suppliers of simpler components and digital entrepreneurs providing high-
end services to automotive companies. An overview of the countries and companies
covered in this book is provided in Table 1.
Table 1 Case studies
MNC headquarters (Chapter 2)

Germany 2 OEMs (125,000/670,000 globally); 2 technology suppliers (8,000/20,000 globally)
MNC affiliates (Chapters 3, 4, 7, 8)
Czechia 3 OEMs (2,248-22,932); 7 Tier 1 suppliers (726-9,000); 2 Tier 2 suppliers (203-848)
Hungary 3 OEMs (1,251-11,803); 7 Tier 1 suppliers (266-4,827)
Poland 3 OEMs (1,876-8,020); 3 Tier 1 suppliers (1,219-7,183)
Spain 1 OEM (4,800); 8 Tier 1 suppliers (50-280)
Italy 4 Tier 1 suppliers (394-10,300); 1 Tier 2 supplier (226)
Domestic companies (Chapters 5, 6 and 7)
Hungary 12 digital entrepreneurs (2-182)
Poland 3 Tier 1 suppliers (200-450); 3 Tier 2 suppliers (50-250); 6 digital entrepreneurs (50-250)
Italy 1 Tier 3 suppliers (413)
Note: employment levels in 2018 in brackets (for Poland: 2017, Spain: 2019)
The book starts with a chapter by Pamela Meil that analyses digital transformation in
the automotive industry from the perspective of the headquarters of major German
automotive MNCs. The volume then covers the impact of digital transformation in
both the old and new peripheries of the automotive industry in Europe. Four chapters

Introduction: Digitalisation and automotive production networks in Europe
address the competitiveness challenge: does digital transformation undermine or

enhance the upgrading prospects of companies participating on the peripheries of the
automotive value chain? In this section, Andrea Szalavetz and Ricardo Aláez-Aller, with
his colleagues, analyse the impact on MNC affiliates in CEE and Spain respectively. The
chapter by Krzysztof Gwosdz and his colleagues contrasts the situation of MNC affiliates,
typically assemblers or tier one suppliers, with that of domestic companies in Poland.
The latter are typically tier two suppliers specialising in simpler products. Andrea
Szalavetz supplements the analysis of domestic companies with a chapter that focuses
on the role of domestic technology leaders engaged in providing software solutions to
automotive companies, investigating the extent to which they could compensate for the
lack of high-value activities in the automotive value chains in peripheral regions.
Finally, there is a section on the jobs challenge. What is the impact of introducing new
technology on working conditions? How does the demand for occupations and skills
change? How have trade unions addressed the challenges? The chapters by Matteo
Gaddi and Monika Martišková present findings on these developments in MNC affiliates
in Italy and CEE, respectively.
The remainder of this chapter discusses some of the key findings. Before addressing
each of the challenges, it provides an overview of the position of CEE and southern
Europe in automotive production networks and the role of Industry 4.0 technologies.
1. Automotive production networks, European industrial

peripheries and Industry 4.0
Production networks in the automotive industry are characterised by a hierarchical
structure in which multinational corporations play a major role. A handful of OEMs,
such as the Volkswagen Group or the PSA Group, develop final products, assemble
vehicles and organise supplier relations in the production network. Moreover, these
carmakers now rely on a relatively small number of large supplier firms that dominate
tier one supply operations and with which they have thus forged close relationships
based on interdependence. They share some research and development functions and
are closely interlinked through the just-in-time, lean production model. As shown in
detail in Chapter 7 by Matteo Gaddi, new technology facilitates horizontal integration
along the value chain, allowing OEMs, or upper-tier operators, to monitor and directly
control production processes in supplier firms to the level of the specific tasks conducted
by individual workers.
There is a complex geography where business relationships often span the globe. At
the same time, a distinct regional division of labour has emerged at the level of world
regions such as Europe. Within these regions, there is a hierarchy between core locations
where the headquarters of MNCs are located and peripheral locations that specialise in
production functions. Importantly, carmakers as well as global suppliers tend to locate
R&D activities related to vehicle development in their core locations; R&D in peripheral
locations is typically geared towards production support. OEMs place assembly and
production functions in the periphery to take advantage of lower labour costs. Given

Jan Drahokoupil
the large sunk costs and dependence on regional supplier networks, assembly activities
tend also to stay in place once labour costs rise. The same applies also to bulky, heavy
and model-specific parts production that needs to be concentrated close to final
assembly plants to assure timely delivery at reasonable cost (for example engines,
transmission, seats and other interior parts). At the same time, lighter, more generic
parts can be produced at a distance and are likely to be relocated to take advantage of
scale economies and low labour costs (for example tyres, batteries and wire harnesses).
In Europe, Germany represents the key core location in automotive production

networks. It is home to major OEMs, most notably the Volkswagen Group, as well as to
global supplier firms such as Bosch. As shown in Table 2, the automotive industry also
plays an important role in the overall economic structure. In 2017, more than 845,000
workers were employed in the narrowly-defined automotive sector in Germany (NACE2
C29 in FTE, Eurostat, sbs_na_ind_r2), but a broader classification would cover about
two million industrial workers. A much lower share of components indicates that
Germany specialises in core functions. Many production activities remain located in
Germany, as a further result of the political sensitivities involved in relocation, but these
production functions are more tightly integrated with R&D functions (Krzywdzinski,
2017). As a core location, Germany specialises in higher-end larger models; at the
same time, the Volkswagen Group enjoys considerable flexibility in locating production
models across its European sites. Importantly, carmakers tend to introduce production
of new electric models in Germany and at other core locations (Drahokoupil et al. 2019,
see also Chapter 2 in this volume).
Table 2 Automotive: value added and employment, 2017
Share in non-financial business economy,* % Share of components in automotive, %

Value added Employment Value added Employment
Slovakia 8.2 50.7 4.8 69.4
Czechia 8.2 50.3 4.8 75.5
Hungary 7.2 47.6 3.6 75.3
Romania 6.4 82.2 4.6 89.1
Germany 6.0 22.7 2.9 31.7
Sweden 3.6 17.9 2.4 26.4
Poland 3.3 67.4 2.2 77.9
North Macedonia 3.2 NA NA NA
EU27 3.2 32.9 1.9 NA
Slovenia 3.1 52.5 2.3 63.4
EU28 2.9 31.7 1.8 50.3
Serbia 2.6 69.7 2.9 85.4
Spain 2.3 41.3 1.3 48.5
France 2.0 36.5 1.4 42.2
Italy 2.0 46.8 1.2 53.1
Notes: EU and candidate countries where automotive share of valued added in non-financial business economy > 2 per cent. Countries
that are considered in this project are in bold. Automotive refers to NACE2 C29: Manufacture of motor vehicles, trailers and semi-
trailers, components; and to NACE2 C293: Manufacture of parts and accessories for motor vehicles.
* Total business economy; repair of computers, personal and household goods; except financial and insurance activities.
Source: calculated from Eurostat [sbs_na_ind_r2, sbs_na_sca_r2]

A large share of components characterises the peripheral producers (see Table 2).
Southern Europe, and Spain in particular, represents the traditional peripheral location
in the European automotive industry. Italy, the home of major carmakers, predominantly
FCA/Fiat, has traditionally enjoyed the status of a core automotive location. However,
with the plight of FCA/Fiat, vehicle production has declined substantially while the
share of components has risen, and many Italian automotive companies now primarily
supply carmakers in western Europe, especially Germany. The product specialisation
of Italy has thus moved towards that of a peripheral producer. The integration of
CEE countries into European production networks in the context of EU enlargement
has changed the geography of production in Europe (Leitner and Stehrer, 2014 see
also Chapter 2 in this volume). CEE countries have become firmly established as key
production locations while automotive employment in all west European countries has
shrunk (Pavlínek 2019). As shown in Table 2, many CEE countries now rely more heavily
on the automotive industry than Germany. While Spain has been largely able to retain
its position as a key assembly location, it has missed out on all greenfield investments
in assembly plants in the last thirty years (Aláez-Aller et al. 2015). In Italy, FCA/Fiat
joined German and French carmakers in opening production locations in eastern
Europe (Poland and Serbia). Moreover, Italian component makers now compete with
CEE companies when supplying German carmakers. Among the CEE countries, the
share of components in automotive employment is highest in Poland. The latter, as
discussed in Chapter 5, somewhat lags behind Slovakia, Hungary, and Czechia as far as
the integration into automotive production networks and technology deployment are
concerned.
Production networks in peripheral locations have a dual structure, with foreign

ownership having the structuring role as far as the nature of value adding activities
is concerned. Foreign-controlled OEMs and upper-tier supplier companies exhibit
higher value added and complexity, and account for most of the R&D (e.g. Radosevic
and Ciampi Stancova 2018; Knell 2017). Domestic companies tend to be integrated into
global production networks as lower-tier suppliers specialising in simpler activities.
At the same time, however, the higher value added functions in foreign subsidiaries
tend to be weakly developed, with most R&D-related activities concentrated in the
core locations (Pavlínek 2016; cf. Drahokoupil and Fabo 2020). Innovation therefore
tends to be restricted to the upgrading of the production process rather than R&D
(Szalavetz 2017a). In Spain, for instance, MNCs do not develop R&D activities related
to product development; such activities can be found only in automotive suppliers with
domestic capital (Aláez-Aller et al. 2015). Two parallel innovation systems can thus be
identified (Radosevic et al. 2010). There is a large FDI-centred system, targeted towards
downstream activities in production such as the development of production processes.
In contrast, domestic innovation activities, however weak, are R&D based: they comprise
a handful of new technology companies specialising in knowledge-intensive services.
Core locations are the first to introduce new technologies into their production
processes. This is not surprising given that they also face higher labour costs and
where the return on automation based on labour saving is thus higher. The level of
automation in manufacturing, measured by the number of multipurpose industrial
robots per 10,000 people employed (see Table 3) is indeed highest in Germany as far

Jan Drahokoupil
as the EU is concerned. Peripheral producers exhibit much lower levels of automation.

Italy ranks very high on this indicator, but this is driven largely by automation outside
the automotive sector (see Table 4). In general, the automotive industry has been at the
forefront of introducing digital technology and automation into production processes.
As shown in Table 4, it accounts for the bulk of industrial robots in Europe. The level of
robotisation in the industry is particularly high in Spain – relative both to comparable
countries and to the rest of its industrial sector, reaching 88 per cent of the German
level.
Table 3 Number of multipurpose industrial robots per 10,000 people

employed in manufacturing industry (ISIC rev.4: C)
2018 Growth 2013-2018

Germany 338 20%
Sweden 247 40%
Denmark 240 36%
Italy 200 15%
Belgium 188 18%
Benelux 184 51%
Netherlands 182 80%
Austria 175 52%
Slovenia 174 91%
Spain 168 12%
Slovakia 165 94%
France 154 22%
Switzerland 146 62%
Finland 140 9%
Czechia 135 99%
United Kingdom 91 32%
Hungary 84 62%
Portugal 68 62%
Norway 56 37%
Poland 42 121%
Greece 23 77%
Romania 21 200%
Estonia 19 217%
Croatia 7 75%
Singapore 831 271%
South Korea 774 80%
Japan 327 3%
China 140 460%
Note: EU and candidate countries where data available and selected comparator countries.
Countries that are considered in this project are in bold.
Source: International Federation of Robotics (World Robotics 2019 – Industrial Robots)

Table 4 Number of multipurpose industrial robots per 10,000 people employed

in automotive industry (ISIC rev.4: C29) and in all other manufacturing
(ISIC rev.4: C excluding C29), 2018
Absolute As per cent of German level

Automotive Other Industries Automotive Automotive Other Industries
to other ratio
Slovenia 1371 91 15.1 108% 47%
Germany 1268 195 6.5 100% 100%
France 1239 102 12.1 98% 52%
Austria 1118 123 9.1 88% 63%
Spain 1110 88 12.6 88% 45%
Slovakia 815 41 19.9 64% 21%
Italy 748 171 4.4 59% 88%
Sweden 718 185 3.9 57% 95%
United Kingdom 687 46 14.9 54% 24%
Portugal 646 38 17.0 51% 19%
Czechia 555 62 9.0 44% 32%
Hungary 369 46 8.0 29% 24%
Poland 189 29 6.5 15% 15%
South Korea 2589 587 4.4 204% 301%
Japan 1165 245 4.8 92% 126%
China 732 70 10.5 58% 36%
Note: EU and candidate countries where data available and selected comparator countries.
Countries that are considered in this project are in bold.
Source: International Federation of Robotics (World Robotics 2019 – Industrial Robots)
The level of robotisation in CEE countries is much lower. At the same time, it has grown
rapidly in the last five years. This can be attributed to rapidly increasing labour costs. In
fact, as discussed in this volume, labour shortages represent the major motivation for
investment in automation in the region. Interestingly, as argued by Monika Martišková
in Chapter 8, when labour costs are taken into account the rate of robotisation in some
CEE countries, Czechia and Slovakia in particular, is higher than average; although, in
contrast, it is lower than expected in Poland, indicating the lower level of technology
deployment which is consistent with the findings in that particular chapter (Chapter 5).
The rate of robotisation in Germany is actually lower than what could be expected
given its labour costs; conversely, it is much higher in South Korea, due largely to this
country’s position as an industrial technology leader.
The use of industrial robots, however, is an imperfect indicator of technologies discussed

under the Industry 4.0 heading. In fact, the use of industrial robots is often better
classified as Industry 3.0. Whereas the latter refers to the automation of manufacturing,
Industry 4.0 entails the increasing digitalisation of the production process. In this
context, as argued in Chapter 3, five stages of digital maturity can be distinguished.
The first stage is Industry 3.0: automation with the use of older generations of fenced-
off robots. In the second stage, more advanced solutions are introduced, but they are
isolated and co-exist with legacy machinery. In the third stage, value adding components

Jan Drahokoupil
are connected for the purposes of digital monitoring. In the fourth stage, production
is controlled through cyber-physical systems. Finally, in the fifth stage, production is
completely automated. Adidas’s Speedfactory represents an example of such a facility.
The key to Industry 4.0 is thus the increased connectivity of production processes
and business functions (stages 3 and 4). The implementation of enterprise resource
planning (ERP) systems is a key tool in achieving such connectivity and this can be
taken as an indicator of Industry 4.0 maturity1. The share of enterprises that implement
the ERP software package to share information between different functional areas in
the countries analysed in chapters in this volume is presented in Table 5. The pattern
largely corresponds to that of robotisation. The automotive sector exhibits a higher
degree of Industry 4.0 maturity than the rest of manufacturing. There is variation
between countries, but Germany does not stand out as it did in terms of robot intensity
– the degree of implementation of ERP systems is similar to that in Spain, Czechia and
Italy. In contrast, Poland and Hungary exhibit a lower degree of Industry 4.0 maturity.
Table 5 Enterprises who have ERP software package to share information between
different functional areas, %, 2019
Manufacturing Automotive
Spain 51 NA
Germany 50 66
Czechia 48 68
Italy 45 62*
Poland 32 48
Hungary 20 44
EU27 47 60
EU28 46 59
Notes: countries covered in this book

10 or more people employed
*2014
Source: Eurostat [isoc_bde15dip]
The aggregate differences outlined here were not reflected in our sample. Respondents
in the headquarters of German MNCs did not see the process of Industry 4.0
implementation in their companies as particularly advanced, indicating that
automation will play a larger role in the future. They also pointed out that some of
the most modern production technology is located in newly-opened plants in CEE. The
case studies of MNC affiliates thus focused on companies that represented leaders in
the implementation of Industry 4.0 technology. In our CEE sample, the most advanced
Industry 4.0 technologies were found in Hungarian affiliates. Overall, MNC affiliates in
CEE operated a mixture of highly-automated and semi-manual activities.
1. The implementation of the manufacturing execution system (MES), rather than that of the ERP, may be
considered a more important indicator of Industry 4.0 maturity on the shop floor level in production locations.
However, there is a lack of comparative data on the use of MES.

Figure 1 provides some examples of the technologies, classified according to Industry

4.0 maturity levels. The most advanced Industry 4.0 solutions involved production
control through cyber-physical systems. These belong to stage 4 in our classification,
but the deployment was rather experimental and none of the companies classified as
having achieved a fully-fledged stage 4 maturity. The technology solutions included,
for instance, automation of data analytics, also achieved through artificial intelligence
solutions, to identify the process parameters having the largest influence on product
quality. The integration of production functions through ERP systems, as discussed
in Chapter 7 on Italian plants, allows the achievement of the objectives of the just-
in-time lean production model in which production planning is pulled by customer
orders, with customers gaining access to data about, and to control of, individual tasks
assigned to workers. Technologies such as ERP systems and cobots have also been
introduced in Spanish plants. At the same time, as argued in Chapter 4, we found that
MNC affiliates exhibited a lack of purposeful strategies to take advantage of the full
potential of stage 4 automation. Finally, as shown in Chapter 5 on domestic suppliers
in Poland, many companies operating in the lower tiers of the supplier network
effectively rely on manual labour and are thus yet to implement even an Industry 3.0
level of automation.
Figure 1 Stages of digital maturity: examples of investments in digital technologies

in surveyed companies
5 Completely automated factory

– Not in our sample
¡ Production control through cyber-physical systems

4 – Manufacturing execution systems;
– Digital production planning, predictive maintenance;
– Inventory management through radio frequency identification technology.
¡ Connection of value adding components; digital monitoring

3 – Visualisation of production status based on real-time data analytics, robotic process automation;
– Advanced internal connectedness.
¡ Advanced solutions, but isolated and co-existing with legacy technology

2 – Collaborative robots, automated material handling, automated guided vehicle;
– Data collection through cyber-physical systems, harmonisation of legacy IT systems.
1 No Industry 4.0: factory automation using older generations of fenced robots
Source: adapted from Andrea Szalavetz (Chapter 3)
There is some coordination of Industry 4.0 projects at MNC level while corporate
headquarters, as discussed in Chapter 2, may have a slight preference for launching
innovative pilot projects at headquarters sites. However, competition between
MNC affiliates is key to understanding the motivation for investing in Industry 4.0
technologies. As discussed in Chapters 3 and 4, MNC affiliates face continuous pressure
from their parent companies to cut costs. They need continually to improve efficiency

Jan Drahokoupil
and flexibility. The initiative to implement Industry 4.0 technologies thus comes from
local managements seeking to improve the competitive position of the affiliates. New
technology may also allow them to achieve some strategic differentiation from other
affiliates that they are competing with for the allocation of production and other
projects. Moreover, in CEE, labour shortages were a key motivation for investing in
automation. These were relevant for MNC affiliates in the upper tier of the supply chain
and for lower-tier domestic companies alike.
A position in the periphery represents a constraint on the adoption of new technologies.

More specifically, our case studies underlined the limited autonomy of MNC affiliates
in capital expenditure discussions, with affiliates needing to seek the approval of their
head office for such investments. It is typically required that any investment in new
technology brings a rate of return that allows the costs to be recuperated within two
years. Moreover, affiliates are normally expected to finance such investment from their
own resources. Lower-tier suppliers, foreign or domestic, may therefore not be able to
finance such investment since lower value added activities do not bring sufficient profit
to plough back in the name of investment. Furthermore, as discussed in Chapter 5 on
Poland, domestic suppliers also face broader barriers in terms of conservative attitudes
towards technology, shortages in terms of competences and skilled staff and a lack of a
culture of cooperation.
2. The competitiveness challenge

The digital transformation may provide an opportunity for peripheral locations to
upgrade, moving to activities which offer higher value added and, hence, revenue. In an
optimistic scenario, as discussed in more detail in Chapter 3, new technology increases
the knowledge intensity of the production activities in which companies in peripheral
locations specialise. Industry 4.0 technologies may facilitate the further decentralisation
of advanced activities including engineering, design and software development across
production networks. Digital transformation thus allows ‘factory economies’ (Baldwin
2013) to accumulate technological and R&D activities that have been traditionally
located in core, headquarters locations.
However, such a scenario can be contrasted with a pessimistic one involving downgrading
and the loss of existing competitive advantages. Accordingly, Industry 4.0 technologies
may automate some of the knowledge-intensive activities performed by engineers in
production locations (Flecker and Schönauer 2016; Szalavetz 2017a). For instance,
process development can be automated through self-optimising solutions embedded in
cyber-physical systems (see Chapter 3). Crucially, as emphasised in Chapter 4, Industry
4.0 undermines the key comparative advantage of peripheral producers: lower labour
costs and labour flexibility. Automation questions the very rationale for relocating
labour-intensive business processes to low-cost countries: automated production can
be profitably employed near ultimate markets or in headquarters locations (Dachs
et al. 2017). Moreover, peripheral locations have also competed via the flexibility of

labour market regulation (Köllő 2019). However, Industry 4.0 solutions also facilitate
production in locations with less flexible labour regulations.2
In fact, our case studies do not indicate that the deployment of Industry 4.0 would
lead to major structural change in the position of peripheral producers. We observed
process upgrading via the use of new technology, but that is rather a continuation of a
long-term process: any supplier needs continuously to upgrade production processes
to stay competitive, while the rates in supplier contracts typically assume continuous
improvements in productivity. However, we did find examples of product, or functional,
upgrading, including the allocation of more complex activities extending also, in some
cases, to R&D activities. Upgrading via the deployment of new technologies improves
the position of affiliates in competing for new product lines, possibly increasing the
commitment of the parent company to developing those units. Importantly, however,
affiliates enjoyed little autonomy in making decisions about the allocation of such
projects. The terms under which they were undertaken were also set by the headquarters.
As a result, argues Andrea Szalavetz in Chapter 3, the value chain position of production
subsidiaries in peripheral locations does not change even where they experience
functional upgrading.
We observed the decentralisation of advanced functions, as anticipated in the optimistic

scenario, but their effect on improving the value captured in peripheral sites, and hence
securing improvements in incomes, was limited. This can be related to two processes
(see also Szalavetz 2017b). First, new technology reduces the strategic importance
of delegated functions. Once advanced tasks, such as testing, process development,
incremental product development, design and simulation, have been supported
by digital technologies, they can be controlled and coordinated from a distance and
delegated to subsidiaries as operative functions. Second, as the terms under which
the more advanced tasks are undertaken is decided by the headquarters, functional
upgrading may not bring higher margins. As affiliates take on additional higher value
tasks, parent companies will cover the costs of their execution, such as the salaries of
the engineers that need to be hired. The parent companies retain full control of internal
prices, so functional upgrading may not increase profitability. The higher value captured
thus comes primarily in the form of higher wages for higher skilled staff.
The overall impact of employing new technologies on the position of the analysed
companies in the value chain seems to be positive, even if the developmental effects are
limited by their peripheral position, namely their dependence on the decisions made
in MNC headquarters. In contrast, as argued in Chapter 5, it is the lack of investment
in technology that represents a serious threat to lower-tier suppliers. At the same time,
however, we observed that Industry 4.0 technologies undermine the key locational
advantages of upper-tier producers (the importance of labour costs and flexibility). The
position of peripheral producers, as discussed in Chapter 3, is thus arguably becoming
more precarious and likely to be tested in the next wave of restructuring.
2. Empirical studies assessing the extent to which there has been a consequent reshoring of activities to higher-
wage locations are discussed in Chapter 4.

Jan Drahokoupil
3. The jobs challenge

The job displacement effects of new technologies have attracted considerable attention.
There is now a cottage industry of research that estimates the distribution of the risk
of automation across jobs and countries (Frey and Osborne 2017; Arntz et al. 2016;
Nedelkoska and Quintini 2018; PwC 2018). Estimates of the actual share of jobs that
are at risk of automation in individual countries differ, but, as discussed in more detail
in Chapters 3 and 8, there is a consensus that industrial employment is particularly
susceptible to automation. The operator in production doing routine work thus
represents the emblematic occupation at risk of automation. In Europe, CEE countries
and Spain rank among those with the highest risk of job losses. This is related to these
countries having a higher share of routine jobs (Keister and Lewandowski 2017).
Moreover, as discussed above, the rate of automation in these countries is lower than in
countries such as Germany that have already made some of the adjustments.
At the same time, the higher productivity achieved through automation is likely to
generate demand for labour in non-automated tasks. This may partially (Acemoglu and
Restrepo 2019) or even fully (Arntz et al. 2016; Dauth et al. 2018) offset the negative
employment effects of the new technology. However, these positive employment effects
take place away from manufacturing, being especially found in the area of business
services. While the employment effects of new technologies drive job polarisation at
the aggregate level (Autor and Dorn 2009; Goos et al. 2014), they also change demand
within manufacturing, with IT and electronics knowledge replacing traditional skills
profiles (Pfeiffer et al. 2016). In this context, it is important to realise that the same
technologies can have different effects on labour (or skill) demand, depending on the
role of the workplace in the production network (Krzywdzinski 2017). More specifically,
when the same process technologies are used, higher skills will be required in factories
that are involved in the roll-out and ramping-up of new products and new process
technologies – that is, as discussed above, typically in the core locations.
There were no redundancies directly related to the introduction of Industry 4.0

technologies in the companies that we analysed. This may be attributable to our
research programme taking place at a time of an upswing in demand in the automotive
sector. Indeed, CEE countries actually experienced severe labour shortages. Any labour
savings effects were thus seen positively by employee representatives too, as they helped
to address labour shortages. Spanish affiliate companies, in turn, were able to relocate
displaced workers.
However, we observed that new technologies did have an impact on the skills profile
of workers. The increased level of automation and digitalisation implied a demand
for new engineering profiles (IT specialisations), a need for programming skills in
existing categories of employees (e.g. maintenance workers) and a minimum level of
occupational training for new recruits extending to operator positions. The changing
skills demand for individual occupations is analysed by Monika Martišková in Chapter
8 on CEE. This shows that education and skill requirements are changing with respect
to most positions, requiring a retraining of the existing workforce. At the same time,
training policies were developed only in OEMs. Even there, however, a systematic

approach aimed at developing skills for all workers whose skill profile was becoming
obsolete was missing, while retraining policies were not on the agenda of CEE trade
unions. In contrast, as discussed in Chapter 2, trade unions in Germany were able to
push for a general policy on re-training, particularly for younger workers with a long-
term perspective at the plants. However, even in the OEMs and leading technology
providers analysed in the German chapter, re-training mainly involved job rotation,
with little re-training provided specifically for Industry 4.0 applications.
At the same time, for some job categories, particularly operators, new technology
entailed deskilling. Here, operators became ‘machine feeders’ monitoring automated
processes. What is sometimes described as increased task complexity in fact refers
to multi-tasking: workers are put in charge of several machines at once, feeding and
monitoring them simultaneously. Work thus becomes less strenuous but rather routine,
requiring a lower level of skill. Chapter 8 provides examples of workers whose jobs had
been automated who were transferred to routine jobs where they do not need to use
the skills that they had previously employed. In this specific case, their wage stayed the
same. However, in the longer term, such a transformation is likely to contribute to job
polarisation by creating routine, low-paid jobs.
Finally, increased efficiency through the deployment of Industry 4.0 goes hand-in-hand
with enhanced monitoring and control over the labour process. Many workers thus lose
autonomy, having to follow pre-optimised procedures designed in an automated way
and possibly in client companies. Increased work intensification was observed broadly,
both in CEE and in southern Europe. Chapter 7, however, analyses the link between
work intensification and the deployment of key Industry 4.0 technologies such as ERP.
In this chapter, Matteo Gaddi also argues that automatic coordination across the value
chain, the core of Industry 4.0, can effectively remove working conditions from the
arena of collective negotiation. Gaddi outlines the way in which CGIL/FIOM, an Italian
metalworking trade union, has sought to address this via the collective bargaining
agenda.
4. Concluding remarks: The Covid-19 challenge

Automation is only one of the major challenges faced by the automotive sector in
Europe. The transition towards electromobility, the increasing importance of digital
rather than mechanical technology in the final product and prospective changes in the
consumer use of automobiles have all put the future of European carmakers in question
(Drahokoupil et al. 2019; Galgóczi 2019). The economic crisis triggered by the Covid-19
pandemic has added an extreme demand shock that is likely to accelerate restructuring
processes, possibly in dramatic ways.
First, the recession will imply a reduction in production capacity in Europe. We have
thus now entered the consolidation phase that is discussed in Chapter 3 in the context of
peripheral locations only as a looming threat. Tight labour markets were an important
contextual factor that both mediated the impact of job-saving technologies and provided
an incentive for automation, particularly in CEE. However, finding workers is not likely

Jan Drahokoupil
to be a problem in the next few years anywhere in Europe. At the same time, the crisis
may delay investment in digital transformation across value chains. Second, the crisis
may accelerate the switch to electric mobility, also because demand stimuli in higher-
income countries are likely to be geared towards supporting more sustainable mobility.
This, as discussed in Chapters 2 and 4, may reinforce a concentration of production in
core locations as well as the opening of new, state-of-the-art production facilities. Finally,
we found that Industry 4.0 technologies were deployed in support of the lean, just-in-
time model. However, the crisis has highlighted the fragility of such interdependent and
geographically-spread systems. A shift towards more resilient production systems may
involve reshoring from more remote locations (Van den Bossche et al. 2020). However,
as argued in Chapter 2, this is not likely to threaten the position of European peripheral
locations. In fact, they may witness an opening of new production facilities that fully
take advantage of digital transformation.
References
Acemoglu D. and Restrepo P. (2019) Automation and new tasks: how technology displaces and
reinstates Labor, Journal of Economic Perspectives, 33 (2), 3-30.
Aláez-Aller R., Gil-Canaleta C. and Ullibarri-Arce M. (2015) FDI in the automotive plants in Spain
during the Great Recession, in Galgóczi B., Drahokoupil J. and Bernaciak M (eds.) Foreign
investment in Eastern and Southern Europe after 2008. Still a Lever of Growth? Brussels, ETUI,
139-170.
Arntz M., Gregory T. and Zierahn U. (2016) The risk of automation for Jobs in OECD Countries,
OECD Social, Employment and Migration Working Papers189, Paris, OECD Publishing.
Autor D. and Dorn D. (2009) This job is “Getting Old”: Measuring changes in job opportunities
using occupational age structure, American Economic Review, 99 (2), 45-51.
Baldwin R. (2013) Global supply chains: why they emerged, why they matter, and where they are
going, in Elms D.K. and Low P. (eds.) Global value chains in a changing world, Geneva, World
Trade Organization, 13-60.
Dachs B., Kinkel S. and Jäger A. (2017) Bringing it all back home? Backshoring of manufacturing
activities and the adoption of Industry 4.0 technologies, MPRA Paper 83167, Munich, Munich
Personal RePEc Archive.
Dauth W., Findeisen S., Suedekum J. and Woessner N. (2018) Adjusting to Robots: worker-level
evidence. Opportunity and Inclusive Growth Institute Working Papers 13, Minneapolis, MN,
Federal Reserve Bank of Minneapolis.
Drahokoupil J. and Fabo B. (2020) The limits of foreign-led growth: Demand for skills by foreign
and domestic firms. Review of International Political Economy, https://doi.org/10.1080/0969
2290.2020.1802323
Drahokoupil J., Guga Ş., Martišková M., Pícl M. and Pogátsa Z. (2019) The future of employment in
the car sector: four country perspectives from Central and Eastern Europe, Prague, Friedrich-
Ebert-Stiftung.
Pfeiffer S., Lee H., Zirnig C. and Suphan A. (2016) Industrie 4.0 – Qualifizierung 2025, Frankfurt
am Main, VDMA. https://www.vdma.org/documents/105628/13417295/VDMA%20
Studie%20Industrie%204.0%20-%20Qualifizierung%202025.pdf/bbfe37d6-f738-4558-
b2b7-1b01a04d166c.

Flecker J. and Schönauer A. (2016) The production of ‘placelessness’: digital service work in
global value chains, in Flecker J. (ed.) Space, place and global digital work, London, Palgrave
Macmillan, 11-30.
Frey C.B. and Osborne M.A. (2017) The future of employment: how susceptible are jobs to
computerisation?, Technological Forecasting and Social Change ,114 (January), 254-280.
Galgóczi B. (ed.) (2019) Towards a just transition: coal, cars and the world of work, Brussels, ETUI.
Goos M., Manning A. and Salomons A. (2014) Explaining job polarization: routine-biased
technological change and offshoring, American Economic Review, 104 (8), 2509-2526.
Keister R. and Lewandowski P. (2017) A routine transition in the digital era? The rise of routine
work in Central and Eastern Europe, Transfer, 23 (3), 263-279.
Knell M. (2017) R&D internationalisation and local innovation in the Visegrad Group after the FDI
peak, in Galgóczi B. and Drahokoupil J. (eds.) Condemned to be left behind? Can Central and
Eastern Europe emerge from its low-wage model?, Brussels, ETUI, 111-131.
Köllő J. (2019) Towards a “Work-Based Society”? in Kovács J.M. and Trencsényi B. (eds.) Brave New
Hungary: mapping the ‘system of national cooperation’, Lanham, Lexington Books, 139-158.
Krzywdzinski M. (2017) Automation, skill requirements and labour-use strategies: high-wage and
low-wage approaches to high-tech manufacturing in the automotive industry, New Technology,
Work and Employment, 32 (3), 247-267.
Leitner S.M. and Stehrer R. (2014) Trade integration, production fragmentation and performance
in Europe - blessing or curse? A comparative analysis of the New Member States and the
EU-15, wiiw Research Reports 397, Vienna, The Vienna Institute for International Economic
Studies.
Nedelkoska L. and Quintini G. (2018) Automation, skills use and training, OECD Social,
Employment and Migration Working Papers 202, Paris, OECD Publishing.
Pavlínek P. (2016) Whose success? The state-foreign capital nexus and the development of the
automotive industry in Slovakia, European Urban and Regional Studies, 23 (4), 571-593.
Pavlínek P. (2020) Restructuring and internationalization of the European automotive industry,
Journal of Economic Geography, 20 (2), 509-541.
Pfeiffer S, Lee H., Zirnig C. and Suphan A. (2016) Industrie 4.0 – Qualifizierung 2025, Frankfurt
b2b7-1b01a04d166c.
PwC (2018) Will robots really steal our jobs? An international analysis of the potential long term
impact of automation, London, PricewaterhouseCoopers.
Radosevic S. and Ciampi Stancova K. (2018) Internationalising smart specialisation: assessment
and issues in the case of EU New Member States, Journal of the Knowledge Economy, 9 (1),
263-293.
Radosevic S., Savic M. and Woodward R. (2010) Knowledge based entrepreneurship in Central
and Eastern Europe: results of a firm level survey, in Malerba F. (ed.) Knowledge intensive
entrepreneurship and innovation systems evidence from Europe, London, Routledge, 198-218.
Szalavetz A. (2017a) Industry 4.0 in ‘factory economies’, in Galgóczi B. and Drahokoupil J. (eds.)
Condemned to be left behind? Can Central and Eastern Europe emerge from its low-wage
model?, Brussels, ETUI, 133-152.
Szalavetz A. (2017b) Upgrading and value capture in global value chains in Hungary: more
complex than what the smile curve suggests, in Szent-Iványi B. (ed.) Foreign direct investment
in Central and Eastern Europe: post-crisis perspectives, Cham, Palgrave Macmillan, 127-150.

Van den Bossche P., Levering B., Castaño Y. and Blaesser B. (2020) Trade war spurs sharp reversal
in 2019 reshoring index, foreshadowing COVID-19 test of supply chain resilience, New York,
Kearney.
WEF (2014) The future of manufacturing: driving capabilities, enabling investments, Geneva, World
Economic Forum.
All links were checked on 08 June 2020.

The headquarters perspective
Chapter 2
Inside looking out: Digital transformation in the German
automobile sector and its effects on the value chain
Pamela Meil
The term Industry 4.0 has been in our discourse for some time now although there
is still some confusion about what it means. It is, however, consensus that digital
technology, together with other factors, is unleashing a transformation of how things
will be produced and new possibilities of what can be produced. One aspect involves a
broad array of IT and digital applications aimed at making industrial production more
productive or efficient, more networked, increasingly monitored and more automated.
Other elements entail potential new product and service innovations, higher quality
in production, the generation of more information and analytics, and greater synergy
between customers and producers.
Germany, in particular, has placed a high emphasis, backed up by large investments, in

Industry 4.0 applications. The hope is to lay a foundation for ongoing competitiveness
given the country’s still-strong industrial base (compared to other western industrialised
economies), especially in the light of the changing production landscape and pressure
from emerging economies. Industry 4.0 has become a kind of catch-all phrase, but
it does have some characteristic elements: new opportunities are seen in the linking
of services and production; there is a stronger connection between pre- and post-
production processes; and the roles of the various actors (customers, for example) are
being increasingly integrated into the production process (Kagermann et al. 2013).
1. A world of global value chains

Another trend characterising the last couple of decades of production in industrial
goods, and increasingly also services, is the distribution of activities across global value
chains. Initially, low-level tasks and production activities were targeted for outsourcing
to lower cost sites (Porter 1985). As time progressed, more and more activities and
functions, aided by developments in digital technology, have been outsourced which
has created complex networks of production and services (Henderson et al. 2002).
The motives for externalising activities has varied: costs; access to markets; access to
resources; access to labour; pressure from shareholders; pressure to reduce wages at
home and push for concessions from local labour forces; and so on (Meil 2019). As a
result of developments in global value chains, intermediate production activities have
grown extensively (Cattaneo et al. 2010), with pieces of the stages of production criss-
crossing the globe and making it increasingly difficult to trace the origins of value added
creation for a particular product (Linden et al. 2011).

Pamela Meil
Research on global value chains (GVCs) has demonstrated that their growth is dynamic
(Gereffi et al. 2005, Gereffi and Kaplinsky 2001). The simple outsourcing of a product
or process from location A to location B, mainly for reasons of cost reductions,
often changes over time as interconnectedness and interdependence grows between
the sites. Thus, the simple outsourcing of low-skill tasks for the purposes of cost
reduction has long since stopped being an accurate characterisation of global value
chains, particularly for industries producing complex products or services such as
automobiles. Over time, actors in value chains shift, suppliers change and chains
lengthen, altogether creating a complex production network in which supply chains are
integrated and interdependently linked (Humphrey and Schmitz 2002).
In this study, the auto industry and its suppliers are a central focus of interest. Complex
value chains, developed over a long period of time between dispersed production sites
and across supplier networks, play an important role in the automobile sector. At
the same time, this sector is one of the main ones in which Industry 4.0 technology
applications are being developed, implemented and deployed. The increased potential
of automation; the development of ‘smart factories’ in which logistical systems
are digitised, connected and streamlined; the use of sensors in assembly lines;
sophisticated person-machine (robot) interactions; artificial intelligence and ‘learning’
machines; and links between external and internal systems of operation: these are all
technologies linked to the term Industry 4.0 and which all have fields of application in
the auto industry (Lichtblau et al. 2015).
The other contributions in this collection examine developments in new digital

transformations in the auto industries of the emerging economies of eastern Europe as
well as in Spain and Italy. The former has strong links to western Europe automobile
OEMs, and many are outsourced production sites or subsidiaries. There are also
subsidiaries of auto manufacturers from other European countries, such as Germany,
in Italy and Spain although Italy in particular also continues to ha its own long-standing
indigenous auto production and engineering plants.
A central question addressed by these studies is: what effect does, or will, new digital
transformations, often subsumed under the rubric Industry 4.0, have on their local or
regional sites? As part of OEM value chains, will production and engineering sites be
upgraded or downgraded? Will the effects be marginal or lead to noticeable shifts in
how and what is produced? And what will be the impact of Industry 4.0 on workforces?
Will there be shifts in the division of labour across value chains? Will employees in
emerging economies be replaced by automation and their workplaces degraded? Or
will the opposite occur: a shift towards more high-skilled tasks and more diverse
activities?
Decisions on the future of sites along the value chain ultimately come from OEM
headquarters, where strategic decisions are made concerning future investments,
site use and capacity as well as the division of labour between sites. This chapter
addresses precisely this side of the issue by asking the following question: what strategic
orientations do OEMs pursue regarding the implementation and planning of new digital
technologies and Industry 4.0 applications at their outsourced sites and subsidiaries?

Inside looking out: Digital transformation in the German automobile sector and its effects on the value chain
Many OEMs and their tier one suppliers in the auto industry are located in Germany
and this is the case described and analysed here.
2. A note on method
Methodologically, the original approach was to identify companies who were part of
the empirical analysis of the other case studies presented in this book. Of those with
German owners or headquarters, the design plan was to conduct analyses of matched
samples linking expressed central company strategy to conditions at production sites
in the countries represented in this collection. This proved impossible to implement
due to difficulty of access, in part because of the saturation of Industry 4.0 studies in
Germany and the consequential refusal of new requests, and ultimately due to impasses
on account of the corona crisis.
In the end, four companies participated in the study: two automobile producers; one
high-end automobile supplier which carries out production and facility planning for a
broad range of customers; and one highly innovative producer of automation systems.
All of the companies have production or engineering development sites dealing with the
auto industry in one or more of the countries in southern or eastern Europe represented
in this book. Eleven interviews, based on an open-ended questionnaire, were carried
out. This information was supplemented with desk research on the companies and
issues related to Industry 4.0 and value chains.
3. The cases
3.1 Case 1: EDAG
An engineering service provider to the auto industry with three divisions: one
specialising in vehicle engineering; one in the development of production solutions
(factory and assembly line planning); and one devoted to electronics development and
in-car IT systems. The company has over 8,000 employees and more than twenty sites
worldwide, usually connected to the sites of auto manufacturing facilities among others
in Italy, Spain, the Czech Republic, Poland and Hungary.
The respondents from production solutions offer a range of services to their auto
manufacturing customers that deal with topics surrounding Industry 4.0 applications
such as the development toward ‘smart factories’, in which digitalisation and networking
for facility, production and logistical planning is conceived and implemented, as well as
the potential development of ‘smart assistants’ (for example, sensors and the use of
augmented and virtual reality).

Pamela Meil
3.2 Case 2: Volkswagen

Volkswagen Group is the holding company for twelve auto manufacturers in seven
European countries which have about 120 production sites in over thirty countries
around the world. The company has some 670,000 employees working in production,
services or sales. Production sites can be found in Spain, Italy, Poland, Hungary, Slovakia
and Slovenia, most of which began as car brands from those countries and later became
part of the VW group. This is important because a certain amount of differentiation and
some autonomy exists based on brand identity and history. This is true within Germany
as well as outside it.
The respondents for this study came mainly from a newly-established unit within the
group: VW Group Components, which emerged out of a long process of converging
individual product groups and brands. This unit is the supplier for all of VW’s car
brands and can also sell to customers outside the group although currently about
95% of its business is within VW. VW Group Components employs 80,000 workers
in 61 production sites across 47 countries. It encompasses a broad range of products
including seats, undercarriage, gearbox and drivetrain, engines, foundry and e-mobility
(batteries for new electric cars, for example).
Auto components production is particularly interesting to study with regard to

digitalisation because, while it includes the most labour-intensive parts of manufacturing
and assembly, it also contains some of the most automated parts of the production
process. Some of the more highly traditional parts of automobile production, and the
oldest machines, are found in component production, at the same time increasingly
alongside some of the newest and most automated due to the production of electric
cars. Electric engines and battery production all fall within the purview of Group
Components.
Generally, the idea behind the concentration of components into a centrally-steered

group was to bundle the production of certain components at particular sites and
to distribute capacity utilisation more efficiently. Obviously, the process of closing
down, shifting and concentrating activities was a contentious one. Some respondents
portrayed the shift to more modern and digital forms of production as a step-by-step
process, particularly because there are production lines and machines that are quite
old, but still in operation, and many of these are in Germany. In fact, some of the ‘most
state-of-the art sites for component production are in eastern Europe simply because
they were built more recently.’ (VW-5)
3.3 Case 3: BMW

BMW is a premium car brand with about 125,000 employees in auto-related activities
worldwide, including the other brands in the group – MINI and Rolls-Royce. There
are thirty production plants in 14 countries. Compared to other carmakers, BMW is
relatively small and its activities are confined to a relatively low number of countries
for actual auto manufacture. Even fewer are involved in higher range activities such

as development and engineering. BMW is currently building a new assembly plant in

Hungary but, otherwise, auto production is confined to its sites in Germany, China, the
US, Britain (due to the takeover of MINI and Rolls-Royce) and Mexico. Unlike VW,
BMW outsources considerably to external suppliers for parts and services. It also enters
into general contracting agreements with large engineering and design companies to
take over pre-production processes for certain projects.
The interviewees from BMW work in prototype development for auto assembly, with
expertise in digital transformation processes in pre-series production.
3.4 Case 4: Festo

Festo is well-known in Germany as a highly innovative company specialising in
automation technology, including interactive robots and VR and AR applications.
An interesting aspect of Festo is that it also has a large business unit specialising in
training and consulting and the development of e-learning systems, including an entire
set of training modules for Industry 4.0 technologies. The company has over 20,000
employees worldwide and an eighty per cent share of turnover is targeted for research
and development activities. Production sites exist in several countries (China and the
US have major sites) but in Europe, outside of Germany, only in Hungary, Poland and
Belgium. There are also smaller sites for research and development in Hungary and
Bulgaria. The export share of what is produced is particularly high: sixty per cent. Most
employment is in research and development and in sales (about seventy per cent), with
only thirty per cent in production, logistics tasks, assembly and manufacturing.
The interview partner is from the department of solutions technology transfer.
4. Areas investigated in the interviews

The questionnaire used in the study contains three main areas of investigation:
1. Background information on company structure and global presence. The

estimated level of Industry 4.0-type applications in the company or the area of
the company which the interviewee represents (necessary to limit the focus for
very large companies). The estimated growth in importance for specific areas and
technological applications of Industry 4.0 for the company or unit, including in a
five-year forward-looking timeframe.
2. Company strategy regarding site use and the role of various factors in decisions
to outsource or invest in sites. Company strategy for investing in Industry
4.0 applications across the value chain. Determinants or reasons for deciding
investments and choosing particular applications. Differences in Industry 4.0
implementation and use between HQ (or home) production sites and remote sites.
Opinions on the occurrence of reshoring as a result of Industry 4.0. Opinions on
applications that should remain at HQ or the home country and not be outsourced.

Pamela Meil
3. Effects of Industry 4.0 on workers and working conditions:
a. increases or decreases in employment
b. new divisions of labour (also across the value chain)

— the influence of Industry 4.0 on the use of sites across the value chain
and whether this differs by region
c. effects on worker profiles and qualifications:

— have new workplaces or tasks been created? Has deskilling or upskilling
occurred?
— the impact of Industry 4.0 on work and tasks: will they become more
standardised and simpler, or more complex and demanding? Which
kinds of tasks are most affected?
— training or programmes that are offered for Industry 4.0 applications –
by unit and, if applicable, by region
It is obvious that the range of products, processes and services deployed in the company
or headquarters unit has an impact on the implementation of Industry 4.0 and its
deployment across their value chains. If they are not very far along at home, they are not
going to be very far ahead abroad. It is expected that the companies will be at different
levels of implementation in different departments and applications. Given this diversity,
the goal here is not to convey a definitive portrayal of strategies for future Industry 4.0
developments in a particular company or sector. Rather, it provides a snapshot, based
on the estimates of interviewees, of the current conditions and possible business models
that German auto producers and tier one suppliers are pursuing for their various sites
across the value chain, at home and abroad.
5. The impact of global value chains

Putting developments in the auto industry in context, McKinsey estimates that what
it calls ‘global innovations’ sectors – industries that include automobiles, computers
and electronics, and machine-building – have given rise to the most valuable, highly
traded and knowledge-intensive of all the value chains involved in the production of
goods (MGI 2019). Spending on R&D and intangible assets averages thirty per cent of
revenues in ‘global innovation’ industries, 2-3 times the figure in other value chains
(MGI 2019: 2). As Sturgeon (and his colleagues) has shown, a lot of trade in auto
manufacturing takes place as intermediate goods (Sturgeon and Memevodic 2011;
Sturgeon and van Briesebroek 2011).
Regionalisation, meaning a concentration of value chains in local and regional networks,

has been increasing. McKinsey sees this trend as most evident for global innovation
value chains, which include the auto industry ‘given their need to closely integrate many
suppliers for just-in-time sequencing’ (MGI 2019: 9). However, a noticeable exception
to this trend is central and eastern Europe, which continues to integrate with western
Europe as Figure 1 below demonstrates.

It appears that industries in the emerging economies of eastern and central Europe
are, in fact, integrating more deeply into the supply chains of OEMs in western Europe.
Many countries have joined the production networks of the large western European
carmakers (especially Germany).
Figure 1 China and emerging Asia are building domestic supply chains, while central
and eastern Europe continues to integrate with western Europe
(reprinted from MGI 2019)
Change in traded intermediate inputs relative to total intermediate inputs
Percentage points
6
Global innovations 2007–17
4 Regional processing
Labor-intensive
2 Resource- goods
intensive Knowledge-
goods intensive Labor-
0
Labor- services intensive
intensive Regional goods Labor-intensive
-2 services processing services Regional
processing
-4 Labor- Labor-
intensive Resource- intensive
intensive goods Knowledge-
-6 services Global intensive
innovations goods
services
Resource-
-8 intensive
goods Knowledge- Global
intensive innovations
-10 services
-12
-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Europe Asia–Pacific China Growth of domestic intermediate output
(excluding China) Compound annual growth rate, %
Source: World Input-Output Database; McKinsey Global Institute analysis
6. Findings
Respondents were asked how they understood Industry 4.0 and to assess the relevance
for their companies of various applications. Almost all were especially careful in their
assessment of the level that their companies had achieved in terms of the implementation
of Industry 4.0. Between a choice of beginner, ‘on the way’ and advanced, most chose ‘on
the way’ to describe where their company stood in the introduction and use of Industry
4.0 technologies. This, in fact, reflects the overall picture that surveys have portrayed
on the progress of Industry 4.0 in Germany (see Figure 2) (Kinkel et al. 2016; Heidling
et al. 2019).

Pamela Meil
Figure 2 Level of digitalisation in machine-building in Germany, 2016

100
80 25
60 20
27 41 26
40 19
16 13
19 13 48
20
20 16 22
14
0
less than 100-499 500-1999 more than 2000 total
100 employees
yes, contracted yes not contracted planned in the next 3 years
Source: Kinkel et al. 2016
Explanations for the assessment of ‘on the way’ revolved around the feeling that only
some areas and research topics involving Industry 4.0 were at an advanced stage. In
general, however, respondents’ perceptions of Industry 4.0 involved highly complex
networking and examples of artificial intelligence and algorithm development, albeit
that these were still in the early phases of pilot projects at their various sites. Their
assessments should therefore be understood as keeping the bar rather high for
determining how far along were Industry 4.0 applications in their respective companies.
They did not understand the term to mean general uses of digital tools or ongoing shifts
towards more automated production.
The implementation of Industry 4.0-type applications was often made initially in the
form of pilot projects. This was done either by identifying areas or processes at different
sites – also outside Germany – where a particular new tool or application would be
especially relevant and then trying to deduce lessons which, eventually, would lead to
implementation at other sites. Both BMW and VW had highly similar approaches in
this type of initial implementation strategy. One respondent from a supplier who has
an overview of several OEMs pointed out that one problem is an abundance of strong
pilot projects, ‘super islands in some factory, in which Industry 4.0 is practised and
demonstrated’ (EDAG 1) but that it never really progresses beyond the pilot stage:
‘The projects never seem to seep into the regular processes but remain little islands of
innovation.’ For VW Group Components, in particular, investments to replace older
equipment were planned to be carried out incrementally. Although there did seem to
be a slight preference for launching innovative pilot projects at German sites – perhaps
also because this is where development engineers tend to be situated – the introduction
of pilots at other sites was not out of the question. Several times VW’s engine plant
in Poland was mentioned as being more modern technologically than some plants
in Germany and, therefore, a prime location for pilot projects or new investments in
technology.

Without actual investment numbers for these specific types of technologies, it is difficult
to know precisely how accurate this picture is. Festo conducts most of its state-of-the-
art R&D in Germany, but it is a high-end supplier geared to exports. BMW seems to
have a number of its pilot programmes in the German headquarters, although some
high-end projects specifically related to quality control are being carried out at foreign
plants. Meanwhile, VW Group Components has been investing heavily in plants in
Chemnitz and Salzgitter; battery research and production, an important future topic for
e-mobility, appeared to be prioritised for its German sites. One respondent from VW
did raise the possibility – hypothetically – that the production of gas and diesel cars
could theoretically be concentrated mainly at older production sites or ones outside
Germany, whereas the production of electric cars could be concentrated at newer sites
in Germany, at least for European markets. Nonetheless, there was no clearly explicit
strategy at any of the companies to invest only in Industry 4.0 technologies at the home
country site.
All in all, it seems that future strategies for implementing Industry 4.0 technologies in
the cases examined here are still in flux, independent of place. It seems quite evident
from the discussions, however, that there is not a clearly articulated strategy to keep
external sites at a low technological level either by not investing in their production
facilities or by slowly fading them out of the value chain.
Respondents were questioned regarding a list of specific technologies associated with

Industry 4.0 and on their view of their relevance for their companies now or in the near
future.
— automation
— cyber-physical systems
— business process automation
— digital knowledge transfer
— intelligent solutions (predictive maintenance, workforce analytics)
— artificial intelligence and learning algorithms
— digital networking with external and internal systems.
Looking in detail at these specific applications that are often linked to Industry 4.0
gives a more concrete picture of the use, or intended use, of Industry 4.0 in the overall
company strategy, even if companies are still not currently at a very advanced stage of
implementation.
All of the respondents indicated that automation would increase and would play a large
role in the future of production. It was not always clear, however, if they shared the
same exact definition of automation: whether production IT, greater use of robots, etc.
With cyber-physical systems (interactions between worker and robot; and the use of
sensors for production monitoring and analytics), the reaction was more mixed. Several
interviewees were not familiar with the term or technology while others characterised it
as ‘buzzword’. Others did see the potential in cyber-physical systems, depending on what
customers (in this case, the OEMs) wanted. Nonetheless, a certain amount of caution

Pamela Meil
or holding back was evident. As one respondent put it, ‘I was contacted nearly every
week about a great sensor that delivers super measurement data or links to clouds. Like
mushrooms popping up after rain. But it must be recognised that such things only work
when there is connection to what is actually being produced.’ (EDAG 1)
Business process automation and intelligent solutions were also seen as extremely
important topics that companies were currently addressing. For several, it also involved
efficiency effects in administration and services, not only in production. And for most,
the topics dealt with streamlining or even standardising processes. To achieve this, it
was necessary to understand the processes, extending right across the value chain, in
great detail and this was considered a ‘central’ challenge for the future. ‘If you have
an – excuse me – ‘shitty’ process, and it is digitised, then you have a digitised ‘shitty’
process.’ (VW-4) ‘If you can’t play the piano, and you buy a really expensive piano, you
still can’t play the piano.’ (EDAG 1) Thus, process optimisation was considered highly
relevant, as long as it was understood what processes were being optimised. Predictive
maintenance, as a sub-category of intelligent solutions, was already being implemented
at many sites. Projects for quality detection using new Industry 4.0 applications and
algorithms for machine learning were underway at all of the companies. It is clear as
regards these two specific technologies that applications that were easily definable and
tightly linked to production were the ones most likely to be deployed.
Knowledge sharing through digital media (Web 2.0, for instance) was not widely
practised.
All the companies were, however, interested in using artificial intelligence and
algorithms, but they were mainly seen as topics for the future. Moreover, they were all
quite careful in utilising the term AI, because it covers a broad range of possibilities and
depends on the available data. The auto companies seemed to be targeting AI for the
identification or evaluation of mistakes or quality problems in production. Ideas and
pilot projects on the subject were in discussion or actually being implemented in both
OEMs in this area. One respondent cautioned that, although everyone was talking about
AI because it was ‘modern’, what was being carried out reflected ‘minimal’ applications.
Nonetheless, even this respondent felt that AI would gain importance over time.
With regard to the issue of digital networking both within and outside the organisation,
the view was that these types of digital links and networking have existed for a long
time, particularly between top suppliers and OEMs. A number of new areas or unsolved
issues remain as future challenges to be addressed – among them intellectual property
and privacy concerns and also the need for, but difficulties with, the standardisation
of systems. For VW Group Components, part of VW Group but nonetheless also
a supplier, the goal is a unified system for its internal customer: VW Production. Of
course, this includes all sites, even outside Germany, falling within VW Group. The issue
seems to be mainly one of achieving transparency, improving logistics and reaching a
better just-in-time availability of components. Consequently, new digital tools will be
developed and implemented, but whether or not this can be categorised as Industry 4.0
is questionable. Digital networking in Industry 4.0 is further-reaching and involves the
links between services and production activities as well as connections with customers.

These types of application seemed less of a concern to respondents than more traditional
IT infrastructures and company internal shared databases.
Another of the central issues surrounding digital transformation discussions is the

explosion in the data and information being generated. What should be done with this
and how should it be used? Where is information collected and analysed, and what are
the systems for reporting? For sites down the value chain, and for the empowerment or
development of workforces, this becomes a central issue in terms of monitoring, control
and autonomy.
The picture that emerges from the interviews is that both local sites and central
headquarters are involved in data analysis based on digital data reporting. ‘One has
to look which data is needed where’ (VW-4). At the moment, if plants have their own
programme planning or manufacturing control systems, then only local sites need
the data. Having said that, there is nonetheless an evident trend that data on the
optimisation of production lines will be collected more centrally and then shared. In
a positive scenario, in fact expressed by respondents here, the data would be used to
improve performance at all the plants. Naturally, the negative scenario is that the data
could be used for benchmarking and performance comparison and, theoretically, to
substantiate reductions or even closures (Meil et al. 2003).
In the strategy department of VW Group Components, for instance, there is a definite

plan to move from a decentralised reporting structure delivering reports to headquarters
towards a more centralised and standardised structure. Here, the notion is to optimise
capacity utilisation at the level of the entire group across all of its sites, and to bundle
products and processes ever more efficiently. Consequently, there has been some
shift away from local autonomy once held at decentralised units to a more centralised
decision-making structure. This development applies to all the units in the value chain,
including the German ones. However, in order to achieve a ‘real time’ reporting flow
from the plants to the centre, investments in new types of competencies and technologies
are necessary in Germany and beyond. A modern engine plant in Poland is currently
seen as the ‘gold standard’ in which remote reporting via iPads located on the shop
floor is available to central monitoring units. On the one hand, this allows increased
potential for control and monitoring; on the other, it could have the effect of raising the
competency levels of the local workforce.
7. Effects on work and workers

This bring us to the issue of what effect developments in Industry 4.0 have on the
organisation of work and working conditions, which is a central focus of the research
presented here. Will digital transformation result in a decrease in employment? Will
the division of labour shift, leading to a downgrading or upgrading of sites? Will more
skilled tasks move to the headquarters or home sites of companies, resulting in remote
sites down the value chain becoming external low-skill workbenches?

Pamela Meil
Given the backgrounds of the interviewees – who come from central headquarters and
from departments for strategy, change management and project development, and who
do not have detailed knowledge of the organisation of work or the conditions of work
at shop floor level – their view is somewhat aggregated and quite abstract. Even in the
cases at the level of production described in the other contributions, interviews did not
take place with workers themselves. Therefore , we can only portray here a broad picture
of possible developments regarding divisions of labour, skills levels and development,
general employment trends, etc.
Most respondents expect a decrease in employment in production as a result of

increased automation leading to ‘efficiencies’ in the long run. However, they also predict
opportunities in other areas – monitoring, programming, data analysis – to increase.
Moreover, most respondents expect a shift in occupational profile, which will affect
most areas including production. There are a number of studies on Industry 4.0 which
also document such shifts (Hirsch-Kreinsen 2014; Pfeiffer et al. 2016; Heidling et al.
2019).
Currently, the view is that most of the decreases can be achieved through attrition
(retirement, etc.) but this is, of course, mainly with regard to Germany. VW is partially
owned by the state of Lower Saxony and there is a ten-year guarantee regarding
employment security, particularly regarding the introduction of new digital technologies.
A main issue accompanying digital transformation is that task areas and jobs will be
changing, sometimes drastically. This is partly due to digital transformation in the auto
sector, but also because of the move to e-mobility which is a major new focus for the
carmakers. If this trend continues, many assembly-related jobs will become obsolete.
Previous research has already shown that Industry 4.0 technologies, as well as other
changes to the organisation of work, induce a shift in qualification profiles even for
skilled workers in production areas (Pfeffer et al. 2018; Heidling et al. 2019; Meil and
Heidling 2010). IT and electronics knowledge are replacing traditional qualification
profiles, such as metalworking, machine-building or other specialities in auto-related
occupations that were designed for the production of cars based on combustion
engines or diesel. With electric motors, subjects which were never considered a part of
automobile production, such as chemical engineering, have even become relevant for
auto production (VW – 5). The change is, therefore, not only about low or high skills,
but also a massive shift in the competence mix of the workforce. This affects all sites,
although perhaps at different speeds and with different emphases.
Interestingly, there seems to be an inclination at the central divisions and strategy

levels to think very much in terms of processes rather than people. Consequently, the
overwhelming focus is on getting the technology, and especially the process, right. The
other things – work organisation or the composition of the workforce – are expected to
adapt or be adapted to meet the new demands. Given that the line of communication
at these levels is mainly with management or engineers, this view is not so surprising.
Certainly, getting processes right across a complex value chain are crucial. Nonetheless,
this fairly typical planning orientation – first the technology, followed by competencies

and the organisation of work – has been the modus operandi in most development
waves and this almost always leads to problems in implementation (Böhle 1998; Böhle
et al. 2002).
The general tenor of the interviews is that some jobs will be lost as a result of Industry
4.0 but others will be gained. There has been some thought about how to enact the
shift, particularly in light of employment protection regulations in Germany and given
that trade unions are a strong factor in this particular industry. The general concept
seems to be of a long-term ‘re-training’ by having younger workforces – those with a
long-term perspective at the plants – rotate through various workstations for stints of
several months. It is unclear if similar arrangements are being offered across all sites
and countries, although itis quite clear that such programmes are not targeted for older
workforces.
It is evident that, other than recruitment and job rotation strategies, there appeared to
be little training taking place specifically for Industry 4.0 applications at any of the sites
in any of the cases.
What will happen to work as a result of the introduction of Industry 4.0? With regard
to effects on work tasks and content, we asked what future developments could be
expected. These findings would then give us additional information on whether work
might be deskilled, and thus easier to automate, or what kinds of reskilling could be
expected and what types of workers would be sought. In line with questionnaires from
the other case studies presented in this book, we asked if activities would become
increasingly standardised simpler, or more demanding and complex.
Interestingly, most respondents understood standardisation as compatibility and not as

a characteristic of a task or activity. Thus, they predicted an increase in standardisation,
but mainly regarding product platforms or systems integration rather than of the
tasks or activities being carried out by workers. Some, however, suggested that tasks
would become simpler in terms of being based on systems and architectures that
would make operating systems easier to use, also as regards shop floor operators, by
using digital technology interfaces. Some loss of competence in certain activities could
also be expected through the deployment of artificial intelligence. Areas mentioned
were evaluation and quality assurance, which currently depended on high levels of
worker experience but which were tasks that could potentially be replaced by artificial
intelligence and machine learning. Otherwise, respondents tended to believe that many
aspects of working in digital networked environments would become more demanding
and interesting. This was the case because the activities would become more diverse
and less repetitive and would include new topics and areas of expertise.
8. Impacts on the value chain

As one interviewee put it, ‘the automobile industry is one of the most globally networked
and logistically optimised industries’ (VW-5). Because of this, all the respondents felt
that, when it came to strategies for introducing digital technologies abroad, it was

Pamela Meil
necessary to include all the actors in order for the system to work the way it should. This
not only goes for the OEMs but the suppliers as well. Thus, the idea is to link not only
the various sites and tier one suppliers, as is the case now, but in the future also the tier
two and even tier three suppliers. Only in this way can the advantages of Industry 4.0
technologies be optimised. EDAG is one of those tier one suppliers, but it also carries
out work as a general contractor for OEMs. The EDAG respondent pointed out that
everyone along the chain must have the same tools and methods. ‘To put it bluntly, I
can’t have my colleagues in India use a drawing board while we work with CATIA V5
[an engineering software program].’ The main argument why digital transformation
will tend to upgrade the individual parts of the value chain is that the desired increases
in efficiency, transparency, monitoring and control systems cannot be achieved without
an overall coordination between all of the actors, including suppliers and remote sites.
It is an economically-driven logic, not one based on empowerment or a strategy for
upgrading.
What implications the resultant upgrading has for the organisation of work, the division
of labour and workers’ competencies and qualifications at individual sites seems to be a
secondary consideration. The local sites have to adapt because the requirements come
from the headquarters to deploy certain technologies and achieve targets.
9. Reshoring?
There is a fair amount of hype in the current discourse in Germany about the potential
for reshoring, in particular in connection with the increased deployment of Industry 4.0
applications. The argument is that new investments in digital technologies that bring
high levels of efficiency and capital utilisation, together with synergy effects concerning
production, services and customers, makes reshoring increasingly attractive (Strange
and Zucchella 2017).
Quite frankly, the evidence for this position is not particularly convincing. For one thing,
the levels of Industry 4.0 deployment in Germany are still so marginal that it is hard
to mount arguments that it is leading to reshoring. Certainly, there has always been a
certain amount of reshoring by companies which underestimated the transaction costs
and the long-term investments involved in outsourcing, especially for complex products
and processes. Levels of reshoring have actually remained amazingly constant over the
last decade (Eurofound 2016). Nevertheless, the extremely high levels of outsourcing
characterising the latter part of the 20th century have slowed down, partly as a result
of saturation levels having been reached and partly as a result of the post 2008-2009
financial crisis.
Companies which have a longer track record of outsourcing, however, a category to

which carmakers belong, have complex value chains which are part of a dynamic process
of growth and change. Once offshoring or outsourcing has taken place, various forms of
upgrading usually occur: products, processes, functions, shifting to new sectors or shifts
in complete value chains – all of which have labour and capital dimensions (Gereffi et
al. 2011; Gereffi and Kaplinsky 2001). In the development of value chain integration, the

movement tends to be from lower to higher level activities within sectors, for instance to
high value added and more knowledge-intensive activities (Meil 2019). The automobile
sector is, naturally, part of the industrial group which has high levels of knowledge-
intensity (MGI 2019).
There are basically three scenarios that can be hypothesised as outcomes for offshored
or outsourced sites as a result of the introduction of Industry 4.0 digital applications:
1. The central sites invest in new technologies at home and thereby pursue a reshoring
strategy. This would mainly result in the downsizing or closure of offshored or
outsourced sites over time
2. Value chains undergo a shift – this can occur in two directions:

— an upgrading of the site through investment and the introduction of new
digital technologies
— a downgrading of the site through a lack of investment in new digital
technologies, thereby reducing it to carrying out low-level tasks with low
levels of skill. The site would service mainly as a cost-cutting destination,
based on standardised products
3. There is no clearly discernible change in the relationship between offshored or

outsourced sites and the headquarters, and roles in the value chain are not affected
by Industry 4.0 technological transformation.
The vast majority of respondents in this sample did not believe that digital transformation
would lead to an increase in reshoring. The consensus was that value chains across
sites were already highly integrated and interconnected and that new investments in
Industry 4.0 were unlikely to change this.
As we have seen, some respondents here did mention the possibility that companies
could decide to concentrate their ‘old’ technology or products, i.e. diesel motors or cars
using the combustion engine, in sites outside Germany while building plants for new
products and applications at home. The opinion was that, given core competencies and
quality considerations, it made sense to keep important processes and products for the
future in Germany. Some companies, such as Festo in this sample, concentrate on high-
end products and in any case manufacture largely within Germany. Indeed, they expect
that, with such new technologies as 3D printing, specialised niche production could be
carried out more easily in domestic plants.
Festo’s position on reshoring was most definitely a minority position among respondents
in this sample, however. Among the rest, a common position did exist that, for new and
proprietary products, it might be worthwhile considering what could be produced at
home rather than be externalised. On the other hand, it was considered highly unlikely
that production that had already been outsourced would be brought back, nor that not
investing in the digital transformation of existing remote sites across the value chain
was a realistic option. In general, the strategy of differentiating between old and new
products in terms of site use does not appear to be particularly clear-cut at present. It

Pamela Meil
would be worthwhile considering which areas of Europe might be most susceptible to

strategies geared to differentiating between old products and new ones.
10. Concluding remarks

OEMs and tier one system suppliers who would, presumably, be the companies
dispersing new digital and Industry 4.0 applications across their value chains are still
in the process of undergoing transformations themselves. Certainly, some applications,
particularly in the automobile industry, are quite advanced: those involving logistics,
ERP systems, automation with standard robots and some shop floor digital interactive
software. Others are currently being introduced in the form of pilot projects. But for
now, it is hard to discern any clear trend in terms of shifts in the division of labour or
jarring changes in the organisation of work.
It does seem clear, however, that change facilitated by digital tools will continue and
that it will have effects as regards job losses and job gains, shifts in occupational and
competence profiles and, possibly, adjustments in the position and role of various sites
across the value chain. For production in the carmaking industry, it does not appear
that there will be major differences among sites further down the chain when re-tooling
for more digital processes and networking of systems. It does not make sense to have
sites at quite different stages of development. However, it is particularly in assembly,
where the lowest-skilled parts of production exist, that the greatest moves toward
automation, and accompanying employment loss, will take place. Although there might
be some time lag in re-tooling between sites – old or new, outsourced or at home – the
planning is for it to occur at all sites along the chain. This does not necessarily mean
that the situation for all sites along the chain will be the same. In times of recession, as
in the current corona crisis, employment reduction will tend to occur in places where
the conditions for labour are more precarious. That is, those with less labour protection
and union organisation.
In terms of decision-making, strategy, steering, etc. there are also differences across the
chain. The companies examined here have their design, development and innovation
management facilities mainly at home. It is at the headquarters where high-end
developments in digital transformation take place and which ultimately decides what
technologies get developed and where they are implemented. It is largely left to local
sites to find ways to adapt to these changes, for instance, in the preparation of their
workforces. Although local units have a say in capacity utilisation and other control
issues relevant for their plants, decision-making is generally becoming more centrally-
managed at company headquarters. This is especially evident for VW which, formerly,
had a large network of different brands with fair amounts of autonomy to operate in
their particular areas. Now, there is a push by Production or Group Components to
streamline the use of sites. In all the cases described here, this has certainly not meant
a phasing-out of sites located outside the home country; at least, not up to now. In fact,
there are sites in eastern Europe that, although small, are some of the most modern in
the company portfolio.

Ultimately, it does seem that there will be a shift in value chains with some upgrading
occurring for those sites that survive the new demands. However, it appears that there
will also be an increase in the monitoring and centralised analysis of data and that the
design, engineering and development tasks linked to Industry 4.0 will remain, mostly,
at home.
Sars – Cov-2
It would be remiss not to try to consider what effects the pandemic – the most significant
event of our time – might have on the issues discussed here.
Some of the interviews and the writing of this chapter took place during the corona
crisis. It should not be expected that this crisis would have a particularly strong impact
on the results of this study but it was, nonetheless, an all-embracing topic on the minds
of participants. Firstly, a great deal of their work, as well as some of the empirical
work carried out for this project, was moved to the digital world. Many respondents
felt this would have a lasting impact on how work is carried out and that much more
would now be done virtually. It also made it all the clearer how inextricably integrated
and interconnected were their supply and production chains within the global world
of the automobile sector. This was not necessarily a soothing thought. They did not
feel this would necessarily change post-corona, but it did bring to the forefront the
interdependence and fragility of systems. Thus, there were some new considerations of
whether proprietary systems or technologies that were of particular significance for the
company’s future should, perhaps, be kept closer to home. However, these cautionary
thoughts seemed to apply less to Europe and rather more for sites further afield.
Nonetheless, the major recession which is confronting the global economy, and which is
hitting the automobile sector as much as any other, is most likely to lead to job loss and
site closure. It is easier to cut employment in countries in which labour protection and
job regulation are low, and additionally in which the role of local markets is not so high.
This could be bad news for the sites of German carmakers in the emerging economies
of eastern Europe. It may be temporary, but it could well lead to a delay in shifts toward
digital transformation across the value chains.

Pamela Meil
References
Böhle F. (1998) Technik und Arbeit - neue Antworten auf „alte Fragen“, Soziale Welt, 49 (3),
233-252.
Böhle F., Bolte A., Drexel I., Dunkel W., Pfeiffer S. and Porschen S. (2002) Umbrüche im
gesellschaftlichen Umgang mit Erfarhrungswissen: theoretische Konzepte, empirische Befunde,
Perspecktiven der Forschung, München, ISF.
Cattaneo O., Gereffi G. and Staritz C. (2010) Global value chains in a postcrisis world: resilience,
consolidation, and shifting end markets, in Cattaneo O., Gereffi G. and Staritz C. (eds.) Global
value chains in a postcrisis world: a development perspective, Washington D.C., World Bank,
3-20.
Eurofound (2016) – ERM annual report 2016: Globalisation slowdown? Recent evidence of
offshoring and reshoring in Europe, Luxembourg, Publication Office of the European Union.
https://www.eurofound.europa.eu/sites/default/files/ef_publication/ field_ef_document/
ef1652en.pdf
Gereffi G., Fernandez-Stark K. and Psilos P. (2011) Skills for upgrading: workforce development and
global value chains in developing countries, Durham, Duke University: Center on Globalization,
Governance & Competitiveness.
Gereffi G., Humphrey J. and Sturgeon T. (2005) The governance of global value chains, Review of
International Political Economy, 12 (1), 78-104.
Gereffi G. and Kaplinsky R. (2001) The value of value chains: spreading the gains from
globalisation, IDS Bulletin, Special issue, 32 (3).
Heidling E., Meil P., Neumer J., Porschen-Hueck S., Schmierl K., Sopp P. and Wagner A. (2019)
Ingenieurinnen und Ingenieure für Industrie 4.0, Frankfurt, IMPULS-Stiftung.
Henderson J., Dicken P., Hess M., Coe N. and Yeung H.W.-C. (2002) Global production networks
and the analysis of economic development, Review of International Political Economy, 9 (3),
436-464.
Hirsch-Kreinsen H.(ed.) (2014) Wandel von Produktionsarbeit – „Industrie 4.0“, Arbeitspapier 38,
Dortmund, Technische Universität Dortmund. https://www.wiwi2.tu-dortmund.de/wiwi/ts/de/
forschung/veroeff/soz_arbeitspapiere/AP-SOZ-38.pdf
Humphrey J. and Schmitz H. (2002) Developing country firms in the global economy: governance
and upgrading in global value chains, INEF-Report 61, Duisburg, INEF.
Kagermann H., Wahleter W. and Helbig J. (2013) Umsetzungsempfehlungen für das
Zukunftsprojekt Industrie 4.0. Abschlussbericht des Arbeitskreises Industrie 4.0, Frankfurt am
Main, Geschäftsstelle der Plattform Industrie 4.0. http://www.acatech.de/
Kagermann H., Wahleter W. and Helbig J. (2013) Securing the future of German manufacturing
industry Recommendations for implementing the strategic initiative INDUSTRIE 4.0 Final report
of the Industrie 4.0 Working Group, Frankfurt am Main, Geschäftsstelle der Plattform Industrie
4.0.
Kinkel S., Rahn J, Rieder B. Lerch C. and Jäger A. (2016) Digital-Vernetzes Denken in der
Produktion, Frankfurt, IMPULS-Stiftung.
Lichtblau K., Stich V., Bertenrath R., Blum M., Bleider M., Millack A., Schmitt K., Schmitz E. and
Schröter AM. (2015) Industrie 4.0 Readiness, Frankfurt, IMPULS-Stiftung.
Linden G., Dedrick J. and Kraemer K. (2011) Innovation and job creation in a global economy: the
case of Apple´s Ipod, Journal of International Commerce and Economics, 3 (1), 223-240.
MGI (2019) Globalization and transition: the future of trade and value chains, McKinsey Global
Institute (MGI).

Meil P. (2019) Wandel der globalen Wertschöpfungsketten und (Rück-)Verlagerungen von

Produktionsstandorten durch Digitalisierungsprozesse und Industrie 4.0, in Hirsch-Kreinsen
H. and Karačić A. (eds.) Neue Chancen für den Industriestandort durch Industrie 4.0?
Rückverlagerung, Regionalisierung und Urbane Produktion, Dokumentation zum Workshop am
16. Juli 2019, Düsseldorf, FGW – Forschungsinstitut für gesellschaftliche Weiterentwicklung,
25-29.
Meil P. and Heidling E. (2010) Der Dienstleistungskreislauf - Neue Organisations- und
Kompetenzanforderungen entlang der Prozesskette, in Heidling E., Böhle F. and Habler T.(eds.)
Produktion und Dienstleistung - Integration als Zukunftschance, München, Rainer Hampp
Verlag, 135-156.
Meil P., Heidling E. and Schmierl K. (2003) Die (un)sichtbare Hand. Nationaler Systeme
der Arbeitsregulierung in die Ära des Shareholder Value – Ein internationaler Vergleich:
Deutschland, Frankreich, USA, München, ISF.
Pfeiffer S, Lee H., Zirnig C. and Suphan A. (2016) Industrie 4.0 – Qualifizierung 2025, Frankfurt
b2b7-1b01a04d166c.
Porter M. (1985) Competitive advantage: creating and sustaining superior performance, New York,
Free Press.
Strange R. and Zucchella A. (2017) Industry 4.0, global value chains and international business,
Multinational Business Review, 25 (3), 174-184.
Sturgeon T. and Memedovic O. (2011) Measuring global value chains: intermediate goods trade,
structural change and compressed development, UNIDO working paper, Vienna, UNIDO.
Sturgeon T. and van Biesebroeck J. (2011) Global value chains in the automotive industry: an
enhanced role for developing countries?, International Journal of Technological Learning,
Innovation and Development, 4 (1/2/3), 181-205.

Digital transformations in factory economies
Chapter 3
Digital transformation and local manufacturing
subsidiaries in central and eastern Europe:
Changing prospects for upgrading?
Andrea Szalavetz
1. Introduction
The digital transformation (DT) of value generation is expected to transform the
drivers of growth, upgrading and modernisation in ‘factory economies’.1 Advanced
robotics, cyber-physical systems and artificial intelligence-powered business process
automation are anticipated to bring about unprecedented technological unemployment
(Brynjolfsson and McAfee 2014; Frey and Osborne 2017), in particular in countries
specialised in activities that are exposed to automation (World Bank 2016). Some
pessimistic observers have contended that these technologies may even induce
a downgrading process in these countries or jeopardise local subsidiaries’ prior
upgrading achievements by automating some relatively knowledge-intensive tasks2
now being performed by local engineers (Flecker and Schönauer 2016; Szalavetz 2017).
Moreover, the very reason for maintaining the current pattern of the global division of
labour (keeping previously relocated labour-intensive business processes in low-cost
countries) might also be questioned since smart factories controlled by a minimum
number of staff can be located anywhere, e.g. close to final markets or in investors’
home countries (Dachs et al. 2017).
However, an opposite scenario is also conceivable in which existing manufacturing

units, representing locally-embedded production capabilities, are upgraded by advanced
manufacturing technologies. Consequently, FDI-hosting factory economies could
undergo further capital deepening, with local manufacturing subsidiaries receiving
further investment in tangible and intangible capital. Moreover, DT might support and
enhance the decentralisation of increasingly advanced activities within organisations,
including engineering, design and software development. This would enable factory
economy actors to accumulate technological and R&D capabilities (Szalavetz 2019a)
and increase the knowledge intensity of their contribution to total value added. In short,
while the first, pessimistic scenario is about factory economy actors’ downgrading and
the loss of previously-acquired competitive advantage, this latter scenario suggests a
DT-driven further modernisation and upgrading of these countries.
1. According to Baldwin’s (2013) categorisation, in international production networks there are headquarter
economies, where economic actors mainly govern the production networks (and carry out business development
and other intangible activities); and factory economies that provide the labour.
2. Examples include: tooling and design – jeopardised by the diffusion of additive manufacturing solutions;
process development – taken over by self-optimisation solutions embedded in cyber-physical production
systems; production planning – superseded by smart planning algorithms; maintenance planning – subsumed
within embedded predictive maintenance solutions; engineering – taken over by virtual engineering (cf. Will-
Zocholl 2016); and other technological support tasks.

Andrea Szalavetz
This research seeks to contrast these two contradictory hypothetical scenarios with initial
empirical evidence drawn from three central and eastern European (CEE) countries:
Czechia, Hungary and Poland. Interview-based case study research was conducted at
a sample of automotive subsidiaries in these countries to explore the developmental
outcomes of DT.
The automotive industry, dominated by foreign-controlled, export-oriented manu

facturing units: subsidiaries of global original equipment manufacturers and their
global suppliers (Pavlínek 2017), was selected as the specific context for the research
since this industry is a forerunner, also in central and eastern Europe, in adopting
digital technologies. With nearly continuous large-scale investment inflows, this
industry has been one of the main drivers of growth, employment and exports in the
‘integrated periphery’ of the European automotive industry (Pavlínek 2018).
The rest of the chapter proceeds as follows. First, some related strands in the literature
are listed and reviewed (section 2). Research design, data collection and data analysis
methods are outlined in section 3, while the results of the data collection exercise are
presented in section 4. Section 5 provides a discussion and some concluding remarks.
2. Related literature
There are at least four strands in the literature that are relevant to this research
(Figure 1).
Figure 1 Research related to digital transformation
GVC theory
upgrading downgrading
Digital transformation
(DT)
DT in CEE DT in automotive
Dunning’s eclectic industry
paradigm
Technology- O
driven L
changes Automotive I
GVCs
CEE within automotive
GVCs
Source: elaborated by the author

Digital transformation and local manufacturing subsidiaries in central and eastern Europe: Changing prospects for upgrading?
The first is the scholarship on global value chains (GVC).3 The GVC method of analysis
is constituted from an analytical approach used to investigate changes in (a) the global
composition of the value adding activities of geographically dispersed, networked and
functionally integrated economic actors; (b) the governance of these activities; and (c)
the global distribution of value added (Dicken 2003; Gereffi et al. 2005; Gereffi and
Fernandez-Stark 2016). It is, in particular, the literature on upgrading – a key construct
in the GVC literature – that guides this research. Upgrading is defined as specialising
in higher value adding activities within GVCs than previously, achieved by enhancing
existing capabilities and/or developing new ones. In Humphrey and Schmitz’s (2002)
classification, upgrading may refer to (better) products; improved and more efficient
processes; higher-skill functions; and/or the shift to new and technologically more
advanced sectors or value chains. At the same time, the opposite tendency – the issue of
downgrading – may also be relevant (cf. Blažek 2016).
Another stream of research deals with the economic and business implications of
digital transformation. Rapid developments in computer science and in information
and communications technologies, the emergence of several enabling technologies4 and
smart applications, and the interplay between manufacturing science and computer
science and technology (Monostori 2015) have all revolutionised manufacturing
operations and business management practices. Digital solutions improve the excellence
of operations, enhance productivity, contribute to resource optimisation and allow for
faster and more substantiated (data-supported) decision-making (Brettel et al. 2014).
Note that most scholars maintain that the revolutionary aspect of DT is not limited to
manufacturing production. DT is, rather, about an across-the-board transformation of
business, implying new business models and new ways of organising, integrating and
controlling value adding activities. Consequently, digital transformation is also referred
to as the fourth industrial revolution, or Industry 4.0 for short (Kagermann et al. 2013;
Manyika et al. 2013; Schwab 2016).
The studies most closely related to the subject of this chapter take a focused perspective,
discussing the specifics of digital transformation in CEE (e.g. Horváth and Szabó 2019;
Prašnikar and Redek 2019; Szalavetz 2017) and/or in the automotive industry. These
latter contributions are concerned not only with the impact of digital technologies on
automotive end-products (vehicles), components, production processes and associated
business functions but they also explore digitalisation-driven changes in business
models and in the composition of and key actors in GVCs (e.g. Burkacky et al. 2019;
Ferràs-Hernández et al. 2017; Xu 2019).
The third strand of the literature on which this research draws originates in Dunning’s
eclectic paradigm (Dunning 1993), applied in particular with regard to the question of
whether any technology-driven changes can be observed in firms’ ownership, location
and internalisation advantages (Strange and Zuchella 2017). For example, the issue of
3. GVCs describe the full range of the tangible and intangible activities carried out to bring a product or service
from its conception to its end use and beyond (Gereffi and Fernandez-Stark 2016).
4. These include the Internet of Things (IoT), cloud computing, 3D printing, artificial intelligence, big data
analytics, virtualisation and augmented reality. Some scholars refer to cyber-physical production systems as the
epitome of the digital transformation of manufacturing (e.g. Monostori et al. 2016).

Andrea Szalavetz
offshoring and backshoring in the Industry 4.0 era (Dachs et al. 2017) can be discussed
within Dunning’s framework: in terms of firms’ evolving competitive and location
strategies (Di Mauro et al. 2018) or in terms of the evolution of governance modes in
international business networks (Alcácer et al. 2016).
Papers in the fourth research strand are concerned with the features of automotive
value chains (e.g. Sturgeon et al. 2008) and with (any changes in) the position and role
of factory economies in CEE within automotive value chains (Pavlínek 2017).
These research strands all convey the message that GVCs are in constant flux, hence
they need to be analysed taking an evolutionary approach. GVC dynamics, manifested
also in the phenomena of actors’ upgrading and downgrading, is driven among others
by external factors (e.g. changing business, institutional and regulatory environments),
lead firms’ adaptation and strategic actions, actors’ capability accumulation and,
most importantly from the point of view of this research, technological progress. New
technologies may transform both the existing organisation of value creation activities
and associated power relations. For example, DT is expected to have a transformational
impact on various dimensions of GVCs, including firm-specific and locational advantages,
geographic scope and governance (Porter and Heppelmann 2014; Rehnberg and Ponte
2018; Strange and Zuchella 2017).
Against this background, we propose that digital technologies have produced an

outwards shift in the production possibility frontier. In line with the theory of GVC
integration-driven catch up (OECD 2013; UNCTAD 2013), in low-cost locations the
local manufacturing subsidiaries of global companies were the first to embrace these
technologies. The integration of these technologies in the production systems of
local subsidiaries brings about an array of opportunities to increase the efficiency of
operations. A pure deployment of new technologies is not sufficient: to exploit these
opportunities, local subsidiaries have to develop their technological capabilities and
make complementary intangible investments, e.g. transforming their processes and
organisational set-up to implement new production methods. Consequently, in addition
to learning-by-doing and process upgrading, digital upgrading also engenders functional
upgrading. Moreover, upgraded production methods and the related increases in
subsidiaries’ competences can substantiate product upgrading; that is, assignments to
manufacture technologically more sophisticated products than previously. Additionally,
since digital transformation increases the complexity and the software-intensity of all
value adding processes, this may incentivise parent companies to delegate partial R&D
tasks to competent subsidiaries.
Altogether, digital upgrading enables both process and functional upgrading and may
also beget product upgrading. Conversely, delays in – or the lack of – digital upgrading
are associated with a rapid loss of competitiveness since the distance of companies with
unchanged technology to the production possibility frontier thereby increases to such
an extent that it makes survival impossible.

3. Research design, data collection and analysis

Since the purpose of this research was to clarify which of the hypothetical scenarios
advanced in the literature on the developmental impact of DT is supported by real-
world evidence, we decided on an exploratory, qualitative approach, drawing on a field-
based data collection method: multiple case study analysis (Eisenhardt 1989; Yin 2014).
We applied the method of purposeful sampling (Patton 1990) and chose companies
representing illuminative cases from the point of view of implementing digital manu
facturing technologies.
We selected companies that differ in their degree of Industry 4.0 maturity. The
literature abounds in measurement models for the maturity of Industry 4.0 (e.g. Mittal
et al. 2018; Nick et al. 2019; Schumacher et al. 2016; Schuh et al. 2017; Scremin et al.
2018). These authors analyse various dimensions of Industry 4.0 readiness, including
the breadth and depth of the utilisation of various Industry 4.0 technologies, the
smartness of products, the digitalisation of transactions (with customers and partners),
the integration of digital technologies in the production process (‘operations’), the
breadth and depth of data-driven decision-making and the extent of integration of
digital technologies in corporate practices, standards and business models. Maturity
models also include indicators quantifying employees’ competencies and readiness to
work in an Industry 4.0 environment and indicators evaluating the sophistication of
management strategy regarding digitalisation.
These studies apply five or six stages describing the levels of Industry 4.0 maturity
ranging from basic level (in the technologies and processes dimensions, this refers
to the isolated deployment of IT-embedded solutions and partial connectivity) to full
implementation (i.e. fully-digitalised production systems featuring horizontal, vertical
and end-to-end integration of processes, functions and activities, and which allow for
self-optimisation and self-adaptation).
It is important to bear in mind that selected dimensions of maturity are not relevant,
or are only partially so, for manufacturing subsidiaries. For example, the dimension
of ‘customers’ (use of customer data, digitalisation of sales) does not apply, since this
belongs to the authority of the HQ. In a similar vein, local subsidiaries have no say in
decisions about (transition to digital) business models. The dimension of ‘products’,
referring to product data collection over the product lifecycle and the creation of
digital product-services systems, applies only partially since the maturity stage in these
dimensions is a function of HQs’ strategic choices concerning whether to transfer the
related activities and know-how to subsidiaries.
The dimensions that are relevant with respect to subsidiary-level Industry 4.0 maturity
are ‘operations’, ‘technology’, ‘management competences’, ‘culture’ (e.g. knowledge
sharing) and ‘people’ (the ICT competences of employees, the openness of employees
to new technology and the autonomy of employees). Note that, as described below, our
interview questions focused only on the ‘technology’ and ‘operations’ dimensions since
the purpose of this research was not to evaluate the maturity of the surveyed companies

Andrea Szalavetz
but rather to explore the impact of investments on subsidiary upgrading. It is, therefore,
beyond the scope of this paper to provide a detailed overview of the development levels
pertaining to each stage of Industry 4.0 maturity. Firms were selected if they displayed
at least stage 2 maturity in any of the indicators of these two considered dimensions.
The sample consists of 28 large, export-oriented companies, subsidiaries of global

automotive companies and tier one suppliers operating in Czechia, Hungary and
Poland.5 Our aim to include local subsidiaries of the same lead companies from each
country was only partially successful: the sample includes two subsidiaries of the
same mother company operating in Poland and in Hungary; two others operating in
Poland and Czechia; and two instances of subsidiaries operating both in Czechia and in
Hungary. Table 1 summarises the specifics of the empirical data.
Table 1 Empirical data collection
Czechia Hungary Poland

Number of firms interviewed 12 10 6
Additional interviews with Representatives of an Representatives of A representative of a trade
employer organizations and employer association and (1) Metalworkers Federation; union federation and a tier
trade unions sectoral unions (2) Association of one supplier (informing
Hungarian Automotive about general Industry 4.0
Component Manufacturers trends and the maturity of
Polish firms)
Interviewees TU (5), IT manager, division CEO, CTO, director of TU (2); director of

manager (logistics), operations; TU (2), HR (2), production/operations (3);
technology officer, Industry other* director of a division
4.0 specialist
* ‘Other’ includes an Industry 4.0 project officer, a digital engineering team leader, a chief information officer and representatives
of the work council
HR = human resources officer; TU = trade union representative; CTO = chief technology officer; CEO = chief executive officer
The interview protocol, consisting mainly of open-ended questions to facilitate

exploration, was designed around three6 main topics: (1) the specifics of the Industry
4.0 technologies adopted by the given companies; (2) the motivations of the surveyed
firms’ investments in advanced manufacturing technologies; and (3) the developmental
outcomes of digital technology implementation. Regarding this latter issue, the
questions were intended to explore whether and how DT fosters upgrading; and whether
it can produce any changes in the GVC role of the given subsidiaries. Finally, we also
asked whether interviewees expect any changes in the location advantages of factory
economies as a result of DT.
5. Data collection and analysis was conducted by Monika Martišková in Czechia, Kristóf Gyódi and Katarzyna
Śledziewska in Poland, and Andrea Szalavetz in Hungary.
6. Only the topics included in this summary chapter are referenced here. There were additional questions with
regard to the impact of digital manufacturing technologies on employment and the nature of work. These
questions and the related findings are discussed by Monika Martišková in chapter 8.

Interviews were conducted between January and March 2018; and, since the
implementation of Industry 4.0 solutions has intensified only recently, a period of five
years (between 2013 and 2017) was selected as the period for which survey data would
be gathered.
Interviews lasted thirty to ninety minutes. Multiple data sources, including press
releases, corporate websites, business press articles, company reports and notes to the
financial statement have been employed in order to triangulate the findings. Detailed
descriptions of each case formed the basis of within-case and cross-case analysis
(Eisenhardt 1989). This made it possible to cross-check interviewees’ remarks regarding
specific issues and identify consistencies or contradictions.
The main limitation of this case study analysis is the small size and the biased nature of
the sample, consisting of companies operating in an industry that is a digital forerunner.
Consequently, although the conclusions drawn from the insights obtained during the
interviews may not be generalisable, the research has considerable value in terms of the
insights it offers into the future for automotive manufacturing subsidiaries located in
CEE under the impact of digitalisation.
4. Results: Descriptive analysis

4.1 Adoption of digital manufacturing technologies
On average, the surveyed companies display a relatively high degree of Industry 4.0
maturity; at least, in the light of the low average performance of business digitalisation
in these countries.7 Nevertheless, the breadth and depth of digital technology adoption
is highly heterogeneous across the sample.
The activity mix of the companies interviewed is a mixture of highly automated and
manual/semi-manual activities. Processing is, in most cases, fully automated and
manual workers load and discharge the machinery. The individual components of the
production system are of a heterogeneous level of technology. Less than half of the
sample companies reported that they employ collaborative robots or driverless in-plant
transport systems (AGVs). The managers interviewed explained the lack of AGVs with
reference to space constraints in their factories and pointed out that new, greenfield
facilities are already designed in a way that would permit extensive robotisation.
Nevertheless, the companies had started to invest in industrial and service robots,
employing them in processing activities (e.g. welding, cutting and painting), assembly,
warehouse management and materials handling.
7. ‘Average performance’ denotes the business digitisation performance score of the Digital Economy and Society
Index, specifically the percentage of enterprises using electronic information sharing, social media, big data
analytics and cloud solutions. According to the most recent data (DESI 2019), Hungary and Poland scored
among the lowest in Europe in terms of the integration of digital technologies (Hungary was 27th, Poland 25th
and Czechia 23rd in the EU-28 (DESI 2019). Hungary scored also quite lowly in terms of the share of enterprises
using industrial or service robots (just 3 per cent) (Eurostat: https://ec.europa.eu/eurostat/web/products-
eurostat-news/-/DDN-20190121-1).

Andrea Szalavetz
Over and above these basic and isolated, albeit spectacular, manifestations of Industry
4.0 technologies, the surveyed companies have all progressed along the stages of
the connectivity of production processes and business functions (such as inventory
management, material flows and maintenance). Process data are extracted and, in the
case of the more developed half of the sample, fed into the manufacturing execution
system. Production status and key performance indicators are visualised and, in about
25 per cent of cases, even analysed (through embedded analytical solutions) for data-
driven decision-making.
On average, the managers interviewed have adequate knowledge of Industry 4.0,8 albeit
the heterogeneity of the sample applies in this respect as well. Accordingly, over and
above robots, they would mention the term cyber-physical system, i.e. mechanisms
to generate, capture, store and process data in order to improve the performance of
operations. Additionally, interviewees reported that some production-related business
functions are digitally supported. Examples of smart solutions include the real-time
tracking of production processes, dashboard-based visualisations of key performance
indicators, intelligent production monitoring systems, data-driven production
scheduling, machine vision-based quality testing and predictive maintenance solutions.
Some informants reported investment in the harmonisation of their own IT systems
and that of their tier one and tier two suppliers so that lead companies could gain a real-
time overview of processes along the whole supply chain.
Most of the respondents pointed out that DT is a long and gradual journey. Currently,
smart technologies are integrated in legacy shop floor environments – in a way to avoid
any disruptions or disturbances in ongoing production that is running at full capacity.
Transforming a ‘running’ production system, however, poses formidable difficulties, as
illustrated by the following interview excerpt.
‘It is not only our inability to finance the costs of investing in digital solutions.
You know, we are running at full capacity and do our best to meet the deadlines
and produce the required volumes. We simply do not have the capacity to engage
in a lengthy exercise of screening our processes, elaborating a DT plan, looking
for technology suppliers, interacting with them, restructuring the processes and
implementing the new solutions.’
Moreover, since different activities are controlled by different software solutions, the
harmonisation of heterogeneous legacy systems is indispensable to enabling data
integration and the interconnection of all activities and processes. This is a precondition
of the transition to Industry 4.0 – from the current ‘Industry 3.0 +’ environment
prevailing in the dominant majority of firms in the sample. As a rule of thumb, it was
found that the newer the production site, the more digitally mature it is.9
8. However, only four of them have an overarching, subsidiary-level DT strategy in place.

9. Some companies have already started to invest in the automation of data analytics and even in the
implementation of artificial intelligence solutions, for example, as a means of identifying the correlation between
the various monitored processing parameters and product quality; or have developed predictive analytics
solutions to avoid machine failures.

Figure 2 summarises sample companies’ investments in Industry 4.0 technologies.

Note that not even the most developed companies can be classified as having achieved
stage 4 maturity. Although these companies are experimenting with, or have introduced,
selected stage 4 solutions, they are still far from displaying the maturity level that
characterises stage 4 companies. Characterised by a compressed development towards
digital maturity, these companies would be implementing both stage 3 and stage 4
investments. Moreover, the dominant majority of sample companies is in the process
of implementing stage 2 and some stage 3 investments. Inter-country differences – the
Hungarian companies in the sample feature much higher Industry 4.0 maturity than do
the Czech or Polish ones – are the result of biased sample selection rather than reflective
of a higher preparedness among Hungarian companies to embrace digital technologies
(cf. Nick et al. 2019).
Figure 2 Examples of investments in digital technologies in surveyed companies,

classified according to associated maturity level
– Production control through cyber-physical systems, manufacturing execution systems;
4 – Paperless shop-floor management, digital production planning, predictive maintenance;
– Inventory management through radio frequency identification technology, digital simulation of
processes for optimisation.
– Visualisation of production status based on real-time data analytics, robotic process automation
3 (e.g. of quality control);
– Advanced internal connectedness.
2 – Factory automation, collaborative robots, automated material handling, automated guided vehicle;
– Data collection through cyber-physical systems, harmonisation of legacy IT systems.
Notes: Stages of digital maturity: 1. No industry 4.0 (only factory automation, including older generations of fenced robots); 2. More
advanced solutions working in an isolated environment co-existing with legacy machinery; 3. Connection of value adding components;
digital monitoring; 4. Production control through cyber-physical systems; 5. Completely automated factory (e.g Adidas’ Speedfactory).
Manufacturing execution systems are software packages used to manage factory floor material flows; track and optimise labour
and machine capacity; provide real-time information about inventory and orders; and optimise production activities. Note that the
integration of shop floor data and those from the enterprise system, implying automated data and information exchange, has been
implemented in only a few companies.
Source: Author’s compilation based on interview insights
4.2 Motivation to invest in digital technologies

Apart from the integration of digitally connected, autonomous robots in the production
system and the automation of selected support functions (robotic process automation),
most of the above-listed digital solutions aim at obtaining insights that support
interventions in complex manufacturing processes and achieving better control of
operations.
Companies in the sample have decided upon the automation of core and the digitalisation
of support functions in an effort to resolve the problem of labour shortages; enhance
the quality, flexibility and transparency of operations; and improve productivity
and process efficiency. Some of these motivations are interdependent. For example,
increased transparency allows a rapid reaction to process anomalies, which improves

Andrea Szalavetz
process efficiency. The real-time measurement and visualisation of process parameters

improves not only transparency, and thus enables data-driven decision-making, but
also allows for process optimisation e.g. through the reduction of internal transport
or of work in progress. In this vein, transparency contributes to process efficiency
improvement.
As the following interview excerpts illustrate, companies adopt nuanced, context-

specific approaches when they decide on investment in digital technologies.
‘Augmented reality tools and virtual simulation? No, we do not have such things
here: it is simply not needed. Factory planning is performed at central locations.
Planners use advanced digital factory planning solutions, such as the virtual
simulation of plant layout and material flows. We simply implement the received
plan, correcting and modifying it if necessary, but this kind of work does not
require advanced digital solutions here.’
‘I visited a partner subsidiary in Italy. It is equipped with the most advanced

production equipment and Industry 4.0 solutions: with everything that we would
just love to have. Obviously, we have to admit that much higher value added
products are manufactured at the Italian subsidiary: net sales per employee are
four times as high as in Hungary! They have the wherewithal to invest in these
technologies.’
‘Previously the only factor we considered when deciding about the automation of a
specific task was the return on investment. Now, over and above costs and return,
we consider many more factors: availability of workforce; operator workload; and
ergonomics.’
Technology upgrading through digital solutions was, in some cases, initiated by parent
companies prescribing that cloud-based solutions or paperless factories should be
implemented throughout the whole corporation. Most often, however, subsidiaries
themselves decided on the specifics of digital technology deployment. Subsidiary
managements face a ‘digitalisation imperative’ in a similar vein to headquarters.
However, in the case of headquarters, DT is about strategic differentiation and
business model innovation, since it strengthens the competitive advantage and enables
additional revenue generation (Szalavetz 2019b); whereas in the case of manufacturing
subsidiaries, the imperative of process upgrading through digital solutions is driven by
parent companies’ non-abating pressure to cut costs, increase efficiency, reduce cycle
time and improve both the flexibility and the excellence of operations. Subsidiaries are
thus encouraged to suggest and deploy digital solutions that would result in quality
improvements and/or cost savings and enable a prompt and flexible response to new
requests.
As these interview excerpts illustrated, subsidiaries have to finance these investments

themselves, which is compounded by the requirement to have their DT projects accepted
by parent companies. A Polish interviewee pointed out that a ‘Catch-22’ situation applies
in this respect: the relatively low local wage level delivers a lower return on investment

than that of DT projects in high-wage economies. Nevertheless, increasing local wages,

growing labour shortages – also in low-cost locations – and customer expectations in
terms of customisation, quality and delivery times make lead companies more inclined,
even in factory economies, to acknowledge the ‘robot dividend’ (cf. Huang and Sharif
2017).
4.3 The developmental outcomes of digital transformation – impact

on subsidiaries
Our interview results indicate that the implementation of digital technologies has
contributed to process upgrading in the surveyed companies – a precondition for
survival amidst inter-subsidiary competition for resources and lead companies’ aims to
streamline their supplier base. Manufacturing subsidiaries have been facing continuous
pressure to increase productivity and resource efficiency and to reduce the costs of their
operations. Above a certain threshold, however, this has proven to be increasingly
difficult to achieve – at least with traditional methods. The deployment of digital
technologies has opened up a whole range of opportunities for the further improvement
of the required indicators.
Moreover, increased digital maturity and the resulting improved efficiency and quality
were ‘rewarded’ by parent companies delegating more sophisticated production
tasks than previously (entailing product upgrading). Note that product upgrading
is the outcome of parent companies’ strategic decisions; subsidiaries have no say in
determining the composition of the product mix they manufacture.
About half the managers we interviewed spoke about DT-related functional upgrading,
highlighting that they have been assigned new and relatively more advanced tasks
than previously. Some local production units have obtained ‘product mandates’ i.e. full
responsibility for the further development of the products (e.g. specific components)
they manufacture and regarding the improvement of the related production processes.
Engineers in these companies have been assigned new tasks, such as product design,
simulation and software development, for example as regards the development of
the manufacturing execution system. They have been involved not only in analysis
of production technology malfunctions but have also been entrusted with process
development. Lead companies have delegated particular R&D activities to subsidiary
level: as corporate global R&D has become increasingly complex and multi-faceted,
subsidiary researchers and engineers have been assigned partial R&D tasks to be rolled
out to partner subsidiaries once completed.
Most of the new functional assignments which had been delegated to subsidiary level
were related to the increased ‘softwarisation’ of production and support processes. New
knowledge-intensive assignments have contributed to subsidiaries’ accumulation of
technological capabilities through learning-by-doing (Szalavetz 2019a).
Despite non-negligible achievements in the field of cost efficiency, operational excellence

and functional upgrading, the value chain position and autonomy of the surveyed

Andrea Szalavetz
subsidiaries have barely changed. These companies were, and remain, manufacturing
units within the global organisation of their parent companies, subject to hierarchical
governance that has not changed. Although some subsidiaries have acquired the
status of a competence centre, local autonomy has failed to increase in a meaningful
way. Investments in digital technologies have been decided upon according to the
same organisational mechanism as previously: a combined top-down and bottom-
up budgeting procedure. Subsidiary initiatives were accepted if, and only if, local
subsidiaries were in a position to cover the associated expenses, including the financing
(i.e. the hiring) of the staff involved in the development and deployment of the new
solutions. This has proved to be a remarkably hard constraint which has, in a number of
cases, hindered subsidiaries’ digital upgrading.
In other instances, the costs of subsidiary initiatives aiming at introducing advanced

digital solutions have (partially) been covered by parent companies; however, only if
the subsidiaries could prove that return on investment would be rapid, usually in less
than one year.
In addition to establishing a complete lack of digital upgrading-driven changes

in subsidiaries’ position in the value chain, it is worth investigating whether their
increased digital maturity, the resulting process upgrading and the accompanying
functional upgrading had any beneficial impact on basic corporate (subsidiary-
level) performance indicators. Our interview results indicate that, although both
employment and revenues grew considerably in the companies in the sample, these
developments were not necessarily associated with investments in digital technologies.
The improvement in performance indicators was, rather, driven by capacity expansion
and explained by the upswing in the business cycle; that is, by increasing demand for
the products manufactured by the local subsidiaries in the survey. Obviously, enhanced
digital maturity contributes to subsidiaries’ ability to cope with higher quantitative and
qualitative requirements. Altogether, it appears that the impact of digital upgrading on
subsidiaries’ performance indicators is beneficial, albeit only in an indirect manner.
5. Discussion and concluding remarks

The insights obtained from the companies interviewed suggest that the probability
of the pessimistic scenario, outlined in the introductory section, is quite low. In the
period covered by our survey, production expanded considerably in the companies in
the sample and this was accompanied by investments in tangible (advanced production
technology) and intangible assets. Since new production equipment already integrates
advanced digital technologies, investments in the harmonisation of the IT system and the
deployment of a manufacturing execution system were also considered indispensable.
Capacity expansion has brought to the fore the pressing labour shortages that local
companies have already been facing for several years. In order to prevent labour
shortages from becoming a bottleneck to further capacity increases, additional
investments have been made in the automation of production and support processes,
i.e. in the deployment of advanced robotic solutions.

These developments have led to upgrading along various dimensions, including process
and product upgrading, as well as functional upgrading driven by parent companies
delegating increasingly advanced tasks to subsidiary engineering teams.
These positive developments notwithstanding, there are some considerations that call
for caution.
Above all, we should note that these developments can, in part, be interpreted as a
lucky coincidence since the period under survey coincided with the longest upswing
in the automotive business cycle (Collie et al. 2019). Rapidly increasing demand
prompted investment in expanding the production capacity of lead companies’
existing manufacturing facilities and driving operational effectiveness through the
implementation of digital solutions. It was partly the path dependence originating in
global automotive companies’ past investment decisions, coupled with the upswing in
the automotive business cycle in the second half of the 2010s, that gave an impetus to the
gradual transition towards higher Industry 4.0 maturity in manufacturing subsidiaries
in CEE.
Furthermore, despite these unambiguously positive developments, the following

paragraphs argue that some of the anticipated DT-driven adverse effects may well
materialise, albeit later and more gradually than the projections of technological
alarmists.
First, further investments are expected that will increase the level of automation in the
subsidiaries we examined. These investments are driven partly by the necessity to keep
up with competitors implementing advanced technology and partly by the decreasing
price and dramatically improved features of robotic solutions. Another reason is
that the existing semi-automated or manual production technology in CEE is aging
towards obsolescence. The next phases in the evolution of the manufacturing facilities
we surveyed will be marked by a gradually increasing share of automated processes,
replacing the current manual or semi-manual, labour-intensive stages in subsidiaries’
production systems. This will by itself have a sizable labour-saving effect, triggering
technological unemployment.
Alternatively, with persisting labour shortages, in particular regard to skilled

maintenance staff, robot programmers and engineers, investors will reconsider the
locational advantages of their existing manufacturing facilities. Note that DT is bound
to reduce the importance of one of the important existing locational advantages for
CEE: the flexibility of the local labour force. Industry 4.0 technologies not only reduce
the labour-intensity of production but they also make existing production systems
adaptive, flexible and reconfigurable (Váncza et al. 2011).10 If technological solutions
10. It is, in particular, the modular organisation of the shop floor, a technological and organisational change
accompanying digital transformation, that has enabled production systems to become flexible and
reconfigurable. Modular organisation at the shop floor refers to the ease of adding new components to, or
subtracting obsolete ones from, the production system without the need to redesign the entire system or the
specific production process.

Andrea Szalavetz
enable production systems to adapt to changes in the external environment without

major increases in costs or reduction in throughput, the importance of labour flexibility
– that is, driven by lenient workplace regulation in CEE – will be reduced.
Revisiting past location decisions seems inevitable also because manufacturing facilities
in headquarter economies are also being upgraded by advanced manufacturing
technologies, and these latter investments are being supported by a variety of generous
policy instruments. Industry 4.0 technology-based capacity expansion in advanced
economies – brand new assets representing advanced digital production technology –
may effectively squeeze out existing low-cost production facilities, while there will be no
need even to backshore the previously relocated, old capacities.
The timing of these developments is difficult to predict.
For example, the timing of the transition to advanced automation and robotic techniques
at the surveyed companies (and at other automotive subsidiaries in CEE), implying a
reduction of labour intensity and, eventually, technological unemployment, is a function
of the depreciation of existing legacy assets. Past investments have created significant
path dependence; consequently, a hasty transition to advanced manufacturing
technologies would involve prohibitively high adjustment costs (in that case, existing
assets would need to be written off).
Apart from physical and technological obsolescence, the timing of asset replacement
is also influenced by the development of adjacent technologies. For example, advances
in materials science call for advanced processing technology: lightweight metal can
be more reliably processed and welded by automated technology. Other moderating
factors include workplace regulation and the intensity of competition. Compliance with
occupational health and safety regulations – or, more broadly, with good manufacturing
practice – requires an increasing use of advanced and smart technologies on the shop
floor (e.g. remotely controlled robots in painting and welding, or collaborative robots
in materials handling). Competition and customers’ ever-increasing expectations,
again, require the implementation of digital technologies to increase flexibility and
responsiveness.
The probability of the other development, according to which modern, automated and
digitally upgraded production facilities in advanced economies render local capacities
obsolete, is a function of three factors: 1) the pace and the direction of the development
of technology; 2) the business cycle; and 3) political pressure for reindustrialisation in
advanced economies compounded with generous policy support.
Regarding the first factor, the emergence of a new dominant design among competing
alternative powertrains may accelerate the obsolescence of some already-outdated
production facilities in CEE. In a similar vein, the imminent automotive downturn
(Collie et al. 2019) is bound to intensify the consolidation of the industry. When
capacities are aligned with demand, under-digitalised and underperforming plants are
the first ones to be closed. Furthermore, support programmes subsidising investment
in smart factories in advanced economies, and the associated political pressure for

reindustrialisation, may effectively shepherd the selection and retention strategies of

lead companies.
Interview evidence also indicates another cause for concern, namely that the structure
of value creation has barely changed in CEE. There are no signs of CEE actors shifting
to a high-road development path in which specialisation in advanced activities and
increasing unit value added would provide a major impetus to growth.
On the one hand, functional upgrading, the uptake of relatively more advanced, higher
value added activities has undoubtedly intensified at some of the companies in our
sample. Functional upgrading has fostered global companies’ local commitment and
their willingness to relocate further and more technology-intensive production to their
manufacturing sites in CEE. The positive effects of previous functional upgrading will
certainly be reinforced by subsidiaries’ implementation of digital technologies.
On the other hand, however, functional upgrading has not given a significant impetus
to local growth (cf. Milberg and Houston 2005). Global companies’ investments in
capacity expansion, upgrading and their relocation of additional production activities
have remained the main engines of growth in the surveyed period, dwarfing the growth
effects of functional upgrading.
In summary, while there are no signs of DT-induced new drivers of growth, the
traditional engines of growth in CEE factory economies are becoming increasingly
prone to erosion.
Consequently, it is safe to argue that the observed beneficial developments cannot

prevent, but only delay, some of the adverse effects of DT becoming manifest. The
surveyed period can best be described as a ‘lull before the storm’.
Interview findings and the resulting considerations have important managerial and
policy implications. The surveyed companies – similarly to other manufacturing
subsidiaries in factory economies – need to navigate between a rock and a hard place.
Evidently, investing now in automation and advanced digital solutions is the better
option, even if it entails some labour shedding, since increased digital maturity is the
precondition (but not the guarantee) of longer-term survival. Holding steady with
unchanged technology may keep the existing workforce in the short-term, but the
looming downturn in the business cycle will probably hasten parent companies’ adverse
location decisions.
At the same time, policy-makers need to recognise that DT-driven devastating

technological unemployment is not fate – not even in those countries that are more
exposed to the disruptive effects of DT than others. Well-conceived public policy
can improve societies’ adaptation to the shifting demand for skills. New approaches
and policy innovations are required in factory economies to enable a higher-road
development trajectory than the one enabled by a simple attraction of efficiency-seeking
foreign direct investment.

Andrea Szalavetz
References
Alcácer J., Cantwell J. and Piscitello, L. (2016) Internationalization in the information age: a new era
for places, firms, and international business networks?, Journal of International Business Studies,
47 (5), 499-512.
Baldwin R. (2013) Global supply chains: why they emerged, why they matter, and where they are
going, in Elms D.K. and Low P. (eds.) Global value chains in a changing world, Geneva, World
Trade Organization, 13-60.
Blažek J. (2016) Towards a typology of repositioning strategies of GVC/GPN suppliers: the case of
functional upgrading and downgrading, Journal of Economic Geography, 16 (4), 849-869.
Brettel M., Friederichsen N., Keller M. and Rosenberg M. (2014) How virtualization, decentralization
and network building change the manufacturing landscape: an industry 4.0 perspective,
International Journal of Mechanical, Aerospace, Industrial and Mechatronics Engineering, 8 (1),
37-44.
Brynjolfsson E. and McAfee A. (2014) The second machine age: work, progress, and prosperity in a
time of brilliant technologies, New York, WW Norton & Company.
Burkacky O., Deichmann J. and Stein J.P. (2019) Automotive software and electronics 2030:
mapping the sector’s future landscape, McKinsey & Company.
Collie B., Van Wyck J., Schaetzberger C. and Milliken K. (2019) An automotive downturn is coming
– it’s time to prepare. https://www.bcg.com/publications/2019/automotive-downturn-coming-
time-to-prepare.aspx.
Dachs B., Kinkel S. and Jäger A. (2017) Bringing it all back home? Backshoring of manufacturing
activities and the adoption of Industry 4.0 technologies, MPRA Paper 83167, Munich. https://
mpra.ub.uni-muenchen.de/83167/.
DESI (2019) Digital Economy and Society Index Report 2019, Integration of Digital Technology,
Brussels, European Commission. https://ec.europa.eu/digital-single-market/en/desi.
Di Mauro C., Fratocchi L., Orzes G. and Sartor M. (2017) Offshoring and backshoring: a multiple case
study analysis, Journal of Purchasing and Supply Management, 24 (2), 108-134.
Dicken P. (2003) Global shift: reshaping the global economic map in the 21st century, London,
Guilford.
Dunning J.H. (1993) Multinational enterprises and the global economy, Wokingham Addison-
Wesley Publishing Company.
Eisenhardt K. M. (1989) Building theories from case study research, Academy of Management
Review, 14 (4), 532-550.
Ferràs-Hernández X., Tarrats-Pons E. and Arimany-Serrat N. (2017) Disruption in the automotive
industry: a Cambrian moment, Business Horizons, 60 (6), 855-863.
Flecker J. and Schönauer A. (2016) The production of ‘placelessness’: digital service work in
global value chains, in Flecker J. (ed.) Space, place and global digital work, London, Palgrave
Macmillan, 11-30.
Frey C. B. and Osborne M. A. (2017) The future of employment: how susceptible are jobs to
computerisation?, Technological Forecasting and Social Change, 114, 254-280.
Gereffi G. and Fernandez-Stark K. (2016) Global value chain analysis: a primer, 2nd edition. https://
dukespace.lib.duke.edu/dspace/bitstream/handle/10161/12488/2016-07-28_GVC%20
Primer%202016_2nd%20edition.pdf?sequence=1_1

Horváth D. and Szabó R. Z. (2019) Driving forces and barriers of Industry 4.0: do multinational and
small and medium-sized companies have equal opportunities?, Technological Forecasting and
Social Change, 146, 119-132.
Huang Y. and Sharif N. (2017) From ‘Labour Dividend’ to ‘Robot Dividend’: technological change
and workers’ power in South China, Agrarian South: Journal of Political Economy, 6 (1), 53-78.
Humphrey J. and Schmitz H. (2002) How does insertion in global value chains affect upgrading in
industrial clusters?, Regional Studies, 36 (9), 1017-1027.
Kagermann H., Helbig J., Hellinger A. and Wahlster W. (2013) Recommendations for implementing
the strategic initiative Industrie 4.0: securing the future of German manufacturing industry: final
report of the Industrie 4.0 working group, Berlin, Forschungsunion.
Manyika J., Chui M., Bughin J., Dobbs R., Bisson P. and Marrs A. (2013) Disruptive technologies:
advances that will transform life, business, and the global economy, Washington, D.C., McKinsey
Global Institute.
Milberg W. and Houston E. (2005) The high road and the low road to international competitiveness:
extending the neo-Schumpeterian trade model beyond technology, International Review of
Applied Economics, 19 (2), 137-162.
Mittal S., Khan M. A., Romero D. and Wuest T. (2018) A critical review of smart manufacturing
& Industry 4.0 maturity models: implications for small and medium-sized enterprises (SMEs),
Journal of Manufacturing Systems, 49, 194-214.
Monostori L. (2015) Cyber-physical production systems: roots from manufacturing science and
technology, at-Automatisierungstechnik, 63 (10), 766-776.
Monostori L., Kádár B., Bauernhansl T., Kondoh S., Kumara S., Reinhart G., Sauer O., Schuh G., Sihn
W. and Ueda, K. (2016) Cyber-physical systems in manufacturing, CIRP Annals-Manufacturing
Technology, 65 (2), 621-641.
Nick G., Szaller Á., Bergmann J. and Várgedő T. (2019) Industry 4.0 readiness in Hungary: model,
and the first results in connection to data application, IFAC PapersOnLine, 52 (13), 289-294.
OECD (2013) Interconnected economies: benefiting from global value chains, Paris, OECD
Publishing.
Pavlínek P. (2017) Dependent growth: foreign investment and the development of the automotive
industry in East-Central Europe, Cham, Springer.
Pavlínek P. (2018) Global production networks, foreign direct investment, and supplier linkages in
the integrated peripheries of the automotive industry, Economic Geography, 94 (2), 141-165.
Porter M. E. and Heppelmann J. E. (2014) How smart, connected products are transforming
competition, Harvard Business Review, 92 (11), 64-88.
Prašnikar J. and Redek T. (2019) Industry 4.0 as an answer to the productivity gap in European
catching-up economies? Ekonomika preduzeća, 67 (3-4), 193-211.
Rehnberg M. and Ponte S. (2018) From smiling to smirking? 3D printing, upgrading and the
restructuring of global value chains, Global Networks, 18 (1), 57-80.
Schuh G., Anderl R., Gausemeier J., ten Hompel M. and Wahlster W. (eds.) (2017) Industry 4.0
maturity index - Managing the digital transformation of companies, Acatech Study, Munich:
Herbert Utz Verlag.
Schumacher A., Erol S. and Sihn, W. (2016) A maturity model for assessing Industry 4.0 readiness
and maturity of manufacturing enterprises, Procedia CIRP, 52, 161-166.
Schwab K. (2016) The fourth industrial revolution, Geneva, World Economic Forum.

Andrea Szalavetz
Scremin L., Armellini F., Brun A., Solar-Pelletier L. and Beaudry C. (2018) Towards a framework
for assessing the maturity of manufacturing companies in Industry 4.0 adoption, in Brunet-
Thornton R. and Martínez F. (eds.) Analyzing the impacts of Industry 4.0 in modern business
environments, IGI Global, 224-254.
Sturgeon T., Van Biesebroeck J. and Gereffi G. (2008) Value chains, networks and clusters: reframing
the global automotive industry, Journal of Economic Geography, 8 (3), 297-321.
Szalavetz A. (2017) Industry 4.0 in ‘factory economies’, in Galgóczi B. and Drahokoupil J. (eds.)
Condemned to be left behind? Can Central and Eastern Europe emerge from its low-wage
model? Brussels, ETUI, 133-152.
Szalavetz A. (2019a) Industry 4.0 and capability development in manufacturing subsidiaries,
Technological Forecasting and Social Change, 145, 384-395.
Szalavetz A. (2019b) Digitalisation, automation and upgrading in global value chains - factory
economy actors versus lead companies, Post-Communist Economies, 31 (5), 646-670.
UNCTAD (2013) World Investment Report, Geneva, UNCTAD.
Váncza J., Monostori L., Lutters D., Kumara S. R., Tseng M., Valckenaers P. and Van Brussel H. (2011)
Cooperative and responsive manufacturing enterprises, CIRP Annals-Manufacturing Technology,
60 (2), 797-820.
Will-Zocholl M. (2016) New topologies of work: informatisation, virtualisation and globalisation
in automotive engineering, in Flecker J. (ed.) Space, place and global digital work, London,
Palgrave Macmillan, 31-51.
World Bank (2016) World development report 2016: digital dividend, Washington, World Bank
Publications.
Xu X. (2019) Value creation for intelligent connected vehicles: an industry value-chain perspective,
in Aagaard A. (ed.) Digital Business Models, Cham, Palgrave Macmillan, 57-79.
Yin R. K. (2014) Case study research: design and methods, 5th Edition, London, Sage.
All links were checked on 28.04.2020.

Chapter 4
Digitalisation and the role of MNC subsidiaries in the
Spanish automotive industry
1. Introduction
There are some earlier references, but it is common to attribute the term ‘Industry 4.0’ to
the report by Kagermann et al. (2013) (see also Kagermann 2015), which was drawn up
to make a diagnosis of German industry and its ability to cope with the new technological
scenario of digitalisation, or digital transformation, as well as of the reforms required
to assimilate it in an optimal way. The name comes from the estimate by the authors
of the report that this digital transformation is, in fact, the beginning of the fourth
industrial revolution, characterised by the development and introduction of cyber-
physical systems (CPS). These are defined as: ‘Systems with embedded software (as part
of devices, buildings, means of transport, transport routes, production systems, medical
processes, logistic processes, coordination processes and management processes)
which: directly record physical data using sensors and affect physical processes using
actuators; evaluate and save recorded data, and actively or reactively interact both with
the physical and digital world; are connected with one another and in global networks
via digital communication facilities (wireless and/or wired, local and/or global); use
globally available data and services; and have a series of dedicated, multimodal human-
machine interfaces.’ (Acatech 2011: 15)
The term ‘Industry 4.0’ has been successful and is already widely used, although it has
met with reticence in the academic world. For example, Valenduc (2018) and Valenduc
and Vendramin (2017) argue, on the basis of the concept of the techno-economic
paradigm (cf. Perez 2010), that it is not really a new paradigm but rather the transition
between the installation and deployment phase of what, according to Perez (2010),
would be the fifth paradigm since the industrial revolution. This began around 1973
and is based on the microprocessor, information and communications technologies
(ICT) and biotechnology. In an earlier document from 2012, the European Commission,
following (for example) Rifkin (2012), was still referring to the ‘third industrial
revolution’ (European Commission 2012: 7). In fact, in the case of Industry 4.0, it is not
so much a question of new technologies as of their application to the production process
in search of greater flexibility, efficiency and competitiveness, so such objections are not
without significance.
In exposing the scenario behind Industry 4.0, ideas and arguments may be repeated,
but the most familiar in the analysis of organisational and technological changes is
that from the 1980s which led from vertically-integrated Fordism to the model known
– among other denominations – as lean production. For example, in Roland Berger
(2016: 5) one can read: ‘It will also allow… a switch from push-production – make and

build up inventory – to pull-production – make to order.’ Aoki (1990), Coriat (1990) and
Womack et al. (1990) are three studies from that time that insist on the idea of switching
from push-production to pull-production, with the model of the Japanese automotive
industry as an example. Dorrenbacher et al. (2018) also refer to a renewed impact of the
principles of lean production on European MNCs (multinational corporations).
Other names referring to the same process are digital transformation, digitalisation,
smart industries or advanced manufacturing, which is the one most used in the United
States (PCAST 2011). Although there are subtle differences in concepts, particularly
between the German and the American visions, what these different names share is an
emphasis on the development of cyber-physical systems and the use of large amounts
of information, both for the operation of intelligent machines in the production process
and for quality improvement, predictive maintenance or adaptation to the needs of
specific customers (customisation and mass customisation). Taking into account
the above-mentioned nuances, in this work, for convenience, we will use the term
‘Industry 4.0’.
The adoption of technologies that could be encompassed by Industry 4.0 is a recent

phenomenon that is expected to have significant economic consequences, both
quantitative and qualitative as regards the demand for labour and possessing a likely
impact concerning the location of production activities. In this sense, the available works
that anticipate such effects are generic, with estimates for a country as a whole and even
for the global economy (see, for example, Acemoglu and Restrepo 2019; McKinsey 2017;
PwC 2018; Roland Berger 2016; WEF 2016). The main result of these kinds of works is
a set of estimates of the possible macroeconomic effects of the spread of Industry 4.0,
which leaves open a whole set of questions concerning the qualitative characteristics of
this process. What is the rate of introduction of the different technologies in Industry
4.0? Do MNCs have a strategy for adopting these new technologies in their different
assembly plants? What effects is the introduction of Industry 4.0 having on employment
in assembly plants? What are the main advantages of the introduction of Industry 4.0 in
the opinion of those agents directly involved?
This chapter aims to provide answers to these questions by obtaining direct information
from qualitative interviews with representatives of a number of automotive assembly
plants located in Spain. Plants in the automotive sector have added interest, since this
sector has been indicated as one of the most susceptible to the introduction of Industry
4.0 technologies and, in addition, MNCs in this sector could be taken as a model for the
possible effects of the expansion of Industry 4.0 on the location of value chains (Dachs
et al. 2019b; Deloitte 2020). In contrast to generic trends, the field work and its results
offer a more realistic picture of the degree of implementation of Industry 4.0 in the
Spanish automotive sector, allowing an understanding of the qualitative aspects related
to intra-company dynamics and competitive pressure.
The information obtained through the field work identifies a higher level of integration
of the production process in a company’s enterprise resource planning (ERP) system,
as the main technology being implemented. Automation is increasing, but no structural
change seems to be detected. Besides, initiative on the introduction of Industry 4.0 in

Digitalisation and the role of MNC subsidiaries in the Spanish automotive industry
MNCs corresponds to the level of the plant and is highly dependent on the motivation
of the personnel directly involved: the introduction of Industry 4.0 technologies is being
carried out with the close collaboration of local agents. On the other hand, the changes
indicated have not, for the moment, led to appreciable cuts in employment, although
there are indications of reassignments of workers to new tasks as well as a demand for
new profiles and new skills in certain jobs.
The chapter is organised thus. Following this general introduction, we detail in section
two the literature on the Industry 4.0 process itself, as well as its impact on employment,
global production and the Spanish automotive industry. We set out our approach to
the field work in section three, while section four analyses our results as regards the
chronology of Industry 4.0, the process of its introduction and its main advantages, the
impact on working conditions and employment, and its effects as regards the positions
of plants within MNC value chains. We end in section five by drawing some conclusions.
2. Literature Review
2.1 Industry 4.0: technological delimitation
In both the United States and Germany, the starting point is the assumption of the
historical strength of the automotive industry, its weight in R&D activity or in the
employment of highly-skilled workers, as well as the decline experienced in recent decades
(PCAST 2011, 2014; Kagermann et al. 2013). The latter has different characteristics in
the two countries, although it is much softer in the German case. Thus, manufacturing
employment in the United States has gone, between 2000 and 2017, from 14 per cent to
ten per cent of total employment and in Germany from 20 per cent to 17 per cent; as for
gross value added, in the United States this has dropped from 16 per cent to 12 per cent
while in Germany it has remained stable at 23 per cent, although in 1991 it was up at
27 per cent (data from the OECD and Statistisches Bundesamt). There is concern about
halting this decline and the fear that not only industrial value added, but also pre- and
post-production services, will end up leaving the country.
The lists of technologies included in the concept of Industry 4.0 are highly similar.
Strange and Zucchella (2017) group them into four categories: the Internet of Things;
big data analytics; robotics; and additive manufacturing (3D printing). Meanwhile, CB
Insights (2019) distinguishes 14 technologies grouped into four categories: necessary
(Internet of Things, industrial sensors, robots/collaborative robots (or ‘cobots’) and
predictive analytics); experimental (edge computing, industrial drones, personalised
manufacturing, augmented reality and virtual reality, wearables and industrial
blockchain); threatening (machine-vision and machines-as-a-service); and transitory
(3D printing and data interoperability).
The world of Industry 4.0 is one of optimisation and flexibility. Central to this is
information, collected and analysed in real time and aimed at connecting all elements
of the factory, performing simulations and obtaining models through virtualisation, as
well as the continuous exploration of new services for potential customers. Information

management must translate into the optimisation of decision-making and, therefore,

improvements in the efficiency of resource use (Kagermann et al. 2013). A large part of
these activities is related to services associated with the development or end-consumer
phases of the relationship and form part of the so-called ‘servitisation’ of the industry
(Vandermerwe and Rada 1988; Raddats and Kowalkowski 2014; Raddats et al. 2019).
Industry 4.0 comprises a transversal technology – that is, it is applicable to a multitude

of productive sectors – although its adoption can be expected to be gradual, with
different rhythms both sectoral and corporate, in a process in which the organisational
culture of the company and the mentality of managers themselves can be decisive.
Given that our work on which this chapter draws focuses on the production process, we
have used in our interviews with companies and social agents the following breakdown
of those Industry 4.0 technologies which are directly related to production:
— logistics solutions for inventory and warehouse management;

— self-guided vehicles (AGVs);
— data extraction systems (CBS): sensors and real-time control;
— augmented reality systems;
— virtual reality systems;
— automation of management procedures: order management, reports, production
programming, remote maintenance, etc.;
— cobots;
— 3D printing;
— functional printing (printed electronics);
— knowledge and information sharing systems (between workers or with suppliers);
— intelligent systems: support for decision-making in production planning, process
optimisation and predictive maintenance;
— industrial drones;
— artificial intelligence and neural networks.
The ultimate goal is improved profitability, which arises both from value creation and
improved asset utilisation rates as well as reduced labour costs (Roland Berger 2016;
BCG 2015). There is a very strong emphasis on planning and logistics, especially supply
and inventory management, but also on reducing maintenance time by incorporating
preventive maintenance.
2.2 Industry 4.0 and employment

One of the most controversial aspects of the implementation of Industry 4.0 is its impact
on employment, although it should be clarified that most of the work does not concern
Industry 4.0 as such but automation. A first estimate, seeking to capture an equivalent
phenomenon and which had a considerable impact, was that of Frey and Osborne,
published as a working paper in 2013 and as an article in 2017, which estimated that
47 per cent of jobs in the United States were at a high risk (i.e. over 70 per cent) of
being automated whereas only 33 per cent of jobs had a less than 30 per cent risk of

being automated (see Frey and Osborne 2013, 2017). Their estimate was based on an
approach which focused on occupations.
Arntz et al. (2016), using a contrasting individual job-oriented approach as a point of

reference, estimate for a sample of 21 OECD countries that around ten per cent of jobs
are likely to be made redundant over the next two decades as a result of technological
progress, although they note that there are significant differences among them (see also
Arnold et al. (2018), who argue that this impact does not have to translate into a similar
increase in unemployment and that, in the long-term, the overall effect on employment
will be positive). For the United States, the figure is nine per cent.
For their part, Nedelkoska and Quintini (2018), applying the criteria of Arntz et al.
(2016) to data from 32 countries and across a wider group of workers, estimate that
14 per cent of jobs are at high risk of automation (i.e. with a probability of automation
greater than 70 per cent), while another 32 per cent would have a probability between
50 and 70 per cent. Countries that, according to Nedelkoska and Quintini (2018),
present a lower risk of automation to the median worker are located in northern Europe
(Norway, Finland, UK, Sweden, Netherlands and Denmark), northern America (United
States and Canada) and New Zealand. In contrast, the countries with the highest risk
are located in southern and eastern Europe (Slovak Republic, Lithuania, Greece and
Spain), in addition to Germany and Japan.
Frey and Osborne (2017: 265) find that ‘A substantial share of employment in services,
sales and construction occupations exhibit high probabilities of computerization,’
although Nedelkoska and Quintini (2018) had concluded that automation mainly
affects jobs in industry and agriculture, with few service industries being at a high risk
of automation. Furthermore, the latter come to the a priori surprising conclusion that
up to 71 per cent of the variation between countries may be explained by intra-sectoral
differences (differences in the organisation of production within the same sector) while
only up to 29 per cent of it may be accounted for by inter-sectoral ones (the industry
mix).
PwC (2018) provide estimates on the percentage of jobs at risk due to automation
for 28 countries, ranging from 22 per cent in Korea to 44 per cent in Slovakia. The
countries most affected are in eastern Europe: in addition to Slovakia, they are Slovenia,
Lithuania and the Czech Republic. However, it obtains a negative relationship between
job automation risk and the density of industrial robots (industrial robots per 10,000
employees in manufacturing industry), by which we can understand that some countries
(such as Korea, Singapore, Japan or Germany) present a lower risk because they have
already made part of the adjustment.
The perception that the impact is distributed by activity, and that the net balance is
positive, is widespread. In addition to the work already mentioned by Arnold et al.
(2018), BCG (2015) forecasts a net increase in employment, at least in Germany. Thus,
it estimates that Industry 4.0 could increase employment between 2015 and 2025
by 350,000 (five per cent), resulting from the creation of 960,000 new jobs and the
disappearance of 610,000, based on generating additional economic growth annually

of one per cent. The reduction in employment will take place in factories, mainly due
to the introduction of robots. This is a conclusion which is also supported by Roland
Berger (2016) which, in a simulation for an automotive supplier, obtains the result of a
reduction of almost half of employment (45 per cent) although it adds: ‘People are still
at the heart of the system’ (Roland Berger 2016: 5).
Meanwhile, McKinsey (2017) estimates that, in a midpoint adoption scenario,

automation could replace, depending on the country, between nine per cent (India)
and 26 per cent (Japan) of employment (from a set of reference countries that also
includes Mexico, China, the United States and Germany). It does not make estimates of
the net balance but, from historical analysis, concludes that technical change generates
net employment.
The way in which automation affects employment is related to its differing impact on
occupations and skill levels. It is generally accepted that automation particularly affects
tasks that require a lower level of skills, especially in production. Thus, Nedelkoska and
Quintini find: ‘A rather monotonic decrease in the risk of automation as a function of
skill level’ (2018: 50). Dauth et al. (2018), for Germany, and Acemoglu and Restrepo
(2017), studying the specific case of robots, come to a similar conclusion.
Acemoglu and Restrepo (2019) went on to establish a theoretical model to analyse

the ways in which automation affects employment. The net impact of automation in
a sector and on added employment is the result of two effects that pull in opposite
directions. First, there is a productivity effect, since automation increases added value
and generates demand for labour in non-automated tasks, which therefore acts in a
positive fashion. Second, however, there is a displacement effect which arises because
automation displaces work from tasks previously assigned and tends therefore to
reduce employment. To the extent that it cannot be ensured that the productivity effect
is greater than the displacement effect, there is no guarantee that the final impact on
employment will be positive. Furthermore, ‘Different technologies are accompanied
by productivity effects of varying magnitudes and hence we cannot assume that one
set of automation technologies will impact labour demand in the same way as others’
(Acemoglu and Restrepo 2019: 11). In their view, this could explain the differences
observed, for example between Germany and the United States, following the
introduction of industrial robots.
The case of industrial robots has received specific attention and there is a consensus
that they contribute to reducing employment in the industries in which they have
been installed. However, differences arise when appreciating the overall effect on the
economy as a whole. In their analysis of the implementation of robots in the United
States, Acemoglu and Restrepo (2017) estimate that, relative to a local labour market
(commuting zone) with no robots, an increase of one robot per thousand workers leads
to a reduction in the employment to population ratio by 0.37 percentage points and
in average wages by 0.73 percentage points. This led them to an estimate of total job
losses for the country as a whole between 1990 and 2007 of 360,000-670,000. The
negative effects are concentrated in manufacturing, while finance, the public sector
and non-robotised manufacturing show positive effects. Negative (or, at best, null)

effects are distributed across all occupations (except managers) and all education
levels. Those with the lowest wages are most affected, which results in increased wage
inequality.
Dauth et al. (2018) analyse the implementation of robots in Germany, concluding

similarly to Acemoglu & Restrepo (2017) in terms of the existence of job losses in the
industries where they have been installed as well as in terms of their impact on wages and
in the widening of the wage gap. However, they also find that the losses in manufacturing
are almost offset by gains in other activities, especially business services. They estimate
that one robot per thousand workers replaces 2.11 jobs in industry (manufacturing jobs)
while generating two jobs in services (the aggregate effect on employment relative to the
population will thus be -0.018 percentage points). In other words, robots change the
composition of employment but not its aggregate level. In addition, they conclude that
part of the adjustment takes place within the factories themselves, with the outcome of
job losses being reflected in fewer jobs for young people.
The relatively greater damage to wages resulting from automation seems to exist in
contradiction to the trend of the polarisation of employment into high and low skill
areas, to the detriment of those in the middle. Representative works on this position
are those of Autor and Dorn (2009) and Goos et al. (2009, 2014). We will not go into
the content of these contributions here, although we should point out that such a result
is, perhaps, greatly influenced by the identification that is made between salaries and
skill level, such that low skill is attributed to low salaries and high skills to high salaries.
By way of hypothesis, it could be ventured that the relative reduction in industrial
employment and the expansion of lower-paid, but not necessarily lower-skilled, service
activities (for example, many care services and feminised occupations), as well as the
trend itself towards lower wages in industry, may have something to do with this wage
depression at the average level.
2.3 Industry 4.0 and the global organisation of production

It has not been explicitly considered in the empirical analysis, but there is one
remaining aspect that seems relevant to consider here: the impact that Industry
4.0 technologies may have on the organisation of production on a global scale and,
particularly, that of the automotive industry. Globalisation is associated with the
fragmentation and geographical dispersion of production, facilitated by information
and communications technologies (Dicken 2015). The result is the configuration of
so-called ‘global production networks’ (GPNs: Yeung and Coe 2015), or ‘global value
chains’ (GVCs: Gereffi et al. 2005; Gereffi and Fernandez-Stark 2016). This has occurred
within the process of extending the operations of MNCs and their establishment of
complex networks of productive and organisational relationships with their suppliers,
both internal and external. However, this offshoring process may be coming to an end
(De Backer and Flaig 2017), with the detection of movements of activities, either of
the company itself or of its suppliers, to the company’s country of origin. This process
began as reshoring, although today the term ‘backshoring’ is widely used. Backshoring
is not necessarily the return of a previously-offshored activity since a company may

have been able to expand by building new production capacity or acquiring companies
in the destination country (Dachs et al. 2019a).
Backshoring has attracted much interest in recent years and studies are beginning
to proliferate which try to quantify it and to establish its impact, although general
theoretical models are still to be developed and the vast majority of empirical analyses
are based on case studies. Barbieri et al. (2018) carry out an exhaustive review of the
literature on reshoring, with a base of 57 documents (53 articles and four book chapters).
Two papers arising from the analysis of a large sample of companies taken from the
2015 European Manufacturing Survey (EMS) have been contributed by Dachs et al.
(2019a, 2019b).
It is not easy to quantify the extent of the phenomenon. Heikkilä et al. (2018), in a study
of Finnish companies, find that 13 per cent of companies had moved production to
Finland between 2010 and 2015 (26 per cent had moved activities outside the country in
the same period, since backshoring and offshoring coexist and the latter remains even
more significant). Johansson and Olhager (2018), in their analysis of the Swedish case
of backshoring, estimate the percentage of companies at 27 per cent between 2010 and
2015. These are high percentages, far removed from the 4.3 per cent obtained by Dachs
et al. (2019a). However, all three studies are in agreement on the more intense impact
of backshoring on high-tech activities, as well as on the reasons cited for engaging in
it. The latter can be summarised as the search for greater flexibility and quality, an
under-utilised capacity problem and aspects related to logistics, such as transport and
coordination.
This leads us directly to the relationship that may exist between Industry 4.0 technologies
and backshoring since, as we have seen, Industry 4.0 facilitates increased flexibility,
adaptability, improved coordination and adaptation to specific customer requirements.
We should also add reductions in costs, particularly labour costs, as well as the
reduction in the share of labour in income (Dauth et al. 2018; Acemoglu and Restrepo
2019) as a result of the change in the capital/labour ratio. Therefore, Industry 4.0 is
directly related to the objectives being pursued under backshoring, while it also makes
it possible to sidestep one of the most powerful reasons for offshoring, which is savings
in labour costs (Di Mauro et al. 2018), as well as the lack of flexibility in labour rules
and laws (Heikkilä et al. 2018). Dachs et al. state: ‘The modernization and innovation of
these home plants by implementing advanced production technologies and accelerating
the digital integration of value adding processes (Industry 4.0) might play an important
role, as economies of scale and high capacity utilization become all the more important
in such high-tech and high-invest lead plants’ (Dachs et al. 2019a: 7).
In their second paper, devoted specifically to the relationship between backshoring and
Industry 4.0, Dachs et al. (2019b) find a positive and significant relationship between
backshoring and investment in Industry 4.0 technology. They estimate that Industry 4.0
brings two benefits to companies: firstly, increased productivity and capacity utilisation,
which translates into lower production costs; and, secondly, greater flexibility and
quality, which enables customised production with very low marginal costs.

In the case of the automotive industry, one more element could act, in addition to
investment in Industry 4.0, to feed backshoring: the switch to electric vehicles and
possible changes in trends in the demand for cars, the types of vehicle ownership and in
mobility patterns all imply far-reaching changes in the industry that may reinforce the
concentration of activity in companies’ countries of origin.
2.4 Spanish automotive industry and Industry 4.0

The Spanish automotive industry is very significant, both for the country’s economy
and in comparative terms with other EU countries. Total employment in the industry
(NACE C29) was 158,000 people in 2017, 8.2 per cent of the country’s manufacturing
sector (data from the National Statistics Institute, INE). It generated ten per cent of
manufacturing value added (€11.3bn) and is also a relevant destination for investment
since, in 2017, it accounted for 15.8 per cent of all manufacturing investment.
However, the overall impact of the automotive industry on the Spanish economy
(including related services and gross fixed capital formation) is much larger. According
to data from ANFAC (Spanish Association of Automobile and Truck Manufacturers),
GDP related to the automotive industry represents about 8.6 per cent of Spanish GDP.
Twelve assembly plants are located in Spain (there are actually two Nissan plants, the
one in Ávila being dedicated, however, to the manufacture of components for Renault);
one plant is owned by Ford while eleven are owned by European manufacturers:
— Ford: Valencia;
— Iveco: Madrid, Valladolid;
— Mercedes-Benz: Vitoria;
— Nissan: Barcelona (Nissan announced the intention to close this plant in May
2020);
— PSA: Madrid, Vigo, Zaragoza;
— Renault: Palencia, Valladolid;
— Volkswagen Group: Barcelona (Seat), Pamplona (Volkswagen).
There have been no greenfield investments in plants assembling vehicles in Spain for
thirty years (Aláez-Aller et al. 2015). In addition, SERNAUTO (Spanish Association of
Equipment and Component Manufacturers) estimates that there are more than 1,000
companies dedicated to the manufacture of components (equipment and spare parts),
belonging to 720 groups. Consequently, it is only possible to find companies that have
Spanish capital at the level of component manufacturers.
The location of operations in the automotive value chain in Europe has been characterised
by two hierarchical structures (Lung 2007; Pavlínek 2015): one for assembly (with high-
end models being assembled mainly in core countries – France and Germany – while the
peripheral states of Europe have become specialised in the assembly of lesser vehicles);
and the other based on functions (R&D has been concentrated in the core regions of the
EU which have become the home for development centres for assemblers).

This double hierarchy determines the type of product assembled in Spain, its role as a
preferential destination for export to European markets and that multinational groups
do not, in general, develop R&D activities related to product development in their
Spanish plants. In general, it could be said that only automotive suppliers with Spanish
capital develop R&D activities in Spain (Aláez-Aller et al. 2015).
In relation to the type of product assembled in Spain, the place occupied by Spanish
automotive producers in the EU value chain has been limited to the assembly of vehicles
with medium/low added value. A breakdown by segment of ANFAC data for passenger
car assembly in Spanish plants reveals that this remained true in 2018, when 362,621
medium SUVs, 672,513 small SUVs, 596,083 small-sized vehicles, 441,562 compacts,
54,486 medium-sized vehicles and 83,029 large vans were assembled in Spain as were,
additionally, 548,467 commercial vehicles and 55,499 industrial vehicles. In 2018,
production was 2.8 million cars (2.2 million, 79 per cent, being passenger cars), of which
2.3 million were exported. With these figures, Spain is the second largest European
manufacturer and the ninth largest in the world, with a 2.9 per cent share of the global
market. Approximately 60 per cent of exports go to four European markets: France,
Germany, United Kingdom and Italy.
The OECD calculates the value added contained in exports and their origin, domestic or
imported. In the case of imports, this can be used as an indicator of the import content
of exports. For 2015, this value was 40.7 per cent in the transport equipment industry
compared to 31.2 per cent for manufacturing as a whole. This indicates a greater intensity
of backward linkages – that is, greater integration in global value chains. In the same
year, the automobile industry was responsible for 30.2 per cent of the imported value
added contained in manufacturing exports (source: OECD).
Little is known about the situation of Industry 4.0 in Spain. There is no study that
quantifies in any way the degree of implementation of such technologies and their
impact. There are reports, usually official ones, on the extent of ICT focused on the
deployment of networks, services, electronic administration, etc. (see, for example,
Ministerio de Hacienda y Administraciones Públicas 2016), but this does not amount
to data on Industry 4.0. Roland Berger (2016) does provide information on specific
aspects of digitalisation, although these are agent assessments. One exception to this
overall picture, however, is the report of the Observatorio ADEI (2017), which carried
out a simulation exercise based on two scenarios: firstly, the convergence of advanced
occupations with the United States, United Kingdom and Germany; and, secondly, the
reduction in the working age population and the structural unemployment rate. On this
basis, the Observatory estimates net job creation by 2030 of 2.4 million: 3.2 million
jobs will be created in advanced occupations (jobs which are adaptable to digitalisation
initiatives), with a further 0.6 million jobs added in occupations not susceptible to
automation; while 1.4 million jobs will be lost.
Furthermore, the IFR (International Federation of Robotics) provides data on robots.

Thus, in 2018 Spain ranked tenth in terms of annual robot installation (annual variations
may be significant) with 5,300 units: far behind China, in first place with 154,000; and
Germany, in fifth place with 26,700. In terms of robot intensity (robots per 10,000

employees), in 2018 Spain ranked 15th with 168 (the global average is 99), far behind
Singapore (831), Korea (774), Germany (338) and Japan (327), which are in the leading
positions, which explains the negative judgement made in reports and studies on the
country’s overall level of digitalisation (Roland Berger 2016; Bondar 2018). However, if
we specifically consider the automotive industry, the situation is comparatively better:
the intensity of robots was 1,110 in 2018 and Spain is in ninth position, in a ranking
led by Korea (2,589) and with much less marked differences. This position is more
in line with the weight of the Spanish automotive sector (second European and ninth
worldwide manufacturer).
Some regional reports have, however, been prepared on the degree of implementation
of Industry 4.0 based on company surveys. For example, AIN (2019) looked at Navarra;
Bilbao, Camino and Intxaurburu (2016) focused on the Basque Country; Xunta de
Galicia (2018) examined Galicia (with detailed reports for different sectors); UGT
(2017) focused on Castilla y León; the Government of Aragon (2018) looked at SMEs in
Aragon; and Hernández et al. (2018) examined Catalonia, which estimates the impact
of automation in terms of a net creation of 13,000 jobs (+0.7 per cent) and a loss in
manufacturing industry of some 12,000 jobs (-3.2 per cent).
3. Field survey
Our work aims to close the data gap by examining the incorporation of Industry 4.0
technologies in the automotive industry and the impact these are having on the industry.
The empirical information comes from original field work which we carried out between
October 2019 and February 2020. Given the absence of prior data, our approach has
been to adopt the method described by Lewis (1998: 456) as ‘iterative triangulation’,
based on ‘systematic iterations between literature review, case evidence and intuition.’
It is not a question of testing a theory, but of constructing one. Therefore, the sample is
not random or stratified but, using the terminology of Eisenhart and Graebner (2007:
27), it is theoretical; that is: ‘Cases are selected because they are particularly suitable for
illuminating and extending relationships and logic among constructs.’
Our sample consists of the Volkswagen plant in Navarra, which manufactures two
car models and is expected soon to start assembling a third; eight supplier plants (see
Table 1); and various social actors consisting of a consultancy firm, a local automotive
cluster and two trade unions.
The questionnaire, focusing on the issues revealed by a review of the literature, was
validated through interviews with industry experts and academics. The effort was made
to eliminate ambiguity in the questions (on the construction and use of questionnaires in
operations management and manufacturing studies, see Flynn et al. 1990 or Synodinos
2003) and, in the end, two different questionnaires were used: one for companies and
one for the social partners.
Information was obtained by conducting thirteen semi-structured interviews (the

questionnaire consisted of structured and semi- or unstructured questions), so that

interviewees could express their point of view in an open way. An e-mail was sent
to firms and to social actors, explaining the purposes of the study and enclosing the
questionnaire in advance so that the most suitable interviewees could be chosen and
information collated beforehand. The interviews were conducted directly by the authors.
This assured a high level of participation and prevented bias due to non-responses.
This method also helped to prevent problems due to respondents misunderstanding
questions, leaving some answers blank or incomplete, or answering inappropriately.
Most of the interviews lasted between sixty and ninety minutes.
The questionnaire comprised 21 questions divided into six parts: identification; adoption
of Industry 4.0 technologies; reasons for the implementation of these technologies;
ways of incorporation; impact on employment and on job content and nature; and
impact on headquarters strategy and the position of the subsidiary.
Our sample of plants (see Table 1) is characterised by an enormous variety of situations

(origin of capital, activity within the automotive sector and size of plant) which limits
the capacity to formulate detailed conclusions on the existence of differences in the
implementation of Industry 4.0 technologies based on such variables. Nevertheless,
the plants interviewed share a series of characteristics that are common to most plants
in the automotive sector in Spain and which, as has been explained in section 2.4 of
this work, are derived from how the Spanish automotive sector has developed: most
of the plants are integrated into foreign capital MNCs; Spanish plants only undertake
product-related R&D activity in very exceptional cases; the main activity of Spanish
plants is assembly; and Europe is the main market for Spanish plants in the automotive
sector.
Table 1 Characteristics of plants interviewed
Plant Number of Person interviewed Origin of capital Main activity Position in the
employees value chain
GKN 180 Quality manager; UK (Melrose) Transmission; original Tier 1

production department equipment (short series);
engineer aftermarket
Vibracoustic 80 Plant manager Germany Chemical treatment of Tier 1

automotive parts Tier 2
SAS 280 Plant manager France (50%)/ Assembly; sequential Tier 1

Germany (50%) deliveries
BENTELER 50 Plant manager Austria Assembly; sequential Tier 1

deliveries
Plásticos Brello 60 Plant manager Spain Plastic parts Tier 1

Tier 2
Flex-n-Gate 100 Product engineer USA Welding; stamping Tier 1

SKF 280 HR officer (training and Sweden Bearings Tier 1
staff recruitment)
Volkswagen 4,800 Process engineers Germany Car assembly OEM

Grupo Aldakin 150 R&D department Spain Industrial automation Tier 1
and robotics

4. Analysis of results
4.1 Chronology of implementation of Industry 4.0 technologies
The companies interviewed recognise that the most relevant change as regards the
digitalisation of their activity has been the integration of their production process with
an ERP (enterprise resource planning) system. Compared to previous iterations, new
model ERPs allow the production process to be directly connected to management
tasks.
Firstly, therefore, the adoption of a new ERP means a considerable effort for plants,
not so much financial as in adapting to the new system (business managers estimated
the period of preparation of the plant before starting to use the new ERP was between
nine and eighteen months). This change began in most of the plants interviewed around
2016; the timing for adoption, as well as the main characteristics, being dependent on
the size of the plant and the origin and characteristics of the company that owns it.
Consequently:
— it was adopted first in the largest plants with multinational capital, with smaller,
locally-owned plants still in the process of implementing an integrated ERP.
Meanwhile, larger companies are in the phase of obtaining greater advantages
from integrated ERP and migrating it to the cloud;
— the type of ERP adopted by plants which are dependent on larger MNCs is usually
SAP and was imposed by the company that owns the plant. Smaller plants and those
with local capital have opted for other ERPs more suited to their characteristics;
— plants recognise the multiple advantages of digitised ERP: real-time process

information; process control and stability; predictive maintenance; 100%
traceability; paperless production; and a significant limitation of human error in
decisions on the production process.
Secondly, the adoption of Industry 4.0 technologies has meant an increase in the degree
of process automation which, according to plant managers, has been characterised by
the following main features and timing (see also Table 2):
— the introduction of conventional robots continues on an upwards trajectory, at least

in quantitative terms. However, technological advances in the field of industrial
robotics are particularly uneven – the software has advanced significantly but the
hardware barely at all. Consequently, there is no obvious technological break when
it comes to traditional robots. This means that the limitations of robot hardware
need, in a practical setting, to be both recognised and worked around before
organisations can take advantage of all the possibilities that have been facilitated
by advances in the software;
— the introduction of cobots is, however, recent (the first examples refer to 2017)
and remains very limited. In some cases, the slowness of cobots (as opposed to

traditional robots) has been mentioned as the main drawback which has limited
their introduction to very specific cases, including the existence of clear ergonomic
problems in a particular workplace;
— the first AGVs (Automated Guided Vehicles) in any of the plants in the sample date
back to 2015, but their generalisation is more recent; in the majority of our plants,
they started to be introduced from 2017 onwards although their use is progressing
very quickly, especially with regard to operations within the plant. The enormous
field for the development of these vehicles can be anticipated in that only one of
the plants interviewed has recognised the use of AGVs in outdoor operations. The
speed of their introduction derives from their advantages in terms of safety and
reliability, the justification which lies behind the investment (a key argument for
obtaining MNC authorisation for capital expenditure on AGVs is that they will
replace active workers) and the financial facilities that AGV suppliers offer for
their acquisition (as an example, it is possible to lease such vehicles);
— the automation of inventory control is spreading and the use of labelling has
become general. However, examples of fully automated warehouses are rare and
the substantial investment involved is only justified in highly particular cases even
though the advantages of a fully automated warehouse are recognised by plant
managers. One plant had invested in an intelligent, fully automated warehouse for
the management of only one key component (an investment made in 2017). Where
there are plans to build new warehouses, these will be digitised and automated but
the automation of existing warehouses is an investment that appears to be lagging
behind.
No direct link is recognised between process integration into the ERP and increased
automation. As an example, in one of our plants, new conventional robots were
purchased in 2014, prior to the adoption in 2016 of a new ERP.
The timing of the adoption of 3D printing follows very different patterns: the larger
plants started using 3D printing around 2012 (they have recently replaced the initial
machines with more sophisticated 3D printers); while smaller plants started using 3D
printing in 2019 (where they are using it at all). All plants agree that 3D printing is not
an option to replace, even partially, conventional production; yet all cite that it offers
huge advantages in the manufacture of specific tools or prototypes.
The use of virtual reality and augmented reality in the production process is limited to
R&D projects. Actual deployments of virtual reality are the case in only one plant (for
the purpose of training workers before starting the assembly of a new product), while
applications of augmented reality are currently being tested in two plants (in respect of
maintenance tasks).
There is unanimity among the plants both on the huge accumulation of data on the
production process resulting from its integration into the ERP as well as on the lack of
strategies to take advantage of the analytical potential of this. Big data analytics is used
in only two plants and for specific projects related to quality problems. For its part,

the use of artificial intelligence has been practically limited to machine-vision in the
identification of defects.
Table 2 Industry 4.0 technologies introduced in automotive industry plants
Industry 4.0 technology Timing
Integration of the production process Since 2016. Still in progress Process control and stability;
with ERP system real-time information;
predictive maintenance;
100% traceability = no paperwork
Automation (cobots) Since 2017. Few examples

AGVs Since 2017. Fast development Being rented. Adoption easy to justify
(cost/benefit)
Inventory control and automation Since 2017. Still in progress

3D printing Initially in 2012; most adopters since Limited use (no production):
2019 prototypes, tools, etc.
Virtual reality Only in R&D Training

Augmented reality Only in R&D
Artificial intelligence Since 2018 Only for very specific problems
Data analysis systems To be developed
4.2 Process of introduction of Industry 4.0 technologies

The process of the introduction of the technologies encompassed by Industry 4.0 into
automotive plants followed some common guidelines in the majority of plants in our
sample. The initiative for introducing a particular technological improvement starts
with the plant in most cases, with original ideas found usually among local suppliers of
capital goods and in local engineering companies that have worked on previous projects
with the plant.
The main problem faced by plants in adopting technological improvements relates

to the need for the approval of capital expenditure by the head office of the company
that owns the plant. Plants have very little autonomy in capital expenditure decisions
and, therefore, have to seek approval by submitting a justification that, in most cases,
involves recovering the planned expenditure over a very short period of time (periods
of one to three years). The probability that the submitted project will be approved
is increased where its implementation involves a reduction in the workforce. In this
process, the plant manager takes on special significance since he or she is the direct link
to MNC decision-making centres, and his or her ability to sell the plant proposal is key
in obtaining final approval for it.
It is not common for the MNCs which own plants to impose the adoption of Industry
4.0 technologies on any plant, except in the case of ERP for which the MNC will have
negotiated licences for use throughout the company. The remaining projects are very
specific to the plant and respond both to the need to solve specific problems and to the
plant’s capacity to realise process improvements.

One of the business managers interviewed considered that the origin of the introduction
of Industry 4.0 technologies could be linked to a change in mentality at the plant around
2012, when the search for solutions to production challenges was directed towards the
outside while, at the same time, collaboration between company departments was
increased. In contrast, the previous path to resolving production problems had been to
rely almost exclusively on internal resources.
According to our interviewees, the internal dynamic of the plant is the key factor in
differentiating between those plants that are actively adopting new technologies
and those that are simply replicating the experiences of neighbours or adopting the
suggestions of MNC headquarters.
4.3 Main advantages of the introduction of industry 4.0 technologies

The main advantages that the plants recognise arising from the adoption of Industry
4.0 technologies are summarised in Table 3. The specific characteristics of the plant
determine the type of technologies that are of special interest to each one. As an
example, suppliers working on sequential deliveries for a vehicle assembly plant must
meet a very strict schedule of cost reduction over the lifetime of the assembled model
and, therefore, feel specific pressure to achieve continuous cost reductions. In addition,
the plant’s ability to meet the planned cost reduction schedule is a key feature of being
able to win new assembly supply contracts when the assembly of a new model is being
negotiated. These sequential delivery plants do not have the capacity to influence the
production awarded and, therefore, the technologies introduced will focus mainly on
cost reduction and on resolving known quality problems.
On the other hand, unlike supplier plants working on sequential deliveries, one of the
plants in our sample is responsible for a very specific product, the demand for which
is met in short runs. Here, the introduction of new technologies has been aimed
fundamentally at achieving a greater degree of flexibility as well as offering an image of
innovative capacity that will encourage the award of new contracts.
In addition to these specific aspects, all the plants also include among the advantages of
adopting these new technologies their positive effects on quality, the health and safety
of workers and the image of the plant. The latter is key not only with regard to customers
but also within the MNC itself, in the context of plants recognising that increased levels
of competition are not only external but also internal as regards other plants within the
company.

Table 3 Main advantages to the introduction of Industry 4.0 technologies
Industry 4.0 technology Cost Quality Productivity Health and Plant Labour
reductions safety at work image shortages
Integration of production ü ü ü ü
process with ERP system
Automation (cobots) ü ü ü ü ü
AGVs ü ü ü
Inventory control and ü ü
automation
3D printing ü ü
4.4 Impact on working conditions and employment

One of the sections of the questionnaire was specifically designed to examine the effects
of the adoption of new technologies on employment and working conditions.
With regard to the quantitative effects on employment, all plant managers agreed
that there have been no redundancies directly linked to the introduction of Industry
4.0 technologies, although they do acknowledge that there have been relocations of
workers due to the further automation of certain phases of the process. In this sense,
new technology is running in an apparently contradictory direction to the general
idea that such technologies fundamentally presuppose the substitution of human
intervention with machines. The only obvious reduction in the number of workers is
associated with the introduction of AGVs, which clearly replace forklifts. Nevertheless,
it is acknowledged that the plants are in a process of reducing the workforce in the
medium- and long-term, and are trying to do so while not generating enforced lay-offs.
With regard to changes in the qualifications and skills of workers, the increase in the
level of digitalisation and automation of plants has meant:
— new skills on the part of maintenance employees, especially with regard to

programming, which complement their traditional mechanical skillsets;
— new profiles of recent hires which recognise the need to have increased IT
specialists among staff;
— new recruits will have a higher level of training than previous intakes, requiring,
even for operator positions, a minimum level of occupational training.
However, suppliers in the chain recognise that the new machines introduced into the
process do not require particularly sophisticated skills since improvements in the human-
machine interface, together with job-specific training, make it easier for an operator to
remain at his or her post and take on other, simpler tasks. The digitalisation of the process
and automation have reduced the contribution of the human factor to key decisions in
the production process, such as what to produce and in what quantities, the diagnosis of
problems, etc., but have not meant the total elimination of the operator who now has to
assume the previously peripheral tasks of monitoring the machine, cleaning, etc.

Our interviews allow us to anticipate a future in which increased automation may

nevertheless reduce the number of operators in a plant while increasing the number
of people employed in maintenance tasks. Simultaneously, the decreasing number
of operators will not require a very high level of training while those occupied in
maintenance will have to improve their level of competence in terms of being able to
carry out programming tasks to take charge of production equipment that is increasingly
sophisticated.
Training activities in plants have, therefore, been partially affected by the introduction
of Industry 4.0 technologies. On the one hand, the introduction of changes in the ERP
and increasing automation have been accompanied by specific training, provided by
external companies as part of the contract for the implementation of new technologies.
On the other, the training of the workforce in general, and of maintenance staff in
particular, must increasingly encompass IT and programming courses.
In any case, there is a general recognition that an increase in automation has meant
improvements in the ergonomic conditions of the workplace.
4.5 Position of plants within MNCs

In the automotive sector, a tendency has been detected for MNCs to encourage
competition between plants so that the allocation of workload for each depends on
its competitive position within the MNC. As has already been pointed out, the plant
managers we interviewed recognise that the initiative for the introduction of new
Industry 4.0 technologies corresponds, for the most part, to the level of the plants
themselves. It is the plants, therefore, that have the autonomy to decide which
technological improvements they would like to carry out, although the authorisation of
capital expenditure is the responsibility of the headquarters. Although the introduction
of Industry 4.0 may be undermining the locational advantage of the availability of
flexible and cheap labour, the plants in our sample do not perceive their position within
the MNC to be affected in this regard. However, they do recognise that they are being
continuously monitored as regards their profitability which, in many cases, depends
directly on the costs of production. Some of the plants in our sample work on sequential
deliveries and, therefore, their current location is completely dependent on the location
of their customer’s plant.
Once a technological improvement has been introduced in one plant, it would be

possible to extend this to the rest of the plants within the MNC. At this point in the
process, plants realise that there are two, opposing effects that the dissemination of
their Industry 4.0 improvements could have on the position of their plant within the
MNC:
— on the one hand, a plant that is very active and successful at introducing
technological innovations could, in disseminating them to the rest of the plants
owned by the MNC, achieve a certain prestige as an agent of technological
improvement, thus strengthening its position within the MNC;

— on the other, the rapid diffusion of such improvements to other plants (normally
competitors within the MNC) could erode the advantages of the plant in terms
of cost and quality. This would recommend the delay of such diffusion in order
to maintain the competitive advantage obtained over the other plants in the
company.
In this sense, our interviews suggest that MNCs have not developed formal systems
which facilitate the dissemination of technological improvements among the plants. In
the interviews we conducted, only one case could be recognised of an MNC that has
developed an incentive system in which plants could share technological improvements.
Within the plants we interviewed, there are instances of the application of an Industry
4.0 technology being first developed in the Spanish plant and then transferred to other
plants in the group. In one case, a Spanish engineering company that had worked with
one of the plants was hired by another MNC plant located in a central and east European
country in respect of the digitalisation of its production process.
In principle, it could be anticipated that MNCs might develop a common services

department that would promote the adoption of Industry 4.0 technologies throughout
the group. In this instance, all the plants supplying the MNC would work under the same
system and with a very similar process, sharing technological problems and solutions.
The extreme case would be represented by those MNCs that operate on the basis of
sequential deliveries, with the geographic location of their assembly plants adjusted to
the location of their customers. Here, it might be acceptable for some plants, depending
on their location, to carry out activities relative to their own locational context, which
may require alternative technological solutions. For example, in one location the
problem might be a poor energy supply, which would encourage the introduction of
more energy-efficient technologies; while in another the main problem might be labour
shortages, which would naturally encourage a greater degree of process automation to
reduce the need for labour. In this sense, the strategy of granting the initiative for the
introduction of Industry 4.0 technologies to the plants themselves would be rational
since the problems faced by each are highly dependent on the economic context of their
location.
Even here, however, this should be complemented with the existence of internal
mechanisms within the MNC that encourage the dissemination of technological
improvements among the plants, granting decision-making capacity on their adoption
to the plants’ own management teams. A strategy that promoted adaptation to the
local environment in terms of collaboration with local companies while, at the same
time, promoting the dissemination of solutions across the group would create a level
of competitive advantage over and above what we detected was the case for most of the
plants in our sample.

5. Conclusions
Since the introduction of the term ‘Industry 4.0’ to refer to a virtual fourth industrial
revolution (Kagermann et al. 2013), numerous papers have been published with the aim
of estimating the effect of the introduction of these new technologies on employment
and, even more relevant for those countries without domestic capital MNCs, to estimate
the possible backshoring effects of assembly activities. Most of the work has focused
on anticipating the macroeconomic effects on the labour market in the medium- and
long-term, but this has not provided information on the qualitative aspects nor on the
degree of implementation of Industry 4.0 and the resistances and complications that
could occur during the adoption process. The work on which this chapter is based has
been intended to contribute to the literature on the subject by providing qualitative
information obtained directly through interviews.
Our area of study has been focused on the automotive sector (one of the most dynamic
in the adoption of Industry 4.0) and plants located in Spain, both that of an assembler
(VW group) and those of eight different tiered suppliers.
Our analysis of the information obtained through field work allows us to recognise a
process of an increasing level of digitalisation (mainly through the integration of the
production process with the company’s ERP), starting about five years ago. At the same
time, although independent of digitalisation, automation is increasing. No structural
change seems to be detectable, but there has been an intensification of this process
(an increase in the number of robots and the introduction of AGVs). In addition, those
we interviewed recognise that greater digitalisation has allowed a huge amount of
information to be accumulated on the production process, but it is also accepted that,
for the moment, all this information is not being used.
The research highlights that initiative over the introduction of Industry 4.0 solutions
lies at plant level and is particularly dependent on the motivation of the personnel
directly involved, including the process engineers (in detecting possible improvements)
and the plant manager (in negotiating to gain authorisation for capital expenditure
from MNC headquarters). The introduction of this technology in Spanish plants is
being carried out with the collaboration of local agents: engineering companies; capital
goods suppliers; technology centres; and, to a lesser extent, universities. The changes
indicated have not, for the moment, led to appreciable cuts in employment, although
there are indications of the reassignment of workers to new tasks and a demand for
new profiles and new skills in certain jobs (for example, maintenance personnel being
required to have programming knowledge).
The increase in the degree of digitalisation is still very recent and the essential
characteristics of the process of introducing these new technologies remain in the
definition phase. However, both the ability to introduce such innovations and to adapt
those that have been developed are key issues in determining the competitive position
of plants and companies. This is even more relevant when the analysis focuses on an
environment as competitive as the automotive sector, in which competition between
regions and countries is determined not only by competition between companies but

also by growing internal competition within the company, manifested in the closure
of some plants and the expansion or opening of new ones in different locations. It was
particularly interesting in the light of this that MNCs do not have formal systems for
disseminating the innovations introduced in a plant to the rest of their plants working
in the same sector.
To this context, which is volatile and in which competitive pressures are acting in
different directions, not only in that of digitalisation, we must add the uncertainties
generated by the switch to electric vehicles and changes in demand behaviour and in the
form of car ownership, as well as mobility patterns. This may condition, and certainly
deepen, some of the trends observed as a result of the implementation of Industry 4.0
technologies.
References
Acatech (German academy of science and engineering) (ed.) (2011) Cyber-physical systems: driving
force for innovation in mobility, health, energy and production, Acatech Position Paper, Berlin,
Springer Verlag.
Acemoglu D. and Restrepo P. (2017) Robots and jobs: evidence from US labor markets, Working
Paper Series 23285, Cambridge, MA., NBER.
Acemoglu D. and Restrepo P. (2019) Automation and new tasks: how technology displaces and
reinstates labor, Journal of Economic Perspectives, 33 (2), 3-30.
AIN (Asociación de la Industria Navarra) (2019) Diagnóstico de la industria 4.0 en Navarra 2019,
Madrid.
Aláez-Aller R., Gil-Canaleta C. and Ullibari-Arce M. (2015) FDI in the automotive plants in Spain
during the Great Recession, in Galgóczi B., Drahokoupil J. and Bernaciak M. (eds.) Foreign
investment in Eastern and Southern Europe after 2008. Still a lever of growth?, Brussels, ETUI,
139-170.
Aoki M. (1990) Toward an economic model of the Japanese firm, Journal of Economic Literature,
28 (1), 1-27.
Arnold D., Arntz M., Gregory T., Steffes S. and Zierahn U. (2018) No need for automation angst,
but automation policies, in Neufeind M., O’Reilly J. and Ranft F. (eds.) Work in the digital age:
challenges of the fourth industrial revolution, London, Rowman & Littlefield Publishers.
Arntz M., Gregory T. and Zierahn U. (2016) The risk of automation for jobs in OECD countries,
OECD Social, Employment and Migration Working Papers 189, Paris, OECD Publishing.
Autor D. and Dorn D. (2009) This job is ‘getting old’: measuring changes in job opportunities using
occupational age structure, American Economic Review: Papers & Proceedings, 99 (2), 45-51.
Barbieri P., Ciabuschi F. and Fratocchi L. (2018) What do we know about manufacturing reshoring?,
Journal of Global Operations and Strategic Sourcing, 11 (1), 79-122.
BCG (Boston Consulting Group) (2015) Man and machine in industry 4.0: how will technology
transform the industrial workforce through 2025? https://www.bcg.com/industries/
engineered-products-infrastructure/man-machine-industry-4.0.aspx
Bilbao Ubillos J., Camino Beldarrain V. and Intxaurburu Clemente M.G. (2016) La administración
pública vasca y la recuperación de la industria: los retos para las estrategias de
reindustrialización inteligentes, IVAP (Herri Arduralaritzaren Euskal Erakundea.

Bondar K. (2018) Challenges and opportunities of industry 4.0 - Spanish experience, International
Journal of Innovation: Management and Technology, 9 (5), 202-208.
CB Insights (2019) Emerging trends: what’s next in advanced manufacturing?. https://www.
cbinsights.com/reports/CB-Insights-Advanced-Manufacturing-2019-Trends.pdf.
Coriat B. (1990) L’atelier et le robot : essai sur le fordisme et la production de masse à l’âge de
l’électronique, Paris, Christian Bourgois Editeur.
Dachs B., Kinkel S., Jäger A. and Palcic I. (2019a) Backshoring of production activities in European
manufacturing. Journal of Purchasing and Supply Management, 25 (3).
Dachs B., Kinkel S. and Jäger A. (2019b) Bringing it all back home? Backshoring of manufacturing
activities and the adoption of Industry 4.0 technologies, Journal of World Business, 54 (6).
Dauth W., Findeisen S., Suedekum J. and Woessner N. (2018) Adjusting to Robots: worker-level
evidence. Opportunity and Inclusive Growth Institute Working Papers 13, Minneapolis, MN,
Federal Reserve Bank of Minneapolis.
De Backer K. and Flaig D. (2017) The future of global value chains: Business as usual or ‘a new
normal’? OECD Science, Technology and Industry Policy Papers 41, Paris, OECD Publishing.
Deloitte (2020) Industry 4.0 in automotive. Digitizing the end-to-end automotive value chain,
Deloitte Insights.
Di Mauro C., Fratocchi L., Orzes G. and Sartor M. (2018) Offshoring and backshoring: a multiple case
study analysis, Journal of Purchasing and Supply Management, 24 (2), 108-134.
Dicken P. (2015) Global shift. Mapping the shanging contours of the world economy, 7th ed.,
London, SAGE.
Dörrenbacher C., Geppert M., Pastush D. and Tomenendal M. (2018) Cross-border standardisation
and reorganisation in European multinational companies, Report 141, Brussels, ETUI.
Eisenhart K.M. and Graebner M.E. (2007) Theory building from cases: opportunities and challenges,
Academy of Management Journal, 50 (1), 25-32.
European Commission (2012) Communication from the Commission to the European Parliament,
the Council, the European Economic and Social Committee and the Committee of the Regions, A
stronger European industry for growth and economic recovery, Industrial Policy Communication
Update, COM (2012) 582 final, Brussels, 10 October 2012.
Flynn B. B., Sakakibara S., Schroeder R. G., Bates K. A. and Flynn E. J. (1990) Empirical research
methods in operations management, Journal of Operations Management, 9 (2), 250-284.
computerisation?, Working Paper Sept. 2013, Oxford, University of Oxford.
computerisation?, Technological Forecasting and Social Change, 114, 254-280.
Gereffi G. and Fernandez-Stark K. (2016) Global value chain analysis: a primer, 2nd ed., Durham,
Center on Globalization, Governance & Competitiveness - Duke University.
Gobierno de Aragón (2018) Guía ara la transformación digital de pymes industriales, Zaragoza,
Instituto Tecnológico de Aragón.
Goos M., Manning A. and Salomons A. (2009) Job polarization in Europe, American Economic
Review, 99 (2), 58-63.
Goos M., Manning A. and Salomons A. (2014) Explaining job polarization: routine-biased
technological change and offshoring, American Economic Review, 104 (8), 2509-2526.

Heikkilä J., Martinsuo M. and Nenonen S. (2018) Backshoring of production in the context of a
small and open Nordic economy, Journal of Manufacturing Technology Management, 29 (4),
658-675.
Hernández Gascón J. M., Fontrodona Francolí J., Morron Salmeron A., Castany Teixidor L., Clavijo
Losada M. and Tascón Alonso B. (2018) L’impacte laboral de la indústria 4.0 a Catalunya.
Barcelona, Generalitat de Catalunya.
Johansson M. and Olhager J. (2018) Manufacturing relocation through offshoring and backshoring:
the case of Sweden, Journal of Manufacturing Technology Management, 29 (4), 637-657.
Kagermann H. (2015) Change through digitization: value creation in the age of the industry 4.0. In
H. Albach, H. Meffert, A. Pinwart, and R. Reichwald (eds.) Management of permanent change,
Wiesbaden, Springer Fachmedien, 23-45.
Kagermann H., Wahlster W. and Helbig J. (2013) Securing the future of German manufacturing
industry: recommendations for implementing the strategic initiative industrie 4.0, final report of
the Industrie 4.0 working group, Acatech (National Academy of Science and Engineering).
Lewis M. W. (1998) Iterative triangulation: a theory development process using existing case
studies, Journal of Operations Management, 16 (4), 455-469.
Lung Y. (ed.) (2007) Coordinating competencies and knowledge in the European automobile
system. Final Report, Brussels, European Commission.
McKinsey (2017) Jobs lost, jobs gained: workforce transitions in a time of automation, Washington,
DC, McKinsey Global Institute.
Ministerio de Hacienda y Administraciones Públicas (2016) Informe anual 2016 de la
Agenda Digital para España. https://www.plantl.gob.es/Seguimiento/Informesanuales/
Informes/2016/Informe-anual-ADpE-2016.pdf.
Nedelkoska L. and Quintini G. (2018) Automation, skills use and training, OECD Social, Employment
and Migration Working Papers 202, Paris, OECD Publishing.
Observatorio ADEI (2017) El trabajo del futuro. Madrid, AFI.
Pavlínek P. (2015) The impact of the 2008-2009 crisis on the automotive industry: global trends
and firm-level effects in Central Europe, European Urban and Regional Studies, 22 (1), 20-40.
PCAST (2011) Report to the President on ensuring American leadership in advanced manufacturing,
Washington, Executive Office of the President, President’s Council of Advisors on Science and
Technology.
PCAST (2014) Report to the President: accelerating U.S. advanced manufacturing, Washington,
Executive Office of the President, President’s Council of Advisors on Science and Technology.
Perez C. (2010) Technological revolutions and techno-economic paradigms, Cambridge Journal of
Economics, 34 (1), 185-202.
PwC (PricewaterhouseCoopers) (2018) Will robots really steal our jobs? An international analysis
of the potential long term impact of automation. https://www.pwc.com/hu/hu/kiadvanyok/
assets/pdf/impact_of_automation_on_jobs.pdf.
Raddats C. and Kowalkowski C. (2014) A reconceptualization of manufacturers’ service strategies,
Journal of Business-to-Business Marketing, 21 (1), 19-34.
Raddats C., Kowalkowski C., Benedettini O., Burton J. and Gebauer H. (2019) Servitization:
a contemporary thematic review of four major research streams, Industrial Marketing
Management, 83, 207-223.
Rifkin, J. (2012) The Third Industrial Revolution: How the internet, green electricity, and 3-D
printing are ushering in a sustainable era of distributed capitalism, The World Financial Review,
March 3.

Roland Berger (2016) The Industrie 4.0 transition quantified: how the fourth industrial revolution is
reshuffling the economic, social and industrial model, Munich, Roland Berger GMBH.
Synodinos N. E. (2003) The ‘art’ of questionnaire construction: some important considerations for
manufacturing studies, Integrated Manufacturing Systems, 14 (3), 221-237.
UGT (2017) Industria 4.0. Análisis, evolución e implicaciones para el empleo en Castilla y León.
Valladolid, GRI.
Valenduc G. (2018) Technological revolutions and societal transitions, Foresight Brief 04, Brussels,
ETUI.
Valenduc G. and Vendramin P. (2017) Digitalisation, between disruption and evolution, Transfer,
23 (2), 121-134.
Vandermerwe S. and Rada J. (1988) Servitization of business: adding value by adding services,
European Management Journal, 6 (4), 314-324.
WEF (World Economic Forum) (2016) The future of jobs: employment, skills and workforce strategy
for the fourth industrial revolution, Geneva, World Economic Forum.
Womack J., Jones D.T. and Roos D. (1990) The machine that changed the world, New York, Rawson
Associates.
Xunta de Galicia (2018) Oportunidades industria 4.0 en Galicia: resumo executivo 2018.
Yeung H. W. and Coe N. M. (2015) Toward a dynamic theory of global production networks,
Economic Geography, 91 (1), 29-58.

Chapter 5
Industry 4.0 and the prospects for domestic automotive
suppliers in Poland
Krzysztof Gwosdz, Grzegorz Micek, Arkadiusz Kocaj,
Agnieszka Sobala-Gwosdz and Agnieszka Świgost-Kapocsi
1. Introduction
Industry 4.0 is claimed to be one of the important triggers of smoother work
organisation, significant modernisation and higher innovation performance. Industrial
policy is currently being modernised in many countries (e.g. France, US, Japan and
China) and aimed at the reorientation of national economies towards the requirements
of the Fourth Industrial Revolution in order to maintain and restore workplaces or
improve competitiveness and added value in the domestic sector (Roland Berger 2016).
However, little is known about the role of Industry 4.0 in the potential upgrading
of companies driven by the implementation of digital solutions (Szalavetz 2019). As
the fourth industrial revolution now underway will dramatically change the way
business is conducted, special attention should be paid to the impact of these new
enabling technologies on domestic small and medium-sized enterprises (SMEs) and
entrepreneurs, who constitute the major part of the local economy in several regions
and form critical assets for their successful and sustainable development. The key
question is whether the adoption of Industry 4.0 may provide opportunities for small
and medium-sized enterprises and entrepreneurs to move up within global value chains
(and avoiding stagnation).
Current radical advances in manufacturing technology are often described under the
umbrella term ‘Industry 4.0’. The fourth industrial revolution relies on a combination
of business and manufacturing processes which, due to the implementation of digital
technologies, should allow the integration of all actors in a company’s value chain
(Rojko 2017; Gracel and Łebkowski 2018). The following solutions and systems are
usually classified as Industry 4.0:
a. cyber-physical systems;
b. various technologically advanced solutions, including:
— smart analytics solutions;
— smart decision-support solutions;
— smart solutions in intra-plant logistics, production scheduling, product
development and testing, etc.
Our research on the role of contemporary changes in the domestic sector and on
digital transformation is focused on the Polish automotive industry, which holds a
significant position in central and eastern Europe (CEE) in terms of its employment
growth and overall potential. Part of the integrated periphery of the EU, Poland attracts
numerous foreign companies involved in various global production networks. The

Krzysztof Gwosdz, Grzegorz Micek, Arkadiusz Kocaj, Agnieszka Sobala-Gwosdz and Agnieszka Świgost-Kapocsi
share of value-added created in foreign-owned companies has reached 90 per cent

in the CEE automotive industry (Pavlínek 2017). Hence, it must be argued that, with
low labour costs and a low share of work in value creation (Pavlínek 2018, 2020), the
CEE automotive industry (and in Poland in particular) remains export dependent. The
extent of domestic companies’ involvement in global production networks is relatively
limited (Guzik et al. 2020).
We have investigated whether, and how, Industry 4.0 technologies are influencing the
position of domestic suppliers (companies with a dominant share of Polish capital)
within value chains in the automotive industry. In this chapter, we argue that a vicious
circle exists in upgrading, illustrated by the various constraints being experienced by
domestic suppliers. It is argued that positive attitudes towards the introduction of
new technologies among managers of domestic automotive companies are rather the
exception than the rule. To sum up, we argue that digital transformation facilitates
neither the internationalisation of local SMEs or global value chain integration in Poland.
Our research is based both on primary and secondary data combined with a literature
review of the position of the Polish automotive industry and the role of domestic
suppliers within it. Fourteen interviews were conducted covering successful automotive
suppliers (six interviews), ‘digital entrepreneurs’ – providers of Industry 3.0 and 4.0
solutions1 – (six interviews) and key public stakeholders (two interviews) representing
institutions promoting the implementation of Industry 4.0 and managers of automotive
clusters (see overview in Appendix). Automotive suppliers in the sample were selected
on the basis of information obtained from, and the recommendations of, key personnel
both among managers of automotive clusters and from IT & automation companies
delivering Industry 4.0 solutions to the market. Data collection has been triangulated
with a variety of secondary sources of information in order to enhance the reliability of
our observations and conclusions.
This chapter consists of three main sections. The next section briefly discusses the Polish
context and concentrates on the position of and current changes in the automotive
sector in Poland, with a particular focus on the role of domestic suppliers. The main
findings of the chapter can be found in section two and these are summarised in the last
section.
2. Setting up the Polish context

2.1 Snapshot of the current state of the automotive industry in Poland
The automotive industry is the second largest manufacturing sector in Poland both in
terms of production and exports. The role of this industry in Poland has been continually
strengthening in the last three decades and, together with other CEE economies, the
country has become one of the world’s fastest growing centres of the automotive industry,
1. Industry 3.0 solutions include robotisation and the automatisation of manufacturing whereas Industry 4.0, as
the next step, further entails the digitalisation of production processes.

Industry 4.0 and the prospects for domestic automotive suppliers in Poland
second only to China.2 Currently, around 1,000 producers are active in Poland, with
total employment exceeding 214 thousand, which makes Poland’s automotive industry
the third largest in the EU after Germany and France. This development has been
driven by massive foreign direct investment. More than 330 new automotive factories
were built in Poland in the years 1989-2017, four-fifths of them by foreign investors
(Domański and Gwosdz 2018). The country specialises particularly in components,
parts and buses, while the dynamics of passenger car manufacturing substantially lag
those of other CEE countries: Slovakia, Hungary and Czechia in particular.
The Polish automotive industry is positioned in the spatial division of labour as an

integrated semi-periphery (Guzik et al. 2020; Pavlínek 2012, 2017, 2018, 2020;
Krzywdziński 2018). Poland (along with other CEE countries) still performs a dual role,
and low value-added and labour intensive products still constitute a substantial part of
total output (one-third, according to estimates by Guzik et al. 2020). Gradual industrial
upgrading is underway; however, the scope of functional upgrading is relatively
limited, especially in comparison with the role of the country in production. Thus, the
dependence on foreign firms and the secondary role of indigenous producers (discussed
in greater detail later in this chapter) is a significant weakness.
2.2 The role and position of domestic suppliers

There are 280 companies with dominant Polish capital in the broadly-defined automotive
sector in Poland (Table 1) and they bring together one-fifth of all employment in the
automotive industry in the country. Less than seven per cent of them can be regarded
as big companies in terms of employment or revenue (above 250 employees or €50m in
revenue). Medium companies represent the core segment of the industry, while small
companies are also numerous (Table 1).
Table 1 Automotive companies in Poland with predominantly domestic capital, by size
Company size As of 2018 Company size As of 2017

(number of No. of Employment (revenue) No. of Total revenues
employees) companies in thousand in €m companies EUR millions
1,000 and more 3 5,800 100 and more 7 2,560.0
500-999 9 5,628 50-99.9 14 969.4
250-499 21 7,198 10-49.9 75 1,565.5
50-249 149 20,136 5-9.9 51 378.4
10-49 56 1,795 Less than 5 86 191.2
No data 42 – No data 47 –
Total 280 40,557 Total 280 5,664.5
Source: authors’ research
2. Dynamic growth and its implications have been widely documented and discussed in the research literature; see,
among others: Domański et al. (2013); Domański et al. (2018); Drahokoupil (2009); Jürgens and Krzywdziński
(2009); Krzywdziński (2018); Nölke and Vliegenthart (2009); Pavlínek (2012, 2017, 2018, 2020); Pavlínek and
Ženka (2016) and Szalavets (2012).

Domestic suppliers represent all segments of component production, with the

largest number being in the manufacture of plastic parts, metal and stamped parts,
technological elements and electric components. Three-quarters of domestic producers
are, predominantly, tier two suppliers. About 30 per cent deliver at least part of
their production directly to OEMs, though only eleven per cent are predominantly
tier one suppliers; production in the rest is focused more on tier two suppliers or for
the aftermarket. Design competence is most common among tier one suppliers (25
per cent of plants) but very rare among lower-tier producers which are, in the main,
subcontractors.
Although there is a clear growth of, and fast learning among, domestic tier one and
tier two component producers, which are capable of providing high product quality
and reliability of delivery, there is only limited progress in design competence among
domestic companies. Only twenty-nine plants have competencies within product design
(according to the requirements of IATF 16949). A relatively new trend is the emergence
of independent start-ups, offering design and R&D services (for example Cadway
Automotive in Rzeszów, CADM Automotive in Kraków and ctrlCAD in Katowice).
The analysis conducted in this research study and also in previous papers (see Domański
and Gwosdz 2009; Pavlinek et al. 2009; Guzik et al. 2020) confirm that there is an
ongoing development of companies that are participating in the automotive supply
chain (reflected in growing revenue, increases in the range of products, development
of tooling shops and construction departments and significant activity in obtaining EU
funds). This has resulted in some companies becoming product specialists.
A recent phenomenon, which started after 2010 and gained momentum in 2015-2017,
is the international expansion of Polish companies. This ‘going international’ trend
(whether through greenfield investments or mergers and acquisitions) is associated
primarily with the opening of a window of opportunity for Polish companies, which
turned out to be the 2008-2009 crisis in the industry (Domański et al. 2013). None
of the Polish producers expanded to core (western European) markets before 2010
and, in the early 2000s, only one Polish manufacturer – Groclin – decided to locate
its activities outside the country, with the dominant motive being cost reduction.
The motivations of companies that began international expansion during the last ten
years are fundamentally different from considerations made prior to the crisis – cost-
driven expansion has not been a particularly relevant factor in any of our investigated
cases. Instead, it has been about the diversification of operations in the entire group
(acquisitions made by Boryszew); expansion within the main activity by entering new
markets (Wielton, Alumetal, Izoblok, Sanok Rubber); or following customer strategy
(greenfield investments by Boryszew and Sanok Rubber). It must be stressed here that
most of the Polish companies which have internationalised had previously achieved
a significant position in the domestic market, be it in the automotive industry or in
other business segments. In this sense, their growth can be significantly interpreted as
a staged development in the internationalisation model (IP model, also called ‘Uppsala
model’) (Johanson and Vahlne, 1977). The only exception is Izoblok, a Chorzów-based
company whose mode of development can better be described by the concept ‘born
global’ (Knight and Cavusgil 2004; Madsen and Servais 1997).

The share of manual operations is still significant in the plants owned by leading Polish
manufacturers. As one of our key public stakeholders remarked, they ‘are currently in
the phase of a difficult leap from Industry 2.0 to 4.0’. Strategies in these companies
mostly aim at the comprehensive implementation of automation and robotisation
(Industry 3.0), especially in the new production departments of companies launched
both in Poland and abroad; however, up to now none of them have clearly declared in
their strategy the desire to show strong leadership towards Industry 4.0.
Barriers to the growth of domestic producers in the (semi)periphery to becoming

European/global suppliers are stronger now than they were ten to twenty years ago.
This is due to: consolidation among the major automotive suppliers; the need to be
present in various regional markets (Asia, America, Europe); the growing complexity of
supply networks; and greater design requirements. The vicious circle which limits the
functional upgrading of domestic suppliers is driven thus: the lack of design competencies
constrains profitability which, in turn, reduces investment and development capabilities.
A small scale of production alongside a low level of R&D competence has made it
impossible to meet OEM expectations – including among the capital and organisational
abilities to follow new client projects. Furthermore, small-scale production and the
lack of design competencies (and, therefore, prospects for contracts for new projects
with higher profit margins) enables the harvest only of ‘transfer projects’, hindering the
possibility to upgrade within value chains. Only a few Polish companies have managed
to break out of this vicious circle. It must also be stressed that some local companies are
quite satisfied with their tier two position, perceiving promotion to tier one status not
as an upgrade in their position in global value chains but rather a substantial risk to the
company’s existence [Interview COMP-08].
Where can the opportunities for domestic automotive firms be found? Four main areas
can be underlined in this regard:
— hybridisation: the involvement of non-automotive segments, or integration with

other sectors. Several Polish companies (especially those producing plastics and
metal components) combine deliveries to the auto industry with production for
home appliances, electronics and construction industries;
— niche products: the ability of Polish companies to acquire competitive advantage

has, after the 2008-2009 crisis, been directly related to the cost benefits and
greater flexibility achieved from meeting specific customer requirements. This
results from the establishment of market niches in the earlier period, above all in
labour intensive segments (various products ranging from trailers or semi-trailers
for special purposes, construction and protective components made of plastics
and aluminium to demanding and technologically-sophisticated production
services);
— developing close cooperation with other SMEs, either directly or via cluster
initiatives enabling full-service supply thanks to the complementary capabilities
of other network participants. This innovative strategy as far as Polish territory is
concerned has been promoted by the members of the Polish Automotive Cluster

(PGM), with the members of the cluster recently initiating the establishment of a
‘PGM’ joint commercial brand;
— the aftermarket.
3. Effects and scale of Industry 4.0 in Poland

Major interest in Industry 4.0 solutions in Poland has been expressed by foreign-owned
companies. A survey conducted on a large sample of manufacturing companies by Astor
in 2017 (ASTOR 2018) shows that a growing number of companies have also started
activities aimed at recognising the possibilities of transformation into Industry 4.0.
However, the Astor research revealed that only seven per cent of factories had started
or partially implemented any Industry 4.0 technologies or solutions (mainly in the form
of Manufacturing Execution Systems). Moreover, not only the level of digitalisation
but also the level of automation remains low. Hence, companies have first to face up
to the third industrial revolution: in 2017, some fourteen per cent of manufacturing
plants had still not entered Industry 3.0 (ASTOR 2018). Interest in the automation
of manufacturing (Industry 3.0) stems from the decreased cost of technologies but,
according to Astor, the leitmotif is pressure from customers to reduce costs and
shortages of highly-skilled employees. The initial trigger (observed especially in SMEs)
is related to price negotiations that do not end up in a deal due to high prices. This is
endangering the position of Polish SMEs in supply chains (ASTOR 2018).
Interviews with managers of automotive companies confirm the above data. ‘Many
automotive companies are still far from automation. Half automation – we are here.
This is a big threat to companies with Polish capital. In order to exist, it is necessary
to automate. Some companies do not know how to do it or why they are doing it. They
look at it, as they did ten years ago, and then it did not make financial sense because
labour costs were cheap. And I have the impression that many companies are still there.
In the end, this is an investment and a risk. The problem is that some have overlooked
the period when it was necessary to get into automation. Companies must go in this
direction and, if they do not go, they will fail. But automation is a change, a challenge
for organisation’ [Interview COMP-07].
In interviews conducted among our automotive companies, suppliers of Industry

4.0 solutions and key public stakeholders, three strategies for domestic automotive
companies regarding digitalisation have been revealed.
First is the most common sceptical attitude which relies on the passive observation
of new solutions. Many Polish companies agree it is necessary to enter the fourth
industrial revolution but, for now, they have decided not to implement Industry 4.0
solutions. This is due to several organisational (managerial), financial, intellectual and
technological barriers (Table 2).

Table 2 Main challenges to implementing Industry 4.0 in companies with domestic

capital in Poland
Domain Main barriers

Management — Low level of openness to cooperation
— Low level of project and process management
— Low level of use of IT tools among managers
Human and social capital — Low social capital; lack of mutual trust between entrepreneurs, and sometimes even
mutual hostility; unwillingness to cooperate; lack of courage in undertaking risky
investments
— Shortage of young automotive specialists; ageing practitioners
— Outwards migration of outstanding specialists
Technology — Too weak involvement of R&D units for cooperation with the SME sector
Financial — Low availability of funds for research and implementation of innovation in automotive
companies
— Low and falling profitability of companies (seven per cent in 2012 to less than four per
cent in 2018)
— Difficult access to external financing for companies in the SME sector
Source: Interviews with managers and key public stakeholders
When Polish managers or owners recognise that, during the implementation of Industry
4.0, there is a need not only to change the technological solution but also to rebuild the
whole company in terms of its management, they resign from this path. This obstacle
is related to the limited competencies of chief engineers and the fear of exchanges
with staff. The following quotation summarises it well: ‘You have to look at this in the
concept of habits; these engineers have always acted in such a way – why should they
change it?’ [Interview ORG-09]. As another respondent claimed: ‘Polish SMEs have
not gone mentally through the development stage – to measure the process and to
improve it’ [Interview COMP-01]. Meanwhile, in other departments of the company,
especially those in which financial analysis is being carried out, there is increasing wage
pressure and a belief that something must be done about introducing automation and
digitalisation.
Secondly, the current stream of EU funding is focused on supporting R&D activities, not
purely on the implementation of digitalisation or automation itself. Hence, medium-
sized companies are gradually implementing Industry 4.0 solutions and technologies
after carrying out R&D activities. The introduction of new solutions is being done on
a step-by-step basis and so the implementation of Industry 4.0 is not holistic. Such
a strategy may even be based simply on a desire to distinguish one’s companies from
other suppliers and for higher prestige among western customers (Table 3).
Thirdly, a long-term corporate strategy aimed at full automation and transition to a

higher level in the value-added chain ought to be a necessity for many firms. However,
radical implementation (a total redesign of manufacturing processes and management)
of Industry 4.0 is uncommon and very rarely takes place even in newly-constructed
factories. One of the interviewees summed this up by arguing that digitalisation and
Industry 4.0 have become slogans and symbols for the introduction of new technologies.
‘But everyone draws from these solutions what they need. Not everyone needs a fully
automated factory’ [Interview COMP-06].

Polish automotive companies which may be classified as tier two suppliers are still far
away from Industry 4.0 and even 3.0. It is not yet sure whether the profile of domestic
suppliers is suitable for Industry 4.0 solutions. Indeed, no domestic automotive suppliers
has gone wholesale into Industry 4.0. To sum up, there is only a limited chance for the
whole automotive industry to enter the digital age of manufacturing. At the current
stage, we can talk rather more about implementation pilots for Industry 4.0 in firms
with predominantly Polish capital. The main method is to implement solutions in small
steps – a systematic implementation of individual solutions in existing production
lines [Interview COMP-10]. Our research shows that companies are especially eager to
implement methods and technologies in the area of predictive maintenance.
An overview of existing pilot schemes in selected Polish automotive companies is

provided in Table 3.
Table 3 Motivations, applied technologies and the results of implementing Industry 4.0
in selected automotive companies with predominantly Polish capital
Motivation behind implementation Main technology applied Results achieved

A long-term strategy of the company CPS data collection and analysis 1. Improvement in the production
aimed at full automation and (PQS); fully automated gas-spring process and in product quality
transition to a higher level in the production line (greater stability of quality)
value-added chain 2. Reducing the number of employees
and increasing their skill level
3. Lower quantity of waste
The will to distinguish the company Fully automated and robotised 1. Greater process control and
among other suppliers; raising the welding line; CPS data collection and optimisation of the product’s
prestige of the company among analysis manufacturing cost
customers in the west; increasing 2. Greater comfort of service work,
production capacity; ensuring increasing the skills of the team
replicable quality in a demanding 3. Authenticating the very high product
technological process quality in the eyes of customers
Implemented as an experiment, the Fully automated processing line for 1. Obtaining the replicability of the
secondary motive was the reduction of thermoplastic materials; IT unit for production cycle
labour costs the central management of injection 2. Adherence to a time operation
moulding processes; advanced quality regime
control systems (AQC); CPS data 3. Technological production stability
collection and analysis 4. Saving on losses, malfunctions and
defects
5. Good information flow (facilitating
responses to complaints)
Source: Interviews with managers and company data
3.1 The impact on human resources

The main factor accelerating the digitalisation of the Polish automotive supply industry
is the current labour market situation, in particular the increase in labour costs
[Interview COMP-09] and staff shortages. This was well summarised by one of the
managers of a global automotive tier one company operating in Poland: ‘When it comes
to Industry 4.0, the conditions in the labour market gave us this gift so we could engage
in 4.0 solutions;’ while another remarked: ‘Industry 4.0 is a direction that we can not

avoid because of the human and economic factor (...) When there is a problem, we want
to know where it is. We want to combine data about production’ [Interview COMP-08].
Introducing Industry 4.0 solutions in a company spurs substantial changes in its skills
breakdown. As one of the managers we interviewed emphasised: ‘The company will
increasingly rely on more skilled employee segments, like constructors as well as IT
specialists’ [Interview COMP-12]. Another interviewee stressed: ‘Because we made a
conscious decision systematically to increase the level of automation and robotisation,
now we need engineers, not production workers/operators. The ideal situation would be
if we did not have operators at all, only automatic lines and engineering staff’ [Interview
COMP-07].
With the push to automate, there are expected to be limited job losses among production
workers in companies introducing Industry 3.0 and 4.0 solutions. However, given
the relatively low technology level among the majority of domestic suppliers, major
job losses may emerge in the future as a result of company failures to survive in the
competitive automotive market. Digitalisation will represent a serious threat to further
competitiveness if technological change turns out to have a disruptive pace and range.
Industry 4.0 requires an interdisciplinary and interdepartmental approach, as well as a

combination of knowledge and skills in several areas (Gracel and Łebkowski 2018). This
combination manifests itself firstly in attitudes towards operational and managerial
change. Our research reveals the existence of intellectual barriers at the managerial and
company ownership levels, but also as regards engineers and production workers. At
the highest management level, there is the problem of accepting innovative solutions
due to high costs while there is, simultaneously, a lack of searching for comprehensive
solutions. There are also, overall, reactive attitudes to changes in the labour market
[Interview COMP-01]. At the mid-management level, the conservative approach to the
training of engineers in Polish companies weakens the possibilities of using Industry
4.0 solutions (ASTOR 2017). At shopfloor level, new solutions increase the scope of
processes that fall on a single employee, thereby increasing responsibility. The shortage
of human resources and position mismatches are one of the major barriers to the
development of Industry 4.0. The ability to learn, unlearn and relearn is relatively low
in Poland.
As emphasised by respondents, there is a significant problem with finding larger groups

of engineers with the required skills to handle larger projects. This gap may, partially,
be filled by transfers from international companies. The second chance for Polish
automotive companies seems to be the use of human resources from related industries.
New areas of technical competence will gain in importance, such as the integration of IT
systems control techniques and the integration of analytical methods in data clouds with
local networks or cyber security. This will lead to the search for highly-skilled engineers
in automation, robotics and software, production engineers, designers of automation
systems and mechatronics, and designers from the virtual reality world or gaming.

3.2 Public policy response

The Strategy for Responsible Development, an instrument launched recently by the
Polish government to manage the main development processes in Poland, identifies five
development traps that Poland faces, including the ‘average product trap’. The authors
of the plan highlight – among others – that R&D expenditure amounts to less than one
per cent of Polish GDP. They also stress the slightly awkward findings that only thirteen
per cent of SMEs innovate (compared to 31 per cent in the EU) while only five per
cent of exports originate in high-tech sectors (https://www.miir.gov.pl/media/14873/
Responsible_Development_Plan.pdf).
This section offers a short look at the current support instruments introduced by the
Polish government which are aimed at accelerating and facilitating the transition of the
Polish economy into Industry 4.0.
The main activity is the Platform for Future Industry (4.0) (Platforma Przemysłu
Przyszłości) which started its operation at the start of 2019 as a foundation under the
Council of Ministers. The role of the Platform for Future Industry (4.0) is to integrate
and accelerate the transformation of the Polish economy towards Industry 4.0. The
Platform’s project also encompasses the integration and coordination of activities for
suppliers, research centres and various entrepreneurs, from both the public and the
private sectors. Moreover, the Platform will help in the setting-up of a cross-linked
business ecosystem (company network) and the coordination of Digital Innovation
Hubs (DIH), and will also provide access to instruments and tools aimed at stimulating
interaction between companies and research institutions (https://www.mpit.gov.pl;
http://przemysl-40.pl).
Andrzej Soldaty, President of the Management Board of the Platform for Future Industry
(4.0) foundation, argues that the government’s involvement in the implementation of
Industry 4.0 extends to measurable effects in the area of the better use of resources
and the better use of market opportunities. In his opinion, ‘Not entering the 4.0 level
and not becoming competitive poses a great threat to the whole economy,’ while the
role of the Platform for Future Industry (4.0) should be to raise the competitiveness
of enterprises by supporting digital transformation. The Platform’s aim is to support
development in four areas: market; business environment; technology; and people.
Soldaty distinguishes two main roles of the Platform: the first is in creating and
recommending, based on developments in knowledge and skills, but also in providing
formal and legal solutions and financial resources; the second is to enable ‘coopetition’
– to encourage competition and cooperation among enterprises by building networks
and establishing thoughts, concepts and common goals in order to raise the level of
advancement.
Assistance in carrying out changes in enterprises will be delivered by regional initiatives

named Competence Centres (Centra Kompetencji) whose role is to offer developed
support instruments, such as workshops and training, and other comprehensive services
for enterprises. Furthermore, the Centres will be responsible for cooperation between
research institutions, technology providers, engineering companies and business

partners. In 2018, three regional Competence Centres were opened in the Mazowieckie,
Wielkopolskie and Śląskie regions (https://www.mpit.gov.pl). In addition, the
Ministry of Entrepreneurship and Technology, together with the three Universities of
Technology – Silesian, Warsaw and Poznań – prepared in 2018 a pilot project named
the ‘Incubator of Industry Leaders 4.0’ (Inkubator Liderów Przemysłu 4.0). The main
aim of the project is to train staff in the Competence Centres and in the Platform for
Future Industry (4.0).
In 2017, a Sectoral Programme of Scientific Research and Development (INNOMOTO)

was launched by the National Centre for Research and Development and the Polish
Chamber of the Automotive Industry. INNOMOTO supports the implementation
of large R&D projects and aims to increase the number of innovative solutions. The
co-financing of projects can help entrepreneurs create new, or expand existing, R&D
departments as well as develop innovative technologies and products. In 2017 and
2018, 57 projects were co-financed by INNOMOTO to a total amount in excess of PLN
326m (https://www.ncbr.gov.pl/; http://innomoto.com.pl).
One of the tools of innovation policy formation has been the revision of the tax law in
Poland, which took place in the second half of 2017. The purchase cost of industrial
robots and 3D printers can now be written off more efficiently. The government hopes
that new incentives will encourage companies, small and medium-sized enterprises in
particular, to invest in new technologies and solutions.
The main weakness of the tools and solutions implemented by the government is
the lack of the necessary flexibility in the case of R&D activities. The specificity of
innovative projects often forces changes in the direction of research during the process,
but programmes do not take into account the complexity of implementing the solution
and the frequent need for cooperation. Representatives of companies in our survey
emphasise that, in the case of projects financed from public funds, it is very difficult
to change the project assumptions during the process and call for greater trust –
‘Entrepreneurs should be given more flexibility and the right to make mistakes as this
is what searching for innovative and effective solutions is about’ [Interview ORG-01].
However, there are promising results among some innovative support programmes
aimed at accelerating cooperation between innovative startups and big companies
in more mature industries. These have been introduced by the Polish Agency for
Entrepreneurship Development (PARP) in partnership with technological parks (i.e.
the PARP KPT ScaleUp accelerator). With the low level of industry 4.0 implementation
in Polish companies, technology parks become an essential channel for transferring
state-of-the-art solutions. It is hard to assume that incubators and technology parks
will be a central pillar for introducing Industry 4.0 solutions among Polish producers,
but they can be an important element in the upgrading of Polish industry. Know-how
in the area of particular unsolved problems is critical to the success of initiatives along
with support for global expansion.
Another example of good practice could be clusters (e.g. Automotive Silesia, the East
Automotive Alliance). The role of a cluster is to support cooperation, spread knowledge

and provide the opportunity to attract potential suppliers. A representative of one of

our organisations [Interview ORG-01] pointed out that (especially foreign) executives
increasingly expect companies to have a range of competences – from design through
to the ability to make a prototype and to develop an implementation-ready solution. In
this context, cooperation between companies is crucial because most local firms are too
small to carry out this kind of imperative alone.
4. Conclusion – the thorny road to Industry 4.0?

Polish automotive suppliers, especially SME companies, are at the very beginning of
the path leading to the implementation of Industry 4.0 solutions. There is a growing
awareness among owners and managers, but few companies have yet started to taste
the waters and, for now, the thorough implementation of smart manufacturing in the
automotive industry is limited to large foreign-owned plants.
At the moment, in automotive suppliers with predominantly Polish capital we can observe
pilot implementations of Industry 4.0 solutions. The companies that are experimenting
with these are mainly seeking a means of better serving their existing customers; very
rarely are they embracing Industry 4.0 as a long-term strategy aimed at upgrading to a
higher level in the value chain. Thus, the argument that digital transformation facilitates
the independent internationalisation of local SMEs and their integration into global
value chains can not be confirmed at the current stage of development of the bulk of
companies in the automotive industry in Poland. There are only a handful of companies
that are consciously introducing a strategy of full implementation of Industry 4.0.
However, a more adequate concept describing such companies would be ‘readiness for
change towards 4.0;’ there is a strategy to reach 4.0 based on a high sense of advance
(the implementation of comprehensive Industry 3.0 solutions), preparing staff and
seeking the availability of capital. Even so, the vast majority of domestic suppliers are
at the transition stage from Industry 2.0 to 3.0 solutions. A specific feature of Polish
suppliers is still a significant amount of production by manual labour, especially in
companies founded in the 1990s. Undoubtedly, the acceleration of the implementation
of solutions in the field of automation and robotics production at the level currently
observed in Polish industry will be a vehicle that will also accelerate the implementation
of Industry 4.0. However, this will be an incremental upgrading because there are no
mechanisms that allow leapfrogging. The interviews we conducted indicate that, at the
current stage in the automotive industry in Poland, companies are mostly interested in
Industry 4.0 solutions in the area of predictive maintenance.
The application of Industry 4.0 solutions in the coming years will not be a factor likely to
improve significantly the position of Polish automotive suppliers in European production
networks. In a perspective of a few years alone, neither will it have a negative impact
on their functioning in the market. This is connected on the one hand with the inertia
of industry (the duration of contracts already concluded) and, on the other, with the
existence of cost and product niches in which local producers may remain competitive.
However, if the technology gap continues, this will represent a serious threat to the
competitiveness of these companies in the medium and long-term.

The main barriers to the development of Industry 4.0 are the underfunding of
technological innovations; intellectual barriers among owners and managers; and a lack
of skilled staff (both among production workers and among management). Moreover,
the lack of cooperation between companies, research centres and government is having
negative effects on the development of innovations (cf. Table 2). Our research also
shows a deeper problem resulting from the characteristics of organisational culture
and social capital in Poland. As one of the managers pointed out: ‘The plants operating
in Poland are focused solely on production, not on data interpretation. Polish SMEs
have not mentally reached the stage “Measure the process and improve it”’ [Interview
COMP-01]. Another interviewee remarked: ‘We have an inadequate organisational
culture in Poland. The basic barrier is low trust among companies and a low degree of
cross-linkages’ [Interview COMP-01].
Support instruments introduced by the Polish government aimed at accelerating

and facilitating the transition of the Polish economy into Industry 4.0 are of a very
recent nature (the first actions were taken in 2018). Therefore, it is too early to draw a
firm conclusion about the effects of the newly-implemented programmes. The public
stakeholders we interviewed are also careful in their opinions, stressing that public
support should not be overestimated in the whole process of speeding toward Industry
4.0 in Poland. Nevertheless, managers assess the activities of the Ministry of Enterprise
and Technology as valuable, mainly in the field of providing support for education and
staff training. What is considered by managers as particularly relevant and effective in
terms of support is the possibility of experimenting and exploring new fields. Without
public financial support, Polish companies could not afford this. There have been some
promising results among some innovative support programmes aimed at accelerating
cooperation between innovative startups and big companies in more mature industries
which have been introduced by the Polish Agency for Entrepreneurship Development
(PARP) in partnership with technological parks and Special Economic Zones. The
major long-term positive effect of current policies may be found in networking activities
among business companies, research centres, universities and other institutions.
Four main factors that have, or could have, an impact on accelerating the implementation
of Industry 4.0 solutions in the Polish automotive industry have been identified. The most
important, according to this survey, is the growing cost and decreasing availability of
employees. Support offered by public authorities could become an important factor but,
because the leading tool (the Platform for Future Industry 4.0) is at the implementation
stage, it is difficult to indicate how effective it will be. It should be remembered that
a large number of countries have implemented similar mechanisms, so it is difficult
to predict ex ante the extent to which the activities of the Polish institutions will be
effective. Given the structural features of Poland (low institutional social capital), one
should be careful in forecasting that it will be a breakthrough factor. Another vehicle
for the implementation of Industry 4.0 solutions might be customer policies, primarily
among OEM and tier one companies towards their tier two and lower level suppliers.
This path, however, is associated with the danger of increased dependence on the
dominant partner.

The growth of Industry 4.0 in Poland may depend not only on the production companies
themselves but also on the development of domestic companies specialising in the
integration of automatic and digital solutions (‘digital entrepreneurs’). Technological
changes related to Industry 4.0 seem to provide a window of opportunity for medium-
sized Polish suppliers of tailor-made technological and software solutions. These
are providers of comprehensive services in the field of industrial automation, PLC
programming, robotics, SCADA visualisation systems, MES class systems, industrial
informatics, data collection and archiving. Thus, we predict that the upgrading
of companies with Polish capital will be more indirectly than directly related to the
development of Industry 4.0 in the automotive industry. The large cluster of automotive
companies in central Europe creates a market for domestic companies offering Industry
4.0 solutions which can build their competence on the basis of cost and responsiveness
factors. The relatively large size of the domestic market facilitates the implementation
of what are ground-breaking solutions for innovative SME companies.
On the other hand, the massive size of the internal market reduces the pressure on
companies to ‘go global’ and take on challenges abroad. Also, the relatively close
headquarters-coordinated structures of foreign-owned subsidiaries hamper any moving-
up in the newcomer value chain. The key to the upgrading of domestic companies is
the engagement of Polish tech companies in developing digital solutions which would
make many indigenous firms experience a ‘leapfrog effect’. However, according to some
experts, ‘Due to the lack of expertise in automation and robotics in Poland, foreign
suppliers will be able to profit more than others from the trend towards automation’
(https://industryeurope.com/polish-automation-gaining-momentum/).
Thus, we may conclude that, at the current stage of research, although the domestic
market has significant limitations (low demand from local companies and limited
decision-making competence among foreign subsidiaries), it does offer some important
growth factors for digital entrepreneurs. This point was well summarised by one of our
interviewees: ‘The level of global OEMs is beyond our reach at present. But between
them and small companies there is a vast space for growth’ [Interview COMP-01].
Taking into account the preliminary conclusions resulting from this exploratory research
among high-tech domestic companies which are providers of Industry 4.0 solutions, a
promising research agenda is emerging which could aim, inter alia, at the identification
of the growth factors appropriate to such companies and in-depth investigation of
the conditions for the promotion of competence within European and global value-
added chains. Future research might also explore the functioning of the ecosystem for
high-tech domestic ‘digital entrepreneurs’; the main mechanisms and factors behind
the upgrading of these companies; and the impact of regional and local features on
their emergence and growth, in particular the extent to which geographic proximity –
both to customers and to other high-flying domestic companies – is important for the
development of such firms.

Appendix
List of interviewed companies and organisations
Interview No Type of company Main activity/products Turnover in PLN millions (M)*
Employment**
COMP-01 Digital entrepreneur Control and automation systems PLN 10-20M

100 employees
COMP-02 Digital entrepreneur Provider of industrial automation and PLN 10-20M

industry 4.0 solutions 50 employees
COMP-03 Digital entrepreneur Provider of automation, IT solutions and PLN 50-100M

industrial robotics for different branch of 60 employees
industry, inluding automotive
COMP-04 Automotive supplier Design & testing services for automotive PLN 50-100M
Tier 1 / Tier 2 industry, special machines-design and 400 employees
production, Conversion of vehicles from
combustion engine to electric drive
COMP-05 Digital entrepreneur Design of vehicles, modules and parts PLN 50-100M
for the transport industry (including 250 employees
automotive)
COMP-06 Digital entrepreneur Design and manufacturing of 3D scanners PLN 10-20M

for the automotive industry 50 employees
COMP-07 Automotive supplier gas springs, ball joints and tie rods PLN 10-50M
Tier 1 250 employees
COMP-08 Automotive supplier plastic components for automotive, PLN 50-100M

Tier 2 electrical enginnering, home appliances 150 employees
and interior decoration industry
COMP-09 Automotive supplier Stamped metal parts PLN 100-200M

COMP-10 Digital entrepreneur Factory automation and processing PLN 100-200M

technologies, drive products, computerised 250 employees
numerical controllers
COMP-11 Automotive supplier Rubber products, plymers and elastomers PLN 5-10 M
Tier 2 combined with metal and plastics for the 50 employees
automotive, household appliances and
electrical enginnering industries
COMP-12 Automotive supplier Starter batteries PLN 200-500M

ORG-01 Business Organisation Association of manufacturers of Over 20 members (1-2 bln of

automotive parts and accessories. revenues and 5000 employees
in 2017 in total)
ORG-02 Business Organisation Support for the ecosystem of high-tech PLN 10-50M
companies: technology incubator and 50 employees
accelerator, certified Living Lab, providers
of hardware, software and network
infrastructure
* Euro (EUR) to Polish zloty (PLN) annual average exchange equalled 4,298 in 2019.
** Employment figures do not include agency or temporary workers.

References
ASTOR (2017) Inżynierowie Przemysłu 4.0: (Nie)gotowi do zmian?, White Paper, Kraków,
ASTOR Publishing. https://www.astor.com.pl/images/Industry_4-0_Przemysl_4-0/ASTOR_
Inzynierowie_4.0_whitepaper.pdf
ASTOR (2018) W stronę przemysłu 4.0. Praktyczny przewodnik, Harvard Business Review Polska
(HBRP). https://www.hbrp.pl/b/w-strone-przemyslu-40-praktyczny-przewodnik-raport/
P1FQ7xwE1
Domański B., Guzik R., Gwosdz K. and Dej M. (2013) The crisis and beyond: the dynamics and
restructuring of automotive industry in Poland, International Journal of Automotive Technology
and Management, 13 (2), 151-166.
Domański B., Guzik R. and Gwosdz K. (2018) The new international division of labour and the
changing role of the periphery: the case of the Polish automotive industry, in Tamasy C. and
Taylor M. (eds.) Globalising worlds and new economic configurations: geographical perspectives,
London, Routledge, 85-100.
Domański B. and Gwosdz K. (2009) Toward a more embedded production system? Automotive
supply networks and localized capabilities in Poland, Growth and Change, 40 (3), 452-482.
Domański B. and Gwosdz K. (2018) Changing geographical patterns of automotive Industry in
Poland, Prace Komisji Geografii Przemysłu Polskiego Towarzystwa Geograficznego [Studies of
the Industrial Geography Commission of the Polish Geographical Society], 32 (4), 193–204.
https://doi.org/10.24917/20801653.324.12
Drahokoupil J. (2009) Globalization and the state in Central and Eastern Europe: the politics of
foreign direct investment, London, Routledge.
Gracel J. and Łebkowski P. (2018) The concept of Industry 4.0 related Manufacturing Technology
Maturity Model (ManuTech Maturity Model, MTMM), Decision Making in Manufacturing and
Services, 12 (1-2), 17–31.
Guzik R., Domański B. and Gwosdz K. (2020) Automotive industry dynamics in Central Europe, in
Covarrubias V. and Ramírez Perez S. (eds. ) New frontiers of the automobile industry, exploring
geographies, technology, and institutional challenges, Basingstoke, Palgrave Macmillan,
377-400.
Johanson J. and Vahlne J.-E. (1977) The internationalization process of the firm: a model of
knowledge development and increasing foreign market commitment, Journal of International
Business Studies, 8 (1), 23-32.
Jürgens U. and Krzywdzinski M. (2009) Changing East-West division of labour in the European
automotive industry, European Urban and Regional Studies, 16 (1), 27-42.
Knight G.A. and Cavusgil S.T. (2004) Innovation, organizational capabilities, and the born-global
firm, Journal of International Business Studies, 35 (2), 124-141.
Krzywdzinski M. (2018) Die Rolle von Niedriglohnperipherien in den globalen
Wertschöpfungsketten der Automobilindustrie: der Fall Mittelosteuropas, PROKLA, 48 (193),
523-544. https://doi.org/10.32387/prokla.v48i193.1144
Lee J., Bagheri B. and Kao H. (2015) A cyber-physical systems architecture for Industry 4.0-based
manufacturing systems, Manufacturing Letters, 3, 18-23.
Leyh C., Bley K., Schäffer T. and Forstenhäusler S. (2016) SIMMI 4.0 - a maturity model for
classifying the enterprise-wide IT and software landscape focusing on Industry 4.0, Proceedings
of the Federated Conference on Computer Science and Information Systems (FedCSIS),
1297-1302. https://annals-csis.org/Volume_8/pliks/478.pdf

Madsen T.K. and Servais P. (1997) The internationalization of Born Globals: an evolutionary
process?, International Business Review, 6 (6), 561-583.
Nölke A. and Vliegenthart A. (2009) Enlarging the varieties of capitalism: the emergence of
dependent market economies in East Central Europe, World Politics, 61 (4), 670-702.
Pavlínek P. (2012) The Internationalization of corporate R&D and the automotive industry R&D of
East-Central Europe, Economic Geography, 88 (3), 279-310.
Pavlínek P. (2017) Dependent growth: foreign investment and the development of the automotive
industry in East-Central Europe, Cham, Springer.
Pavlínek P. (2018) Global production networks, foreign direct investment, and supplier linkages in
the integrated peripheries of the automotive industry, Economic Geography, 94 (2), 141-165.
Journal of Economic Geography, 20 (2), 509-541. https://doi.org/10.1093/jeg/lby070
Pavlínek P., Domański B. and Guzik R. (2009) Industrial upgrading through foreign direct
investment in Central European automotive manufacturing, European Urban and Regional
Studies, 16 (1), 43-63.
Pavlínek P. and Ženka J. (2016) Value creation and value capture in the automotive industry:
empirical evidence from Czechia, Environment and Planning A: Economy and Space, 48 (5),
937-959.
Rodak A. (2019) Cztery ważne pytania o Przemysł 4.0, na które menedżer powinien znać
odpowiedź, 30 August 2019. https://www.astor.com.pl/o-nas/artykuly/biznes/10125-4-
wazne-pytania-o-przemysl-4-0-na-ktore-menedzer-powinien-znac-odpowiedz.html
Roland Berger (2016) The Industrie 4.0 transition quantified: how the fourth industrial revolution is
reshuffling the economic, social and industrial model, Munich, Roland Berger GmbH.
https://www.rolandberger.com/en/Publications/The-Industrie-4.0-transition-quantified.html
Rojko A. (2017) Industry 4.0 concept: background and overview, International Journal of Interactive
Mobile Technologies, 11 (5), 77-90. https://doi.org/10.3991/ijim.v11i5.7072
Szalavetz A. (2012) The Hungarian automotive sector: a comparative CEE perspective with special
emphasis on structural change, in Welfens P. (ed.) Clusters in automotive and information &
communication technology, Berlin, Springer, 241-270.
Szalavetz A. (2019) Industry 4.0 and capability development in manufacturing subsidiaries,
Technological Forecasting and Social Change, 145 (C), 384-395.
The barometer of the automotive parts production sector in Poland, research conducted by the
SDCM and Frost & Sullivan Q1 2018. [Quoted in Automotive Sector in Poland 2019. Figures,
Summaries, Analyses, SDCM, Warsaw; 2018 JETRO Survey on Business Conditions of Japanese
Companies in Europe].

Chapter 6
Digital entrepreneurs in factory economies:
Evidence from the automotive industry in Hungary
Andrea Szalavetz
1. Introduction
Digital entrepreneurs1 are perceived as being innovative in the Schumpeterian (1934)
sense. Their offerings rely on, embody or are embodied in digital technologies (Lyytinen
et al. 2016) that are bringing about a multiplicity of new product-service combinations
and revolutionising the patterns of value-adding activities. Consequently, digital
entrepreneurs are considered to have a transformative impact. Their activities disrupt
some industries, rendering them obsolete, create new ones and transform the business
practices and models of actors in related industries (Vial 2019). Note that, since digital
technologies are general purpose ones, practically all industries are ‘related’.
Digital entrepreneurial ventures have a large potential impact not only in a technological
sense but also in an economic one: their high growth potential is demonstrated by the
rapidly-growing number of business unicorns based on digital technologies.2 Given this
double impact, it is no surprise that digital entrepreneurship is currently deemed of
paramount importance to economic development (Nambisan et al. 2019).
Digitalisation is expected to herald a new era in entrepreneurship (Nambisan 2017), not

only in advanced economies, although the development benefits of digital technologies
are not evenly distributed (World Bank 2016). Yet, digital entrepreneurship may become
a new, qualitative source of economic growth, intensifying the catching-up of countries
that are prepared to exploit the much-praised capacity of digital technologies, namely
that they ‘democratise innovation and entrepreneurship’ (e.g. Aldrich 2014; Nambisan
2017).3
However, in line with the scholarship which posits that not all entrepreneurs are equal
(e.g. Henrekson and Sanandaji 2019; Lafuente et al. 2019) and, furthermore, that there
are non-negligible differences among digital entrepreneurs themselves (e.g. Sussan
and Ács 2017; von Briel et al. 2018), it is essential to explore the features of digital
entrepreneurs outside the centres of digital technology production. Uncovering the
differences between advanced economies and less developed ones in the features and
prospects of digital entrepreneurs may extend our understanding of the differences in
the potential of these agents to become levers of growth and upgrading.
1. Digital entrepreneurs are considered in this chapter in the narrow sense of ‘digital technology entrepreneurs’
(Giones and Brem 2017).
2. ’Unicorns’ denote companies valued at $1bn or more: https://www.cbinsights.com/research-unicorn-
companies.
3. For a review and a comprehensive critique of this view, see Dy (2019).

Andrea Szalavetz
Complementing a large body of studies focusing on the nature and implications of

digital entrepreneurship in advanced and in high-performing emerging economies (e.g.
China), there is an emerging literature analysing the features and the practices of digital
entrepreneurs in economic peripheries, in particular in Africa (e.g. Graham 2019).
By contrast, there is scarce empirical evidence on the specifics of digital entrepreneurs

in central and eastern European dependent market economies (CEE).4
The purpose of this chapter is to address this gap by drawing on insights gathered from
interviews with twelve Hungarian digital entrepreneurs operating in the automotive
technology ecosystem. We analyse the particularities of digital entrepreneurs in CEE;
that is, whether the surveyed companies display the features described in the academic
literature on digital entrepreneurs. This allows for a consideration of the impact of
digital entrepreneurship on the dependent position of the region; specifically, whether
these important agents of innovation represent a strategic opportunity to shift CEE
economies to a relatively higher-road trajectory of economic development. Can digital
entrepreneurs enable these countries to break out of the dependent model?
The rest of this chapter is structured as follows. The introductory section is followed
by a brief review of the literature on the specific features of digital entrepreneurs.
Subsequently, the method of empirical data collection is outlined and the empirical
findings presented. The final section discusses the findings and concludes with some
propositions regarding the ways of interpreting and improving the developmental
outcomes of these particular species of companies.
2. Digital entrepreneurs: a particular species driving high-road

development
Digital entrepreneurship is defined as the setting up of entrepreneurial ventures with
offerings (products, services or product-service systems) that embody, or are embodied
in or enabled by, digital technologies (Lyytinen et al. 2016). Prior research associates
digital technology-based new ventures with knowledge-intensive, Schumpeterian
entrepreneurship, and postulates that these companies have a high growth potential
(Henrekson and Sanandaji 2019; Huang et al. 2017; Lassen et al. 2018). The activity of
digital entrepreneurs is expected to bring about meaningful economic gains in terms of
innovation, productivity, growth and employment (Lafuente et al. 2019).
Scholarly analyses list a number of additional distinctive characteristics that apply to

digital entrepreneurs (Figure 1).
Besides the two most common catchwords (Schumpeterian and disruptive) referring to
their innovativeness, important distinctive features of digital entrepreneurs include a
4. A notable exception is Skala (2019). See also a companion paper prepared in the framework of this project
(Szalavetz 2020).

Digital entrepreneurs in factory economies: Evidence from the automotive industry in Hungary
‘lean start-up’ mode of market entry5 (Blank 2013; Ries 2011) and a higher than average
speed of scaling-up (Autio and Cao 2019; Huang et al. 2017). In Nambisan’s (2017:
1035) wording, digital technologies allow entrepreneurial processes to ‘unfold in a non-
linear fashion across time and space’ (italics added).
Figure 1 Characteristics of digital entrepreneurs
Industry Lean
Disruptive agnostic start-up
Easy Short time

market to market
Innovative entry
Participate
in a digital Digital entrepreneurs Rapid
ecosystem scale-up
Rapid
Lack of employment
hierarchical growth
Rapid
governance revenue
Easy access growth
to finance
Relational Internationa-
governance lisation
Source: author’s compilation based on her survey of the literature
Since digital technologies allow for low-cost experimentation with entrepreneurial

ideas, entry barriers are lower and market entry is easier while the time to market is
shorter for digital than for conventional entrepreneurs (Autio and Cao 2019; Nambisan
2017).
Digital entrepreneurs’ rapid internationalisation is facilitated by digital technologies

themselves. Digital infrastructures and platforms bridge distance and enable larger
than average market reach. Moreover, if the number and needs of users or customers
escalate, these can be met without adding proportionately more resources (Zhang et
al. 2015). Consequently, the value created and appropriated by entrepreneurs can
grow rapidly – this is referred to by Nambisan (2017) as the non-linearity of digital
entrepreneurs’ growth.
Scaling-up is also enabled by digital entrepreneurs’ relatively easy access to finance.

It is claimed that digital entrepreneurs are able to overcome resource constraints and
5. Instead of entering the market with a product deemed ‘perfect’, as a result of large-scale upfront development,
lean start-ups would launch ‘minimum viable products’ or offerings that are intentionally incomplete (Nambisan
2017), relying on customers’ feedback for further development.

Andrea Szalavetz
obtain funding for their expansion relatively easily, for two reasons: firstly, because they
are able to harness digital technologies that reduce the information asymmetries which
hinder conventional lending processes (Estrin et al. 2018); and, secondly, because
they are major beneficiaries of the intensifying interest of ‘BigTech’ companies (the
best capitalised, largest technology companies) in financial services provision (Frost
et al. 2019), and/or are recipients of corporate venture capital investment by large,
established, non-digital firms trying to integrate digital offerings in their core products.6
Digital entrepreneurs are considered industry agnostic (Autio and Cao 2019), targeting
customers in virtually any sector. This substantiates the claim that digitalisation has
transformed the nature and degree of openness in innovation and entrepreneurship
(Nambisan et al. 2019). Compared to conventional start-ups, it is easier for digital
entrepreneurs to acquire large established companies as customers, since these latter
need to adapt to the ‘digitalisation imperative’ to streamline their operations, improve
their processes and create new business models (Crittenden et al. 2019). Additionally,
digital entrepreneurs can benefit from strong public incentives supporting their growth,
among others by subsidies for the adoption of new digital solutions.
Over and above being integrated in particular value chains, the business environment
for digital entrepreneurs can rather be described as a digital ecosystem, i.e. a network of
interdependent and collaborating organisations that use digital infrastructure to create
value jointly (Sussan and Ács 2017; Valdez-de-Leon 2019).
Another noteworthy feature characterising digital entrepreneurs is that their inter-

organisational exchanges are characterised either by relational governance based
on trust, collaborative problem solving and information sharing (Gereffi et al. 2005);
or by ecosystem governance in which the rules of participation and the distribution
of revenues among the partners are clearly established.7 Compared to the captive or
hierarchical governance modes characterising the transactions of physical product
suppliers or manufacturing subsidiaries in factory economies, this feature suggests
that local digital entrepreneurs rely on a high level of technological knowledge for
their integration in global value chains and that their contribution involves knowledge-
intensive, high value-adding activities.
3. Research design, data collection and analysis

Since digital entrepreneurship by domestic-owned actors in factory economies is a
nearly uncharted territory of academic research (Szerb et al. 2018), this chapter employs
an exploratory research design, based on corporate interviews, to obtain insights on the
ways digital entrepreneurs exploit the specifics of cyber technologies (Eisenhardt 1989).
6. For example, Sandler (2017) provides a survey of the top venture capital investment providers in the automotive
technology sector and shows that there are several established OEMs among them.
7. Being embedded in digital ecosystems, i.e. in loose networks of digitally connected and interacting organisations
that are not managed by a hierarchical authority (Valdez-de-Leon 2019), characterises an increasing number
of digital entrepreneurs.

Articles in the business press and reports by management consultancy firms abound
in success stories describing the evolution of some highly-valued digital empires.
Although the Global Unicorn Club contains barely any companies from peripheral
factory economies,8 local observers, also in ‘low/moderate performer’ dependent market
economies, find it relatively easy to identify a couple of local high-flying, entrepreneurial
companies specialised in today’s paradigm-changing, digital technologies.
The context of this study is Hungary, a typical dependent market economy (Farkas
2011, 2016) in which both innovation performance (European Innovation Scoreboard
2018) and business digitalisation performance are particularly weak.9
The sample was selected on the basis of two criteria. The selected companies were:
(1) domestic-owned entrepreneurial ventures specialised in the provision of digital
solutions; and (2) involved in supplying automotive companies. The context of one
single industry, the digital automotive technology ecosystem, was selected in an effort
to homogenise the sample – at least partially. The automotive industry proved to be a
good choice, since the digital intensity of value-adding activities is among the highest
in automotive value chains (Calvino et al. 2018). Furthermore, given Hungary’s strong
specialisation in this industry10 and the dominance of foreign-owned manufacturing
units, this industry accordingly exemplifies Hungary’s dependent market economy
status and its exposure to developments in the automotive industry and to the strategic
decisions of lead companies.
The method of purposeful sampling (Patton 1990) has been applied and companies
whose cases seemed promisingly information-rich were chosen. This was made possible
by the author’s database of a collection of business press and technology press articles
describing the achievements of Hungarian companies in terms of digital transformation
and digital innovation.11
Twelve domestic-owned entrepreneurial ventures were interviewed between January

and April 2019. Interviews lasted 90 minutes on average and were guided by an interview
protocol consisting mainly of open-ended questions to facilitate exploration. The
questions were organised around three topics: the history of the venture; its business
strategy; and the factors enabling its integration in highly- concentrated automotive
value chains.
8. In August 2019 the ’Club’ had 393 members, with US and Chinese unicorns accounting for the dominant
majority of listed companies. The new member states of the European Union were represented by one firm from
Estonia and one from Malta.
9. According to the business digitalisation pillar of the composite Digital Economy and Society Index, Hungary
scores the second lowest in the EU-28, ahead only of Romania (DESI 2018). Hungary’s position in international
rankings of entrepreneurial capabilities is also much lower than those of its CEE counterparts (Hungary was
50th in the 2018 edition of the Global Entrepreneurship and Development Index; in contrast, Poland was 30th,
Slovakia 36th and the Czech Republic 38th (Ács et al. 2018: 28-29).
10. This industry accounted for more than a quarter (27.1 per cent) of total manufacturing production in 2018
(source: author’s calculation from Central Statistical Office data).
11. See companion paper (Szalavetz 2020) for details.

Andrea Szalavetz
The empirical data obtained during the interviews have been analysed in two pieces
of work. The main focus of this book chapter is the specifics of the surveyed firms,
their offerings and their business strategy; while a companion paper (Szalavetz 2020)
is concerned with the factors enabling the integration of digital solution providers in
automotive value chains.
The qualitative data obtained from individual interviews have been analysed content-
wise, involving the identification of the key commonalities that facilitate interpretation.
Analysis was conducted using standard within-case and cross-case analysis techniques
(Eisenhardt 1989). We applied the constant comparative method for data analysis
(Glaser 1965), collecting and analysing data simultaneously. This allowed us to cross-
check the emerging patterns in subsequent interviews and/or contrast interviewees’
remarks with those gained in prior interviews.
4. Results
To set the context, we first asked about the specifics of the surveyed firm’s products
and/or solutions. We asked our interviewees to recount the history and how their
offerings had been developed. The interviews had been preceded by the compilation of
secondary source data (press releases and business press articles about the company,
public profit and loss accounts and notes to the financial statement). These documents
disclosed important basic data on the firms in question and were useful also in terms
of triangulating interview information. The basic data of the surveyed firms are
summarised in Table 1.
The detailed descriptions in Table 1 highlight that the offerings of the sample companies
show great diversity, reflecting the multiplicity of entrepreneurial opportunities
stemming from conceivable product-service combinations. Notwithstanding this
diversity, some commonalities allowed for the classification of our sample companies into
two groups. Based on the accounts of the interviewees, we have grouped the solutions of
the surveyed firms into a 2x2 matrix according to hardware/software intensity and the
customer specificity of the given solution (Figure 2). Hardware-intensity is obviously
considered in a relative sense since the solutions of all companies are highly software-
intensive or, in a broader sense, intangibles- and knowledge-intensive.
Figure 2 reflects that the distribution of the sample is skewed, since the dominant
majority of the companies create and deliver custom-tailored digital solutions (industrial
cyber-physical product-service systems). These technology providers integrate digital
technologies in customers’ production/business systems to enhance the efficiency of,
and the value of the data generated by, customers’ production/business processes. By
contrast, the companies in the bottom left quadrant (BLQ) offer productised (rapidly
scalable) digital solutions.

Table 1 Overview of sample firm characteristics (data for 2018)
No. Product € Employment Year of Interviewee

foundation
1 A self-driving software stack. ~5m 182* 2015 Marketing officer

A simulation solution for testing autonomous vehicles (the
purpose-built virtual representation of the environment,
allowing the recreation of problems in vehicles’
environments so that simulation and validation exercises
can be carried out).
A power-efficient hardware IP core to accelerate the
deployment of artificial intelligence-based self-driving
software that solves the problems associated with the
current excessively high power consumption of the
hardware that accelerates AI-based automated driving
solutions.
A highway autopilot solution for autonomous highway
driving.
2 Business intelligence: provision of big data, data 9.5m 136 2006 Communications
visualisation and analytics-based solutions of company- officer
specific problems; strategic consulting relying on data
science approaches.
3 Connected car vehicle-to-everything (V2X) solutions: a ~1m 29 2012 Technology

software stack allowing for V2X communications to be officer
integrated in on-board units or roadside units.
4 Integrated digital ergonomics system, i.e. a motion ~7k 4 2014 Managing

digitising and evaluating device that captures, measures, director
records and analyses data related to assembly workers’
motion, to be used for ergonomic analyses and testing.
5 Engineering services: ** development and implementation 3.9m 31 1990 Business unit

of production tracking systems, barcode and RFID manager
solutions for production logistics and warehousing, self-
developed real-time location system.
6 Engineering services:** development and implementation 766k 10 2015 Founder

of visual inspection solutions (camera-based or 3D
scanning-based) for quality control in manufacturing
production; industrial software development e.g.
traceability systems and MES.
7 Immersive virtual reality system, i.e. a 3D educational and 101k 2 2013 Founder
virtual collaboration platform to be used (among others)
by students specialised in automotive engineering or to
be applied for training new employees in automotive
companies. Furthermore, this platform integrates various
online collaborative tools connecting multiple users: used
for example in new product development.
8 Development, manufacturing, deployment and 7.1m 51 2002 Founder

commissioning of custom-tailored production machinery
combined with smart solutions.
Analysis and solution of specific technological problems
related to customers’ product and process development
and engineering activities.
R&D in the field of simulation methods and finite element
analysis.

Andrea Szalavetz
No. Product € Employment Year of Interviewee

foundation
9 Engineering services:** development and deployment 5.5m 46 1991 Business

of cyber-physical production systems (CPPS), robotic development
systems integration, development of CPPS-based manager
functional solutions (e.g. quality control, process
automation, production monitoring and optimisation,
etc.). R&D on collaborative robots and the development of
demonstration use cases for collaborative robots.
10 Conceptual design and implementation of customised 508k 14 2012 Founder

special purpose machinery for factory automation;
systems integration services (robotics, computer vision,
measurement systems, data acquisition and processing).
11 Design and implementation of cyber-physical systems 67k 2 2013 Founder

and analytics solutions for manufacturing companies.
Consultancy about the ways and methods of digital
transformation and implementation of smart factory
solutions. Data-driven and AI-powered business process
re-engineering and optimisation, solution of technological
problems.
12 Industrial Internet of Things (IIoT) platform for smart ~25k 10 2017 Founder
factories, based on big data technologies and machine
learning. The platform is capable of implementing
machine learning-powered process optimisation. The
platform supports smart factory applications. Design and
implementation of smart factory solutions on the basis of
this platform.
€ = net sales in EUR (the exchange rate used for conversion from HUF was 319); k = thousand, m = million, employment = number of
employees, MES = manufacturing execution system.
* In addition to 182 employees in Hungary, the company has dozens of employees abroad.
** Engineering services include assessment of the customer’s processes; identification of bottlenecks; conceptual design of a solution;
procurement, deployment, installation (commissioning) and, in some cases, the servicing and maintenance of system-specific hardware
e.g. machinery or track and tracing infrastructure, cameras, sensors or other data capture tools, user interfaces and other system
components; together with the development and deployment of the related software e.g. reporting algorithms, mobile applications and
system integration services.
Figure 2 The classification of sample companies’ products & solutions
Custom-tailored
Productised solutions
digital solutions
Hardware-intensive 4, 5, 6, 8, 9, 10
Software-intensive 1, 3, 7 2, 11, 12
Source: elaborated by the author, based on information from the interviews
Although it is challenging even for technical experts to determine the technological

novelty of specific solutions, in order to guide our analysis we have grouped the solutions
of the surveyed firms also according to the novelty of the technology (Figure 3). In

categorising individual solutions, we relied both on the opinion of the managers

interviewed and the concepts of technological novelty outlined in the literature.12
On the one hand, Figure 3 confirms the claim that (most) digital entrepreneurs are
industry agnostic: their solutions can be used by customers in any sector (Autio and
Cao 2019). The customer portfolios of most of the surveyed firms are not limited to
automotive industry actors; nevertheless, automotive companies represent a large
share of their customers. This demonstrates the pioneering status of the automotive
industry in the field of digital transformation.
Figure 3 Some additional features of sample companies’ products & solutions13

Novelty and science-intensity of the technology
High Medium-high
Established, rapidly
Nascent digital technology
maturing digital technology
High
Industry specificity
Automotive 1, 3
industry-specific
Broad-based 4 5, 6, 8, 9,
(can be applied also in
12 8 7 2 10, 11,12
other industries)
Low
On the other hand, Figure 3 also suggests that the offerings of the majority of firms in
the sample are neither disruptive nor radical innovations based on nascent technology.
The solutions of firms in the bottom right quadrant (BRQ) of the matrix rely on already-
existing, and rapidly maturing, digital technologies, e.g. cyber-physical systems, factory
12. In order to determine novelty, Abernathy and Clark (1985), for example, analyse the capacity of an innovation
to influence the established production system and customer base, classifying innovations as incremental or
radical. Radical innovations make existing production systems obsolete, destroy the value of existing expertise,
demand new procedures and/or create new markets. In a similar vein, Tushman and Anderson (1986) classify
technologies as competence-enhancing or competence-destroying – the latter is characterised by a higher degree
of novelty. Other scholars in the innovation literature rely on concepts of (a) technological uncertainty, e.g.
regarding the means to accomplish certain tasks (e.g. Fleming 2001); or (b) familiarity and previous experience
with the product and process technologies employed to create the desired new product or solution. In this latter
sense, Henderson and Clark (1990) consider a technological invention radically new if, compared to existing
technologies/solutions, it is based on different scientific and engineering principles.
13. Companies 8 and 12 are represented in multiple quadrants. This refers to different products/activities. For
example, besides designing and implementing custom-tailored and smart solutions embedded in special
machinery, No. 8 is also engaged in the solution of product development-related technical problems and
conducts basic research to develop material science-specific simulation methods – used by global automotive
companies aiming at reducing the weight of selected components. No. 12 is specialised in basic research-
intensive IIoT development, which is represented in the bottom left quadrant of the matrix. Additionally,
however, it designs and implements industry 4.0 projects for Hungary-based manufacturing companies (mainly
automotive ones). This latter activity is classified in the bottom right quadrant of the matrix. Note that custom-
tailored individual solutions of companies in the bottom right quadrant are highly heterogeneous also in terms
of the technological and R&D capabilities required to design and implement the given solution.

Andrea Szalavetz
automation, simulations, digital twins and analytics. These technologies are applied in
company-specific combinations and enable adopters’ digital transformation to achieve
improvements in their existing production systems and/or solve particular technological
or business problems.
Irrespective of the deployment of smart factory-specific digital technologies requiring

extensive software development and systems integration capabilities, these solutions no
longer convey nascent technologies. Smart factory-specific or ‘industry 4.0’ solutions are
becoming more and more mature and established. Considering that the term ‘industry
4.0’ was officially introduced less than a decade ago, in 2011, at the Hannover trade
show, this reflects the acceleration of technology innovation cycles.
As the following interview excerpt demonstrates, the entrepreneurial strategies and

practices of BRQ companies have not changed, they have simply grown digital.
‘The activities we perform have not changed radically; we simply integrated digital
technologies both in our activities and in our offerings.’ (No. 5, 6, 9)
Notwithstanding that these offerings have no ‘transformative’ impact, i.e. they are
not expected to bring about creative destruction, they are evidently innovative in
a Schumpeterian sense, representing ‘new product-service combinations’ and/
or ‘reform[ing] or revolutioni[sing] the pattern of production by exploiting […] an
untried technological possibility for producing […] existing commodities in a new way’
(Schumpeter 1943: 132).
In contrast, the offerings of companies in the ‘nascent technology’ column can be

regarded as radical novelties. Interview data confirm that these born digital companies
introduced their offerings in the market as lean enterprise-specific, incomplete,
‘minimum viable products’ (Ries 2011). The solutions of this group of companies are
at different stages of R&D and commercialisation, and all have undergone continuous
evolution ever since the first versions were introduced. Although the managers
interviewed (No. 1, 3, 4, 7 and 12) have all underscored that their offerings require
several years of further development, the ‘still incomplete’ products of these companies
are generating, in some cases, revenues that are already non-negligible.
Investigating the association between the novelty of the technology and business
performance, our data indicate that there is no meaningful relationship between these
variables (Figure 4). For example, although the offerings of companies in the BLQ of the
matrix (No. 4, 7 and 12 – the IIoT platform in this latter case) represent radical novelty,
the impact of these companies in terms of revenues is lower than that of companies in
the BRQ.14
14. Note that a simple comparison of turnover data without considering the cost of goods sold may provide a
distorted picture. This item may be quite large in the case of companies supplying smart factory solutions
together with systems integration services since it may include purchased special purpose machinery.

Figure 4 Turnover and employment in sample companies

Technological novelty
High Medium-high
Established,
Nascent
maturing
High
Automotive 5m (182)
specific 1m (30)
Specificity
44k 10m 2,5m

General
(5) (136) (23)
Low
Note: data in circles are averages for respective groups
More importantly, the growth performance of nascent technology companies does not
unambiguously validate the assumption that digital entrepreneurial ventures have a
high growth potential. The level of ‘adequate’ performance in terms of revenues can
barely be determined in the case of nascent technology companies, whose offerings
represent radical novelty, while it is also hard to fathom how long it takes to reach the
tipping point after which sales performance ‘explodes’ – this is highly heterogeneous
across digital entrepreneurs. Nevertheless, it is clear that the turnover data of companies
4, 7 and 12 (this latter is a very young company) leave a lot to be desired.15
One reason for their failure to scale is that they have not been able to overcome the usual
financial constraints faced by entrepreneurs. Although several companies obtained
either venture capital investment or research grants, the managers interviewed
considered the low level of external funding as one of their main obstacles to growth.
Companies 1, 2 and 3 are the only ones to represent a textbook case of Schumpeterian,
high-impact, rapidly-growing digital entrepreneurs specialised in nascent technology
and offering born-global products.
As for entrepreneurs specialised in hard-to-scale, custom-tailored digital solutions,

the main determinant of growth is, in principle, their business development capability.
However, as the following interview excerpt illustrates, business development was not
an issue for companies in the right-hand column of Figure 2 since demand for their
offerings was growing rapidly.
‘There is such a high demand for our specialised expertise in digital engineering
services provision that we do not have to make substantial investments in business
development – we have more assignments than what we can reliably accomplish.’
(No. 8)
15. This finding is consistent with the literature on business gazelles and high-impact firms (e.g. Ács 2011), positing
that the average high-impact firm is not a new start-up.

Andrea Szalavetz
Nevertheless, neither have these companies experienced rapid growth: in their cases,
growth has been rather moderate, albeit sustained. The main bottleneck limiting growth
in this latter group is the lack of skilled software developers and engineers, exacerbated
by fierce competition for talent both from the better-capitalised local subsidiaries of
global companies and from foreign labour markets.
The employment data of companies in the sample also seem disappointing, especially in
the light of the literature emphasising the strong positive impact of entrepreneurship on
job creation (see survey in Haltiwanger et al. 2013). The companies in the sample had,
on average, been operating for ten years in 2019; even so, only two of them have more
than 100 employees. The average number of employees is 43 across the sample and,
without the two outliers, it is only twenty.16
Interviews reveal that the market orientation of the surveyed firms is closely related
to the specifics of their offerings. The providers of production-related digital services
or product-service systems have not internationalised:17 they have remained local,
targeting Hungary-based manufacturing firms that were, in most cases, the local
subsidiaries of global companies.18 Companies 2 and 4, and those in the BLQ of
Figure 2 offering productised solutions, are predominantly export-oriented. Some of
the exporting companies have even established sales offices in Silicon Valley and in
emerging Asian economies.
Regarding the governance modes characterising the transactions of the surveyed

companies, our interview information confirms the prevalence of relational governance.
Relational governance is justified in cases where the planning and implementation
of custom-tailored digital solutions require close collaboration between technology
providers and adopters. This collaboration is based on trust and the sharing of knowledge
between the two parties. Solution provision is not a one-off activity: the technology
providers demonstrate their capabilities, build trust and accumulate knowledge about
customers’ problems in the course of the initial projects. Subsequent assignments by
the same contractors are usually broader and deeper. Another explanatory factor of the
prevailing mode of governance is the uniqueness of knowledge, precluding price-based
competition and hierarchical governance.
‘It’s a kind of joint experimentation with our main customer to improve our
offering further. It is not a market-based transaction where price matters.’ (No. 4)
‘It is not the price of our services that matters. What matters is achieving the trust
of prospective customers so that they believe in our capabilities, that we can solve
16. Note that, instead of hiring new employees, several small companies (with fewer than ten employees) would,
from time to time, resort to independent contractors (freelance software developers) providing software
development services to accomplish specific projects. They would do so because orders were volatile.
Consequently, company-level employment data do not precisely reflect the real employment impact of these
ventures.
17. This finding is consistent with the Polish experience; see the chapter in this book by Gwosdz et al. (2020).
18. Company No. 2 is an exception: it offers business intelligence services, supporting business management rather
than solving production-related technological problems. Its customers are mainly international, including some
Fortune 500 companies.

their problems.’ (No. 2); ‘[It’s not the price of our services but] what matters is
being involved in internationally-funded research projects.’ (No. 9)
Ecosystem governance was relevant in the cases of two companies in the sample (No. 1
and 3) and, occasionally (i.e. in some projects), also for Nos. 8, 9 and 12.
‘We collaborate with our future customers in a number of research and

demonstration projects funded by foreign stakeholders, research funds, local
municipalities or EU-programmes. A non-negligible share of our revenues stems
from these collaborations. You see, our competitiveness is based on the reputation
we have built so far. Our [ecosystem] partners trust that we are able to contribute.’
(Nos. 1 and 3)
5. Discussion and policy implications

From these results, we can conclude that the specifics of the surveyed digital
entrepreneurs do not fully and unambiguously conform to those described in the
literature (Figure 5).
Figure 5 Results
ü
Industry
ü
Lean
Disruptive agnostic
ü start-up
Easy Short time

market to market
Innovative entry
ü
Participate
in a digital Digital entrepreneurs Rapid
ecosystem in CEE scale-up
Rapid
employment
Lack of
hierarchical
ü growth
Rapid
governance revenue
Easy access growth
to finance
Relational Internationa-
governance lisation
ü
Indeed, the offerings of most of the surveyed companies prove to be industry agnostic
while the governance mode characterising their transactions is relational or ecosystem-
based, not hierarchical. Half of the firms in the sample have, indeed, introduced ‘still

Andrea Szalavetz
incomplete’ products, to be further developed according to customers’ feedback, which

confirmed the lean enterprise-specific mode of digital entrepreneurs’ market entry. On
the other hand, the custom-tailored solutions offered by the other half of the sample
have also attained ready-to-launch form following an iterative process of joint fine-
tuning by teams in both the vendor and the customer. In that sense, the examples of
the surveyed firms would all confirm the ‘lean start-up’ feature characterising digital
entrepreneurs.
However, although the companies in the sample are all innovative in a Schumpeterian
sense, their offerings were disruptive in few cases. Instead of a ‘transformative impact’,
the solutions of companies in the right-hand column of Figure 2 have enabled adopters
to perform their traditional core activities more efficiently than previously.
Instead of explosive growth, most companies have experienced only a more or less
modest increase in revenues and employment. For most, access to finance has proven
to be one of the key obstacles to scaling-up.
Furthermore, contrary to the alleged rapid internationalisation of digital entrepreneurs,

the majority of the surveyed companies – those in the BRQ of Figure 3 – have remained
local.
Most of the differences we identified are related to the specifics of the offerings. Note
that companies with productised offerings were under-represented in the sample while
the providers of customised digital solutions for manufacturing plants were over-
represented. Further research is required to determine whether the distribution of
digital entrepreneurs is significantly different in dependent market economies from that
of advanced economies, i.e. in terms of a higher than average share of entrepreneurs
offering production-related digital solutions to the local manufacturing subsidiaries of
global companies. Intuition suggests that this is the case; however, the small size of the
sample does not allow for general conclusions in this respect.
From another perspective, it is obvious that, in a country where innovation and business
digitalisation performance are weak, and labour productivity and entrepreneurial
performance low, all kinds of digital entrepreneurs matter – not only the high-impact
ones that display explosive growth. Whether their products are disruptive or not,
digital entrepreneurs play a crucial role in improving these performance indicators.
They contribute to the upgrading of local technology, since the adoption of digital
technologies improves adopters’ productivity and competitiveness. Consequently,
all kinds of digital entrepreneurs – not only high-growth ventures with disruptive
offerings based on radical innovation – can assist dependent market economies’ efforts
to progress towards a high-road trajectory of economic development. The surveyed
companies should be acknowledged as drivers of productivity- and innovation-driven,
high local value-added, qualitative development.
Nevertheless, in the dependent market economies of CEE, the extent to which digital
entrepreneurs generate economic gains for their countries of origin is dwarfed by that
of efficiency-seeking foreign direct investment in export-oriented manufacturing. For

example, the performance of even some high-flying companies in the sample appears
insignificant in comparison with that of traditional automotive subsidiaries.19
Altogether, local digital entrepreneurs are, currently, barely able to improve the
dependent position of CEE economies: their number and economic impact are too small
to bring about the required qualitative shift in the development trajectories of these
countries. Digital entrepreneurship could become a statistically more significant source
of GDP growth only where two conditions are fulfilled. On the one hand, a critical mass
of digital entrepreneurs is indispensable: their number needs to increase rapidly. On
the other, digital entrepreneurs need to be able to access the inputs necessary for their
growth in terms of finance, business development know-how and adequately skilled
labour.
Our results call for a fostering of digital entrepreneurship, as an avenue to qualitative

economic development and upgrading. This demands no radical policy innovations:
traditional policy instruments20 are required, promoting the accumulation of digital
competencies and subsidising investments that increase companies’ digital maturity.
This latter promises to kill two birds with one stone: in addition to improving technology
adopters’ total factor productivity, it also offers new business opportunities to local
digital entrepreneurs.
References
Abernathy W.J. and Clark K.B. (1985) Innovation: mapping the winds of creative destruction,
Research Policy, 14 (1), 3-22.
Ács Z.J. (2011) High-impact firms: gazelles revisited, in Fritsch M. (ed.) Handbook of research on
entrepreneurship and regional development: national and regional perspectives, Cheltenham,
Edward Elgar, 133-174.
Ács Z.J., Szerb L., Lafuente E. and Lloyd A. (2018) Global entrepreneurship and development index:
2018, Cham, Springer.
Autio E. and Cao Z. (2019) Fostering digital start-ups: structural model of entrepreneurial
ecosystems, Proceedings of the 52nd Hawaii International Conference on System Sciences.
https://scholarspace.manoa.hawaii.edu/bitstream/10125/59979/0539.pdf
Blank S. (2013) Why the lean start-up changes everything, Harvard Business Review, 91 (5), 63-72.
Branstetter L.G., Glennon B. and Jensen J.B. (2019) The IT revolution and the globalization of R&D,
Innovation Policy and the Economy, 19 (1), 1-37.
Calvino F., Criscuolo C., Marcolin L. and Squicciarini M. (2018) A taxonomy of digital intensive
sectors, OECD Science, Technology, and Industry Working Papers 2018/14, Paris, OECD
Publishing.
19. In 2018, the turnover of Magyar Suzuki was higher than that of No. 3 by a factor of 2,000, and that of a relatively
smaller local automotive subsidiary, Hankook Tire, by a factor of 700. (Source: author’s calculation from data of
the Top 50 Hungarian non-financial companies, HVG, 25th July 2019.)
20. At the same time, the experiences of the surveyed companies also call for new policy mechanisms, e.g.
accelerator programmes that foster the scaling-up and the internationalisation of digital ventures offering
productised solutions.

Andrea Szalavetz
Crittenden A.B., Crittenden V.L. and Crittenden W.F. (2019) The digitalization triumvirate: how
incumbents survive, Business Horizons, 62 (2), 259-266.
DESI (2018) Digital Economy and Society Index report 2018: integration of digital technologies.
https://ec.europa.eu/digital-single-market/en/desi
Dy A.M. (2019) Levelling the playing field? Towards a critical-social perspective on digital
entrepreneurship, Futures. https://doi.org/10.1016/j.futures.2019.102438
Eisenhardt K.M. (1989) Building theories from case study research, Academy of Management
Review, 14 (4), 532-550.
Estrin S., Gozman D. and Khavul S. (2018) The evolution and adoption of equity crowdfunding:
entrepreneur and investor entry into a new market, Small Business Economics, 51 (2), 425-439.
European Innovation Scoreboard (2018) https://ec.europa.eu/docsroom/documents/30684
Farkas B. (2011) The Central and Eastern European model of capitalism, Post-Communist
Economies, 23 (1), 15-34.
Farkas B. (2016) Models of capitalism in the European Union: post-crisis perspectives, London,
Palgrave Macmillan.
Fleming L. (2001) Recombinant uncertainty in technological search, Management Science, 47 (1),
117-132.
Frost J., Gambacorta L., Huang Y., Shin H.S. and Zbinden P. (2019) BigTech and the changing
structure of financial intermediation, BIS Working Papers 779, Basel, Bank of International
Settlements.
Giones F. and Brem A. (2017) Digital technology entrepreneurship: a definition and research
agenda, Technology Innovation Management Review, 7 (5), 44-51.
Glaser B.G. (1965) The constant comparative method of qualitative analysis, Social Problems,
12 (4), 436-445.
Graham M. (ed.) (2019) Digital economies at global margins, Cambridge MA, MIT Press.
Haltiwanger J., Jarmin R.S. and Miranda J. (2013) Who creates jobs? Small versus large versus
young, Review of Economics and Statistics, 95 (2), 347-361.
Henderson R.M. and Clark K.B. (1990) Architectural innovation: the reconfiguration of existing
product technologies and the failure of established firms, Administrative Science Quarterly,
35 (1), 9-30.
Henrekson M. and Sanandaji T. (2019) Measuring entrepreneurship: do established metrics
capture Schumpeterian entrepreneurship?, Entrepreneurship Theory and Practice. https://doi.
org/10.1177/1042258719844500
Huang J., Henfridsson O., Liu M.J. and Newell S. (2017) Growing on steroids: rapidly scaling the
user base of digital ventures through digital innovation, MIS Quarterly, 41 (1), 301-314.
Lafuente E., Ács Z.J., Sanders M. and Szerb L. (2019) The global technology frontier: productivity
growth and the relevance of Kirznerian and Schumpeterian entrepreneurship, Small Business
Economics. https://doi.org/10.1007/s11187-019-00140-1
Lassen A.H., Ljungberg D. and McKelvey M. (2018) Digital disruptors: on the potentials and
characteristics of digital knowledge intensive entrepreneurial ventures, in International Society
for Professional Innovation Management (ed.) Proceedings ‘ISPIM Innovation Conference:
innovation, the name of the game’, 1-14.
Lyytinen K., Yoo Y. and Boland Jr. R.J. (2016) Digital product innovation within four classes of
innovation networks, Information Systems Journal, 26 (1), 47-75.

Nambisan S. (2017) Digital entrepreneurship: toward a digital technology perspective of

entrepreneurship, Entrepreneurship Theory and Practice, 41 (6), 1029-1055.
Nambisan S., Wright M. and Feldman M. (2019) The digital transformation of innovation and
entrepreneurship: progress, challenges and key themes, Research Policy, 48 (8).
Patton M.Q. (1990) Qualitative evaluation and research methods, London, Sage.
Ries E. (2011) The lean startup: how today’s entrepreneurs use continuous innovation to create
radically successful business, New York, Crown Business.
Sandler E. (2017) Ten VC firms investing in the mobility of the future. https://www.
autofinancenews.net/10-vc-firms-investing-in-the-mobility-of-the-future/
Schumpeter J.A. (1934) The theory of economic development: an inquiry into profits, capital, credit,
interest, and the business cycle, Cambridge MA, Harvard University Press.
Schumpeter J.A. (1943) Capitalism, socialism & democracy, London, Routledge. https://eet.pixel-
online.org/files/etranslation/original/Schumpeter,%20Capitalism,%20Socialism%20and%20
Democracy.pdf
Skala A. (2019) Digital startups in transition economies: challenges for management,
entrepreneurship and education, Cham, Palgrave Macmillan.
Sussan F. and Ács Z.J. (2017) The digital entrepreneurial ecosystem, Small Business Economics,
49 (1), 55-73.
Szalavetz A. (2020) Digital transformation – enabling factory economy actors’ entrepreneurial
integration in global value chains?, Post-Communist Economies, 32 (6), 771-792.
Szerb L., Vörös Z., Páger B., Ács Z.J. and Komlósi É. (2018) Case study on the Hungarian new
tech entrepreneurial ecosystem, Project FIRES deliverable, Work Package 5.8. http://www.
projectfires.eu/wp-content/uploads/2018/02/D5-8-Case-Study-REVISED.pdf
Tushman M.L. and Anderson P. (1986) Technological discontinuities and organizational
environments, Administrative Science Quarterly, 31 (3), 439-465.
Valdez-De-Leon O. (2019) How to develop a digital ecosystem: a practical framework, Technology
Innovation Management Review, 9 (8), 43-54.
Vial G. (2019) Understanding digital transformation: a review and a research agenda, Journal of
Strategic Information Systems, 28 (2), 118-144.
von Briel F., Recker J. and Davidsson P. (2018) Not all digital venture ideas are created equal:
implications for venture creation processes, Journal of Strategic Information Systems, 27 (4),
278-295.
World Bank (2016) World development report 2016: digital dividends, Washington DC, World Bank
Publications.
Zhang J., Lichtenstein Y. and Gander J. (2015) Designing scalable digital business models, in
Baden-Fuller C. and Mangematin V. (eds.) Advances in strategic management, Bingley, Emerald
Publishing, 241-277.

Impact on workers
Chapter 7
Technological and organisational innovation under
Industry 4.0 – Impact on working conditions in the Italian
automotive supply sector
Matteo Gaddi
1. Introduction: research aims and summary of the impact on

working conditions of current technologies
This chapter focuses on the main technological and organisational changes that are
taking place in automotive supply companies located in Italy and the consequences
these are having for working conditions.
As suppliers of parts and components to original equipment manufacturers (OEMs),

these companies form part of a series of production networks which transcend national
borders. The linkage with OEMs plays a decisive role, especially from the point of view
of just-in-time supply practices: in this sense, organisational models (Lean Production
extended throughout a fragmented production chain) as well as new technologies
(Industry 4.0) are being implemented simultaneously to ensure close coordination and
synchronisation within the entire production chain. Specifically, this encompasses:
a) planning of production processes; b) transmission of production orders; and c)
control of the progress of production in order to meet delivery dates. Furthermore, the
organisation of work (i.e. working time and rhythms, workloads, etc.) is almost entirely
determined by the needs of the OEMs which are requiring their suppliers to adopt
particular organisational and technological tools.
At the same time, automotive supply companies located in Italy are subject to
competitive pressure from companies in countries with low labour costs. Or, rather,
it is the OEMs themselves who are creating this type of competition in order to lower
their supply costs, putting western plants in a position of competing with plants in low
labour cost countries. This is resulting in the adoption of production models by plants
located in Italy based on seeking to maximise profitability through sizable increases
in productivity, achieved through a particular organisation of work facilitated by
applications of new technology.
These issues – and specifically the connection between Lean Production and Industry
4.0 – are playing an important role in the development of the working conditions of
employees in supply companies. Explicitly, the result is a general intensification of work
through a significant increase in the degree of exploitation of the workforce.
In Italy, the National Industry Plan 4.0, which has provided strong tax incentives
for companies to make investments in 4.0 technologies and presented by the Italian
government in 2016, was received with great enthusiasm by the larger part of public
opinion, as well as by the majority of business organisations, political parties, etc.

Matteo Gaddi
The research programme of Fondazione Claudio Sabattini has been aimed towards
developing an understanding of the specific consequences that these innovations are
having for working conditions. This chapter presents the main results of our findings as
regards the automotive supply sector.
Our research questions are related to:
— the interweaving of Industry 4.0 technologies and Lean Production models

(Butollo et al. 2018; Sanders et al. 2016; Sony 2018; Wagner et al. 2017);
— the type of technologies, especially in information and communications
technologies (ICT), that companies in this sector are implementing (Otzemel et al.
2018; Qin et al. 2016; Thoben et al. 2016; Brettel et al. 2014; Tzafestas 2018);
— what consequences these technological and organisational innovations are having
for working conditions from the perspective of: work cadences and rhythms;
workloads; work content; human-machine relationships; and control over the
performance of work.
From the point of view of the influence on working conditions, the main results that our
research has highlighted can be set out as follows.
The companies involved in our research are paying general, and serious, attention to
investments in ICT. Enterprise resource planning (ERP), manufacturing execution
systems (MES) and internet-based forms of connectivity are widespread and being used
for the management of all aspects related to: a) the planning of activities; b) relations
between suppliers and customers; c) the planning and scheduling of different stages in
the production process; and d) monitoring and control.
There is no lack of investment in fixed capital (machinery, robots, plant, etc.); rather,
there is a new wave of automation underway in which the greater share of investment
is in products and services characterised by connectivity and ICT capabilities. Under
these investments, and within the Industry 4.0 process as a whole, companies have
the objective of increasing productivity with consequences for employment that are not
likely to translate in the near future into redundancies and lay-offs (at least, not mass
redundancies), but rather into no expansion of the workforce even though production
volumes may be increasing.
Companies’ aims here are being made possible by technologies which are able to trace
the beginning and the end of each single task: data relating to all operations within
the production process can be recorded, collected and monitored thanks to computer
systems. Furthermore, machines and plants are generating a quantity of data related
both to volumes produced, the phases carried out and the processing of each batch
and any problems that are limiting functionality and causing downtime (breakdowns,
setting-up, controls, lack of materials, etc.). In the case of manufacturing operations, ICT
tools – more specifically, the apps which are established within them – achieve this by
reading barcodes connected to work orders via optical readers, personal tablets and on-
board PCs embedded within a machine or a line. The barcodes indicate the work order
being performed, the machines and components used and the production process being

Technological and organisational innovation under Industry 4.0
carried out, and associate it with the ID of a particular operator. All the data obtained
are immediately uploaded to the company’s information system servers via ERP or
MES, and made visible in real time to the offices responsible for control and monitoring,
while the MES communicates work orders directly back to each workstation.
A count of the exact time taken by a particular task is thereby commenced in which a
company is able to develop a clear and precisely detailed understanding of the time
taken up by the work cycle – but, of course, not only that one: the ability to record
time does not only affect individual task operations but the entire production cycle
(right from the acquisition of orders to the delivery of product). These cycle times are
incorporated into work orders defined by engineering departments by means of the
ERP software that plans the production and defines the scheduling. Indeed, a strict
definition of working time is, therefore, a prerequisite for the true coordination of the
various production phases.
In this way, a cycle time can be imposed within which an operator must conclude
each particular task, taking it away from the knowledge and control of workers while,
at the same time, allowing real-time and remote control over work performance. The
constraint on workers can therefore be understood as the obligation to adapt their work
rhythms and cadences to cycle times, not only concerning the speed at which a semi-
finished product moves between work phases but right the way across each single phase
within the production process as a whole.
Thus, these technological investments possess a clear labour-saving character:

i.e. companies are benefiting from the ability to do more with less. This increase in
productivity has occurred mainly under an intensification of work cadences and a
marked reduction in the time assigned to machine operators for each task/operation
within the wider production process. In consequence, the level of ‘saturation’, i.e. the
ratio between shift working time and the quantity of work actually done during that shift,
is deteriorating significantly. Working times are becoming, in many cases, extremely
difficult for workers to meet, due also to the high degree of variability in workloads and
production mixes (Gaddi 2018, Gaddi 2019).
This intensification of workloads (and the saturation of the work process) has, in our
view, at least three causes: a) operations carried out by workers are often complementary
and subordinate to those carried out by machines (since workers are conditioned by
the cycle time of the machine); b) under the pretext of automating the toughest tasks,
workers are now in charge of operating more than one machine at the same time; and
c) workers are in charge of a number of operations – self-checking, quality control, etc.
– which were previously the responsibility of others. In the context of (a), we should
note in addition that the worker is becoming forced to act as a mere appendix of the
machine while, furthermore, the ready appearance of data on performance in control
and monitoring offices – which may be located elsewhere and even internationally – is
an additional source of pressure within the workplace.
Through these control systems, companies are able to compare internal costs with prices
from external suppliers: in this way, companies are able to achieve more objectives: to

Matteo Gaddi
calculate production costs; to calculate the cost of each worker; and to decide whether or
not to outsource a certain production phase. In this way, competition between ‘internal’
and ‘external’ workers is also being created, which also adds to the pressures being put
on employees.
To try to conceal these negative effects of a more highly saturated work process,
companies are attempting to portray work cycle times as in some way ‘objective’, or
possessed of a scientific quality determined by the technology, instead of them appearing
for what they are, i.e. the social decisions of companies. This means, moreover, that
they are being hidden from the perceptions of workers and, thereby, removed from
formal and informal negotiation.
The same things are occurring with the use of dispatch tools in the field of maintenance/
assistance/repair, etc. By combining the use of scheduling software with devices
(tablets, smartphones, etc.), operators are being provided with a list of maintenance/
assistance interventions to be carried out, including the time to be spent on them and
with the aggravating circumstance that workers’ locations are also being recorded and
controlled by the technology.
Neither is office work exempt from these operations: in addition to the classic
mechanisms for recording the start and end of operations, software is capable of tracking
the various ‘clicked’ functions and of checking for any errors, overlaps or repetitions,
activities which have no added value, etc.
For these reasons, our contention is that Industry 4.0 technologies, in guaranteeing
the minute traceability of each single operation and its progress status, are facilitating
the real-time extraction of information which allows the pervasive and real-time
control of the performance of each employee. We believe, therefore, that Industry 4.0
technologies are fulfilling the so-far incomplete manufacturing revolution developed
under Lean Production; and that it is the coming together of both which is having the
most deleterious consequences for working conditions. The ‘brilliant factory’ is just
that: a powerful device for controlling workers.
This chapter is organised as follows. Following this brief introduction and summary of
theme, section 2 summarises the application of Lean Production models and Industry
4.0 technologies within the sector. Section 3 introduces some essential characteristics
of the contemporary Italian automotive sector; and then goes on to document the field
research carried out by the Fondazione Claudio Sabattini, with a series of case studies
used specifically to illustrate various themes on the influence that these models and
technologies are having on the shape of the sector and on working conditions within it.
Finally, section 4 provides some conclusions.

2. Applications of Lean Production and Industry 4.0

Lean Production, of which the production system in Toyota is the foundation, aims to
increase productivity by eliminating waste, resulting in a tightly controlled production
flow in which all elements, including suppliers, are strictly synchronised. Overproduction
(anything which is not necessary for the following production stage) is seen as waste
and should be eliminated while waiting times need to be minimised. Lead times are
compared with production times, eliminating synchronisation errors, material delays,
sudden queues, failures, lack of operators, tooling times, etc.
Meanwhile, Industry 4.0 factories are ‘smart’ ones in which a set of technologies -
communication tools, connectivity, data collection and processing - allows to connect
work tools,equipments, plants, and products so that they can communicate directly with
each other and with centralized systems, at such a speed that they can do so continuously
and in real time: the increase in the computerization of manufacturing systems and the
use of network and ICT technologies allows to integrate and synchronize all parts of
the system in an information network (Forschungsunion 2013; European Parliament
2016).
Industry 4.0 technologies are, therefore, a perfect match for the objectives of Lean
Production: digital technologies can play a decisive role in shortening waiting times
and contributing to a reduction in plant reset times.
Lean Production systems and Industry 4.0 technologies are used widely in automotive
production, both by car manufacturers and suppliers – the latter often being compelled
by the former to introduce them. In both cases, this is translating into a heavy
intensification of the pace of work and of workers’ performance.
Our research has shown that applications of Lean Production and Industry 4.0
technologies are closely connected, even sharing the same ‘philosophy’. In our opinion,
Industry 4.0 is not fully understandable unless we also take into account that close
relationship with Lean Production. In particular, Industry 4.0 technologies facilitate
the full implementation of Lean Production, overcoming the technical constraints
that previously limited its application. Furthermore, it is the intertwining of these two
elements that is having the most critical consequences for working conditions.
Essentially, Lean Production implies the passage from a ‘push’ to a ‘pull’ logic: whereas
formerly production planning was ‘pushed’ by sales forecasts, now it is being ‘pulled’ by
customer orders – i.e. from another company, or even from another department within
the same one. Thus, it is orders which trigger the entire production chain. One of the
pillars of Lean Production is just-in-time: if the entire production process is ‘pulled’ by
customer orders, nothing must be produced upstream that is not required downstream.
Each piece must be produced at the time a downstream workstation requests it through
kanban: the visual instruments which transmit information and instructions on the
materials to be supplied by storage areas – ‘picking kanban’ – and the components to
be produced – ‘production kanban’. New technologies, when applied to the kanban
system, enable the strict synchronisation of different production phases and the

Matteo Gaddi
management of departmental and workstation demand in real time. The entire process
can be constantly monitored from the screen of any connected device.
OEMs control their suppliers even though the latter are formally responsible for their
own internal processes. This translates into OEMs having a substantial degree of control
over work performance.
The attached figure highlights the typical integration architecture adopted by companies
in the automotive sector in Italy – the so-called ‘pyramid of automation’. The first level
in this architecture is represented by enterprise resource planning, which constitutes a
set of planning tools for order acquisition and processing, supply chain management,
the management of human resources and the production capacity of plants, production
engineering, etc. The second level is the manufacturing execution system, which
performs scheduling functions, the dispatch of production orders, resource allocation,
product and workforce tracking, performance analysis, production reporting, etc. The
MES, starting from the general process planning generated by the ERP, deals more
specifically with its scheduling and with the dispatch of production orders to each
department and/or workstation. Subsequent operational levels are based on SCADA
(Supervisory Control and Data Acquisition, which monitors and supervises machinery
and devices on the factory floor), as well as technologies such as PLCs (Programmable
Logic Controllers contained on single pieces of machinery or plant) or other, similar,
tools and apps. The final level is represented by sensors and other data collection tools
in the field.
Figure 1 Automation pyramid
ERP
Management
& planning
MES
Scheduling
(productionprocesses)
SCADA/PLC
Supervision & control (operational level)
Field Devices (sensors etc.)

Data collection
Industry 4.0 technologies allow the (vertical) integration (or vertical networking) of this
architecture, reducing the number of steps between decisions and system control and,
hence, flattening the pyramid.
In fact, Industry 4.0 combines the digitalisation of manufacturing processes with real-
time data acquisition, processed and analysed via server and edge (cloud) computing as
a means of optimising industrial processes (Akerman 2018; Rojko 2017).

The whole process is started with a Data Acquisition Module, facilitating the statistical
analysis of the data that has been acquired. The different nodes of the network
(products, machinery, controllers, etc.) exchange information through technologies
developed through IoT (Internet of Things) applications. The data acquired in this way
are processed not only by cloud computing and Big Data analytics, but are also used
by CPSs – Cyber Physical Systems, i.e. virtual simulation tools for physical processes.
Thereafter, feedback is transmitted to PLCs, the MES and ERP to make production
not only flexible but also highly reconfigurable, and the latter within a very short
timescale.
This means that planning, scheduling, work order dispatch and plant operation can be
continuously redefined – compelling the workforce to adapt continuously. Integration
of all levels within CPS facilitates close cooperation between all departments and
individual phases. In this just-in-time environment, increases in productivity are
based on the strong intensification of work performance, i.e. a greater degree of labour
exploitation.
This system connects the entire production cycle right from the end to the very
beginning: i.e. backwards from the end of the assembly line – the point at which a
customer’s production orders can be dispatched to that customer – one stage after
another, workstation after workstation. From the point of view of Lean Production,
this eliminates overproduction (components are supplied in the exact number in which
they are required) and minimises waiting times (components get to each station at the
exact moment required, and without loss of time).
Industry 4.0 technologies also provide crucial support for the full implementation of
the logic of kanban, deployments of which are becoming progressively electronic and
which make sending, receiving, recording, etc. both easier and faster. Starting from
general production planning via ERP, and hence from production scheduling by times
and workstations via MES, electronic kanbans can be generated and transmitted to
connected devices embedded within each workstation. When the requests of a kanban
have been met, the electronic recording system shows the progress in order to allow
management to monitor it in real time and step in immediately when synchronisation
adjustments are needed.
Respect for assigned times is central to Lean Production, in which the pace of production
is determined by ‘takt time’ – the time taken up by the production of one unit of a
specific output. Industry 4.0 technologies allow the real-time monitoring of takt times
through connected recording devices which immediately upload data to company
information systems which compare actual and planned times. In this way, takt time
determines working times across all lines and at each workstation, imposing rhythms
and working systems to meet the standards set by the company. They also facilitate the
levelling of workstation saturation, or heijunka: once takt times are defined, workloads
to be allocated to the various workstations are computed automatically, taking into
account the availability of facilities and staff.

Matteo Gaddi
Strict adherence to takt time and the obsession of Lean systems with reducing lead
times means that tools which are presented as aids for workers – in carrying out their
assigned tasks, they do not have to ‘waste time’ in thinking, checking, verifying, etc.
– or otherwise as ways of reducing operator anxiety about possible errors are thereby
transformed into tools for intensifying work performance. The implications of poka-
yoke (‘foolproof’) systems, which provide detailed instructions to each workstation,
or automatically assign combinations of batch-machine-component codes, renders
workers less and less autonomous: MES communicates with all the connected machines
and guides workers through the various operations, indicating the components to be
used and the sequence of operations to be carried out.
The integration between Lean Production and Industry 4.0 is also very important in
terms of logistics: lines are automatically supplied by electronic kanbans, with products
delivered both to the warehouse as well as to external suppliers via the same software
tools and computer systems. In this case, the provision of everything needed for
production directly to workstations, far from being an aid to the operator, constitutes
yet another way to cancel any form of ‘waste’ in terms of ‘non value added activities’.
This classification of activities into ‘value added’ and ‘non value added’ (NVAA) is
another critical element of Lean Production – the former being the only activities that,
in progressively transforming inputs, add value to them. In contrast, non value added
activities do not directly add value to inputs and must, therefore, be compressed as
far as possible since they are viewed as ‘downtime’. NVAAs usually include all such
activities connected with arranging/predisposing, searching, placing, pushing, pulling,
dividing, cleaning, walking, all types of waiting, all types of stoppages, measuring,
counting, controlling, sending/transferring, etc.
World Class Manufacturing (WCM) is an evolution of Lean Production and is also

strongly integrated with Industry 4.0: under this approach, all operational activities
and production support must be continuously improved in order to generate a flow of
value added without waste and with the fewest possible losses. In other words, workers
must strive for a workflow that takes place at maximum speed and at minimum cost.
WCM defines waste as leakages of value due to some kind of overproduction (stocks
awaiting processing, defective parts, plants stopped as a result of failure rates, etc.); while
losses are the cost of not allocating a resource to its optimal alternative use, measured
as the loss of value added associated with such a misallocation – an economist would
call these ‘opportunity costs’. A worker awaiting instructions, materials, equipment,
the restart of an idle machine, etc. generates not only a waste but also a loss because
the waiting time could, alternatively, be used to carry out activities that produce value
added.
WCM assigns special importance to so-called Total Preventive Maintenance (TPM):

production workers have to keep their workstations clean and tidy, perform daily
maintenance of machines and equipment, etc. – all of which are non value added
activities in the WCM logic. In other words, companies are assigning to production
workers those activities which were previously carried out by maintenance workers.
This meets a two-fold objective: a) to saturate production workers as far as possible;

and b) to ensure the correct and continuous operation of plants within the logic of a
tightly-controlled flow of production.
Actual operating times depend on faults, settings, adjustments, etc., the resolution of
which increases the rate of utilisation of plants as well as their performance and the
quality of the final product. These three factors together – rate of utilisation, performance
and product quality – determine the overall equipment effectiveness (OEE) i.e. the total
efficiency of the plant. Industry 4.0 technologies allow OEE to be computed in real time
and data to be transmitted to the offices which are in charge of monitoring and control.
Improving these indicators also has enormous consequences for workers in terms of the
monitoring of performance, the intensification of working time and the introduction of
bonuses related to actual performance.
As in all Lean systems, therefore, identifying and eliminating NVAAs is crucial to

WCM and its technical pillars: ‘cost deployment’; ‘focused improvement’; ‘autonomous
maintenance’; and ‘workplace organisation’. In particular, the cost deployment pillar
takes place in each area, identifying the losses and wastes inherent in all manufacturing
processes and sub-processes, quantifying them in terms of cost and defining action
programmes aimed at their reduction. For example, the cost associated with NVAAs is
the number of minutes taken up by workers engaged in such activities and multiplied by
their wage per minute – thus identifying these as a cost to be cut.
WCM introduces distinctions between resulting losses, which can be observed, and
causal losses, the latter being the origin of the former. Total costs, therefore, can be
traced back by identifying the resulting losses and summing up the corresponding causal
losses. For example, a worker performing NVAAs implies lost production – which, in
turn, is associated with fixed costs, the depreciation of investment items, energy costs,
etc. – but also in terms of keeping downstream workers waiting: i.e. lengthening lead
times and desaturating the work process, etc.
The precise quantification of all these cost items is why each workstation is required to
be equipped with on-board connective devices (sensors, monitors, optical readers, etc.)
capable of recording and transmitting any imperfections, irregularities or diversions of
the flow. In this way, it is possible to single out any potentially disposable NVAA, which
now takes on a very broad definition based on the causal chain of all the resulting losses
and wastes. This is also what allows the allocation of workloads and cycle times to be
presented as entirely ‘scientific’, as the result of the software processing of variables
that are, by definition, objectively measured. This puts companies into a coveted and
advantageous position since, where these are not questionable, this makes formal and
informal negotiation on workloads, labour organisation and staffing levels increasingly
hard to achieve.

Matteo Gaddi
3. The field research: Italian companies in the automotive supply

sector
3.1 The shape of the automotive supply sector
In order to understand the practical impact of Lean Production and Industry 4.0, it is
necessary to explore briefly the contemporary shape of the sector.
We should stress that the Italian automotive industry is undergoing a process of

transformation characterised by an absolute drop in the production of vehicles (from
1998 to 2018: the number of vehicles produced dropped by 1.3m), accompanied by
a relative rise in the production of parts and components. This implies a significant
change in the structure of employment. In 1998, forty per cent of employment was
devoted to parts and components and 52 per cent to final (vehicle) production; but, by
2018, these proportions had been reversed, to 53 per cent and 41 per cent respectively.
Italy is, in some respects, moving towards a production specialisation similar to that of
some central and east European countries which specialise in the supply of components
to western (mainly German) car manufacturers. However, Italy is characterised by
higher wages, which raises doubts about the permanence of this kind of production
model. In fact, two models of supply are emerging: that of countries with low labour
costs (Poland, Czechia, Slovakia and Hungary), characterised by state funding and
investment incentives, located close to Germany and at the service of German OEMs;
and that of western countries in which the production of parts and components is
still inextricably linked to the national manufacturers who absorb it. However, the
disappearance of a national carmaker in Italy makes the situation in this country
increasingly uncertain: if Fiat Chrysler Automobile’s production volumes are too low
to absorb the parts and components produced in Italy by multinational companies,
these could make drastic decisions about their Italian sites, preferring those located in
other countries with lower labour costs and closer geographical proximity to final sites
of assembly. Component manufacturers located in Italy are, therefore, increasingly
re-orienting their production to supply foreign, in particular German and French, car
manufacturers.
Furthermore, it is clear that supplier companies are being conditioned by two aspects:
a subordinate position in the production chain; and competition with companies in the
low-cost countries of central and eastern Europe.
As supplier companies, their production conditions are significantly determined by

the OEMs which have the power to decide costs, timing and conditions. For example,
they must continuously provide just-in-time supplies to OEMs, with rigid planning and
scheduling of their own production processes and a strict synchronisation of internal
production with both their customer(s) and their own suppliers.
Moreover, in facing up to competition from low-cost countries, they will try both to
reduce production costs further, with negative consequences for wage levels; and to
increase labour productivity, with negative consequences for working conditions as

regards the intensification of performance, the reduction of assigned cycle times and
increases in workload saturation.
Industry 4.0 technologies and Lean Production models are the main elements that are
determining working conditions in the automotive supply sector. In the next section,
we highlight the close interweaving of these two elements. It should be stressed that
these two elements are being applied not only at the level of an individual company, but
throughout the supply chain as a whole.
3.2 Outline of the research

To highlight what is being determined by the technological and organisational
transformations that we have described above, we have examined several companies
operating in Italy in the automotive supply sector (see Appendix for details):
— Midac (Verona) – battery manufacturer and a tier one supplier;

— Fiamm Hitachi (Verona) – also a battery manufacturer and a tier one supplier;
— Cebi Motors (Padua) – a manufacturer of micro-motors and gearmotors, and a
tier two supplier;
— Fonderie Montorso (Vicenza) – a castings manufacturer and tier three supplier;
— Magneti Marelli (Corbetta, Milan) – a manufacturer of dashboards, inverters and
control units, and a tier one supplier;
— ST Microelectronics – a semiconductor manufacturer and tier one supplier.
In each company, a series of interviews were carried out with workers from different
departments in order to develop our views on how the entire production cycle is being
affected by these transformations. The plants involved in the research are all unionized.
The interviews involved full-time union officials, shop stewards, and even simple
workers who are members of Fiom Cgil (metalworkers’ union). The research, in fact,
was carried out in collaboration with Fiom Cgil, which is interested in understanding
the spread of Industria 4.0 in the industrial metalworking sectors and its consequences
on working conditions. For each plant, in-depth interviews were carried out (from a
minimum of 5 to a maximum of 15) with the figures mentioned above, each lasting
about one hour. Subsequently, further analyses were carried out through the analysis
of: the companies’ Industrial Plans, the investments made and planned, the layouts of
the production departments (plants, machinery, lines). The analysis of these documents
was followed by further in-depth analysis through interviews. A draft-report was
prepared for each plant and sent to full-time union officials and shop stewards so that
it could be discussed collectively (a sort of Focus Group) and then supplemented with
further information and details that emerged during these meetings. At some plants it
was also possible to visit the production departments, depending on whether or not the
company management were willing to authorise this possibility. Interviews and factory
visits took place in the second half of 2018 and during 2019.
Thus the approach to the research questions is qualitative.

Matteo Gaddi
In the rest of this section, analysis of our field material has been organised as follows:
first of all (in section 3.3), suppliers’ relationships with their client companies have
been studied to highlight how the latter’s role is decisive in determining production
conditions; and then the knock-on effects of this on the network of supply companies
that the suppliers have, in turn, created (since they, too, have decentralised significant
parts of their production). We then focus on internal conditions in the plants from
the point of view of the role played by planning and scheduling tools in determining
production processes and workloads (section 3.4); and the impact these have in terms
of working conditions in production lines and departments (section 3.5). The latter
highlights, from our examples, that the tools of Lean Production and Industry 4.0
technologies are responsible for the deterioration in working conditions.
3.3 Case studies in supplier relations

This sub-section demonstrates how OEM suppliers closely control their own suppliers,
creating a complex and closely integrated production network the boundaries of
which may also extend internationally. Supplier companies seek to build very close
relationships within their supply network in order closely to synchronise all steps.
Obviously, this close level of synchronisation and control over supply conditions heavily
affects the working conditions of workers in companies in the subsequent links in the
chain.
The fragmentation of production is facilitated by tools that allow the continuous

transmission of production orders, even several times a day (forcing workers in
decentralised companies to speed up and become more flexible to meet these work
orders) and, at the same time, to monitor how the supplier company is working (in
respect of assigned times, production progress, etc.).
Thus, the relationship with the companies to which they are a supplier is a central
preoccupation for companies in the supply chain: customers may impose on them
the application both of organisational innovations (models of Lean Production, thus
extended to the entire production chain) and choices of technological systems. We
should recall here that, in the Lean Production mindset, it is the final customer – i.e.
the OEM car manufacturer – that pulls the whole chain, in the process determining
working conditions in all the downstream links. Companies in the automotive supply
chain must therefore operate according to the same just-in-time principles and systems
dictated by those to which they supply. Industry 4.0 made it possible to establish the
main principles of work organisation (i.e. Lean Production model) throughout the
entire supply chain.
In our case studies, Industry 4.0 is operating in the direction of horizontal integration:1
i.e. the integration of different factories belonging to the same production chain. The
main Industry 4.0 technologies used for this purpose are: ERP extended to the whole
1. As opposed to vertical integration, i.e. the integration of different functions within the same factory.

chain; production order transmission tools; electronic data interchange tools (EDI)
for document sharing; and production progress monitoring tools (MES). Our case
studies show that the intensification of work performance and of control over it is the
direct consequence of the decisions of the car manufacturers whose demands supplier
companies cannot afford to ignore otherwise they would lose supply contracts. This
applies throughout the process, including tier one, tier two and tier three suppliers, and
beyond. The model facilitates a substantial reduction in the possibility of trade unions
to negotiate working conditions: everything is presented as fixed since it is dictated by
the requirements of the customer.
Case studies: OEMs and tier one suppliers
Magneti Marelli (MM) – within its plant at Corbetta, Milan – produces dashboards,
control units, inverters, etc. for manufacturers from other countries such as Germany
(Porsche, Audi and Volkswagen). The way in which the car manufacturers submit their
orders, and the time allotted to Magneti Marelli for supply, have a very significant
impact on the organisation of work and on working conditions.
For instance, Porsche has adopted a ‘herringbone’ production model: the customer can
configure an online order, customising the choice of car. This order is managed by a
central information system which requires all the actors involved in the delivery chain
to be synchronised. The MM Corbetta plant periodically receives production orders
through an EDI tool called Value Added Network (a hosted service offering secure
data transmission between partners). Just five days before the assembly of the car in
Germany, Magneti Marelli receives an order via VAN to start production on exactly the
sequence of on-board instruments which must be assembled in the Porsche plant to
meet the order. Porsche’s orders, therefore, have a daily character and Magneti Marelli’s
production process (and hence its workers) must adapt to them.
Another example of the closeness of tier one suppliers to OEMs can be found among
the battery manufacturers. In order to meet the significant levels of coordination
dictated by the just-in-time principle, planning and production phases in both Fiamm
and Midac must be carefully synchronised, defined and implemented on a just-in-time
basis. Battery charging programmes depend strictly on customer orders: the battery
is charged only when it has to leave the factory otherwise it risks a loss of charge.
This aspect obliges OEMs to provide themselves with a logistical service amongst its
suppliers that is functional to production and just-in-time delivery. No delay, in fact,
can be allowed: the time allotted must be strictly adhered to.
Case study: tier two suppliers
Cebi is a tier two supplier; in this way, it is subject to double pressure: the first is that
from the OEM; while the second is that from the tier one supplier.
In Cebi, the acquisition of customer orders is carried out by ICT. The tier one supplier
is Brose (a German firm), which now accounts for eighty per cent of Cebi’s production
and for which Cebi is the only supplier. Brose makes a call based on its daily needs and

Matteo Gaddi
transmits this order to Cebi automatically via an EDI tool; this allows Brose to provide
a daily frequency of orders and Cebi to verify the stocks and the quantities of materials
which must be allocated to that customer. Cebi, in fact, operates with a deposit account
at a customer plant (Consignement Stock), from which it withdraws materials daily
according to its production needs. In this way, Cebi’s system works on the basis of
daily flows, updating the stock requirements and planning the production needed to
rebuild stock at the German plant. Through this way of managing customer orders, the
organisation of production at Cebi is modulated according to Brose’s requests.
In general, the management system at Brose continually processes the flow of orders
that it receives from car manufacturers before turning them over as orders to Cebi.
The role of the tier one supplier can be particularly invasive: here, Brose demands that,
every four hours, Cebi production operators fill in a control card in connection with the
amount of production they have realised. This data is monitored both by the internal
offices in Cebi and by Brose itself.
Case study: use of new technologies and organisation of the supplier network
amongst tier two suppliers
Cebi is itself digitally connected with some of its suppliers. In this way, Cebi’s suppliers,
before shipping materials, can provide data notifications concerning each job and the
control parameters which apply (already performed and self-certified by the supplier
in question). This system allows materials to arrive directly on Cebi’s production lines
which, based on a ‘free pass’ model, saves testing time while also facilitating data
collection.
Cebi’s programming department develops data with the aid of material requirements
planning (MRP) software and sends out its supply plans on this basis. The lead time for
suppliers do differ: from China it is one month; while from local network suppliers is
daily: for this reason some of them having machines and processes dedicated to Cebi and
which work in a continuous cycle. Supplies from faraway low-cost countries represent
an element of system fragility whose effects are passed on to Cebi’s workers who have
to deal with any delays by working overtime or on Saturdays (or even during holidays).
Case study: use of new technologies and organisation of the supplier network
amongst tier three suppliers
In the case of Fonderie di Montorso, ‘core’ suppliers directly receive orders generated by
the management software, together with a technical data sheet detailing the specifications
and the cycle times which must be observed. The relationship with subcontractors is
based on special methods of control by FM through the use of computer tools. By virtue
of its control system, FM can outsource many stages in the production process. The
management system developed by FM comprises a SAP dashboard with an ERP tool
which allows the integration of external contractors in the system.
The subcontractors therefore have their own workstations equipped with PCs on which
are installed the web dashboard that interfaces with the company system; in this way,

production status is displayed directly. The system was developed and supplied by FM,
together with the technology and training (made by FM technicians).
FM can therefore control – in real time – the production feeds of its suppliers and also
verify that they are complying with the production plan.
3.4 Impact on workers: planning and scheduling of work orders

We have seen that planning and scheduling tools constitute fundamental elements
in the automation architecture aimed at maximising the productivity of the entire
production process.
Once orders are acquired, they are immediately processed via ICT tools (ERP) to define
production programmes. The latter, in turn, are immediately scheduled (via MES) to
define the individual work orders which are transmitted to each workstation. In this way,
companies are able to remove from collective bargaining both the overall production
volumes (and the question of the corresponding numbers of workers needed to deal
with them) and the workloads assigned to each worker. It must always be remembered
that the work orders transmitted to each worker have already been defined from the
point of view of the time assigned (and which is, in turn, getting tighter and tighter as a
result of competition).
The purpose of the companies here is, as we have highlighted, to ‘objectify’ the assigned
workloads and working times so as to make them indisputable. Regardless of the state
of technological progress, the consequences for working conditions, both in terms of the
intensification of pace and control of the process, are very similar.
How process planning/scheduling translates into work orders (workloads) and

performance control
MM uses TESAR’s tools, such as MOTIS1 and MOTIS2, amounting to an integrated

system for planning and controlling production. These work thanks to real-time data
collection terminals directly connected to production machines. MOTIS1 constitutes
software for the planning and scheduling of industrial production: it optimises the
workloads of machines and workers as well as the performance and productivity of the
entire company. MOTIS2 is the software for production management, data collection
and monitoring (MES): this allows for the control and management of the production
process, considering both the declarations of operations performed and the automatic
monitoring of the production parameters of any machine/plant. Moreover, the
interactive management of these declarations feeds a complete and powerful system of
real-time supervision, statistics, indicators and reports.
Through these items of software, production is not only programmed but also scheduled
as regards the assignments for each operator and the equipment required to carry
out the workload within each shift on the basis of acquired orders, relative priorities,
availability of materials and plant, etc. In this way, the organisation of production is

Matteo Gaddi
rigidly determined, with little room either for the autonomy of workers or for bargaining
over workloads. For MM, it is also decisive that this planning/scheduling of activities
is continuously monitored, thus introducing forms of control over the performance of
work.
In the lines dedicated to the production of dashboards, daily production plans are
indicated on work orders bearing job codes which also correspond to the operating
programs of the machines and the types and quantities of products to be manufactured.
Components to be inserted on electronic circuit boards are also indicated by the
production plan; once they have been ‘called’ digitally, the pick-and-place machine is
automatically activated to place the boards correctly. During the day, the production
program may vary, requiring the resetting of the entire line to be done in the shortest
possible time. All these operations are recorded. The planning of activities and their
recording are two closely intertwined aspects of the organisation of work, whose
integration is facilitated by the ICT tools available. In MM, the application of WCM
techniques has also resulted in pressure to reduce downtime.
In STM, Industry 4.0 is being implemented as a means of guaranteeing the maximum

use of the plant given that it has been the subject of sizable investment.
Within the production process, silicon is processed through different machines, following
a process flow defined by the R&D department. This flow is based on infrastructure
provided by Workstream (a type of MES), which provides operators with elementary
information confined to the path that the batch must take. Through Workstream,
operators carry out a double process: on the one hand, following automated scripts
appearing on computers embedded within the machines, they move the batch between
the various machines; on the other, they keep track of the work process. The set-up for
each machine, identifying the particular operations to be carried out, is downloaded
separately. FTP communication protocols allow the machines to access a server from
which Workstream ‘picks’ the necessary recipe. Each batch is, in fact, associated
with specific tooling: Workstream ‘reads’ the batch (through a barcode reader) and
automatically selects the set-ups to be downloaded. The process is highly automatic:
for each machine, Workstream (a) identifies and extracts the tooling set-ups associated
with the batch; (b) records each operation; and (c) indicates to the operator the next
step. Workstream allows the batches to be traced and processed using information
contained in the barcode.
In each case, the tooling set-up is prepared by ICT engineers on the instructions of the
R&D department.
This has led to the occupational de-skilling of workers: previously, they designed the re-
tooling and thus knew the whole process; now, they simply load and unload the batch
because many of the steps are being managed directly by the software.
How process planning and scheduling interact with machinery and tools
With regard to the battery companies, Midac wants to introduce between three and
five new robots on top of those that already exist; currently, the machines are partially

automated but the operator is always needed for loading/unloading. Despite the
company stressing that robotisation will not have consequences for employment,
union officials and shop stewards highlight that even partial automation produces
some effects: it is possible to make increases in production volumes in the presence of
constant or even slightly reduced staff numbers (via the non-replacement of retiring
staff, etc.). This means that there will not be a wave of redundancies but that, if there
is an increase in production volumes, the level of employment will not correspondingly
increase: i.e. the increase in productivity is of a labour-saving character.
The battery production processes of Midac and Fiamm are very similar. The first
part relates to lead smelting and rolling, processes which are governed by a screen
displaying the production parameters. Battery fluid is prepared by a machine that
works automatically, using production data entered on a PC. The program has already
been installed on the machine: in this way, it is only run by the operator.
In Fiamm, the line is managed by a master panel and the operation of the lines tends
to be modular so that, in the event of breakdowns, only single parts need to be reset
instead of the entire line. Both in Fiamm and Midac, the communications system
between the machines is determined by a ‘master’ Programmable Logic Controller
(PLC) that controls the ‘slave’ PLCs: for example, the speed is determined by the master
potentiometer; a system that, when connected to individual PLCs, monitors whether
the machine is working or not. In Fiamm, a SAP-developed ERP tool calculates the
production volumes achieved and those that are lost in the event of machine breakdowns
and requirements for maintenance. In addition, the worker, at the beginning of each
duty, must enter a personal code on the production line and ultimately record its end.
Both these aspects are tools to control worker performance.
The final process of assembling the batteries, both in Midac and Fiamm, takes place
through different machines; the whole process is highly automated; the role of the
operators is that of the loading, control and activation of the line. The constraints on
workers, therefore, are determined by the cycle times of the machines. Each machine
has a PLC that communicates with the others, while there is a central system that
records all the items that have been realised during the shift, allowing the company to
exercise real-time monitoring of workers’ performance.
In Cebi, the operators find out the work order at the workstation by means of reading
a barcode with an optical scanner: the PC automatically shows the volume to be
produced, the line to be used and the composition of the work team (each operator’s
ID must be inserted). The machinery is automatic, being loaded and started by the
operator through a PC embedded on the machine using a standard code (the machine
is already set according to the general planning defined earlier by ICT tools). Therefore,
the operator has only to load/unload the machinery and intervenes only in the case of
stops and for process controls. In the latest generation of machines, unloading takes
place by a robot, eliminating a work task.
A screen visible to the line supervisor collects data on operating production, making
visible the number of wastes by type. Here, a control card is filled in and entered into

Matteo Gaddi
the PC by an operator, who enters the numbers of good pieces, repairable pieces and
those which are rejected (with reasons for the rejection) at the end of each duty. Within
the most robotised department, a series of robots weld the flanges, recording and
transmitting the production data automatically and in real time.
Finally, the assembly of micro-motors is also carried out using machines that work
automatically; a robot provides the pieces for the first manufacturing station and then
each subsequent one. In this case also, the operators only have to load/unload, start
up the machines and check stops and malfunctions (through the data displayed on the
monitors).
At another FM plant, in Crevalcore (Bologna), MES has been implemented: the

machining centres are connected to a networked PC (and therefore to a centralised
system), so that manufacturing declarations take place through log-in and log-out
operations carried out by the worker who records the start and end of each duty via his
or her personal ID. A scanner reads the barcode for each operation – at which point,
personal ID and work order are associated so that, at any time, the operator may be
identified as having been logged on to a particular machining centre – identifying the
particular phase of the production and on the particular work order. At the end of the
duty, the operator inserts into the ICT system data on the number of pieces processed.
Even if FM’s Vicenza plant does not yet have MES, the monitoring of production is still
done in real time: the worker records the number of realised pieces on the department
PC; in this way, the pieces that have been machined are visible to the ICT system as being
available for the next phase. The equipment and production processes are governed by
software: all operations are controlled from the point of view of relative costs through
ERP; every ten minutes, via another piece of business intelligence software, the
operations performed by workers and robots are monitored. The business intelligence
software in question is QlikView, which is a reporting tool. It can also be used as an app
and is accessible through the company network (by those authorised to access it). The
equipment generates data that is collected, stored, processed and classified: in this way,
reports are created and control is continuous.
3.5 Impact on workers: production lines and departments

Production planning and scheduling results in work orders being executed on the line
or at workstations. Even in the practical execution of these tasks, technology has a very
advanced role in determining working conditions and allowing companies to exercise
control in real time. However, there are also deskilling effects. The aim of companies is
to achieve the highest possible saturation of the workforce and machinery; frequently,
the two things coincide because, in order to achieve the maximum saturation of
machines, the rhythms of work are thereby intensified. Very often, the times assigned
to workers depend on the cycle times of the software embedded in the machines or
the instructions sent to them, sometimes remotely, via ICT networks. The cycle times
incorporated in the machines and tools are not ‘objective’, but depend in practice on
the social choices of companies to increase work saturation. Sometimes, it is the case

that the imperatives of production over-ride the times when a line should be stopped
to check that a particular machine is functioning properly and delivering satisfactory
product quality. This demonstrates a “formal” tension within different aspects of world
class manufacturing that, in reality, confirms the priority order between production
numbers and quality, but which is also to the detriment of the saturation of work.
Condition of work and saturation of the work process are dictated

by machines and technologies
On the dashboard production lines at MM, the circuit board production process is
completely automated: only the production code is entered manually and thus the
machine (these are numerical control (NC) machines) is activated by running a program.
Usually, the first operation to be performed every day is to load the program on all the
machines; then, for each step, the program is run for each machine: products move
between one machine and the next on conveyors according to predetermined timings.
There is a system for collecting data generated by the operation of the machines, installed
a few years ago, which allows the entire line to be monitored. On each line, there is a
screen controlled by technologists from their workstations which collects production
data for each machine and each line. This screen displays all the process information,
including any problems and faults.
Each machine then marks out the beginning and end of each processing stage and
carries out its operations based on the time established according to the type of product.
The installed systems mean that it is possible to trace the entire production process and
verify whether the cycle times have been observed.
Subsequently, the circuit boards are transported to the line where the dashboard
assembly phase starts. Here also, it is the case that operator intervention consists only
of inserting the board and components, while the rest is automated. It is clear that the
work of the operators is strongly constrained by the computer programs which are
simply run to activate automated processes that have predetermined cycle times.
The working time of workers on these lines, besides being constrained by the cycle times
of the programs, is inevitably linked to the quantity of production to be carried out. This
quantitative objective is conditioned by the level of operability of the machines; if a
machine is not working properly, the cycle should be stopped to request maintenance.
However, the imperatives of production frequently prevail over everything else. This
consideration confirms that the application of production systems, such as WCM,
is exclusively aimed at maximising production via a shortening of cycle times and
regardless of the quality aspects that, formally, should represent one of the cornerstones
of these models. In fact, the workers we interviewed stressed that, over the years, cycle
times and rhythms have continually intensified, in particular via the elimination of all
downtime and the intensification of work rhythms.

Matteo Gaddi
In MM’s power train department, the production cycle is quite similar to that of
dashboard assembly. The circuit boards are not loaded manually but by robots. Part
of the supply from the warehouse is automated, with two robots pulling a trolley: these
move along magnetic strips and are programmed by the warehouseman to stop at the
various kanban stations. The material is deposited and the worker takes it to carry
out operations. All the production lines have kanbans for material storage, organised
according to production sequences to try to erase as much waste time as possible.
The level of automation is very high; for example, a robot takes care of the positioning
of the control units. The task of the operator is to receive the materials to be assembled,
then assemble them and put them into a machine that carries out riveting, welding,
electrical testing and labelling. Once these operations have been completed, the control
unit returns to the worker for visual inspection and subsequent packaging. Here also,
the production times are, essentially, the cycle times of the machines; the line screen
shows the cycle times of the machines and these must be observed by the workers. It is
also clearly the case in this department that the constraints exercised by the machinery
(and the programs which run them) are crucial.
The traceability of production is guaranteed by the requirement imposed on the operator

to scan the cover label (every 12 pieces) with an optical reader. In this way, when the
label is read, the time spent on production is recorded.
Saturation and possible deskilling
In the manufacturing department of STM, the influences which technology is having on

working conditions are also characterised by possible deskilling. Prior to the introduction
of Workstream, the operator followed and knew all the various phases of the production
cycle. Now, manufacturing phases are carried out by machines which are programmed
to perform all the operations. These machines have been designed to incorporate FTP
communications protocols that manage the entire download of the tooling set-ups from
the server which provide the machine with its operating instructions.
On some lines, the level of automation is currently such that it is possible to operate
them remotely, i.e. to launch the setting-up process directly from offices. Therefore: a)
the steps that a lithographic machine must take are programmed; b) there is a double
programming: of the flow, that is the sequence of machines; as well as of the processes
that each single machine must carry out; and c) the programming encompasses the
ability to activate each machine either from within the department (that is, on each
machine) or remotely. The machines are very complex; so much so that the supplier
companies have groups of workers who work permanently with STM employees at the
Milan plant as only they know all the details of operation and programming.
This has strong consequences for the human-machine relationship: the machinery, in
fact, operates on the basis of programs written by people who are not the operators who
use them, working on the basis of logic that has no room for the understanding, and
therefore the control, of STM’s workers.

Intervention in industrial processes to achieve tightly-controlled flows of production
In both Midac and Fiamm there are processes whose times are dictated by chemical and
physical constraints (paste preparation, casting, drying, etc.). However, the companies
have been intervening in all these phases to shorten the times: in Fiamm, to shorten
curing times (from 15 days to 24 hours), a special oven has been installed; while, to
avoid loss of time due to production changes, new machinery (auxiliary ovens) etc. have
been installed. Above all, Fiamm has compressed machine times on its ironing line: the
production volumes of the line have been increased from 22 metres per second to 32 by
dint of speeding up machine movements with the use of inverter motors. In this way,
all the machines on the line have been speeded up to make the line speed uniform: by
changing the general setting of the system (on the master panel) the speed of individual
machinery may be changed.
In Midac, the installation of PLCs on all equipment allows real-time signals to be sent to
the maintenance department (via monitors and smartphones) to guarantee immediate
intervention in the event of faults, minimise machine downtime, obtain the necessary
spare parts and realise scheduled maintenance.
Meanwhile in Fiamm, in order to guarantee that the machinery is fully operational,

significant pressure is exerted on maintenance technicians: they receive calls on a
mobile phone and must enter every alert on Geocall. This is an ICT system that includes
many functions: receiving reports directly from equipment; creating work packages to
be performed with scheduled machine downtime; assigning interventions to technicians
or teams based on production shifts; reporting works and opening new requests for
intervention through an operational workflow; defining the check list of controls;
and checking activities in real time. It is a highly automated system which allows the
company to exercise a very strong form of control of the times and performance of
maintenance technicians whose job is to restore operation as quickly as possible. The
maintenance technician has to insert a personal ID on the PC embedded on the machine
to track the start and end of activities; for each intervention, a report must be compiled
that is checked by the head of department and which evaluates the time spent and the
quantity/quality of the performance.
Re-tooling machinery to reduce downtime
In Cebi, the automated lines produce 500-550 pieces an hour; to feed these lines
continuously, the flanges are welded by robots.
Thanks to the robotized welding of the flanges, these parts are supplied to the assembly
line at intervals (cadencies) dictated by the operating programs of the robots, forcing
the workers to adapt.
In addition, also on the assembly line there is a constraint in operation dictated by

technology. The machine, once loaded, runs autonomously and its operation time is
used by workers for control, restore and registration activities; primarily, however, the
constraint on workers reflects the need to unload the machine because, in the absence of

Matteo Gaddi
this stage, the equipment would stop. In the event of a product changeover, line retooling
times have been drastically reduced from 45 minutes to 15 using a PC that recalls the
set-up corresponding to the new production batch. Re-tooling and maintenance have
been put in the hands of the operators, increasing their workloads, and the company
no longer hires maintenance technicians. Two particular elements of Lean Production
are in play here: SMED (Single-Minute Exchange of Dies), which seeks to reduce plant
setting times and eliminate waste; and Total Preventive Maintenance whose aim is the
optimisation of capacity utilisation of the lines through the progressive reduction of
unplanned extraordinary interventions.
It is also the case in Cebi, therefore, that technological and organisational innovations
have led to the possibility of achieving higher production volumes with the same number
of staff, whose jobs have become characterised by great flexibility and variability.
One of the main investments in connection with Industry 4.0 at Cebi concerns the
wiring-up of some machines in order to have data accessible in real time via PCs or
other devices, reducing time and steps. Each machine will, as a result of this greater
connectivity, be able to communicate and send alerts to the person in charge of the
production process so that he or she can intervene immediately, avoiding machine
downtime. This will contribute to the improvement of OEE, in terms of the degree of
utilisation of equipment. Some lines will be equipped with a central PC that collects data
from PLCs and other devices, data which will then be passed to the MES. Other lines
will have a direct PLC-network connection, with direct data transmission to the MES.
In this way, the assessment of the OEE parameter is used to put the group’s plants, and
therefore the workers, in competition with each other.
Effects on white collar staff
In STM’s design offices, design automation has been implemented through the use of
increasingly complex design and simulation software. In particular, digital design offers
much more advanced support for CAD (Computer Aided Design) than analogue design.
The change in the design sector has been significant and extends to the deskilling of the
designers. The design process is very complex and involves several steps; inevitably,
the objective of the company is to shorten the times for these steps, and their number,
as far as possible. For this reason, STM has defined a ‘process design kit’: a set of
software tools that allows the automation of as much of the design as is conceivable. The
definition of precise rules corresponds to the objective of minimising change to reduce
costs. To achieve this goal, STM addressed its requirements to companies specialised in
the supply of CAD, such as Cadence, Mentor and Synopsys which provide about ninety
per cent of the tools used in design. Their software is able to count how many times a
function is pressed, how many clicks are made, etc. and thus its ability to control the
performance of the designers is, therefore, extremely pervasive.
STM also wants to document all the steps taken in design – which, in addition to being
an extra tool for control, entails additional workload. Greater workloads and complex
processes result in a highly stressful situation for workers. Moreover, STM can exert
control thanks to the traceability of all its processes: when problems occur, the company

is able to identify which operators have been responsible and this is adding to the
stressful climate.
Meanwhile, the designers feel less specialist and skilled than before because the software
has ‘absorbed’ many of their skills.
3.6 A possible trade union bargaining agenda

The findings of this research (and other similar ones carried out in collaboration with
employees organisations of Cgil, Fiom in particular), have made it possible to begin to
define an agenda of bargaining issues.
First of all, the theme of information rights. The organisational and technological
innovations that are applied in the plants derive from precise choices made by the
management and ownership of the companies that must be communicated, in good
time, to the workers and their representative organisations. The National Collective
Agreement for Metalworkers provides that the management of companies with at least
50 employees must provide the trade union representatives with a series of information
concerning, among others, a) the strategic choices of the companies on the production
activity, b) the changes in the production system that affect the technologies adopted,
the overall organization of work and the type of production; c) the outsourcing of phases
of the production activity. Unfortunately, the information system that companies
implement with regard to workers’ organizations often works very partially and
late. The research has shown that the extent of the technological and organizational
transformations that companies are implementing has such an impact on workers
that it is necessary to fully implement the information rights provided by the National
Collective Agreement, so as to have, in good time, useful information to negotiate with
the company the actions it intends to implement.
The research revealed the need for the union to regain the opportunity to discuss work
organisation. This means the possibility to discuss, in the first instance, cycle times,
saturation and workloads with companies. With full-time union officials and shop
stewards, objectives were discussed in relation to these issues which have become
part of the union’s bargaining agenda, despite the fact that Italian companies refuse
to recognise the right of the union to negotiate work organisation issues on which they
claim to be able to take decisions unilaterally. Secondly, the possibility of discussing
work organisation also involves bargaining on broader aspects, such as production
volumes and workforce size.
Negotiation of cycle times, saturation and workloads, for example, can lead to a
lower workload for each worker, thus determining the need to expand the workforce
to maintain overall production volumes. This need is particularly strong in those
departments where the pressure on workers is very hard.
The organisational and technological innovations implemented by companies, aimed at

maximizing productivity through a greater degree of work exploitation, are also causing

Matteo Gaddi
problems in terms of health and safety. These problems affect both classic musculoskeletal
disorders and forms of work-related stress. The Italian law obliges companies to adopt all
measures to protect workers, including the respect of ergonomic principles in the design
of workplaces, in the choice of production equipment and in the definition of work and
production methods. The implementation of these safety measures must take place with
the involvement of the trade union: in many situations, the discussion with companies
on health and safety problems allows for the involvement of work organisation issues
which, as said before, companies would like to determine unilaterally.
Levels of stress and pressure on workers have also worsened as a result of the use of
technological systems to control work performance: trade union bargaining is trying to
limit and regulate the use of these tools. This issue is not only limited to privacy but is
functional to a certain model of work organisation.
Last but not least, the issues of outsourcing and production chains. In order to limit
the negative consequences of these corporate strategies, CGIL has defined the objective
of practising so-called «inclusive bargaining»: i.e. bargaining that also involves the
workers of sub-contractor companies and supplier companies. This is a very ambitious
and difficult objective, but it can no longer be postponed in the light of the concrete form
taken by the industrial structure in Italy (and in Europe).
4. Conclusions
The structure of the European automotive industry is characterised by two main aspects:
the production chain is highly fragmented and dispersed across different countries; but,
at the same time, supplies must be sent to OEMs on a just-in-time basis, so all the stages
of the entire production chain must be closely synchronised. Industry 4.0 technologies
make it possible to coordinate these increasingly complex and fragmented chains, as a
result of the use of ICT tools and apps that make it possible to manage and monitor all
the phases of the process in real time.
Italian automotive supply chains are under pressure from two entwined phenomena:
the reduction of vehicle production in Italy, which has obliged Italian suppliers to
increase the volumes of components supplied to foreign car manufacturers; and the
competition exercised in this area by low-cost plants in central and eastern Europe that
are supplying the western European automotive industry.
Italian supplier companies are responding to this pressure by intensifying the exploitation
of the workforce.
Technological (Industry 4.0) and organisational (Lean Production) innovations are

closely connected and, as a result of these connections, are leading to new models of work
organisation. These new models are having a serious impact on working conditions,
symbolised by the intensification of the pace and rhythms of work; the saturation of
workloads; the real-time control of work performance (in turn leading to workers being
subjected to increased work-related stress); and often also to occupational deskilling.

A different use of technologies, i.e. to support workers rather than to establish the
ground for their further exploitation, is certainly possible: but this requires trade
union intervention for a different organisation of work and production. It is clear that
technologies cannot be negotiated only from the point of view of their use in practice,
but starting from their conception and design: only in this way will it be possible to plan
interventions which are in favour of workers (Dina 1982; Noble 1979; Panzieri 1961).
References
Akerman M. (2018) Implementing shop floor IT for Industry 4.0. Doctoral thesis, Gothenburg,
Chalmers University of Technology.
Brettel M., Friederichsen N., Keller M. and Rosenberg M. (2014) How virtualization, decentralization
and network building change the manufacturing landscape: an Industry 4.0 perspective,
International Journal of Mechanical, Aerospace, Industrial and Mechatronics Engineering, 8 (1),
37-44.
Butollo F., Jürgens U. and Krzywdzinski M. (2018) From lean production to Industry 4.0. More
autonomy for employees? WZB Discussion Paper SP III 2018–303, Berlin, Berlin Social Science
Center.
Dina A. (1982) Tecnologia e lavoro. Richiami storici e problemi attuali, Classe, 22.
European Parliament (2016) Industry 4.0, Brussels, European Parliament.
Forschungsunion and Acatech (2013) Securing the future of German manufacturing industry.
Recommendations for implementing the strategic initiative Industrie 4.0: final report of the
Industrie 4.0 working group, Bonn, Federal Ministry of Education and Research.
Gaddi M. (2018) Industria 4.0 e il lavoro. Una ricerca nelle fabbriche del Veneto, Milano, Edizioni
Punto Rosso.
Gaddi M. (2019) Industria 4.0. più liberi o più sfruttati? L’industria 4.0 vista da chi lavora nelle
aziende metalmeccaniche di Milano e provincia, Milano, Edizioni Punto Rosso.
Noble D.F. (1979) Social choice in machine design: the case of automatically controlled machine
tools, in Zimbalist A. (ed.) Case studies on the labor process, New York, Monthly Review Press.
Oztemel E. and Gursev S. (2018) Literature review of Industry 4.0 and related technologies, Journal
of Intelligent Manufacturing, 31, 127-182.
Panzieri R. (1961) Sull’uso capitalistico delle macchine nel neocapitalismo, Quaderni Rossi, (1),
53-72.
Qin J., Liu Y. and Grosvenor R. (2016) A categorical framework of manufacturing for Industry 4.0
and beyond, Procedia CIRP, 52, 173-178.
Rojko A. (2017) Industry 4.0 concept: background and overview, International Journal of Interactive
Mobile Technologies, 11 (5), 77-90.
Sanders A., Elangeswaran C. and Wulfsberg J. (2016) Industry 4.0 implies lean manufacturing:
research activities in Industry 4.0 functions as enablers for lean manufacturing, Journal of
Industrial Engineering and Management, 9 (3), 811-833.
Sony M. (2018) Industry 4.0 and lean management: a proposed integration model and research
proposition, Production and Manufacturing Research, 6 (1), 416-432.
Thoben K.D., Wiesner S. and Wuest T. (2016) ‘Industrie 4.0’ and smart manufacturing: a review of
research issues and application examples, International Journal of Automation Technology,
11 (1), 4-19.

Matteo Gaddi
Tzafestas S. (2018) The Internet of things: a conceptual guided tour, European Journal of Advances
in Engineering and Technology, 5 (10), 745-767.
Wagner T., Herrman C. and Thiede S. (2017) Industry 4.0 impacts on lean production systems,
Procedia CIRP, 63, 125-131.
Appendix
Overview of case studies
Name Italian employees Multinational
Cebi Motors 226 This company is part of Cebi Group: this is a Group headquartered
(Tier 2 supplier) in Luxemburg, with more than 3,000 worldwide employees and 11
plants in eight different Countries
Fiamm Energy Technology 394 Fiamm was bought by Hitachi, so it became an Italian subsidiary of
(Tier 1 supplier) this multinational company
Midac 485 Italian company with worldwide presence (this company had
(Tier 1 supplier) subsidiaries in France, Germany, UK, Sweden, Nederland, Australia)
Fonderie di Montorso 413 Italian company

(Tier 3 supplier)
Magneti Marelli 5101 Magneti Marelli has been sold by Fiat (FCA) to the Japanese
(Tier 1 supplier) group Calsonic Kansei: both groups (i.e. both Magneti Marelli and
Calsonic Kansei) have a worldwide presence of production plants.
Together they have created a company (Marelli) that has over 170
sites worldwide (production plants and research and development
centers).
STMicroelectronics 10300 This Group has a worldwide presence and a total of 46,000
(Tier 1 supplier) employees. The company involved in the research is the Italian
subsidiary.

Chapter 8
The transformation of jobs and working conditions:
Towards a policy response
Monika Martišková
1. Introduction
The introduction of new technologies and their impact on workers is not a new topic
among scholars; innovation is perceived as an embedded feature of capitalism and nec
essary for capital renewal (Hall 2010). However, there are some aspects which make the
current changes different from previous waves of technological revolutions: the speed
of innovation and its destructive potential regarding technologies currently in use but
which are quickly becoming obsolete (Komlos 2016); their association with jobless
growth (Brynjolfsson and McAfee 2012); and their facilitation of new business models
that reach across the globe with minimum physical capital and with a very low number
of employed workers, which is especially relevant in the IT sector (Soete 2018). In pro-
duction areas, new technologies are used to reduce costs by limiting the input of labour
while preserving or even increasing production levels. The transformation of working
conditions and the reduction in job opportunities are consequences of the deployment
of new technologies in the production process which deserve researchers’ attention.
In particular, there is an urgent need to understand the ongoing changes caused by new
technologies in order to provide relevant policy responses to prevent a deterioration
in working conditions for workers in production. One of the biggest challenges
facing the implementation of Industry 4.0 is the adaptation of workers’ skills to new
technologies. Business leaders in the three countries that we have studied in the central
and east European (CEE) region considered the lack of skilled workers to be the biggest
obstacle to the implementation of Industry 4.0. However, the policy responses at
company, regional and national level are uncoordinated and unsystematic. Effective
reskilling and retraining policies are costly and require both personal engagement and
a plausible institutional framework whose parameters are set by collective bargaining
and/or by public institutions. We argue that, in the context of CEE countries, effective
policy responses to protect workers from negative impacts or to provide reskilling are
missing. In this chapter, we discuss the reasons for these attitudes but also the potential
consequences of inaction.
We have studied the automotive industry as a model sector for the introduction of new
technologies and the impact these are having on working conditions in three CEE countries
– Czechia, Hungary and Poland. The aim is to depict the transformation of working
conditions related to the introduction of new technologies based on in-depth interviews
with managers, trade unionists and workers at specific automotive industry sites and to
analyse their policy responses. Interviews with actors representing sectoral organisations
of trade unions and employers were also conducted. Our interviews focused on various

Monika Martišková
aspects of changes related to the deployment of new technologies in production including:

the installation of robots; the use of new electronic devices; the implementation of systems
that allow production synchronisation, such as Manufacturing Execution Systems (MES)
or Enterprise Resource Planning (ERP); the automation of internal logistics; and some
advanced technologies such as 3D printing or virtual reality at shop floor level. We were
interested in the impact of these technologies on work schedules, changes in workloads,
the impact on workers’ autonomy when performing particular tasks and changes in the
character of the tasks themselves (whether they are more routine or more non-routine),
as well as the changing requirements for workers’ levels of education.
Our qualitative approach has allowed us to understand the perceptions of local

employees and stakeholders on the deployment of new technologies and their impact
in transforming jobs. Particular emphasis was given in the research to the working
conditions of shop floor workers who are expected to experience the most dramatic
changes in both job content and job opportunities and in their prospects for retraining.
Despite trade unions being expected to play an important role in the promotion of
retraining policies at company level (Jolly 2018), our research results suggest that this is
not a collective bargaining priority in the majority of companies we studied. The reason
is that, despite the implementation of labour-saving technologies, a displacement of
labour has not occurred since, in recent years, the region has suffered significantly from
labour shortages which have further postponed policy responses. We conclude that
limited recognition of the negative impacts of new technologies on working conditions
and employability has prevented the establishment of mitigation strategies at company,
regional, sectoral and national level.
In this chapter we first discuss job transformation in the CEE region and the observed
impacts on the quality and quantity of jobs, and then we present our methodology. In
the third section we investigate the impact of new technologies in terms of the numbers
of jobs in CEE automotive plants and then discuss aspects of the transformation in job
quality. We then comment on the policy responses of trade unions in the three CEE
countries, with special emphasis on reskilling policies. In the last section we summarise
our findings.
2. Impact of new technologies on working conditions and job

opportunities
Workers in the automotive industry are expected to be confronted with change in two
different, but not mutually exclusive, ways. One relates to the restructuring of supplier
chains and the varying location strategies of multinational corporations (Pereira
and Romero 2017; Kagermann 2014) which Andrea Szalavetz in chapter 3 discusses
in detail in this book. The second relates to the introduction of new technologies in
the workplace which might have an impact on job opportunities and also modify job
content (Bonekamp and Sure 2015). In this chapter, we devote attention to the second
aspect of job transformation at plant level, focusing on the impact on those workers
directly involved in production processes.

The transformation of jobs and working conditions: Towards a policy response
Employment levels are expected to shrink mostly among manual workers whose
opportunities to participate in additional training may be more limited, while educated
non-manual workers will have higher probabilities of adaptation to job transformation
(Arnold et al. 2018). Especially for lesser educated workers, the job reduction effect of
technologies is highlighted in a study by Acemoglu and Restrepo (2017) who estimate
that each additional robot reduced between three and six jobs between the 1990s and
2007 in the US. For six EU countries (Finland, France, Germany, Italy, Spain and
Sweden), the recently estimated job reduction effect was, over a similar period, also
negative; with each additional robot introduced leading to two job losses (Chiacchio et
al. 2018). The job reduction effect is, however, not distributed evenly because young
people, lesser educated workers and those working in industry are most exposed while,
for instance, technician employment levels have actually increased with the higher
deployment of new technologies in the most recent decades (ibid.).
Scholars have distinguished between the displacement and the productivity effects
of new technologies (Dauth et al. 2017). While the displacement effect refers to the
reduction of jobs because of the introduction of new technologies, the productivity
effect refers to increases in the level of productivity leading to increases in employment.
A recent study from Germany suggests that the displacement effect of technologies
may be counterbalanced by a productivity effect in other sectors, leading to increases
in employment levels elsewhere and making an economy seemingly unaffected by
the introduction of technologies (Dauth et al. 2017). However, even where there is no
overall effect on employment levels, one may question the quality of jobs created under
the productivity effect as these are, mostly, created in services and are both less secure
and more precarious (Novta and Pugacheva 2018). Moreover, Erturk (2019) suggests
that previous experience might well not be repeated during the fourth industrial
revolution. For instance, the demand for technicians and other high-skilled workers
may decrease in forthcoming years on account of individual productivity growth thanks
to the possibilities provided by advanced technologies; therefore, new technologies
might have a net negative effect on employment levels overall, as well as among manual
workers in particular.
Besides the reduction of job opportunities, workers are expected to face a deskilling
effect as regards their own jobs. A deskilling effect occurs when technologies deconstruct
complex tasks into simple steps which make work much easier but which also decrease
the requirement for workers’ skills (Attewell 1987). Evidence suggests that it is mainly
workers involved directly in production who have experienced a deskilling effect in the
last fifty years (Kunst 2019). Another aspect of deskilling, as recognised in the labour
sociology literature, is workers’ decreasing autonomy in decision-making (Agnew et
al. 1997; Erturk 2019), which Industry 4.0 technologies are expected only to intensify
(Butollo et al. 2019). Crouch (2018) also recognised the threat of increased control
within the labour process through the implementation of sensors, chips or wearables –
various devices attached to one’s clothing or body, serving to monitor movements and
improve performance – which contribute to losses in workers’ autonomy and an increase
in the level of control to which they are subject. New technologies implementation also
contributes to increased levels of stress of employees (Körner et al. 2019).

Monika Martišková
In contrast to the deskilling effect of new technologies, increasing task complexity is

perceived to mean the upgrading of workers because only the more difficult and complex
tasks will be performed by humans while new technologies will handle repetitive or
physically difficult tasks (Kergroach 2017). Several authors suggest that increasing task
complexity and the related requirements for better educated workers are an inevitable
part of the technological revolution (Bonekamp and Sure 2015; Porter and Heppelmann
2014). This is the reason why high expectations are assigned to Industry 4.0 technologies
when it comes to the prospect for workers’ upskilling, as workers are expected to be
accomplishing more complicated and advanced tasks in the future (Hirsch-Kreinsen
2016). On the other hand, workers’ upskilling requires comprehensive retraining
policies which are not always provided in the workplace or by public institutions.
Retraining being provided to workers in the workplace is considered to form one aspect
of decent work because, through retraining, an employee is given further chances to
be employable in the future (ILO 2019). Although employers’ motivation to provide
retraining is limited, according to Cappelli (2004) most employers whose business
relies on social capital are willing to provide more retraining. Similarly, Benhamou
(2018) distinguishes the learning and the lean organisation, pointing out that lean
organisations provide less training to workers than learning organisations. Retraining
policies are further dependent on the size of company, capital intensity and unionisation
rate. In the case of the automotive sector, an important factor is also the position of the
company in the production chain, with lower tier suppliers having fewer resources to
devote to employee training and reskilling than ultimate manufacturers (OEMs). Trade
unions, through partnerships with employers, may employ effective policies which
help workers reskill. A good example is provided by the employment security councils
operated in Sweden by employers and trade unions and which provide redundant
workers with assistance in reskilling and employment (Engblom 2019). In the changing
labour market, retraining seems to be a crucial aspect of labour market policy in which
trade unions should play an important role (Jolly 2018; Bamber 1989).
We will conceptualise our empirical evidence based on the effects that introductions
of technology may have for working conditions and jobs. Our empirical investigation
concentrated on the displacement effect and the deskilling effect of new technologies,
looking also at current policies dealing with the retraining and reskilling of workers
as an important element of mitigation strategies. In the next sub-section, we briefly
describe the context of the CEE region and then, in the subsequent section, introduce
the sample of companies we interviewed and our empirical findings.
2.1 Job transformation in central and east European countries

Looking at the prospects for CEE countries, job transformation is expected to hit the
region significantly with new technologies having a prevailing displacement effect.
The main reason is the high share of manufacturing jobs in these countries. In the last
twenty years, the CEE region has experienced a growth in manufacturing jobs, especially
through job relocations from west European countries to central and east European
ones in the automotive and related industries (Pavlínek 2019). A large part of these

manufacturing jobs are based on routine tasks that are expected to be easily automated
in the near future (Keister and Lewandowski 2017). Despite the CEE region not being
a frontrunner in the introduction of new technologies (Krzywdzinski 2019), there are
various predictions that many manual and routine jobs are expected to disappear, or
change substantially: one OECD study predicts for Czechia that around 15 per cent of
jobs are at a high risk of automation while another thirty per cent are at a significant risk
(Nedelkoska and Quintini 2018); while the prediction of Chmelař et al. (2015) is that
ten per cent of jobs will be lost and another 35 per cent transformed.
The reason for such a high threat of job losses in CEE countries is their high share of
manual workers. While plant and machine operators, according to the ISCO classification,
take up 14 per cent of the total workforce and 36 per cent of manufacturing in Czechia,
in Germany their overall share in employment is 6.3 per cent and, in manufacturing,
13 per cent. The figures for Hungary and Poland are even higher than for Czechia (see
Table 1). This is a consequence of the intensive relocation of jobs by west European
companies to CEE countries in the last twenty years. About 387,000 jobs have been
destroyed in the automotive industry in western countries while almost the same
number have been created in CEE countries: about 329,000 between 2005 and 2016
(Pavlínek 2019). The main drivers have been low labour costs and state subsidies, which
have created largely medium-skilled and routine-intensive jobs in the region (Keister
and Lewandowski 2017).
Table 1 Share of plant and machine operators in manufacturing
Plant and Plant and Total Total % plant and % plant and %
machine machine employment manufac- machine machine manufacturing
operators operators in turing operators operators in in total
TOTAL manufacturing in total manufacturing employment
employment in total
manufacturing
employment
HU 666.4 378.4 4,373.4 984.4 15.2% 38.4% 22.5%

PL 1,685 737.8 16,078.8 3,380.4 10.5% 21.8% 21.0%
CZ 727 395.9 5,093.9 1,438 14.3% 27.5% 28.2%
SK 370 223.8 2,502.1 620.6 14.8% 36.1% 24.8%
DE 2,560 1,016.3 40,481.6 7,804.3 6.3% 13.0% 19.3%
Source: Eurostat [lfsa_egised, lfsa_eisn2], own calculation
The level of labour costs in CEE countries may be hampering the introduction of labour-
saving technologies since the rate of return on investment is negatively correlated with
the level of labour costs. However, other factors such as attitudes and the general level
of investment and competition in the sector and in the country play an important role.
When looking at Czechia, Hungary and Poland, and comparing them with Germany
(their most usual business counterpart and/or the owner of many sites), the relationship
between the level of labour costs and the introduction of industrial robots confirms
the negative correlation between labour costs and the number of implemented robots.
While labour costs in automotive in all four Visegrad countries oscillates between

Monika Martišková
10,92 EUR per hour in Poland to 15.12 EUR per hour in Czechia, in Germany hourly
labour cost is 48 EUR per hour.1
We have compared robot adoption rates with actual labour costs to see how the
introduction of robots corresponds to the level of wages in a given country. We followed
Atkinson’s (2018) methodology, assuming that a new industrial robot costs $250,000
and would replace two workers in two shifts working fifty weeks per year, on the basis
of average labour costs compiled by Eurostat. Computed return rate (see Figure 1) for
automotive is 23.6 months for Czechia, 30.6 months for Hungary and 32.6 months
for Poland while for Germany the replacement rate is 7.4 months. From these rates
of return, we can derive an expected figure for the number of robots introduced into
production per ten thousand employees and compare the actual rate of introduction of
industrial robots with the worldwide average. Interestingly, we arrive at the conclusion
that Czechia and Slovakia have a higher than expected number of robots introduced in
production given their labour costs, while Poland and Germany are below the expected
figure. Hungary’s rate of industrial robots in production corresponds to its level of
labour costs.
The higher than expected introduction of robots in Czechia and Slovakia indicates
that we need to take into account other factors than simply-assumed labour costs in
the given country. Despite not being considered in Atkinson’s methodology, decision-
making on the introduction of robots is influenced also by labour costs in the home
countries of multinationals. The wage difference between western and eastern Europe
still provides incentives for multinationals to remain in eastern Europe. At the same
time, rising labour costs in eastern countries has, in recent years, motivated companies
to invest in labour-saving technologies there.
Significant role in investment strategies is also played by the level of competitiveness,

not only in the sector but also between different sites within the same company. Some
level of discretion over individual investment decisions is retained in particular plants,
which sites use to enhance their competitiveness. Another factor in how investment
decisions are made is customers’ control of production in suppliers within the chain,
including specifying what equipment should be used to produce particular parts, which
often also entails new investment in more advanced technologies. All these factors
are relevant in the case of our three studied countries (see also chapter 3 by Andrea
Szalavetz).
1. Eurostat, labour costs data for 2018 for NACE C manufacturing, increased by 20% for automotive [lc_lci_lev].

Figure 1 Actual and expected number of robots implemented in automotive on average

labour costs in five countries
1,800 45
1,600 1539 40
1,400 35
1268 32.6
1,200 30.6 30
1,000 24.5 25
23.6
815
800 20
600 555 15
465 485
400 350 369 373 10
7.4 189
200 5
0 0
SK CZ DE PL HU
Robots per 10000 employees Expected no. of robots Return rate in months
Source: own compilation, based on Atkinson (2018) (data on labour costs: Eurostat lc_lci_lev; robots: IFR 2019)
3. Transformation of jobs in Czechia, Hungary and Poland

While the level of advance in the introduction of technologies can be characterised as
heterogenous in the companies we visited (see chapter 3), at least some effort towards
using, or trialling, new technologies were reported in all of them. The companies in
our sample face multiple pressures in the sector which has resulted in the deployment
of new technologies. First, there are constant cost competition pressures between
the subsidiaries of multinationals. Second, a tight labour market in each of the three
countries has proved to be a serious obstacle to satisfying increased demand in recent
years. Third, there are rising expectations among customers of the increased variability
of products which can be attained through innovation in production technologies and
processes. As a result, they are being constantly forced to increase the effectiveness of
their production capacities through the introduction of new technologies as a means of
increasing the efficiency of production and decreasing production costs.
Our respondents were most familiar with the introduction of robots on production lines
and/or of automated systems in internal logistics. In some cases, they also mentioned
elements of cyber-physical systems, such as production digitalisation and data analysis.
Installation of advanced technologies which improve workers’ performance, such as 3D
vision, wearables or data glasses, were mentioned mostly in OEMs.
In our empirical research, we concentrated on evidence for the transformation of jobs

associated with new technologies. The implementation of labour-saving technologies
embodied in industrial robots mostly affects shop floor workers as regards the quality
of their work as well as job opportunities in general. When investigating the impact
of new technologies on the working conditions of manual workers, we stick to four

Monika Martišková
aspects widely discussed in the literature and in public debate: (1) the deskilling effect,
demonstrated by decreases in job difficulty, increases in job cadence and a decreasing
level of autonomy and control; (2) improving health and safety at work; (3) changes in
job opportunities; and (4) the retraining prospects of workers in our sample companies.
Our research was guided by a series of questions related to the impact of new technologies
on working conditions:
a. Changes in the number of jobs: To what extent have new technologies reduced
jobs?
b. Deskilling: To what extent are workers experiencing the simplification of work
tasks and the loss of autonomy in work performance?
— Changes in physical job difficulty and task complexity: Do we observe at
plant level a decrease in physically difficult jobs? Are the new tasks being
assigned to workers more or less complex?
— Changes in job cadence: To what extent are new technologies increasing the
speed of work?
— Level of control: Are employees exposed to increased levels of control by
employers?
c. Access to training: Bearing in mind knowledge of the importance of retraining
in an era of the introduction of disruptive technologies, are retraining policies at
company level being promoted by trade unions or employers?
Our sample consists of 28 companies, of which nine are OEMs, 17 are tier one producers
and two are tier two producers (see Table 2).2 Our sample contains 34 respondents
from 28 operations in Czechia, Hungary and Poland, of which 17 are trade union
representatives at company level (or are works councillors) while 17 are various
management representatives from different fields (four Industry 4.0 managers; four
production managers; three IT managers; two CEOs; two HR managers; one logistics
manager; and one representative of an education facility). Additionally, in Czechia and
Hungary, five representatives of trade unions and employers at sectoral level were also
interviewed. Semi-structured interviews, based on a pre-defined questionnaire, lasted
between sixty and ninety minutes and were recorded following the consent of the
respondent. Interviews were conducted in the course of 2018 in Hungary and Poland
and in 2018 and 2019 in Czechia. Building on the 39 interviews we conducted in the
three Visegrád countries, we have also conducted three focus groups with 13 employees
working at shop floor level in internal logistics and on assembly lines in two Czech tier
one suppliers.
2. Data collection and analysis was conducted by Monika Martišková in Czechia, Kristóf Gyódi and Katarzyna
Śledziewska in Poland, and Andrea Szalavetz in Hungary.

Table 2 Respondents in our research from company level
Type of respondent and country OEM TIER1 TIER2 Total Tota

CZ 5 9 2 16
Education facility employees 1 1
Industry 4.0 manager 1 1 2
IT manager 1 1
Logistics manager 1 1
TU representative 3 6 2 11
HU 3 8 11
CEO 1 1
HR manager 2 2
Industry 4.0 manager 2 2
IT manager 2 2
Production manager 1 1
TU representative 2 1 3
PL 4 3 7
CEO 1 1
Production manager 2 1 3
TU representative 2 1 3
Total 12 20 2 34
Source: own data
During the period of our research, between 2018 and 2019, the majority of the companies
which we visited were expanding their production capacities while simultaneously
introducing labour-saving technologies. The trend of absolute decreases in the number
of workers was observed in only some of the companies in our sample: out of the 28
companies, only three experienced a decrease in employment between 2013 and 2018
while in 19 employment increased by more than five per cent (see Table 3). Many of
these increases at company level are attributed to the inflow of foreign workers into
the countries. In Czechia, the number of foreign workers reached 580,000 while, in
manufacturing positions (ISCO 8 – Plant and machine operators and assemblers),
their number tripled between 2014 and 2018 from 42,000 to 124,000 (see Figure 2).
In Hungary, the number of foreign workers grew from 140,000 to 180,000 between
2013 and 2018;3 while, in Poland 400,000 foreign workers were registered in 2018.4
As a result, we could not observe that new technologies had resulted in a displacement
effect in the period under study because, despite such introductions, the expansion of
production had led to increases in employment levels in most of the companies.
3. https://www.ksh.hu/docs/eng/xstadat/xstadat_annual/i_wnvn001b.html.
4. https://udsc.gov.pl/400-tys-cudzoziemcow-z-waznymi-zezwoleniami-na-pobyt/

Monika Martišková
Table 3 Revenues and employment changes in the sample companies
Global production network position Country Employees in 2018 Average change between 2013 and 2018
OEM CZ 3,287 0.0%
OEM CZ 2,248 0.3%
OEM CZ 22,932 4.8%*
OEM PL 1,876 -5.5%**
OEM PL 8,020 2.9%**
OEM PL 3,245 8.8%**
OEM HU 3,550 6.9%
OEM HU 1,251 10.7%
OEM HU 11,803 4.3%
TIER1 CZ 3,764 10.4%
TIER1 CZ 1,396 15.8%
TIER1 CZ 1,141 9.1%
TIER1 CZ 9,000 6.8%*
TIER1 CZ 726 6.2%*
TIER1 CZ 1,872 2.6%
TIER1 CZ 1,908 9.2%
TIER1 PL 8,499 15.9%
TIER1 PL 7,183 10.0%**
TIER1 PL 1,219 15.0%
TIER1 HU 3,731 5.7%
TIER1 HU 4,827 7.2%
TIER1 HU 266 -10.1%
TIER1 HU 1,343 16.3%
TIER1 HU 2,394 14.9%
TIER1 HU 1,033 5.1%
TIER1 HU 718 7.3%
TIER2 CZ 203 -2.4%
TIER2 CZ 848 7.9%
* Averages between 2013 and 2017

** Change only between 2017 and 2018
Source: own compilation based on data from companies’ final accounts for Czechia, Hungary and Poland

Figure 2 Employment of foreign workers in Czechia
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
2004 2005 2006 2007 2008 2009 2010 2011 2014 2015 2016 2017 2018
Total EU Other
Source: Czech statistical office, https://www.czso.cz/csu/czso/foreigners-in-the-czech-republic-uvmo2pjmg2
3.1 Deskilling effect of new technologies

In this sub-section, we aim to assess the deskilling effect of the introduction of new
technologies in the companies in our sample. Here, we will focus on changes in physical
difficulty and task complexity, in job cadence and in the level of control to which
employees are subject.
Respondents confirmed in our interviews that work automation had a positive impact
on the physical difficulty of manual labour. In painting and welding shops in particular,
significant improvements were reached through the introduction of new technologies.
On production lines, the partial automation of the most demanding and difficult tasks
occurred, e.g. in lifting heavy parts or improving ergonomics. A decrease in the manual
difficulty of work was also recorded in internal logistics through the introduction of
automated guided vehicles (AGVs). Improved health and safety in workplaces marked
previously by physically demanding work was one of the impacts of job automation
which was most appreciated by our respondents.
Decreasing the physical demands of work was, however, outweighed in many workplaces
by increases in work cadence and the implementation of advanced lean production
principles. Respondents who were able to compare longer periods highlighted, in
particular, the increased work cadence on lines, e.g. in welding or assembly. When
examining the specific contribution of new technologies to increased workloads,
respondents revealed, however, that the recent introduction of new technologies is not
the only reason for workload increases and that this process is rather continuous, being
associated with lean production principles.
Despite the theory that new technologies are expected to leave humans mostly more
complex tasks to deal with, with repetitive and physically difficult tasks being picked

Monika Martišková
up instead by machines, our evidence on job transformation suggests that the greater
deployment of technologies in the production process does not necessary lead to
increased work complexity and workers’ upgrading but, rather, to workers being asked
to multi-task. In other words, workers being relieved from hard and repetitive work
which is now being performed by a machine does not mean that the worker experiences
upgrading; instead, he or she is now being made responsible for feeding and controlling
the positions and performance of several machines at once. For instance, in Polish
companies, respondents confirmed that the introduction of robots implied for each
employee the responsibility for a longer fragment of production and an increased
variability of the tasks that one worker should know; with, at the same time, increased
exposure to time pressures with upgrading not being the result.
An emerging issue from our interviews was the implementation of advanced control
and traceability techniques in production, with critical attitudes to increased work
pace and control also appearing in our interviews: ‘I must acknowledge, operators’
work has become more stressful than it used to be. With smart sensors measuring
every movement, it is easy to trace which operator committed a mistake. Or else, some
workers used to work faster than the average and spared some time to have a rest.
Now, this is impossible: every processing step has a predetermined processing time;
you should neither work faster nor slower.’ (Hungarian respondent)
As an illustrative example, we can examine the machine-feeding jobs which are

spreading in production areas in the companies we visited. These robots standardise
working tasks to such an extent that workers have become a mere appendix to the
machine, because the worker’s task is only to put the right components in the right
place, with the rest being done by the robot. This also contributes to a significant loss
of workers’ autonomy when performing the task, as one of our respondents suggested:
‘Operators simply monitor the equipment. If a green light is blinking, they needn’t do
anything; if, however, the light turns to red, they must follow a predetermined, simple
protocol. Otherwise, they have little autonomy to intervene: they can stop the machinery
or call the line supervisor or maintenance staff. Previously [in the early 2000s], they
were allowed to repair the fault if they had adequate skills or creative ideas; now, with
modern production machinery, it is strictly forbidden’ (Hungarian respondent). Loss
of autonomy was also highlighted by the Czech workers in our focus group. When a
problem in production occurs, workers are obliged to call technical support, which may
take some time and which decreases production pace and, potentially also, the wages
of those workers. ‘They do not allow us to touch the screen despite it seeming trivial
what the technician does there. I would appreciate to learn how to handle at least the
basics in these new robots, but we are not required to know it’ (focus group participant
in Czechia, 2019).
Our interview evidence suggests that workers are being exposed to new technologies
having deskilling effects, mostly in terms of work simplification and the loss of
autonomy within production while, at the same time, experiencing rising work takt
(the time for the production of one unit of output, measured from the start of one unit
and concluding at the start of the next) as well as requests for multi-tasking. Industry
4.0 technologies further advance the principles of lean production in the companies

adopting them, so our respondents did not perceive the changes as radical but rather as
evolutionary. This contributes to their lack of a critical attitude towards the deployment
of new technologies in production.
We will return to a discussion of the attitudes of trade unions to new technologies and
their policy responses in a later section.
3.2 Changes in skills requirements and jobs in the automotive sector

Changing working conditions will have consequences for the composition of the
workforce in terms of educational level and the complexity of work tasks. In this sub-
section, we aim to assess the impact of new technologies on job requirements and
opportunities in the automotive industry for shop floor workers. For that purpose, we
consider traditional occupations in workplaces based on the European Automotive Skills
Council (EASC) report and our own evidence in respect of: maintenance technicians;
CNC operator/tool and die makers; paint technician/motor vehicle painters; assembly
line operative/ assemblers; materials planning analysts; welders; and logistics assistants.
In the accompanying Table 4, we have provided an overview of current educational
requirements according to ISCED levels, summarising how new technologies transform
job content and how they contribute to developments in the number of jobs. In the final
column, we indicated changes in skill requirements regarding each position based on
our interviews.
Logistics assistants are considered the lowest-skilled among production workers.

They are exposed to decreasing job opportunities wherever internal logistics is being
automated. This, however, varied in the companies we visited occasionally because of
premises being unsuitable for automation (e.g. in one company respondents indicated
that outdated production premises, with steps on the floor, had prevented the introduction
of AGVs). However, where the automation of internal logistics is implemented, there is
a decreasing demand for low-skilled workers in this field; nevertheless, they are usually
relocated to other production facilities within the company and are not as yet facing
redundancy.
Welders’ work is being continuously automated, primarily because of the higher

reliability of welds performed by robots; and, secondly, because of their scarcity on the
labour market, according to statistics.5 Even though these workers might experience
decreasing job opportunities in particular companies where the share of robots is
growing significantly, welders are still professionals in demand on the labour market,
especially those who are able to use different welding techniques. Nevertheless, our
findings also suggest that task standardisation and automation have contributed to the
relative deskilling of welders. This was highlighted in two Czech companies producing
electronic components and fuel combustion engines. Both respondents observed
5. For instance, in Czechia in January 2020 there were 8,244 jobs for welders being offered by employers but only
867 registered welders were available in the register of the Employment office of Czechia (MPSV 2020, https://
www.mpsv.cz/web/cz/analyza-poptavky-po-pracovni-sile-a-nabidky-pracovni-sily).

Monika Martišková
diminishing work opportunities for medium-skilled manual workers. ‘We have

difficulties in adequately remunerating these medium-skilled people and finding them
appropriate positions in the company. Some of them have retrained as CNC machine
specialists, and some of them remain in the control and repairs department, but some
have been downgraded in their position. There is simply no position where they can
show their skills’ (trade union representative, Czechia).
Automation is undoubtedly affecting assembly line operators in automotive companies

since the majority of the companies we visited have introduced robots and/or
collaborative robots to their assembly lines. Despite many of these changes decreasing
the number of workers required for specific tasks, this has not been mirrored in an overall
decrease in the demand for workers. Furthermore, since assembly incorporates various
tasks and positions, there is no clear trend in the transformation of working conditions.
In some cases, workers have lost autonomy and their tasks have been deconstructed
following the introduction of robots, whereas others have needed to learn new roles and
gain a deeper understanding of new processes. The introduction of robots has also led
to assembly workers being exposed to increased requirements for multi-tasking when
responsibility for a longer fragment of the production line is assigned to them, but this
does not result in their upskilling.
Painting is being rapidly automated in our companies, mainly because of the dirty and
dangerous character of the activity. Nevertheless, a significant number of painters, who
undergo lifelong learning because of new technologies and materials, are still required.
Maintenance technicians are an in-demand profession on the labour market as a result

of the increasing number of machines being implemented in production. Additional
training, and even increases in the requirements for formal education, are typical of
these positions although, on the other hand, the standardisation of repair tasks arising
from the introduction of new technologies may, in some cases, contribute to the
deskilling of workers. However, the prevailing perception of such positions is that they
are in high demand and require formal qualification at least at the level of secondary
education.
The work of materials planning analysts is expected to be transformed significantly

as a result of digitalisation and the implementation of artificial intelligence (AI) in
resource planning. That is why their numbers might decrease in production, although
the requirements for educational qualifications are likely to increase.
Among new positions identified in our interviews, one respondent mentioned the role of
an ‘electrician lite’ category expected to be needed in the future in the assembly of electric
cars. This role would encompass a more narrowly-defined skillset for electricians, but
would not require full qualification of electricians. This might have significant impact
on demand for electricians, formerly the most skilled among manufacturing workers.
There is also increasing demand for various engineers and programming specialists.

Table 4 Changing skills requirements and job opportunities
Traditional occupations Minimum Observed changes Impact of Changes in

in the automotive sector education level based on interviews automation on educational
required labour demand requirements
Logistics assistant Primary — If logistics is automated, the Decreasing No changes

education position disappears (workers
(ISCED 1) mostly transferred to other
low-skilled positions)
Welder Upper secondary — Welding being automated Expected to Increasing,

education continuously; ongoing decrease, but because of various
(ISCED 3 + automation in lower tier currently in techniques
compulsory suppliers demand, driven applied in welding
training) — Many welders need to retrain by production
or else experience deskilling increases
Assembler Secondary — If assisting robots, limited Expected to Increasing for

vocational value added decrease, but more specialised
education — Decreasing physical demands currently in positions
(ISCED 2 and 3) of work demand, driven BUT decreasing
— Decreasing error rates as a by production for ordinary
result of new technologies, increases operators
facilitating the employment of in robotised
lesser-skilled personnel workplaces
— Increasing work pace
— Increasing complexity of
work because of increased
variability of production
— In some cases, the need to
understand the basics of
machine operation
— Increasing importance of
product quality checks
Paint technician Secondary — Automation has brought Decreasing Remaining

vocational significant improvement in the same or
education working conditions increasing, with
(ISCED 3) — Workers remain at the more advanced
workplace or are transferred to positions in
other workplaces handling painting
machinery
Maintenance technician Secondary — Predictive maintenance Increasing Increasing

vocational requires reskilling and (because more (acquired through
education retraining as there is a need machines are training and/or
(ISCED 3 and for better understanding of deployed in formal educational
4); advanced maintenance (but, in some production) qualifications)
positions also cases, technologies may make
ISCED 5 the work easier, especially if
some maintenance tasks are
standardised)
CNC operator/tool and Secondary — Tool prototypes in 3D printing Steady Increasing

die maker vocational require new knowledge (acquired through
education (mostly in OEMs) training)
(ISCED 3 and 4) — Or require retraining on
upgraded machines

Monika Martišková
Traditional occupations Education level Observed changes Impact of Changes in

in the automotive sector required based on interviews automation on educational
labour demand requirements
Materials planning University — Increased educational Decreasing Increasing

analyst education qualifications needed as a (acquired through
(ISCED 6 and 7) result of SAP (introductions training and/or
of ERP) and expected AI formal educational
elements implemented in qualification)
planning
New positions
‘Electrician lite’ Post-secondary — New category of ‘electrician Expected to -
education (from lite’ needed if mass electric car emerge
ISCED 4) production emerges
Engineering positions From short- — Creative new thinking Increasing Increasing

cycle tertiary expected (acquired through
education — Importance of understanding formal educational
(ISCED 5) possibilities and their effective qualification)
implementation in practice
Programming specialists From short- — Importance increases in Increasing Increasing
cycle tertiary coordinating all production (acquired through
education functions, understanding formal educational
(ISCED 5) processes, suggesting qualification)
solutions
Source: own compilation based on interviews and EASC report (2018)
Our observations tell us that educational requirements are changing for most of these
positions and retraining is needed. However, the level of retraining provided is usually
not enough to deliver an upgrading of workers’ skills. Moreover, retraining policies in
the companies in our sample differ significantly. At OEMs, retraining policies have been
developed and offer opportunities to keep pace with new technologies and professional
development but, in lower tier suppliers, depending on the training system present in
the company, shop floor workers have either received short training on the use of a new
machine or, sometimes, not even that: respondents in some companies (in Poland and
Czechia) revealed that there is no time for training at all.
Interestingly, workers’ access to reskilling and retraining is, based on our observations,
highly dependent on the individual’s cognitive skills and their interest in learning and
their ability to learn: ‘When someone´s task is robotised, he or she can attain training
which will allow them to operate the machine at the basic level, but this still presupposes
some interest in understanding new technologies by that worker’ (employee in an OEM
training facility, Czechia).
‘This is difficult; much is dependent on the individual’s cognitive skills and ability to learn.
Otherwise, with new technologies, people are either required to learn new information
or, if not, are increasingly becoming appendices to the machines’ (tier one trade union
representative, Czechia). This mostly relates to assembly positions where robotisation
might lead both to upgrading and to downgrading. While the former is substantially
possible, given individuals’ interest in learning and a company’s retraining schemes in
practice, the latter may be observed in the case of older workers and among employees
less interested in reskilling and/or retraining. ‘Older people, but not necessarily

inexperienced ones, are placed in these machine-feeding positions. You can see them
there, where they are basically taking a rest while receiving the same wage as when they
worked in a more difficult job position’ (tier one trade union representative, Czechia). In
general, trade union representatives perceived robotised workplaces offering ‘machine-
feeding jobs’ as advantageous for workers and were not pushing for general retraining
and reskilling; neither did they oppose these types of positions.
On the other hand, respondents reported increasing requirements for attained education
by employees, basically in all the traditional positions recognised in Table 4. Interviewees
in Hungary confirmed that workers need a deeper understanding of production processes.
‘Operators today are not your grandfather’s blue collar factory workers! There are more
engineers on the shop floor than simple manual workers. To get hired, you would at
least have to possess a general certificate of secondary education or, rather, a certificate
of vocational education in automotive technology. You can find manual workers mainly
in in-plant materials transportation and in machine loading, or elsewhere, in assembly
firms.’ Evidence from a Polish company suggests that workers in automated production
are also facing an increase in responsibilities since they need to ensure production in
more complex processes in which human error is much more costly.
Various managers also pointed out that a better skilled workforce is significantly lacking,
although it is not always clear what employers mean by a ‘better skilled’. In many cases,
it might be simply the manual skills required on assembly lines although, regarding
new technologies, some managers explicitly revealed that they need workers who are
sufficiently educated as to recognise the possibilities of new technologies and to design
their efficient implementation in production processes. Nevertheless, a lack of better
skilled workers, in both senses, is considered a major obstacle to companies’ upgrading.
At the same time, in the companies we visited, incentives and complex reskilling and
retraining schemes that would allow employees to ‘upgrade’ from low-skilled shop floor
workers to a skilled workforce able to work with new technologies are missing. In our
sample, in only one case in Poland have obsolete blue collar workers at an OEM been
retrained to carry out office jobs. In general, this transition from shop floor to office jobs
is scarce and retraining schemes do not encourage this form of transformation.
4. Trade union responses

The purpose of our interviews was also to understand trade union strategies at company
level as regards the changing working conditions associated with automation and
robotisation in workplaces. In our interviews, trade union representatives articulated
the mostly positive impacts of automation while only a few revealed negative impacts on
working conditions and workers’ prospects. All the respondents had experience of the
introduction of new technologies, in production in particular, although the extent of the
experience obviously differed depending on their company’s position in the production
chain, with OEMs and tier one suppliers being leaders in implementation. Nevertheless,
all trade union respondents perceived machines and robots as representing continuous
improvements drawn from lean production processes and related to the production of
new and upgraded products.

Monika Martišková
Given the dependent position in global production chains of local companies in all three
countries, any new investment in equipment and in improvements to the production
process was perceived by respondents as a confirmation of the company’s intention
to remain in the location and the ability of local management to propose changes and
attain some level of upgrading. From this perspective, trade unions did not oppose
the implementation of new technologies, nor did they question the changing working
conditions of workers.
Moreover, the use of various robots, sensors and visualisations on assembly lines that
are aimed at decreasing workers’ error rates, but which also shrink their autonomy, were
welcomed by our respondents. Trade union representatives interpreted automation and
task standardisation as tools to help workers perform their work better. In the context of
local remuneration systems, perfect work performance is an important issue for workers
for whom, on average, twenty per cent of the wage is dependent on performance. At
the same time, this positive attitude must be interpreted in the context of the severe
labour shortages being experienced by companies in the automotive sector. Trade
unions considered that the introduction of robots highlighted the increasing workloads
to which the remaining workers were exposed as a result of absent colleagues.
Our research also revealed that trade unions possess limited rights of co-determination
when it comes to the introduction of new technologies in our companies. Also, they
gained only limited information from management about the intended changes. The
majority of trade unions claimed they had the chance of information and consultation on
changes with the management, but almost everywhere only following their own request
and often after the decision had been made. Despite the ability to acquire knowledge of
what is going on, this has a purely informative role and there is no possibility to reverse
it through bargaining, while pressure in the form of protests and strike action has not
been recorded so far.
For some trade unions, European works council meetings have proved to be a good
source of information about future changes at global level although these, however, often
remain confidential until actually implemented. Not only are trade union members not
able to reveal this information beforehand, again, neither can they subsequently reverse
it (e.g. if a global announcement on the introduction of technologies is announced, or
when an announcement is made about planned redundancies in some locations).
As we have observed, trade unions have not thus far developed a comprehensive
strategy on how to approach new technologies and workers’ reskilling. Trade union
representatives claimed that their primary role is to protect workers against lay-offs,
while upgrading is not on their agenda either at sectoral or at company level. Up to now,
trade unions have applied standard strategies to protect workers which encompass the
management of redundancies through retirements and through voluntary leavers and,
where involuntary lay-offs are inevitable, they try to apply careful and clear criteria. At
the same time, they are paying increased attention to workers’ protection in terms of the
health and safety aspect of working conditions in the workplace, while paying limited
attention to deskilling and retraining strategies at company level.

Interestingly, some of our respondents claimed that they do not want workers to be
pushed towards reskilling, especially older people, i.e. of pre-retirement age, who
compose a significant proportion of trade union members in many companies. In the
case of younger people, they leave it up to individuals’ decisions regarding their interest
in retraining and their ability to benefit from it. Trade union leaders encourage the
participation of selected individuals in retraining, but mostly those who ‘seem to be
capable’, supporting the individual self-selection of workers for reskilling. As a result,
trade union representatives do not recognise the need to conduct comprehensive
training policies for manual workers at company level, nor do they demand that
employers introduce them. At the same time, employers have proven to be reluctant
to introduce any retraining and reskilling strategies into the collective agreement.
According to our evidence, in only one OEM did a trade union propose this as a topic
for collective bargaining, but it was unsuccessful.
What we have observed at company level is applicable to upper levels as well. When it
comes to sector-level strategies on how to cope with employers increasing educational
requirements and decreasing demand for manual labour, trade unions are not even
involved in the discussion on strategies about the future of the automotive industry in
their respective countries. The Action Plan on the future of the automotive industry
for Czechia,6 for instance, does not indicate any development strategies for current
employees. The Plan addresses the transformation of education programmes in
schools, based on the needs of employers, but the reskilling of workers already on the
labour market is not even mentioned. Similar may be observed at company level, with
employers expecting newcomers to be better trained than their current employees, while
the training of employees is carried out at a level which allows them to keep pace with
current technologies but which does not encourage them to upgrade once employed in
the company.
5. Conclusion: Towards a policy response

Through our research in Czechia, Hungary and Poland we have gathered evidence
which suggests that new technologies are having an impact on working conditions in
terms of changing job opportunities and in job requirements. However, we did not
observe them having any displacement effect, either at company or at sectoral level,
mostly because of labour shortages in the region and increases in production capacities
in many companies. Nevertheless, our research confirms the deskilling effect of the
introduction of new technologies as regards manual workers, as well as the increasing
demand for skilled workers.
Compared to the findings of a similar investigation by Matteo Gaddi into Italian

companies presented in chapter 7 (see also Gaddi et al. 2018), our results suggest that
workers and trade union representatives in CEE countries do not consider changes
6. (https://www.mpo.cz/cz/prumysl/zpracovatelsky-prumysl/automobilovy-prumysl/memorandum-o-
budoucnosti-automobiloveho-prumyslu-v-cr-a-akcni-plan-o-budoucnosti-automobiloveho-prumyslu-v-
cr---232552/ page 36, measure P3).

Monika Martišková
associated with the introduction of new technologies to be something which worsens

working conditions even though, in Italian companies, the level of angst about new
technologies is much more pronounced. It is important to relate this more limited
perception of ongoing changes among our stakeholder interviewees to the context in
which CEE countries find themselves. Here, the introduction of new technologies is
perceived above all as an expression of the owners’ intention to remain in the location,
which is important for subsidiaries in the integrated peripheries of global production
networks. Second, the motivation to invest in new technologies is driven by the lack of an
available labour force which, again, contributes to an appreciation of the role of labour-
saving technologies in production as well as to the lack of evidence of new technologies
having a displacement effect in our companies. Third, the limited recognition of the
negative impacts of the introduction of new technologies is blocking more critical
appraisals and the creation of mitigation strategies at company, regional and national
level. Moreover, in the Italian case, the increasing level of control over the tasks done by
workers is pronounced and can be demonstrated with several detailed examples, while
these aspects were mentioned only rarely by our respondents. This also suggests that
there are different levels of advance in the introduction of new technologies between
Italy and the companies in the countries we studied.
Trade unions have, up to now, not developed strategies to tackle the reskilling of possibly
redundant workers. The only way that workers attain the upgrading of their skills lies
in self-selection and individual plug-ins into company-level retraining. However, in
the future, if workers are considered redundant, their prospects of reskilling will be
determined only by their individual ability to seek retraining and to finance it if no
policies at sectoral level, based on cooperation between employers and the involvement
of public institutions, have been developed by that point. This might contribute to the
further polarisation of job skills, leaving behind older workers, the less flexible and
the young and inexperienced who might have difficulties in establishing retraining or
reskilling opportunities.
The temporary circumstances concerning the labour markets of CEE countries are also
contributing towards a postponement of policy responses regarding the introduction of
new technologies. There is no effort to prevent structural unemployment because, for
now (in 2019), very few people remain unemployed and thus the urgency of the need
to conduct reskilling programmes is very low. Moreover, efficient retraining policies
are simply not on the agenda of trade unions and other stakeholders. This comprises
a striking question of the prospects for manual workers in the labour markets of the
future in which digitalisation and automation will be highly developed and where jobs
for many manual workers will have diminished. The expected consequence is that these
workers will face unemployment with limited likelihoods of finding new jobs. Moreover,
the COVID-19 pandemic may accelerate most of these processes and CEE region will
face structural unemployment in very near future.
This might remind us of the similar development which CEE countries underwent in the
1990s, when the collapse of the Soviet Union caused a rapid increase in unemployment
rates. The main cure that CEE countries applied at that time was to attract foreign investors
who created low and medium-skilled manufacturing workplaces. Unemployment rates

thereby decreased without any significant effort to encompass reskilling strategies, nor
even to talk about industry strategies towards this. This, in turn, reduced the scope
for public policies to deal with unemployment, confining this to the provision of
support for foreign investors through generous state subsidies. However, in contrast to
previous experience, an escape strategy in the form of foreign investors creating manual
labour-intensive workplaces may not be available in the expected forthcoming wave of
structural unemployment. Such changes will require more comprehensive strategies
on how to bring the unemployed back to the labour market and this is something that
actors in the CEE region will have to learn from scratch.
References
Acemoglu D. and Restrepo P. (2017) Robots and jobs: evidence from US labor markets, Working
Paper Series 23285, Cambridge, MA., NBER.
Agnew A., Forrester P., Hassard J. and Procter S. (1997) Deskilling and reskilling within the labour
process: the case of computer integrated manufacturing, International Journal of Production
Economics, 52 (3), 317-324.
Arnold D., Arntz M., Gregory T., Steffes S. and Zierahn U. (2018) No need for automation angst, but
automation policies, in Neufeind M., O´Reilly J.and Ranft F. Work in the digital age: challenges
of the fourth industrial revolution, London, Rowman & Littlefield, 75-87.
Atkinson R. D. (2018) Which Nations really lead in industrial robot adoption?, Information
Technology & Innovation Foundation. https://ssrn.com/abstract=3324659.
Attewell P. (1987) The deskilling controversy, Work and occupations, 14 (3), 323-346.
Bamber, G. (1989) Unions and new technology, in Gladstone A., Landsbury R, Stieber J., Treu T. and
Weiss M. (eds.) Current issues in labour relations: an international perspective, Berlin, Walter de
Gruyter GmbH & Co,17-28
Benhamou S. (2018) The world of work in 2030: four scenarios, in Neufeind M., O´Reilly J. and
Ranft F. Work in the digital age: challenges of the fourth industrial revolution, London, Rowman
& Littlefield, 259-270.
Bonekamp L. and Sure M. (2015) Consequences of Industry 4.0 on human labour and work
organisation, Journal of Business and Media Psychology, 6 (1), 33-40.
Brynjolfsson E. and McAfee A. (2012) Thriving in the automated economy, World Future Society,
27-31.
Butollo F., Jürgens U. and Krzywdzinski M. (2019) From lean production to industrie 4.0: more
autonomy for employees?, in Meyer U., Schaupp S. and Seibt D. (eds.) Digitalization in Industry:
between domination and emancipation, Springer, 61-80.
Cappelli P. (2004) Why do employers retrain at risk workers? The role of social capital, Industrial
Relations: A Journal of Economy and Society, 43 (2), 421-447.
Chiacchio F., Petropoulos G. and Pichler D. (2018) The impact of industrial robots on EU
employment and wages: a local labour market approach, Working Paper 2018/02, Brussels,
Bruegel.
Chmelař A. et al. (2015) Dopady digitalizace v Česku [Impacts of digitalisation in Czechia], OSTEU
Discussion paper 2/2015. https://www.vlada.cz/assets/evropske-zalezitosti/analyzy-EU/
Dopady-digitalizace-na-trh-prace-CR-a-EU.pdf.

Monika Martišková
Crouch C. (2018) Redefining labour relations and capital in the digital age, in Neufeind M.,
O´Reilly J. and Ranft F. Work in the digital age: challenges of the fourth industrial revolution,
London, Rowman & Littlefield, 187-197.
Dauth W., Findeisen S., Südekum J. and Woessner N. (2017) German robots - the impact of
industrial robots on workers, CEPR Discussion Paper DP12306, Washington, DC , CEPR. https://
ssrn.com/abstract=3039031.
EASC report (2018) European sector skill council. automotive industry: Report. https://ec.europa.
eu/social/BlobServlet?docId=18795&langId=en.
Engblom S. (2019) Employment protection, collective bargaining and labour market adaptability.
the Swedish employment security councils in a comparative context, Giornale di Diritto del
Lavoro e di Relazioni Industriali, (160), 887-905.
Ertürk K. A. (2019) Where did good jobs go? Acemoglu and Marx on induced (skill replacing)
technical change, Working paper 2019-02, University of Utah, Department of Economics.
Gaddi M., Garbellini N. and Garibaldo F. (eds.) (2018) Industry 4.0 and its consequences for work
and labour: field research report on the implementation of Industry 4.0 in a sample of Italian
companies, Bologna, Fondazione Claudio Sabattini.
Hall B. H. and Rosenberg N. (2010) Introduction to the handbook, Handbook of the economics of
innovation, 1, 3-9. https://doi.org/10.1016/S0169-7218(10)01001-4
Hirsch-Kreinsen H. (2016) Digitization of industrial work: development paths and prospects, Journal
for Labour Market Research, 49 (1), 1-14.
ILO (2019) Global Commission on the future of work: work for a brighter future, Geneva, ILO.
Jolly C. (2018) Collective action and bargaining in the digital era, in Neufeind M., O´Reilly J. and
Ranft F. Work in the digital age: challenges of the fourth industrial revolution, London, Rowman
& Littlefield, 209-222.
Kagermann H. (2015) Change through digitization: value creation in the age of the industry 4.0,
in Albach H., Meffert H., Pinwart A. and Reichwald R. (eds.) Management of permanent change,
Wiesbaden, Springer Fachmedien, 23-45.
Keister R. and Lewandowski P. (2017) A routine transition in the digital era? The rise of routine
work in Central and Eastern Europe, Transfer, 23 (3), 263-279.
Kergroach S. (2017) Industry 4.0: new challenges and opportunities for the labour market, Foresight
and STI Governance, 11 (4), 6–8. DOI: 10.17323/2500-2597.2017.4.6.8.
Komlos J. (2016) Has creative destruction become more destructive?, The BE Journal of Economic
Analysis & Policy, 16 (4), 1-12.
Körner U., Müller Thur K., Lunau T., Dragano N., Angerer P. and Buchner A. (2019) Perceived stress
in human-machine interaction in modern manufacturing environments – Results of a qualitative
interview study, Stress and Health, 35 (2), 187-199. https://doi.org/10.1002/smi.2853.
Krzywdzinski M. (2019) Globalisation, decarbonisation and technological change: challenges for the
German and CEE automotive supplier industry, in Galgóczi B. (2019) Towards a just transition:
coal, cars and the world of work, Brussels, ETUI, 215-241.
Kunst D. (2019) Deskilling among manufacturing production workers, Discussion Paper 2019-050/
VI, Amsterdam, Tinbergen Institute Amsterdam.
Nedelkoska L. and Quintini G. (2018) Automation, skills use and training, OECD Social, Employment
and Migration Working Papers 202, Paris, OECD Publishing.
Novta N. and Pugacheva E. (2018) Manufacturing jobs: implications for income inequality, LIS
Working Paper Series 736, Luxembourg, Luxembourg Income Study.


Journal of Economic Geography, 20 (2), 509-541.
Pereira A. C., and Romero F. (2017) A review of the meanings and the implications of the Industry
4.0 concept, Procedia Manufacturing, 13, 1206-1214.
Porter M. E. and Heppelmann J. E. (2014) How smart, connected products are transforming
competition, Harvard Business Review, (11), 1-23.
Soete L. (2018) Destructive creation. Explaining the productivity paradox in the digital age. in
Neufeind M., O´Reilly J.and Ranft F. Work in the digital age: challenges of the fourth industrial
revolution, London, Rowman & Littlefield, 29-47.

List of contributors
Ricardo Aláez-Aller is senior lecturer in economics at the Public University of

Navarre in Spain. His research focuses on wage differences, the automotive industry
and economic policy.
Carlos Gil-Canaleta is lecturer in economics at the Public University of Navarre. His

publications include contributions on regional economics and public administration.
Jan Drahokoupil is a senior researcher at the ETUI where he coordinates research

on digitalisation and the future of work. His research also covers the political economy
of international business, the organisation and strategies of multinational corporations
and foreign direct investment.
Matteo Gaddi is a member of the Scientific Committee of the Claudio Sabattini

Foundation and a trade union official at CGIL in Reggio Emilia. He carries out research
and training activities on the themes of work organisation, new technologies, and
economic and industrial policies. Among his recent publications are: ‘Industry 4.0:
freer or more exploited?’ (2019) and ‘Automotive and electric mobility’ (2019).
Krzysztof Gwosdz is associate professor in the Institute of Geography and Spatial

Management of Jagiellonian University in Kraków. His research interests are focused
on issues of local and regional development, foreign direct investment, the social and
economic problems of urban areas, and the restructuring and regeneration of old
industrial towns and regions.
Agnieszka Świgost-Kapocsi is assistant professor at Jagiellonian University. Her

research is focused on regional and local development with a particular emphasis on the
role of women. She is additionally interested in industrial cities, urban geography and
social participation (local initiatives and innovations).
Arkadiusz Kocaj is a researcher in the Institute of Geography and Spatial Management

of Jagiellonian University. His research agenda focuses on economic geography, urban
revitalisation and the geography of manufacturing and traditional industries. He is now
working mostly on medium-tech industries (domestic appliances and the possibilities
for upgrading in semi-peripheral countries) and the fashion industry (its relationship
with sustainability growth).
Juan-Carlos Longas-Garcia is senior lecturer in economics at the Public University

of Navarre. His research interests are the automotive industry, regional economics and
policy, wage differences and industrial organisation.

List of contributors
Monika Martišková is a researcher at the Central European Labour Studies Institute

(CELSI) in Bratislava, Slovakia, and a PhD candidate in the Department of Social
Geography and Regional Development of Charles University in Prague, Czechia. Her
research interests lie in working conditions and industrial relations in the automotive
sector in CEE countries.
Grzegorz Micek is associate professor in the Institute of Geography and Spatial

Management of Jagiellonian University. Currently, he heads a research project funded
by the Polish National Science Centre on the local impact of co-working spaces. He
participated in several EU projects (including the 6th FP project on the delocalisation of
labour-intensive industries). He is a Fellow of the Regional Studies Association.
Pamela Meil is a sociologist of work and a senior research fellow at the Institute for
Social Science Research (ISF München). Her current research includes the impact
of work organisation and skill on hybrid value added systems, the development of a
European perspective on work and employment conditions for slash workers and the
effects of Industry 4.0 and digital transformation on value chains.
Agnieszka Sobala-Gwosdz is associate professor in the Institute of Technical

Engineering at the Bronisław Markiewicz State Higher School of Technology and
Economics in Jarosław and a senior researcher in the Institute of Urban and Regional
Development. Her research interests cover issues of local and regional development
and she has undertaken research mainly in peripheral and frontier regions, especially in
southeastern Poland. She focuses particularly on the growth poles theory, foreign direct
investment and urban regeneration.
Andrea Szalavetz works as research advisor at the Institute of World Economics,

Budapest. Her main research fields include the economics of innovation, digital
transformation, Industry 4.0 and global value chains. Her publications are available at
www.szalavetz.com.
Miren Ullibarri-Arce is lecturer in economics at the Public University of Navarre in

Spain. Her main research interest is wage differences, with a particular stress on the
gender wage gap.

The Challenge of Digital Transformation in The Automotive Industry-2020

Uploaded by

Copyright:

Available Formats

The Challenge of Digital Transformation in The Automotive Industry-2020

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Challenge of Digital Transformation in The Automotive Industry-2020

Uploaded by

Copyright:

Available Formats

The challenge of digital

The headquarters perspective .............................................................................................................................. 23

Digital transformations in factory economies ............................................................................................. 45

Krzysztof Gwosdz, Grzegorz Micek, Arkadiusz Kocaj, Agnieszka Sobala-Gwosdz and

The challenge of digital transformation in the automotive industry 5

List of contributors ..................................................................................................................................................... 177

6 The challenge of digital transformation in the automotive industry

The digital transformation of production is associated with so-called Industry 4.0

The challenge of digital transformation in the automotive industry 7

extent to which digital transformation represents an opportunity, or a challenge, for

The automotive industry in Europe is characterised by a division of labour – organised

Table 1 Case studies

MNC headquarters (Chapter 2)

8 The challenge of digital transformation in the automotive industry

address the competitiveness challenge: does digital transformation undermine or

1. Automotive production networks, European industrial

The challenge of digital transformation in the automotive industry 9

In Europe, Germany represents the key core location in automotive production

Table 2 Automotive: value added and employment, 2017

Share in non-financial business economy,* % Share of components in automotive, %

10 The challenge of digital transformation in the automotive industry

Production networks in peripheral locations have a dual structure, with foreign

The challenge of digital transformation in the automotive industry 11

as the EU is concerned. Peripheral producers exhibit much lower levels of automation.

Table 3 Number of multipurpose industrial robots per 10,000 people

2018 Growth 2013-2018

12 The challenge of digital transformation in the automotive industry

Table 4 Number of multipurpose industrial robots per 10,000 people employed

Absolute As per cent of German level

The use of industrial robots, however, is an imperfect indicator of technologies discussed

The challenge of digital transformation in the automotive industry 13

Notes: countries covered in this book

14 The challenge of digital transformation in the automotive industry

Figure 1 provides some examples of the technologies, classified according to Industry

Figure 1 Stages of digital maturity: examples of investments in digital technologies

5 Completely automated factory

¡ Production control through cyber-physical systems

¡ Connection of value adding components; digital monitoring

¡ Advanced solutions, but isolated and co-existing with legacy technology

1 No Industry 4.0: factory automation using older generations of fenced robots

Source: adapted from Andrea Szalavetz (Chapter 3)

The challenge of digital transformation in the automotive industry 15

A position in the periphery represents a constraint on the adoption of new technologies.

2. The competitiveness challenge

16 The challenge of digital transformation in the automotive industry

We observed the decentralisation of advanced functions, as anticipated in the optimistic

The challenge of digital transformation in the automotive industry 17

3. The jobs challenge

There were no redundancies directly related to the introduction of Industry 4.0

18 The challenge of digital transformation in the automotive industry

4. Concluding remarks: The Covid-19 challenge

The challenge of digital transformation in the automotive industry 19

20 The challenge of digital transformation in the automotive industry

The challenge of digital transformation in the automotive industry 21

All links were checked on 08 June 2020.

Germany, in particular, has placed a high emphasis, backed up by large investments, in

1. A world of global value chains

The challenge of digital transformation in the automotive industry 25

The other contributions in this collection examine developments in new digital

The automotive industry, dominated by foreign-controlled, export-oriented manu