See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/341122026

The Performance of Port Clusters

Article  in  Journal of International Logistics and Trade · September 2004

DOI: 10.24006/jilt.2004.2.1.47

CITATIONS READS

3 98

1 author:

Peter W. De Langen
Copenhagen Business School
109 PUBLICATIONS   3,970 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Ocean container and intermodal hinterland logistics View project

Ports & regional development View project

All content following this page was uploaded by Peter W. De Langen on 22 December 2020.

The user has requested enhancement of the downloaded file.

JOURNAL OF INTERNATIONAL LOGISTICS AND TRADE 47
Volume 2, Number I, 2004, pp. 47-56

The Performance of Port Clusters

Peter W. de Langen·

ABSTRACT

This paper deals with the performance of port clusters. Port clusters are
analyzed using a framework that draws from different schools that deal with
clusters (see De Langen, 2004). Central to the framework is the identification of
eight variables of cluster performance. Four of those-agglomeration and dis-
agglomeration forces, internal competition, heterogeneity of the cluster and the
level of entry and exit barriers-are related to the structure of a cluster and four-
the presence of trust, the presence of intermediaries, the presence of leader firms
and the quality of collective action regimes-are related to the governance of
clusters. The validity of these variables is confirmed in three case studies, of the
port clusters of Rotterdam, Durban, and the lower Mississippi. The strengths and
weaknesses of the three port clusters, the importance of the variables discussed
above and opportunities for policy and management to improve the performance of
clusters are discussed. The results of this study are relevant for cluster scholars
and for scholars specializing in port studies and, since implications of this study
for policy and management in (port) clusters are discussed, the study is also
relevant for (port) cluster managers and for managers affirms in (port) clusters.

Keywords: Agglomeration, Clustering, Durban, Port Clusters, Rotterdam,

Lower Mississippi Ports.

I. INTRODUCTION

Economists frequently use the cluster concept to analyze the economic

development of regions. Because of increasing interaction and competition between
regions, the specialization of regions increases. This leads to the development of
clusters: spatial concentrations of interrelated firms. Competitive clusters can
contribute substantially to the economic development of regions. Therefore, the
question 'what determines the performance of clusters' is relevant.
The cluster concept is frequently applied, but hardly ever to seaports, in spite of
the fact that seaports are clear examples of clustering. Virtually all ports attract fmns
related to the arrival of goods and ships.
The subsequent paper consists of three parts. Part ll briefly discusses insights on
the performance of clusters, derived from various schools and presents a framework

· Port Economist, Department of Regional, Port and Transport Economics, Erasmus University
Rotterdam P.O. Box 1738 3000 DR Rotterdam, The Netherlands. E-mail: [email protected]
48 PETER W. DE LANGEN

to analyze clusters. Then Part III applies these insights to seaports and Part IV
presents conclusions.

II. A FRAMEWORK TO ANALYZE CLUSTER PERFORMANCE

A widely accepted theory on relevant variables of cluster performance is lacking.

At least four relevant schools provide insights that can be are used to develop a sound
analytical framework for examining cluster performance (see De Langen, 2004). The
variables for the performance of clusters are derived from these four schools.
A distinction is made between 'governance variables' and 'structure variables'.
The first group includes all variables that are directly related to the behavior of
organizations in the cluster; and the second group includes all variables for which this
is not the case. Four 'structure variables' -agglomeration and dispersion forces,
internal competition, cluster barriers and cluster heterogeneity-and four governance
variables-trust, leader firms, intermediaries and collective action regimes-are
identified. The effects of all these variables on the performance of clusters are
summarized in the Tables 1 and 2.

Table 1. The Effects of Cluster Structure on Cluster Performance

Element of cluster Effect on cluster performance

structure
A shared labor pool attracts firms to the cluster.
Agglomeration The presence of customers and suppliers attracts ftrms to the cluster.
economies The presence of knowledge (spillovers) attracts firms to the cluster.
(see Krugman, 1991) Land scarcity and high land prices 'disperse' firms from the cluster.
Con~estion disperses firms from the cluster.
Internal Internal competition prevents monopoly pricin~.
competition Internal competition leads to specialization.
(see Porter, 1990) Internal competition promotes innovation.
Entry barriers (such as inaccessible networks) and start-up barriers (such as non-
availability of local venture capital) reduce competitive pressure and prevent the inflow of
Cluster barriers (human) capital.
Exit barriers (such as 'sticky labor' and cluster specific investments) reduce uncertainty
for firms in the cluster.
Cluster Cluster heterogeneity enhances opportunities for innovation.
heterogeneity Cluster heterogeneity enhances opportunities for cooperation.
(see Metcalfe, 1994) Cluster hetero~eneity reduces vulnerability for external shocks.
THE PERFORMANCE OF PORT CLUSTERS 49

Table 2. Variables for the Quality of Cluster Governance

Elements of cluster Effects on cluster performance

governance
The presence of trust Trust lowers coordination costs because costs to specify contracts decrease.
(see Nooteboom, Trust increases the scope of coordination beyond price, because the risk of free
1999) riding decreases.
The presence of Intermediaries lower coordination costs and increase the scope of coordination
intermediaries (see beyond price because they specialize in managing coordination.
Nooteboom, 1999)
The presence of Leader firms generate positive external effects for firms in their network, mainly
leader firms (see by encouraging innovation and promoting internationalization.
De Langen and Leader firms generate positive external effects for flrms in the cluster, mainly by
Nijdam, 2003) organizing investments in the training and education infrastructure, the
innovation infrastructure and the infrastructure for collective action.
The more resources are invested in the collective action regimes, the better the
Quality of
performance of a cluster. Five variables influence the amount of resources
collective action
invested: the role of leader firms, the role of public organizations, the presence
regimes (see Olson,
1971) of an infrastructure for collective action, the presence of a community argument
and the use of voice.

The insights derived from these cluster studies are applied to seaport clusters. The
results of this study are discussed in the following section.

ill. THE PERFORMANCE OF PORT CLUSTERS: EMPIRICAL FINDINGS

Rarely, has the cluster concept has been applied to seaports. Relevant studies are
those of Haezendock (2001) on the strengths and weaknesses of Antwerp's port
cluster, Van Klink (1995) on the development of port networks, and Slack (1989),
among others, on the location behavior of port service industries. The empirical
research in this study contributes to port studies in a number of ways: the application
of new insights from cluster theories to ports, attention to the issue of governance in
seaports, a comparison between different port clusters, and an analysis of implications
of using the cluster perspective for policy and management in seaports.
The arrival of ships and cargo is central to port clusters. Incoming goods and
ships attract different economic activities that can be distilled into five components.
The first component is cargo handling, comprising all activities necessary to load and
unload ships, such as terminal operators, pilots, towage and mooring. The second
component is transport and includes all firms that facilitate the movement of goods,
such as shipping lines, shipbrokers and trucking firms. The third component is
logistics and includes all activities that add value to the transported products, such as
storage, re-packing and blending. The fourth component consists of a specific set of
manufacturing activities. The most important manufacturing activities are oil refining,
chemicals production and steel plants. The fifth component consists of specific
trading companies. Trade activities related to commodities that are frequently stored
50 PETER W. DE LANGEN

in ports (such as oil and grain) are included in the port cluster.
The case studies-Rotterdam, Durban and the Lower Mississippi port cluster
(LMPC)-have been selected because: they are the largest ports on their respective
continents; they have a large gateway function; and they are important clusters in
their particular region. The empirical results are based on desk research, interviews
and a survey. In each port cluster, a list of port experts was made, and these experts
were asked to participate, by making time for an interview and answering a set of
survey questions. These survey questions were answered during the interview, so that
questions and answers could be explained. Forty-three experts participated in
Rotterdam, thirty-four in Durban and thirty-one in the LMPC. This is over four-fifths
of all selected port experts, so the results from the survey and interviews are reliable.
The survey questions address the validity of the variables, the importance of the
variables, and the 'score' of each port cluster compared to a competing port cluster,
for each variable. Rotterdam is compared with Antwerp and Hamburg; Durban with
Richard's Bay and the LMPC with Houston. For all survey results, tests of statistical
significance have been made.
As a first step of the case study research, the size of the three port clusters is
analyzed. The available firm data in Rotterdam led to a small overestimation of the
number of firms, since legal or fiscal ventures are included. Based on these data,
Rotterdam's port cluster encompasses about 3,560 firms, 36% of which is active in
transport and 45% in logistics. The number of manufacturing fums is limited (about
90) but they generate a substantial part of all value added in the cluster. The relevant
cluster region consists of sixteen municipalities, of which Rotterdam and Dordrecht
are the most important ones. In Durban, firm data are lacking, so the cluster
population is identified on the basis of expert opinions and the Yellow Pages
directory. This leads to a minor underestimation of the size of the cluster. Durban's
port cluster consists of about 680 firms, to a large extent transport firms (48%). The
relevant region consists of six municipalities, of which Durban is by far the most
important. The Lower Mississippi Port Cluster (LMPC) can be defmed very precisely.
The cluster consists of 1,168 firms. Transport is dominant and trade relatively well
represented (16%). The cluster region comprises twelve municipalities.
As the second part of the case studies research, the validity of the various variables
is tested and the influence of the variables is further analyzed. The main results are
given below, without presenting the underlying analysis of the survey results. These
can be found in De Langen (2004).

Agglomeration Economies

The presence of customers and suppliers is a clear force towards concentration in

all three ports. Furthermore, the experts in al three cases regard it as the most
important agglomeration force. The presence of knowledge is also regarded as a force
towards clustering in seaports.
The presence of labor was regarded as an agglomeration force in Rotterdam and
the LMPC, but not in Durban, because of high unemployment and limited training
efforts in this port. About half of the respondents in Rotterdam indicate that the high
THE PERFORMANCE OF PORT CLUSTERS 51

wage level in Rotterdam offsets the presence of labor. Congestion in seaports is not
regarded as a force opposing agglomeration, but towards the dispersion of port
activities away from the port. This also holds for high land prices. Thus,
agglomeration economies dominate.
In all three cases, the port cluster would benefit from more establishments in
logistics, manufacturing and trade. These activities cannot be attracted with only
traditional location factors, such as accessibility and land availability. The two most
important 'new' location factors for port clusters are 'the quality of life' in a port
region, and the presence of a good knowledge infrastructure. Reducing the negative
effects of transport flows, for instance by creating dedicated solutions for freight road
transport, can enhance the quality of life. Furthermore, the redevelopment of old port
areas is an opportunity to improve the quality of life.
The knowledge infrastructure increases the attractiveness of the port city for firms
that employ high-skilled people. The knowledge requirement differs per port: in
Rotterdam, petro-chemical knowledge infrastructure would fit, whereas in the LMPC
and Durban, a logistics knowledge base would fit better.

Internal Competition

The presence of internal competition (competition between firms located in the

port cluster) contributes to the performance of the port cluster, because:
• Internal competition prevents monopoly pricing, because the costs of switching
between suppliers in the same port are low,
• Internal competition leads to specialization of firms in the cluster,
• Internal competition fosters innovation.
These conclusions hold in all three cases. The fierceness of internal competition is
moderate in cargo handling and port services such as pilotage and towage. Only in the
LMPC is internal competition in some cargo handling segments (break-bulk and dry
bulk) fierce. This is partly explained by competition between port authorities in the
LMPC, who have accommodated (and even invested in) redundant facilities.
With regard to the lack of competition in cargo handling, the challenge-in
general terms-is to reconcile scale economies with competition. An arrangement
could be to lease a terminal to two operators. In principle, two independent operators .
can use the same terminal. This arrangement is unstable if the two operators are head-
on competitors. However, when one operator offers 'multi-user services' while the
other is self-handling a large liner shipping company, the arrangement could be stable.
This arrangement increases competitive pressure, especially for the multi-user
terminal since the dedicated terminal operator could start to offer services for third
parties. Such an arrangement was discussed with the port experts in Durban and was
regarded as a good arrangement to prevent monopolistic behavior.

Cluster Barriers

The presence of barriers to entry or barriers to exit limits the performance of port
clusters. The most relevant entry barrier is the inaccessibility of knowledge and
52 PETER W. DE LANGEN

networks. The unavailability of 'local capital' is an entry barrier in Durban and the
LMPC, but not in Rotterdam. The only relevant exit barrier is unrecoverable 'port
specific investments' . This fmding suggests a strategy to 'tie' frrms to the cluster does
not contribute to performance; rather, exit barriers should be reduced, for instance by
leasing assets to frrms in the port cluster.

The level of specific investments that have to be made depends on the possibilities
to develop arrangements where specialized actors (including the port authority) make
specific, long-term investments invest in assets that are leased to start-ups and
entrants. Such arrangements include:
• Developing land and infrastructure and leasing this to the private sector. This
activity is widely regarded as the primary role of port authorities, and
essentially is nothing more than making specific investments to lower entry
barriers for private frrms. This policy is common practice in the port industry
and is done in Rotterdam, the LMPC and Durban.
• Investing in specific assets, such as cranes, warehouses and special cargo
facilities. These investments imply port authorities invest in the 'tool' and lease
these to the private sector. This strategy leads to a reduction of entry barriers
and can improve the performance of the port cluster. The port authorities in the
LMPC and Rotterdam make such investments.
• The provision of office space in the port area for port related small and
medium-sized enterprises (SMEs). 'Micro-clustering' of these frrms in the same
area has advantages, such as the presence of knowledge and networks. For these
reason, port authorities can invest in a self-sustaining provider of office space
for port related SMEs. Such an involvement is only valuable when the provision
of adequate office space by the market is not effective, for instance because real
estate investors are not willing to make initial investments. Rotterdam is the
only one of the three ports where the port authority is involved in providing
office space.

Heterogeneity of the Cluster

Heterogeneity contributes to performance, because it creates opportunities for

innovation and cooperation. Furthermore, it reduces vulnerability for external shocks.
A diverse mix of local frrms, foreign owned frrms and cluster based multinationals
(diversity of international scope) and a diverse mix of economic activities are
important in this respect. The LMPC is less diverse than Rotterdam and Durban.
An opportunity to improve the diversity of the cluster is through attracting new
growth activities. In Rotterdam, the following growth activities have been identified
(Welters and De Langen, 2003):
• Logistics, especially 'supply chain management' (Haynes et al, 1997).
• Industrial tourism.
• Waste processing and recycling.
• Maritime services.
THE PERFORMANCE OF PORT CLUSTERS 53

• Offshore construction and decomposition (RMPM, 2002).

• 'Postponed manufacturing' .
• The manufacturing of high value chemicals, such as medicine and bio-tech
products.
• Port leisure.
• Education in port and transport related activities.
Coalitions--comprising both public and private actors-that aim to attract firms
from these industries can be effective. In Rotterdam, such coalitions have developed
to attract new manufacturing activities and offshore decomposition. However,
initiating such coalitions is difficult because of the need to continue efforts for a long
period of time, say up to 10 years, in the absence of short-term results.

Trust

Trust in the cluster contributes to performance. It is the most important

'governance variable'. Both the LMPC and Rotterdam are not 'high trust clusters',
compared to their competitors. The level of trust in the port cluster in Durban is
higher than in Richard's Bay. Since trust is based on a 'social relation' , improving the
social embedding of individuals in the port cluster is a method to improve trust.
However, most experts are skeptical about initiatives to enhance the level of trust.
Once a climate of trust has developed, it can be sustained.
One approach that could make sense is to try to develop 'community spirit' among
young professionals in the port. In Rotterdam, a 'young roundtable' was developed,
where young professionals meet about six times a year and discuss common themes.
One of the objectives of this initiative is to make these professionals aware of the
importance of common themes and to prepare them for later involvement in
organizations (especially associations) that promote common interests. In the LMPC
and Durban, initiatives to build trust are absent.

I ntermedwries

The presence of intermediaries contributes to the performance of the port cluster,

but is relatively unimportant compared to the other governance variables. The
forwarder is regarded as the most important intermediary. Rotterdam is endowed with
a large number of intermediaries. In the LMPC and Durban, intermediaries are not
well represented.

Leader Firms

Leader firms are firms with the ability and incentives to make investments with
benefits for other firms in the cluster. Leader firms generally have a good market
position, an international scope and innovative capabilities. The presence of such
firms in a cluster contributes to its performance. Rotterdam is endowed with relatively
many leader firms, but the same is true for its competitors Antwerp and Hamburg.
54 PETER W. DE LANGEN

Durban has more leader firms than Richard's Bay, the LMPC lags behind Houston in
this respect.

Leader firm involvement is in most cases a top management issue. Typical

activities where a contribution of leader firms would make sense include:
• Attract fmancial resources for joint projects.
• Ensure political support for projects with substantial political involvement.
• Provide management expertise and best practices information.

Provision of a 'support infrastructure' can improve the ability of firms to act as

leaders. This professional support infrastructure guarantees effective project
management. In some cases, associations can provide this support infrastructure; in
other cases setting up specific fmn-to-fmn linkages is more effective.
In Rotterdam, the support infrastructure is relatively good. This is one of the
reasons for the relatively positive evaluation of the involvement of leader fmns in the
collective action regimes in Rotterdam. In the LMPC, the maritime cluster initiative,
initiated by Metrovision, provides the support infrastructure that enables two leader
fmns to become involved. In Durban, the support infrastructure is not well developed.

Collective Action Regimes

Collective action regimes are required because individual fmns tend to 'free ride'
and consequently under invest in projects with collective benefits. This problem is
relevant for investments in training and education, marketing, innovation, and
hinterland access . All these regimes are important for the performance of port
clusters. The following conclusions with regard to the quality of these regimes can be
drawn:
• The hinterland access regime in Rotterdam is better than this regime in the
LMPC and Durban . This is explained by the involvement of public
organizations (especially the port authority) and leader fmns in this regime (see
De Langen and Chouly, 2003).
• The marketing and promotion regime in Durban is not effective, both
governments and private fmns are unwilling to contribute to the quality of the
regime. In the LMPC, this regime depends largely on public actors, whereas in
Rotterdam a good public-private arrangement has been created.
• The training and education regime is especially bad in the LMPC, leader fmn
involvement is lacking in this cluster. The regime is good in Rotterdam, because
of the role of leader fmns and the effective organizational infrastructure.
• The innovation regime is modest in all three ports, in Rotterdam because there
is no 'community spirit'. The internationalization regime is also modest in all
three cases.
• Leader fmn involvement is an advantage for Rotterdam 'across the board'.
 THE PERFORMANCE OF PORT CLUSTERS 55

All fifteen regimes that have been analyzed have shortcomings. Experts have
. indicated opportunities for improving all regimes. Table 3 shows the most frequently
encountered opportunities to improve the regimes.

Table 3. Opportunities to improve collective action regimes

Regime Opportunity

Improve market intelligence

Hinterland access Improve intermodal accessibility
Develop a network with hinterland nodes
Marketing and Developing a collective organization for marketing and promotion, with indirect
promotion financing from the port authority
Strategic cooperation to improve the quality of the training and education
Training and
infrastructure
education
Re-training to increase job mobility of employees in declining industries
Knowledge transfer between small and large firms
Innovation
A high volumes, low nuisance, freight road transport system
Internationalization Port representatives in important markets

IV. CONCLUSIONS

This paper has presented the results of applying cluster theories to ports. It was
shown that existing cluster theories yield many relevant insights that can fruitfully be
applied to ports. A framework in which the most relevant insights are incorporated
was presented. Furthermore, it was argued that ports are indeed cluster of economic
activities related to the arrival of cargo and ships in seaports.
Finally, the results of analyzing the effects of the various relevant variables of the
cluster's performance were presented, and implications for port management
discussed. These implications for port managers are manifold and provide an
interesting area for further research.

REFERENCES

Haezendonck, E. (2001) Essays on Strategy Analysis for Seaports, Leuven, Garant.

Haynes, K. E., King, Y. M. and Stough, R. R. (1997) "Regional port dynamics in the
global economy: the case of Kaoshiung, Taiwan", Maritime Policy and Management,
24 (1) 1995 pp. 93-113.
Klink, H.A. van (1995) Towards the borderless Mainport Rotterdam-an analysis of
functional, spatial and administrative dynamics in port systems, Amsterdam: Thesis
publishers.
Krugman, P. (1991) Geography and Trade, Cambridge Massachusetts, The MIT Press
 56 PETER W. DE LANGEN

Langen, P.W. de (2004) The performance of seaport clusters; a framework to analyze

cluster performance and an application to the seaport clusters of Durban, Rotterdam
and the Lower Mississippi, ERIM and TRAIL thesis series
Langen, P.W. de, and Nijdam, M.H. (2003) Leader firms in de Nederlandse
Maritieme cluster; theorie en praktijk, Delft University Press, uitgave in de Nederland
Maritiem Land serie
Langen, P.W. de, and Chouly, A. (2003) "Regimes portuares et acces a l'arriere-pays",
Les Cahiers Scientifiques du Transport, 44 pp77-95
Metcalfe (1994) Competition, Fisher's Principle and Increasing Returns in the
Selection Process, Journal of Evolutionary Economics, 4(4) pp. 327-346.
Nooteboom, B. (1999) Interfirm Alliances: Analysis and Design, London, Routledge.
Olson, M. (1971) The Logic of Collective Action: Public Goods and the Theory of
Groups, Cambridge Massachusetts, Harvard University Press.
Porter, M.E. (1990) The Competitive Advantage of Nations, London, MacMillan.
RMPM (2002) Annual Report, Rotterdam.
Slack, B. (1989) "Port services, ports, and the urban hierarchy", Tijdschrift voor
Economische en Sociale Geografie, 80 pp. 236-243.
Welters, H.W.H. en Langen, P.W. de (eds) (2003) Het onzekere voor het zekere
nemen, een visie vanuit de wetenschap op de ontwikkeling van het haven- en
industriecomplex Rotterdam op de lange termijn, Elsevier, Rotterdam.

View publication stats