CHAPTER I

GLOBAL
INVESTMENT
TRENDS AND
PROSPECTS
 A. CURRENT FDI TRENDS

1. Global trends

Global foreign direct investment (FDI) flows continued their slide in 2018, falling by 13
per cent to $1.3 trillion from a revised $1.5 trillion in 2017 (figure I.1).1 The decline – the
third consecutive fall in FDI – was mainly due to large repatriations of accumulated foreign
earnings by United States multinational enterprises (MNEs) in the first two quarters of 2018,
following tax reforms introduced at the end of 2017, and insufficient compensation from
upward trends in the second half of the year.

The fall took place despite an 18 per cent rise in cross-border merger and acquisitions
(M&As) (from $694 billion in 2017 to $816 billion in 2018). The negative trend is also in
contrast to a 41 per cent jump in announced greenfield investment values (from $698 billion
to $981 billion).

FDI flows declined sharply in developed countries and economies in transition while those
to developing countries remained stable, rising by 2 per cent. As a result, developing
economies accounted for a growing share of global FDI, at 54 per cent, from 46
per cent in 2017.

Repatriations of United States multinationals’ foreign earnings abated in the second half of
2018. The lifting of tax liabilities on accumulated foreign earnings of United States MNEs
may have contributed to the M&A boom recorded in the last quarter, limiting the global
FDI decline for the year, after projections based on the first six months had estimated that
annual inflows would be down by more than 40 per cent.

Figure I.1. FDI inflows, global and by economic group, 2007–2018 (Billions of dollars and per cent)
34
-28%
World total Developing economies
Developed economies Transition economies

2 500
54%
706
+2%
$1297 557
-27%
-13%

2 000

1 500

1 000

500

0
2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).

2 World Investment Report 2019 Special Economic Zones

Even disregarding the fluctuations caused by the tax reform and the increase in cross-
border M&As, the underlying FDI trend – which discounts the volatility caused by one-off
transactions and swings in intra-firm financial flows – was still negative. Average annual
growth in the underlying trend, which was above 10 per cent until a decade ago, has since
stagnated at less than 1 per cent. That weak underlying trend will continue to affect FDI
prospects (see section I.B.)

2. Trends by geography

a. FDI inflows
FDI flows to developed economies reached their lowest point since 2004, declining
by 27 per cent (figure I.2). Flows to Europe more than halved to $172 billion while those
to North America were more resilient, declining by 4 per cent to $291 billion. Although
cross-border M&A deal making remained active, rising by 21 per cent in value, it was not
enough to compensate for the negative outward FDI from the United States caused by
the tax reforms.

In Europe, a few important host countries, such as Ireland and Switzerland, registered
negative inflows of -$66 billion and -$87 billion, respectively. FDI flows to the United
Kingdom also declined, by 36 per cent to $64 billion, as new equity investments halved.
Despite the repatriations, the completion of a number of megadeals resulted in higher
flows to the Netherlands (up 20 per cent to $70 billion) and Spain (where inflows doubled
to $44 billion).

In the United States, FDI inflows declined by 9 per

cent, to $252 billion, mainly due to a fall of one third
in cross-border M&A sales. Australia’s FDI inflows FDI inflows, by region, 2017–2018
Figure I.2.
reached $60 billion – a record level – as foreign (Billions of dollars and per cent)
affiliates reinvested a record $25 billion of their profits
Per cent
in the country.
1 297 -13
World 1 497
FDI flows to developing economies remained
stable, rising by 2 per cent to $706 billion, with
Developed 557 -27
759
significant differences among regions. Developing economies
Asia and Africa recorded higher FDI inflows in 172
Europe 384
-55
2018, while FDI contracted in Latin America and
the Caribbean. 291
North America 302
-4
Developing Asia, already the largest recipient region
of FDI flows, registered an FDI rise of 4 per cent to Developing 706 +2
economies 691
$512 billion in 2018, with positive growth occurring
in all subregions. China, the largest developing- Africa 46 +11
41
economy FDI recipient, attracted $139 billion, an
increase of 4 per cent. Flows to South-East Asia Latin America and 147 -6
the Caribbean 155
rose – for the third consecutive year – by 3 per cent
to a new record level ($149 billion). Asia 512 +4
493

FDI flows to Africa expanded by 11 per cent to

Transition 34 -28
$46 billion, still below the annual average of the economies 48

last 10 years (at about $50 billion). The rise in flows

2018 2017
was mainly due to the continuation of resource-
seeking investments, slowly expanding diversified Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).

Chapter I Global Investment Trends and Prospects 3

 investments in a few economies, and a more than doubling of FDI flows to South Africa
(from $2 billion to $5.3 billion).

FDI in Latin America and the Caribbean was 6 per cent lower ($147 billion) in 2018, failing to
maintain momentum after the 2017 increase (which followed five years of negative growth).
In South America, FDI declined due to lower flows to Brazil and Colombia; in Central
America inflows remained stable.

After a plunge in 2017, FDI flows to transition economies continued their downward trend
in 2018, declining by 28 per cent to $34 billion. The contraction was driven by a halving of
flows to the Russian Federation, by far the biggest economy and largest FDI recipient in
the group, from $26 billion to $13 billion. Part of the decline was due to re-domiciliation of
overseas entities that hold assets in the Russian Federation.

Half of the top 20 host economies in the world continue to be developing and transition
economies (figure I.3). Despite the FDI decline, the United States remained the largest
recipient of FDI, followed by China, Hong Kong (China) and Singapore.

Looking at FDI to selected regional and interregional economic groups, flows remained
relatively stable (figure I.4).

Figure I.3. FDI inflows, top 20 host economies, 2017 and 2018 (Billions of dollars)

(x) = 2017 ranking

252
United States (1) 277
139
China (2) 134
116
Hong Kong, China (3) 111
78
Singapore (5) 76
70
Netherlands (7) 58
64
United Kingdom (4) 101
61
Brazil (6) 68
60
Australia (8) 42
44
Spain (17) 21
42
India (9) 40
40
Canada (15) 25
37
France (13) 30
32
Mexico (12) 32
26
Germany (11) 37
24
Italy (16) 22
22
Indonesia (18) 21
22
Israel (19) 18
16
Viet Nam (21) 14 Developed economies
14 2018 2017
Korea, Republic of (20) 18
13 Developing and transition economies
Russian Federation (14) 26 2018 2017
Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).

4 World Investment Report 2019 Special Economic Zones

Figure I.4. FDI inflows to selected groups, 2017 and 2018 (Billions of dollars and per cent)
Share in world Inward Share in world Share of world
Selected groups FDI inflows FDI inflows (%) FDI stock inward FDI stock (%) GDP (%)

871 67 17 957 56 78
G20
909 61 18 359 56 78

839 65 16 876 52 60
APEC 830 55 17 157 53 60

333 26 6 558 20 14
Commonwealth
334 22 6 571 20 14

323 25 8 844 27 28
USMCA 334 22 9 408 29 28

261 20 3 234 10 24
BRICS 270 18 3 086 9 23

39 3 760 2 2
ACP
36 2 758 2 2

2018 2017
Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).
Note: Data for G20 do not include the European Union.

b. FDI outflows
In 2018, MNEs from developed countries reduced their investments abroad by 40
per cent to $558 billion. As a result, their share in global outward FDI dropped to 55 per
cent – the lowest ever recorded (figure I.5). The significant decline was less a reflection of real
investment intentions than of the impact of the large-scale repatriations of accumulated
foreign earnings by United States MNEs, which resulted in negative outflows. In the first half
of 2018, the reinvested earnings of United States MNEs slumped by a net $367 billion and
turned sharply negative, at -$200 billion, compared with a positive $168 billion in the same
period in 2017. Although reinvested earnings in the second half of the year reverted to a
positive value, FDI outflows from the United States for the full year still declined sharply, to
-$64 billion, compared with $300 billion in 2017. In addition to the immediate repatriation
effect, the tax reforms resolved the tax liability overhang on overseas assets, which may
have contributed to a jump in cross-border M&A purchases by United States MNEs to
$253 billion – a record high. Almost half of those purchases were registered in the fourth
quarter of 2018. The majority of acquisitions took place in the EU, mainly in the United
Kingdom and Germany, but also in India and Japan.

Outflows from European MNEs rose by 11 per cent to $418 billion. French MNEs invested
more than 100 billion in 2018, all in equity investment, becoming the third largest investor
country in the world. Outflows from Ireland and Switzerland, both of which had recorded
negative outflows in 2017, turned positive, reaching $13 billion (up $52 billion) and $27
billion (up $62 billion) respectively.

In contrast, outflows from the United Kingdom declined to $50 billion from $118 billion in
2017 despite a significant rise in cross-border M&As. Investment from German MNEs also
declined by 16 per cent to $77 billion. Although the value of their net M&A purchases more

Chapter I Global Investment Trends and Prospects 5

 Developed economies: FDI outflows, and share in world outflows, 2005−2018
Figure I.5.
(Billions of dollars and per cent)

Value Value of outflows Share in world outflows Share

2 000 90

1 800 80

1 600
70
1 400
60
1 200
50
1 000
40
800
30
600
20
400

200 10

0 0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).

than doubled to $73 billion due to the merger of Bayer with Monsanto (United States) for
$57 billion – the largest deal in 2018 – large negative flows of intracompany loans netted
out much of the increase in equity investment.

Japanese MNEs became the largest investors in the world, despite a decline in outward FDI
of 11 per cent to $143 billion. The slow-down in the overall M&A activity of Japanese MNEs
was the result of a 40 per cent decline in their outward FDI in developed countries, mainly
in the United States but also in the United Kingdom. Their investment in Asia increased by
31 per cent to $49 billion, mainly in China, India and the Republic of Korea.

Outward investment by MNEs from developing economies declined by 10 per cent

to $418 billion. Outflows from developing Asia fell by 3 per cent to $401 billion. Investment
from Chinese MNEs declined for the second consecutive year – by 18 per cent – to $130
billion, as a result of government policies to curb overseas investment, as well as increased
screening of inward investment in the United States and Europe. The country, nonetheless,
was the second largest investor in the world after Japan (figure I.6).

Outward FDI from West Asia reached a historic high of $49 billion in 2018, with MNEs from
Saudi Arabia, the United Arab Emirates and Turkey mainly responsible for the increase.
FDI from Saudi Arabia almost tripled to $21 billion, mainly in technology, finance and
infrastructure activities. Turkish companies are increasingly investing in Africa.

Outward investment by Latin American MNEs plunged in 2018 to a record low of $7 billion,
heavily influenced by negative outflows from Brazil and decreased investments from Chile.
Outflows from Brazil fell to -$13 billion, as foreign affiliates continued funneling financial
resources (often raised in overseas capital markets) back to their parents. MNEs from
Mexico increased their outward FDI to $6.9 billion.

At $38 billion, FDI outflows from transition economies were unchanged in 2018. The
Russian Federation accounts for the bulk of the outward FDI in this group (95 per cent). The
country’s outflows rose by 7 per cent to $36 billion, driven mainly by reinvested earnings
and the extension of intracompany loans to established affiliates.

6 World Investment Report 2019 Special Economic Zones

Figure I.6. FDI outflows, top 20 home economies, 2017 and 2018 (Billions of dollars)
(x) = 2017 ranking
143
Japan (2) 160
130
China (3) 158

102
France (9) 41
85
Hong Kong, China (6) 87
77
Germany (5) 92

59
Netherlands (14) 28
50
Canada (7) 80
50
United Kingdom (4) 118
39
Korea, Republic of (13) 34
37
Singapore (8) 44
36
Russian Federation (12) 34
32
Spain (10) 40
27
Switzerland (156) -35
21
Saudi Arabia (28) 7
21
Italy (15) 26
20
Sweden (17) 23
18
Taiwan Province of China (21) 12
18
Thailand (18) 17 Developed economies
15 2018 2017
United Arab Emirates (20) 14
13
Developing and transition economies
Ireland (157) -39 2018 2017
Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).

3. Trends in cross-border M&As and greenfield projects by sector

In 2018, the values of net cross-border M&As and announced FDI greenfield projects
increased (figure I.7). The value of net cross-border M&As rose 18 per cent to $816 billion,
recovering ground after the 22 per cent fall in 2017. The increase was driven by large deal
sizes, especially in the chemicals industry and the services sector, while the number of
deals actually declined.

The value of announced greenfield projects rose by 41 per cent to $981 billion. Also
here, the average project size was the main driver of the increase, as investment activity
measured by the number of projects increased by only 7 per cent. The gains in value were
mostly in extractive and processing industries, and in construction.

a. M&A trends
The value of global net M&As expressed as a percentage of FDI inflows reached 62 per cent,
the highest level since the height of the dotcom boom in 2000. In developed economies,

Chapter I Global Investment Trends and Prospects 7

 Value and number of net cross-border M&As and announced greenfield FDI projects,
Figure I.7.
2009–2018 (Billions of dollars and numbers)
a. Value b. Number

1 400 18 000 7%

16 000 17 567
1 200
14 000
1 000 981
41%
12 000
816
800 10 000

600 18% 8 000 6 821

6 000
400 -2%
4 000
200
2 000
0 0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

Value of announced FDI greenfield projects Number of announced FDI greenfield projects
Value of net cross-border M&As Number of net cross-border M&A deals

Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics) and information from the Financial Times Ltd, fDi Markets (www.fDimarkets.com) for announced
greenfield projects.

net M&A sales rose by 21 per cent to $689 billion, 84 per cent of the global total. In
developing and transition economies, net M&A sales remained steady at $127 billion.

The increase was driven mainly by a doubling of acquisitions by United States MNEs,
with the jump concentrated in the second half of 2018. The removal of tax liabilities on
accumulated retained earnings overseas following the 2017 tax reforms may have
contributed to the boom. Domestic M&A activity in the United States grew at an even faster
pace than cross-border M&As.

In the primary sector, the largest deal was the acquisition of the oil and gas producer Maersk
Olie og Gas (Denmark) by Total (France) for $7.4 billion as part of continued restructuring
in the sector.

In manufacturing, net M&A sales at the global level remained close to the 2017 level. Deal
making in the pharmaceutical industry, which reached $113 billion in 2015, declined for the
third successive year to $28 billion. The chemical industry made up for the decline through
megadeals, as M&A sales more than doubled to $149 billion. They included the merger
of Bayer (Germany) with Monsanto (United States), worth $57 billion, and that of Praxair
(United States) with the industrial gases group Linde (Germany), worth $32 billion.

In services, net M&A sales rose by over one third to $469 billion. The main driver was
the increase in value of M&As in the financial industry, which almost doubled to $108
billion. Within this industry, M&As involving real estate investment trusts were particularly
numerous. Separately, net M&A sales in real estate activities (part of business activities in
table I.1) were worth $57 billion in 2018. Real estate-related investments thus formed a
sizeable part of cross-border M&As in 2018. Almost all the deals in real estate investment
trusts and three quarters of the deals in real estate targeted assets in developed economies.

b. Greenfield investment trends

The global total value of announced greenfield projects in the primary sector doubled to
$41 billion (table I.2), mostly due to projects in metals mining, which trebled in value to
$20 billion in 2018, the highest level since 2011. Karo Resources (Cyprus) announced a

8 World Investment Report 2019 Special Economic Zones

 Value and number of net cross-border M&As, by sector and selected industries,
Table I.1.
2017–2018
Value
Number
(Billions of dollars) Growth rate Growth rate
Sector/industry 2017 2018 (%) 2017 2018 (%)
Total 694 816 18 6 967 6 821 -2
Primary 24 39 60 550 406 -26
Manufacturing 327 307 -6 1 690 1 600 -5
Services 343 469 37 4 727 4 815 2

Top 10 industries in value terms:

Chemicals and chemical products 65 149 129 198 211 7
Business activities 107 112 5 1 817 1 848 2
Financial and insurance activities 59 108 84 617 599 -3
Information and communication 39 90 131 611 612 0.2
Food, beverages and tobacco 88 55 -37 227 205 -10
Transportation and storage 23 47 109 306 269 -12
Electrical and electronic equipment 26 42 65 307 257 -16
Mining, quarrying and petroleum 23 38 70 466 329 -29
Electricity, gas and water 54 38 -30 171 191 12
Trade 12 35 188 486 501 3
Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics).

Value and number of announced FDI greenfield projects, by sector and selected
Table I.2.
industries, 2017–2018
Value
Number
(Billions of dollars) Growth rate Growth rate
Sector/industry 2017 2018 (%) 2017 2018 (%)
Total 698 981 41 16 350 17 567 7
Primary 21 41 101 83 122 47
Manufacturing 345 466 35 7 855 8 049 2
Services 332 473 43 8 412 9 396 12

Top 10 industries in value terms:

Construction 61 113 84 279 475 70
Electricity, gas and water 90 111 23 302 429 42
Coke and refined petroleum products 15 86 480 75 87 16
Business services 61 78 28 4 419 4 686 6
Motor vehicles and other transport equipment 61 74 20 1 123 1 131 1
Chemicals and chemical products 54 66 21 588 569 -3
Electrical and electronic equipment 60 58 -3 996 1 046 5
Hotels and restaurants 17 49 189 163 422 159
Transport, storage and communications 39 48 24 936 1 018 9
Mining, quarrying and petroleum 20 41 102 79 118 49
Source: UNCTAD, based on information from the Financial Times Ltd, fDi Markets (www.fDimarkets.com).

project worth $4.3 billion in a platinum mine in Zimbabwe, supported by the Africa Finance
Corporation. Large projects were also announced in Chile and Peru.

Announced greenfield projects in manufacturing increased by 35 per cent to $466 billion.

In line with higher investments in extractive industries, the processing of natural resources
was a big driver of the increased investment in manufacturing. Projects in coke, petroleum
products and nuclear fuel increased six-fold to $86 billlion. A project by Shell Canada, a
joint venture of Shell, Petronas, PetroChina, Mitsubishi Corp. and Korea Gas, to build a
liquefied natural gas export facility in Canada was the largest project, with planned capital
expenditures totalling $30 billion.

Chapter I Global Investment Trends and Prospects 9

 In developing economies, the value of announced projects in manufacturing – of critical
importance for industrial development – rose by 68 per cent to $271 billion, halting the
downward trend of recent years (figure I.8). However, projects remained concentrated in
Asia, where announced greenfield investments in manufacturing doubled to $212 billion. In
a positive sign, manufacturing investments also jumped 60 per cent in Africa, to $33 billion.
However, those in Latin America and the Caribbean declined.

The number of manufacturing projects in developing countries rose by a more modest

12 per cent, suggesting that announcements of relatively few large-scale projects explain
the increase in value. For instance, the five largest manufacturing projects in China had a
combined value of $33 billion, accounting for much of the value of announced projects in
China, which doubled from 2017 to $80 billion. The largest announced project was the
plan by BASF (Germany) to invest $10 billion in a new chemical manufacturing base in
Zhanjiang. In the same industry, ExxonMobil (United States) announced plans to build a $7
billion ethylene plant in Zhoushan.

In East Asia, the largest increases in greenfield projects were in higher-skilled industries. In
addition to the mega projects in the chemicals industry, a series of projects in automotive
manufacturing as well as in electrical and electronic equipment boosted the value of
announced projects in China. In East Asia as a whole, the value of projects in the chemicals
industry trebled to $24 billion, that in electrical and electronic equipment rose by half to $25
billion, and that in motor vehicles and other transport equipment also trebled to $25 billion.

The processing of natural resources was a key part of the upturn in West Asia and South-
East Asia and, to a lesser extent, South Asia. In Saudi Arabia, for example, Total (France)
signed a memorandum of understanding with Saudi Aramco to develop a petrochemical
complex in Jubail in a project worth $9 billion. In India, CPC (Taiwan Province of China)
announced its plan to invest $6.6 billion in a petrochemical project in Paradip. As a result,
projects in this industry almost quadrupled to $25 billion in West Asia, those in South Asia

Value of announced FDI greenfield projects in manufacturing, 2005–2018

Figure I.8.
(Billions of dollars and per cent)

Higher-skill industries Natural resources-related industries Lower-skill industries

220

200

180

160 70%
140

120

100

80

60
-16%
40 84%
20

0
2005–2014 2015–2017 2018 2005–2014 2015–2017 2018 2005–2014 2015–2017 2018

Africa Latin America and Asia

the Caribbean

Source: UNCTAD, based on information from the Financial Times Ltd, fDi Markets (www.fDimarkets.com).
Note: Natural resources-related industries include (i) coke, petroleum products and nuclear fuel; (ii) metals and metal products; (iii) non-metallic mineral products; and (iv)
wood and wood products. Lower-skill industries include (i) food, beverages and tobacco and (ii) textiles, clothing and leather; higher-skill industries include all other
manufacturing industries.

10 World Investment Report 2019 Special Economic Zones

increased to $8 billion. In South-East Asia, metal processing attracted investment, more
than doubling the value of announced projects to $12 billion from the value in 2017.

In contrast to the higher-skill and natural resource-related industries, the trend in announced
projects in lower-skill industries was generally lacklustre, not only in Asia but also in
other developing regions. While the value of projects in food, beverages and tobacco in
developing economies rose by 29 per cent to $16 billion, those in textiles declined by 36
per cent to $7 billion. For low-income countries, especially in Africa, the decline in projects
in typical early-industrialization industries is a concern. The need for developing countries
to attract more FDI in these industries to support their structural transformation remains
urgent, explaining the proliferation of industrial policies (WIR18) and special economic
zones (SEZs; see chapter IV).

The global total of announced greenfield projects in services rose by 43 per cent to $473
billion. There were large increases in both construction and power generation. Projects
in construction rose by 84 per cent to $113 billion. Projects in industrial building were
subdued after the 2008 economic crisis, but there has been a revival since the mid-2010s.
Some of these projects are related to the construction of SEZs. For instance, in 2015,
Thailand-based Rojana Industrial Park, a subsidiary of Nippon Steel and Sumikin Bussan
(Japan), announced the project to develop the Dawei Special Economic Zone in Myanmar.
In 2016, Wei Yu Engineering (Taiwan Province of China) announced plans to invest $2.5
billion in the Vung Ang Economic Zone in Viet Nam to construct docks with logistics areas
and agricultural areas. In 2018, the textile manufacturer Shandong Ruyi Technology (China)
announced its project to invest $830 million to establish a textile industrial zone in the Suez
Canal Economic Zone in Egypt.

Greenfield projects in power generation rose by 23 per cent in 2018, to $110 billion,
accounting for almost all projects in utilities. Whereas total investment, including domestic
investment, in power generation is only slowly reducing its reliance on fossil fuels,
international investment through greenfield FDI is focused predominantly on renewable
energy. In the past decade, the value of greenfield projects in renewable electricity exceeded
that of fossil fuel-based electricity generation every
year. In 2018, announced capital expenditures in Global cross-border capital flows,
renewable electricity totalled $78 billion and in fossil Figure I.9.
2014–2018 (Per cent of GDP)
fuel-based electricity only $27 billion (see chapter
II.C). The positive trend in international greenfield
investment in this sector should be put in context. 1.6 2.3

In developing economies, announced greenfield

capital expenditures on power generation projects 2.0 1.7 2.0
(all types) came to $70 billion. This compares with an 3.5 3.4
annual investment gap of over $500 billion to achieve 1.0
the United Nations SDGs, as estimated in WIR14. 1.9

4.1
3.2
2.4 2.3 2.0
4. FDI and other cross-
border capital flows 2014 -0.9 2016 2017 2018

2015
The decline in global FDI flows was in line with the
trend in other cross-border capital flows. Together FDI Portfolio investment Other investment
(mainly bank loans)
FDI, portfolio flows and other investment (mostly
Source: UNCTAD, based on IMF World Economic Outlook Database.
bank loans) amounted to $5 trillion, or 5.9 per cent
Note: The percentages presented here are based on available data from 187
of global GDP in 2018, a decline of more than 20 per economies. The IMF World Economic Outlook database tracks FDI flows
measured according to the asset/liability principle. Hence, the value of
cent from 2017 (figure I.9). FDI flows is not directly comparable with UNCTAD’s FDI data presented
elsewhere in this report.

Chapter I Global Investment Trends and Prospects 11

 While all three categories of capital flows fell, the decline was the largest in portfolio
investment (down 40 per cent). Portfolio flows are closely linked to financial market
performance, as well as interest rate and currency movements. They are also more sensitive
to geopolitical tensions and country-specific political uncertainty.

Developing economies received just over one third of global cross-border capital flows.
Compared with flows to developed economies, which declined by 27 per cent, flows to
developing economies were more resilient, declining by only 8 per cent, because FDI – the
more stable type of finance – represents a larger share of their capital inflows. Portfolio
inflows and other investment in developing economies declined by 30 per cent and 14
per cent, respectively. Declines in portfolio flows were particularly large in Latin America
and in West Asia. Policy uncertainty and currency instability in major regional recipients of
portfolio flows, including Argentina, Mexico and Turkey, contributed to the declines. In those
countries, too, FDI inflows proved more stable and actually increased in 2018 (chapter II).

The size and relative stability of FDI makes it the most importance source of external
finance for developing economies (figure I.10). Preliminary data for official development
assistance (ODA) (bilateral and multilateral) show an increase of 1.5 per cent to $149 billion.
Preliminary data for remittances show an increase of 9.6 per cent to $529 billion.

However, capital flows to developing economies remain concentrated in a relatively small

number of countries. Asia receives three quarters of capital flows to developing economies.
Portfolio investment and other investment flows are even more skewed towards that
region. The least developed countries (LDCs), with a combined population of 1 billion,
receive just 3 per cent of those cross-border capital flows. For these countries, remittances
remain substantially higher than FDI. They increased by 11 per cent to $40 billion in 2018,
compared with FDI inflows worth $24 billion.

Developing economies: sources of external finance, 2009–2018

Figure I.10.
(Billions of dollars)

800

FDI (directional)
600
Remittances

400 Other investment

Portfolio investment
200
ODA
0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

-200

-400

Source: UNCTAD, based on KNOMAD (for remittances), UNCTAD (for FDI), IMF World Economic Dataset (for portfolio investment and other
investment) and OECD (for ODA).
Note: Remittances and ODA are approximated by flows to low- and middle-income countries, as grouped by the World Bank.

12 World Investment Report 2019 Special Economic Zones

 B. FDI PROSPECTS

Global investment is expected to see a modest recovery of 10 per cent in 2019. This
expectation is based on current forecasts for a number of macroeconomic indicators,
UNCTAD’s econometric forecasting model of FDI inflows and its underlying trend analysis,
and preliminary 2019 data for cross-border M&As and announced greenfield projects. It is
complemented by UNCTAD’s survey of investment promotion agencies (IPAs).

1. Short-term prospects
Projections for FDI in 2019 point to a 10 per cent increase to almost $1.5 trillion – still
below the average of the last 10 years. The main factor driving up expectations is the likely
rebound from anomalously low levels of FDI in developed countries in 2018. Following
the subsiding of repatriations of foreign earnings of United States multinationals in the
second half of 2018, developed-country inflows are likely to revert to prior levels, implying
a significant jump in some countries that normally receive sizeable inflows. The expected
increase of FDI flows in 2019 is also apparent in the 41 per cent jump in greenfield project
announcements (planned expenditures) from their low levels in 2017.

Despite these upward-pointing signs the size of the expected increase in FDI is relatively
limited because the long-term underlying FDI trend remains weak (section I.B.2). M&A data
for the first four months of 2019 confirm the need for caution; the value of cross-border
M&As was about $180 billion, 10 per cent lower than the same period in 2018.

The likelihood of an increase in global FDI is further tempered by a series of risk factors.
Geopolitical risks, trade tensions and concerns about a shift towards more protectionist
policies could have a negative impact on FDI in 2019. Moreover, longer-term forecasts for
macroeconomic variables contain important downsides (table I.3).

The projected increase of FDI flows is highest in developed economies, with Europe
expected to see an increase of more than 60 per cent (recovering but remaining at only
about half of 2016 values) (table I.4). Flows to developing economies are expected to hold
steady, with projections showing a marginal increase of about 5 per cent. Among developing
regions, FDI in Africa is likely to increase by 15 per cent, in view of an expected acceleration
of economic growth and advances in regional integration. Prospects for developing Asia
are cautiously optimistic, especially in South-East Asia and South Asia, with flows rising

Table I.3. Real growth rates of GDP and gross fixed capital formation (GFCF), 2016–2020
(Per cent)

Variable Region 2016 2017 2018 2019 2020

World 3.4 3.8 3.6 3.3 3.6
GDP growth rate Advanced economiesa 1.7 2.4 2.2 1.8 1.7
Emerging and developing economiesa 4.6 4.8 4.5 4.4 4.8

World 2.8 4.1 4.0 3.7 4.1

GFCF growth rate Advanced economiesa 2.0 3.8 3.1 2.5 2.4
Emerging and developing economiesa 3.3 4.3 4.6 4.5 5.3
Source: UNCTAD, based on IMF (2019).
Note: GFCF = gross fixed capital formation.
a
IMF’s classifications of advanced, emerging and developing economies are not the same as the United Nations’ classifications of developed and developing economies.

Chapter I Global Investment Trends and Prospects 13

 FDI inflows, projections, by group of economies and region,
Table I.4.
2016–2018, and projections, 2019 (Billions of dollars and per cent)
Projections
Group of economies/region 2016 2017 2018 2019
World 1 919 1 497 1 297 1 370 to 1 500
Developed economies 1 198 759 557 640 to 720
Europe 612 384 172 330
North America 508 302 291 310
Developing economies 656 691 706 700 to 740
Africa 46 41 46 52
Asia 473 493 512 530
Latin America and the Caribbean 135 155 147 140
Transition economies 65 48 34 45 to 55

Memorandum: annual growth rate (per cent)

World -6 -22 -13 (5 to 15)
Developed economies -6 -37 -27 (15 to 30)
Europe -14 -37 -55 ~ 65
North America -1 -41 -4 ~5
Developing economies -10 5 2 (0 to 5)
Africa -18 -11 11 ~ 15
Asia -8 4 4 ~5
Latin America and the Caribbean -13 15 -6 ~ -5
Transition economies 78 -26 -28 (40 to 50)
Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics).
Note: Percentages are rounded.

slightly (by 5 per cent) thanks to a favourable economic outlook and improving investment
climate. Flows to Latin America and the Caribbean are expected to remain relatively stable,
with a projected decline of about 5 per cent, while in transition economies flows are likely
to see a recovery in 2019, reaching $50 billion.

2. Long-term trends
The relatively modest increase in global FDI projected for 2019 is in line with the slow
growth over recent years in the underlying trend. That trend – net of fluctuations driven by
one-off factors such as tax reforms, megadeals and volatile financial flows included in FDI
– has shown anemic growth since the global financial crisis (figure I.11). Key drivers for the
long-term slowdown in FDI include policy, economic and business factors.

Policy factors. The gradual opening of emerging markets worldwide that spurred FDI
growth until the late 2000s is no longer fueling FDI to the same extent. In the last few years,
restrictions on foreign ownership, based on national security considerations or strategic
technologies, have again been front of mind for policymakers (chapter III). Uncertainty over
the development of the international policy frameworks for trade and investment is also not
supporting investor confidence.

Economic factors. Declining rates of return on FDI are a key factor behind the long-term
slowdown (table I.5). In 2018, the global rate of return on inward FDI was down to 6.8
per cent, from 8 per cent in 2010. Although rates of return remain higher on average
in developing and transition economies, most regions have not escaped the erosion.
In Africa, for example, return on investment dropped from 11.9 per cent in 2010 to 6.5
per cent in 2018.

14 World Investment Report 2019 Special Economic Zones

Figure I.11. FDI inflows and the underlying trend, 1990–2018 (Indexed, 2010 = 100)

FDI underlying trend,

average annual growth rate 1990s: 21% 2000–2007: 8% Post-crisis: 1%

150

FDI

100

FDI underlying trend

50

0
1990 2000 2008 2018

Source: UNCTAD, FDI/MNE database (www.unctad.org/fdistatistics); UNCTAD estimates.

Note: The FDI underlying trend is a composite index (incorporating balance of payments and other variables), constructed by removing the effect on FDI of fluctuations in M&As,
intracompany loans and offshore financial flows through appropriate smoothing techniques.

Table I.5. Inward FDI rates of return, 2010–2018 (Per cent)

Region 2010 2011 2012 2013 2014 2015 2016 2017 2018
World 8.0 8.5 7.7 7.5 7.6 6.9 6.8 6.8 6.8
Developed economies 6.4 6.7 6.1 5.9 6.4 6.0 5.9 5.9 6.0
Developing economies 11.0 11.5 10.1 9.9 9.5 8.4 8.2 8.1 7.8
Africa 11.9 12.0 11.7 11.4 9.6 6.5 5.0 6.0 6.5
Latin America and the Caribbean 9.7 9.8 8.5 7.0 6.3 4.5 5.4 6.2 6.2
Asia 11.4 12.2 10.6 10.8 10.7 10.0 9.6 9.0 8.5
East and South-East Asia 12.5 13.4 11.6 11.9 11.8 11.1 10.4 9.9 9.4
South Asia 8.9 7.6 7.2 6.7 6.1 5.5 6.4 5.6 5.3
West Asia 6.0 6.8 5.6 5.5 5.0 4.7 4.8 3.5 3.4
Transition economies 12.1 14.8 14.6 13.2 13.2 9.0 10.2 11.6 12.4
Source: UNCTAD based on data from IMF Balance of Payments database.
Note: Annual rates of return are measured as annual FDI income for year t divided by the average of the end-of-year FDI positions for years t and t - 1 at book values.

Business factors. Structural changes in the nature of international production are also at
work. The adoption of digital technologies in global supply chains across many industries
is causing a shift towards intangibles and increasingly asset-light forms of international
production, as reaching global markets and exploiting efficiencies from cross-border
operations no longer requires heavy asset footprints (WIR17). The trend is visible in the
divergence of key international production indicators – on a scale from tangible to intangible
– with a substantially flat trend for FDI and trade in goods and much faster growth for
both trade in services and international payments for intangibles (royalties and licensing
fees) (figure I.12).

Chapter I Global Investment Trends and Prospects 15

 Figure I.12. Indicators of international production, tangible and intangible, 2010–2018 (Indexed, 2010 = 100)

180 Intangible

160 Royalties and licensing fees

Trade in services
140

120
Trade in goods

100 FDI underlying trend

Tangible
80

60
2010 2011 2012 2013 2014 2015 2016 2017 2018

Source: UNCTAD.

3. IPAs’ expectations
Despite the third consecutive decrease in global FDI in 2018 and the weak underlying
trend, UNCTAD’s survey of investment promotion agencies (IPAs) shows continued
optimism on the part of IPAs. Their expectations for FDI flows into their own countries to
2021 remain high. However, expectations were more tempered at the global level (figure
I.13). Only 45 per cent of respondents expect global FDI flows to increase, indicating that
IPAs acknowledge the challenges of and competition for the attraction of FDI in the current
global investment climate.

Comparing IPAs’ perceptions for global FDI

prospects between 2016 and 2019 shows
that expectations have been progressively less
optimistic in every year of the survey (figure I.14).
IPAs’ expectations for changes in FDI,
Figure I.13. IPAs rank the United States and China – in
2019–2021 (Percentage of respondents)
joint first place – as the most likely sources of
foreign investment to their countries. Three large
Increase 45 European economies – the United Kingdom,
76
Germany and France – were considered the
25
next most important sources of FDI. India and
No change
12 the United Arab Emirates, not traditionally in the
top 20 outward investor countries, were also
30
Decrease considered as among the top 10 most important
12
sources of FDI for the 2019 to 2021 period.
Globally For own country IPAs in developed economies expect
most investment to go to information and
Source: UNCTAD Investment Promotion Agencies Survey. communications industries, followed by
Note: The survey was conducted from February to April 2019. Results are based professional services, and finance and insurance.
on information from 114 respondents.
In developing and transition economies, IPAs
expect more investment in agriculture, followed

16 World Investment Report 2019 Special Economic Zones

by food and beverages, and information and
IPAs expecting an increase in
communication (figure I.15). More and more Figure I.14. global FDI flows, 2016–2019
countries are looking to attract investment in (Per cent of respondents)
digital technologies and innovation as key drivers
of economic growth. The high ranking of the ICT
sector for FDI prospects is also a reflection of the
investment promotion efforts of IPAs in this sector.
75
The selection of agriculture and food processing 61 56
45
among the most promising sectors in developing
and transition economies indicates that IPAs in
2016 2017 2018 2019
those economies expect a significant share of FDI
to remain connected to natural resources for the Source: UNCTAD Investment Promotion Agencies Surveys (2016–2019).
Note: Percentages reflect survey results of each year.
foreseeable future.

IPAs’ selection of most promising industry for attracting FDI in their own economy,
Figure I.15.
by region, 2018 (Per cent of respondents)
Developing and
Developed economies transition economies

Information and communication 75 Agriculture 45

Professional services 46 Food and beverages 41

Finance and insurance 46 Information and communication 40

Manufacture of chemicals 25 Accommodation and food services 31

Food and beverages 25 Construction 29

Real estate 22 Mining 29

Source: Source: UNCTAD, Investment Promotion Agencies Survey.

Chapter I Global Investment Trends and Prospects 17

 C. INTERNATIONAL
PRODUCTION

1. Key indicators of international production

International production continues to expand. Estimated values for sales and value
added of MNEs’ foreign affiliates rose in 2018 by 3 per cent and 8 per cent, respectively.
Employment of foreign affiliates reached 76 million, at an annual growth rate of about 3 per
cent (table I.6).

Relatively fast growth in value added, compared with sales, suggest that foreign affiliates
of MNEs are able to extract increasing value from their operations. At the same time, more
modest growth in employment appears to indicate a gradual shift in the distribution of value
added between production factors towards capital rather than labour. This is consistent
with the ongoing trend of international production shifting towards digital and intangible
activity (see WIR17).

Intangibles also play an important role in the significant growth of foreign assets over the
past decades. The trend towards asset-light operations documented in WIR17 and the
increasing importance of non-equity modes of international operations (including licensing

Table I.6. Selected indicators of FDI and international production, 2018 and selected years
Value at current prices (Billions of dollars)
Item 2005–2007
1990 2015 2016 2017 2018
(pre-crisis average)
FDI inflows 205 1 414 2 034 1 919 1 497 1 297
FDI outflows 244 1 451 1 683 1 550 1 425 1 014
FDI inward stock 2 196 14 475 26 313 28 243 32 624 32 272
FDI outward stock 2 255 15 182 26 260 27 621 32 383 30 975
Income on inward FDIa 82 1 028 1 513 1 553 1 691 1 799
Rate of return on inward FDI b 5.3 8.6 6.9 6.8 6.8 6.8
Income on outward FDIa 128 1 102 1 476 1 478 1 661 1 792
Rate of return on outward FDI b 8.0 9.6 6.3 6.1 6.3 6.4
Net cross-border M&As 98 729 735 887 694 816
Sales of foreign affiliates 7 136 24 621 26 019 25 649 26 580c 27 247c
Value added (product) of foreign affiliates 1 335 5 325 6 002 5 919 6 711c 7 257c
Total assets of foreign affiliates 6 202 50 747 91 261 95 540 104 915c 110 468c
Employment by foreign affiliates (thousands) 28 558 59 011 69 533 70 470 73 571c 75 897c
Memorandum
GDPd 23 439 52 366 74 664 75 709 80 118 84 713
Gross fixed capital formationd 5 820 12 472 18 731 18 781 20 039 21 378
Royalties and licence fee receipts 31 174 321 325 355 370
Source: UNCTAD.
Note: Not included in this table are the value of worldwide sales by foreign affiliates associated with their parent firms through non-equity relationships and of the sales of the parent
firms themselves. Worldwide sales, gross product, total assets, exports and employment of foreign affiliates are estimated by extrapolating the worldwide data of foreign
affiliates of MNEs from Australia, Austria, Belgium, Canada, Czechia, Finland, France, Germany, Greece, Israel, Italy, Japan, Latvia, Lithuania, Luxembourg, Portugal, Slovenia,
Sweden, and the United States for sales; those from Czechia, France, Israel, Japan, Portugal, Slovenia, Sweden, and the United States for value-added (product); those from
Austria, Germany, Japan and the United States for assets; and those from Australia, Austria, Belgium, Canada, Czechia, Finland, France, Germany, Italy, Japan, Latvia, Lithuania,
Luxembourg, Macao (China), Portugal, Slovenia, Sweden, Switzerland, and the United States for employment, on the basis of three-year average shares of those countries in
worldwide outward FDI stock.
a
Based on data from 165 countries for income on inward FDI and 144 countries for income on outward FDI in 2018, in both cases representing more than 90 per cent of global inward
and outward stocks.
b
Calculated only for countries with both FDI income and stock data.
c
Data for 2017 and 2018 are estimated based on a fixed-effects panel regression of each variable against outward stock and a lagged dependent variable for the period 1980–2016.
d
Data from IMF (2019).

18 World Investment Report 2019 Special Economic Zones

and contract manufacturing) suggests that fixed assets are not the driver of this growth.
The growth of total assets relative to sales over the last decade is in line with the trend in
assets-to-sales ratios of the S&P500.

The rate of return on inward FDI generated by foreign affiliates in host economies remained
at 6.8 per cent in 2018. After a pronounced gradual decline since 2010 it appears to have
reached a plateau in the last three years, at 6.8 per cent of total FDI stock.

2. Internationalization trends of the largest MNEs

In 2018, seven companies entered the UNCTAD ranking of the top 100 MNEs. Three
companies entered following cross-border mergers: Atlantia Spa (Italy), a construction
company, which bought Spanish competitor Albertis; the new Linde Plc (United Kingdom),
which emerged from the merger of two industrial gas companies, Praxair (United States)
and Linde AG (Germany); and Takeda Pharmaceuticals (Japan), which acquired Shire Plc
(Ireland). Four MNEs from developing countries entered the list: three Chinese State-owned
MNEs (SO-MNEs), Chem China, State Grid and China MinMetals, and Tata Motors from
India. Broadcom Inc exited the top 100 because of its decision to move its headquarters
from Singapore to the United States, where most of its operations are based. A second
MNE exited because of financial difficulties: HNA Group (China) entered a severe liquidity
crisis in the second half of 2017 and has since shed more than $40 billion in assets as it
tried to pay off debt accumulated during a spree of acquisitions in the preceding years.
Other companies at the bottom of the ranking slid out as the threshold of foreign assets
continued to increase.

The average level of internationalization of the top 100 MNEs (the ratio of foreign over
domestic assets) decreased in 2018 (table I.7). This was caused by the new Chinese
entries (with large domestic operations), by a number of mergers that boosted domestic
operations, and by the divestment of foreign operations by a few MNEs.

The presence of technology companies in the top 100 MNEs from developing
countries is increasing. New entries in 2017 included the electrical appliance
manufacturer Midea Group (China), following three major acquisitions in 2016: the home
appliances business of Toshiba (Japan), the German robotics company KUKA, and Eureka,
a floorcare brand, from Electrolux (Sweden). During 2018, many semiconductor MNEs
from emerging economies entered joint ventures or increased investment in production
capacity, with some poised to enter the list next year (e.g. SK Hynix, ASE Technologies,
TWC). SK Hynix (Republic of Korea) plans to invest almost $150 billion over the next 10
years into its semiconductor business to maintain its position as one of the world’s largest
chipmakers. Also, last year, Advanced Semiconductor Engineering (Taiwan Province of
China) and Siliconware Precision Industries formed a new holding company, as part of the
consolidation in the global semiconductor industry.

The top 100 MNEs from developing and transition economies also saw the relative growth
of their foreign operations slow, on average, although the absolute growth of their foreign
sales, assets and employees remained significantly higher than that of the firms in the
global top 100. For both top 100 groups, foreign sales are growing faster than foreign
assets and employees, in line with the increasing importance of intangibles, asset-light
operations and non-equity modes of international production.

Since 2010 the number of (non-automotive) industrial MNEs in the top 100 ranking
has dropped by half, from 20 to 10 in 2018. Figure I.16 shows the acquisitions and
divestments of top industrial corporations (excluding automotive firms, which saw little

Chapter I Global Investment Trends and Prospects 19

 Internationalization statistics of the top 100 non-financial MNEs, global and from
Table I.7. developing and transition economies, 2016 and 2017
(Billions of dollars, thousands of employees and per cent)

Top 100 MNEs from developing

Global top 100 MNEs
and transition economies
Variable
2016–2017 2017–2018
2016a 2017a 2018b 2016a 2017 Change (%)
Change (%) Change (%)
Assets (billions of dollars)
Foreign 8 337 8 996 7.9 9 231 2.8 1 895 2 119 11.8
Domestic 4 894 5 538 13.2 6 262 14.8 5 100 5 613 10.1
Total 13 231 14 534 9.8 15 492 7.2 6 995 7 732 10.5
Foreign as share of total (%) 63 62 -1.1 60 -2.3 27 27 0.3

Sales (billions of dollars)

Foreign 4 765 5 200 9.1 5 587 8.1 1 535 1 897 23.6
Domestic 2 737 2 817 2.9 3 790 35.5 2 066 2 537 22.8
Total 7 502 8 017 6.9 9 377 18.1 3 601 4 433 23.1
Foreign as share of total (%) 64 65 1.3 60 -5.3 43 43 0.2

Employment (thousands)
Foreign 9 535 9 662 1.3 9 611 0.8 4 618 4 521 -2.1
Domestic 6 920 7 037 1.7 7 876 13.8 8 622 8 652 0.4
Total 16 455 16 699 1.5 17 488 6.3 13 240 13 174 -0.5
Foreign as share of total (%) 58 58 -0.1 55 -2.9 35 34 -0.6
Source: UNCTAD.
Note: Data refer to fiscal year results reported between 1 April of the base year and 31 March of the following year. Complete 2018 data for the top 100 MNEs from developing and
transition economies are not yet available.
a
Revised results
b
Preliminary results

change) that were in the top 100 ranking in 2010, those that are still in the ranking (above
the line) and those that dropped out (below the line).

The decline in the number of industrial MNEs in the ranking is only partly the result of the
growing presence of technology and digital companies. It is also driven by the scaling-down
of industrial conglomerates. Industrial MNEs disappearing from the top ranking or losing
positions are often undergoing restructuring programmes to focus on their core business.
Of those that left the ranking, ThyssenKrupp (Germany) – after a series of divestments –
announced that it will spin off its lift business. Similarly, ABB (Switzerland) announced the
sale of its power-grid division to Hitachi (Japan) in December.

Other industrial MNEs are still in the 2018 ranking, often as a result of M&As.
Examples of mergers between traditional industrial companies include the new Linde Plc
(United Kingdom), DowDuPont (United States) and LafargeHolcim (Switzerland). Others
acquired major competitors: in 2018 Bayer Ag (Germany) purchased Monsanto (United
States), and United Technologies Corp (United States) bought Rockwell Collins (United
States). Post-merger moves to shed non-core businesses or to realize synergies could
negatively affect the ranking in the top 100 of these companies. For example, United
Technologies already announced it will split into three companies, with the aviation
business remaining the largest. Similarly, DowDuPont (merged in 2017) is splitting this year
into three more focused companies. LafargeHolcim (merged in 2015) has already sold its
business in Indonesia and plans to sell assets in South-East Asia for $2 billion over the
next five years.

The downsizing of industrial MNEs appears to be a general trend. For example, Siemens
(Germany) floated its medical equipment business to attract investors for businesses
outside its core industrial engineering operations, and it separated its wind power
operations. In 2018, Siemens announced that it will spin off its gas and power operations
into an independent company to be listed next year. The most dramatic restructuring is

20 World Investment Report 2019 Special Economic Zones

 Top industrial MNEs’ total divestments and investments (foreign and domestic),
Figure I.16.
cumulative 2010–2018 (Billions of dollars)

Divestments Acquisitions

General Electric Co -120

Bayer AG

United Technologies Corp

Johnson Controls International Plc

Siemens AG

Air Liquide International SA

Cie de Saint-Gobain SA

DowDuPont (since 2017)

Merged MNEs LafargeHolcim (since 2015)

Linde PLC (since 2018) IN

Hewlett Packard Enterprise Co
OUT
Alstom SA

Ferrovial SA

ThyssenKrupp AG

Cemex SAB de CV

Koninklijke Philips NV

Caterpillar Inc

ABB Ltd

BAE Systems Plc

-30 0 30 60 90 120
Source: UNCTAD, based on information from Refinitiv Eikon.
Note: The figure lists non-automotive industrial firms in the 2010 ranking of the global top 100 MNEs. Firms above the line are still in the 2019 ranking. In 2010, in place of the
three merged companies there were either one company (Dow Chemical, Linde AG) or two (Lafarge and Holcim). Caterpillar Inc (United States) and ABB Ltd (Switzerland)
exited the ranking despite acquisitions as these were either domestic or not large enough to stay above the threshold level of foreign assets for the top 100 list.

represented by General Electric (United States), which was at the top of the ranking for
many years and is now sliding down the list following a series of divestments totalling
more than $120 billion at the end of 2018. These divestments started in 2016 with its
financial services division, which until then provided about half of the group’s profits, and
will ultimately reduce the company’s sectors of operation from more than 10 to just two:
aviation and power.

The shedding of non-core businesses by industrial conglomerates in the top 100

has also been the result of pressure from shareholders. Conglomerates’ shares are
no longer commanding a premium as in the past but are trading at a discount. Active
hedge fund managers have been playing a key role behind the trend, as in the case of
Cevian pushing for the break-up of ThyssenKrupp, and ABB and Third Point influencing
United Technologies.

In 2018, top global companies invested more than $350 billion in R&D, representing
over a third of business-funded R&D worldwide. The top 100 list includes global leaders

Chapter I Global Investment Trends and Prospects 21

 in the key industries contributing to R&D: ICT, pharmaceuticals and automotive. The top
three R&D investors were all from technology and digital industries: Amazon.com (United
States) with almost $29 billion of expenditures in 2018, followed by Alphabet (United States)
with $21 billion, and Samsung Electronics (Republic of Korea) with $17 billion. Including in
the sample the top 100 MNEs from developing and transition economies produces a list
of the top 20 R&D investors that captures a large part of innovation expenditures across
the world. The top innovators are concentrated among technology MNEs from the United
States and a few emerging economies (mainly the Republic of Korea and China), followed
by developed-economy pharmaceutical and automotive firms (table I.8). Among the top
MNEs, global international traders, utilities and extractive companies invested the least
in R&D. Top R&D investors from emerging economies were – after Samsung Electronics
– Huawei Technologies (China) with $15 billion, and China Mobile (China) with $6 billion.2

Given the differences in size between MNEs, the absolute value of R&D expenditures is
not a reliable guide to the importance of R&D in maintaining a company’s competitive
edge. For example, the oil company Sinopec (China) invested $1.2 billion in R&D in 2018,
representing only 0.3 per cent of its revenues. Thus, especially for the ranking of MNEs
from developing and transition economies, it is more indicative to look at R&D expenditure
as a percentage of total revenue (i.e. R&D intensity). This changes the ranking among
industries, with pharmaceuticals showing the highest intensities.

In the top 100 MNEs from developing and transition economies, only a few spend
more than 5 per cent of sales on R&D. This is due mostly to the industry composition
of the list and the prevalence of big industrial or extractive conglomerates (table I.9).
However, even comparing like for like industries, the R&D expenditures by companies from
developing countries remain lower. For example, comparing the R&D intensity in the
automotive industry shows an average of 1.2 per cent for the two companies in the
developing-country list (Hyundai and Tata Motors), compared with 4.7 per cent in the
global list (11 companies).

Top 20 R&D investors from the top 100 MNEs (global and developing and transition
Table I.8.
economies), by expenditure, 2018 (Billions of dollars, R&D intensity)
R&D
R&D
Ranking Company Country Industry expenditures
intensity
($ billion)
1 Amazon.com, Inc United States Tech 28.8 12.4
2 Alphabet Inc United States Tech 21.4 15.7
3 Samsung Electronics Co, Ltd Korea, Rep. of Tech 16.5 7.5
4 Huawei Technologies China Tech 15.3 14.1
5 Microsoft Corp United States Tech 14.7 13.3
6 Apple Inc United States Tech 14.2 5.4
7 Intel Corp United States Tech 13.5 19.1
8 Roche Holding AG Switzerland Pharmaceuticals 12.3 20.3
9 Johnson & Johnson United States Pharmaceuticals 10.8 13.2
10 Toyota Motor Corpa Japan Automotive 10.0 3.6
11 Volkswagen AG Germany Automotive 9.6 3.4
12 Novartis AG Switzerland Pharmaceuticals 9.1 16.5
13 Robert Bosch GmbH Germany Automotive 8.7 9.2
14 Ford Motor Co United States Automotive 8.2 5.1
15 Pfizer Inc United States Pharmaceuticals 8.0 14.9
16 General Motors Co United States Automotive 7.8 5.3
17 Daimler AG Germany Automotive 7.5 3.9
18 Honda Motor Co Ltd Japan Automotive 7.3 5.1
19 Sanofi France Pharmaceuticals 6.7 16.0
20 Siemens AG Germany Industrial 6.4 6.7
Source: UNCTAD, based on information from Refinitiv Eikon and Orbis.
a
2017 data.

22 World Investment Report 2019 Special Economic Zones

 Top 15 R&D investors among the top 100 MNEs from developing and transition
Table I.9.
economies, 2017 (Millions of dollars, R&D intensity)

Ranking Company Country Industry R&D expenditures R&D intensity

1 Huawei China Tech 15 300 14.1
2 United Microelectronics Corp Taiwan Province of China Tech 424 8.5
3 Samsung Electronics Co, Ltd Korea, Rep. of Tech 16 451 7.5
4 Tencent Holdings Ltd China Tech 3 465 7.3
5 China Mobile Ltd China Telecom 6 421 5.9
6 SK Hynix Inc Korea, Rep. of Tech 2 047 5.6
7 Cheng Shin Rubber Industry Co, Ltd Taiwan Province of China Industrial 173 4.8
8 Advanced Semiconductor Engineering Inc Taiwan Province of China Tech 394 4.0
9 Midea Group Co Ltd China Tech 1 218 3.1
10 Lenovo Group Ltd China Tech 1 274 2.8
11 Qingdao Haier Co Ltd China Industrial 739 2.7
12 Oil and Natural Gas Corp Ltd India Extractives 1 236 2.2
13 POU Chen Corp Taiwan Province of China Industrial 203 2.1
14 China Communications Construction Co Ltd China Construction 1 457 2.0
15 Wistron Corp Taiwan Province of China Tech 469 1.6
Source: UNCTAD, based on information from Refinitiv Eikon and Orbis.

FDI in R&D activities is growing. MNEs establish R&D activities abroad to locate close
to markets, to access pools of skilled resources, or to cluster near knowledge centres.
R&D-related greenfield investment projects are significant in number and growing. During
the last five years 5,300 R&D projects were announced, representing about 6 per cent of
all investment projects, and up from 4,000 in the previous five years. For pharmaceutical
companies, R&D-related projects can account for as much as 17 per cent of all greenfield
projects (figure I.17). Software and IT services follow, with about 15 per cent of their
greenfield projects related to R&D.

R&D-related projects as a share of total announced projects,

Figure I.17.
by industry, 2010–2018 (Per cent of projects)

Pharmaceutical 17

Software and IT services 15

Chemicals 14

Electrical and electronics 11

Automotive and aerospace 9

Average 6

Machinery and equipment 6

Food, beverages and tobacco 5

Rubber and plastic 5

Other manufacturing 4

Extractives 3

Services, non-tech 2

Source: UNCTAD, based on information from Financial Times Ltd fDi Markets, (www.fdimarkets.com).

Chapter I Global Investment Trends and Prospects 23

 The majority of R&D-related FDI projects is in relatively lower value added design,
development and testing activities, rather than basic research. These activities are also
driving most of the growth in R&D projects abroad. Such projects may seek to access
lower-cost skilled resources or to locate closer to markets where the research phase is
aimed at adapting products to different consumer needs.

Developing and transition economies capture 45 per cent of all innovation-related FDI.
Projects in developing Asia are transforming some economies, including Singapore, Hong
Kong (China), India and Malaysia, into global hubs of applied research. The share of R&D
projects directed towards other developing regions is smaller (figure I.18).

R&D-related announced greenfield FDI projects, by type and region,

Figure I.18.
cumulative 2010–2018 (Number and per cent)

Design, development and testing R&D Share of R&D projects in total

6 2 7 4 3

767

404

3 992

2 537

79
18 38
136 424 169

Developed Africa Developing Latin America Transition

countries Asia and the Caribbean economies

Source: UNCTAD, based on Financial Times Ltd, fDi Markets (www.fdimarkets.com).

3. State-owned multinational enterprises

The total number of SO-MNEs3 is stable. The 2019 update of UNCTAD’s database
of SO-MNEs includes close to 1,500 firms, as in 2017. Last year, three new SO-MNEs –
ChemChina, State Grid of China and China Minmetals – entered the top 100 MNEs ranking,
bringing the number of SO-MNEs in the top 100 to 16 in all, one more than in 2017. These
SO-MNEs include five from China and 11 with developed-country shareholdings (table I.10).

Overall, about 10 per cent of companies in the database are new entrants. In the majority
of cases, the new entrants are SO-MNEs from major emerging markets that have newly
opened subsidiaries abroad. These have replaced an equal number of SO-MNEs that left
the data set for various reasons:
• State ownership shrank below 10 per cent. An example is the French utilities company
Veolia Environment.
• The SO-MNE dissolved or went bankrupt. Examples include Italian terminal services
company Alitalia Servizi and Russian aircraft company Oboronprom.
• The SO-MNE merged or was taken over by other companies. For example, CPFL
Energia from Brazil was acquired by another SO-MNE, State Grid of China. Another

24 World Investment Report 2019 Special Economic Zones

 Table I.10. SO-MNEs in the UNCTAD ranking of the top 100 MNEs, 2017 and 2019
Ranking in Ranking in
WIR19 WIR17 Company Home economy Industry
6 (6) Volkswagen Group Germany Motor vehicles
18 (18) Enel SpA Italy Electricity, gas and water
28 (27) Deutsche Telekom AG Germany Telecommunication
30 (33) EDF SA France Electricity, gas and water
32 (23) Eni SpA Italy Petroleum refining and related industries
40 (81) China COSCO Shipping Corp Ltd China Transport and storage
42 (54) Nippon Telegraph & Telephone Corp Japan Telecommunication
50 (46) Airbus SE France Aircraft
51 (37) Engie France Electricity, gas and water
52 (52) Orange SA France Telecommunication
56 (44) China National Offshore Oil Corp (CNOOC) China Mining, quarrying and petroleum
59 (55) Equinor ASA Norway Petroleum refining and related industries
62 (..) State Grid Corp of China China Electricity, gas and water
67 (..) China National Chemical Corp (ChemChina) China Chemicals and allied products
69 (68) Renault SA France Motor vehicles
97 (..) China Minmetals Corp (CMC) China Metals and metal products
Source: UNCTAD.

Figure I.19. Distribution of SO-MNEs by ownership, governance and size, 2018 (Per cent)

Share of SO-MNEs 7 10 Transition economies

26 Other developing
35 economies
Indirect

18 23
11 China
4 Other developed
Governance

economies
31

2 49 Europe
Direct

9 50
24

Large Small
Minority Majority majority owned majority owned
Ownership
Size
Source: UNCTAD.
Note: Majority-owned shares in voting rights greater than 50 per cent; minority includes golden shares; large have total assets over $5 billion.

example involves Tri-ring Group, a Chinese provincial SO-MNE, which was purchased by
a private company, the Wuhan Kingold Industrial Group.

The resulting geographical distribution of SO-MNEs did not change significantly compared
with that reported in WIR17. European SO-MNEs accounted for a little more than a third of
all SO-MNEs, and another 45 per cent were in China and other developing Asian economies.

SO-MNEs vary considerably:

Ownership: The influence governments can exercise on companies varies significantly

according to their shareholding, from minority participation (or golden share) to majority (or
total ownership). Although it is possible for governments holding a minority stake or a golden

Chapter I Global Investment Trends and Prospects 25

 share to exercise significant control over SOEs, their influence is felt more when they hold a
majority shareholding; 73 per cent of SO-MNEs are majority owned (figure I.19).

Governance: State ownership can be exercised either directly through share ownership
by the government, or indirectly when shares are held by State-owned entities such as
sovereign wealth funds, government pension funds or central banks. Indirect participations
are often smaller. In some cases, such as in Malaysia, Singapore and West Asian countries,
sovereign wealth or investment funds can own majority participations. Some sovereign
wealth funds, such as Norway’s Government Pension Fund Global, can be very influential
even through minor shareholdings (Cuervo-Cazurra, 2018). Finally, State ownership is
increasingly exercised through multiple shareholders, combining sovereign wealth funds,
pension funds and other SOEs.

Size and transnationality: Many smaller SO-MNEs have few foreign affiliates, often in
neighbouring countries, and their overseas presence remains stable over time. Large SO-
MNEs have in recent years more actively invested and expanded abroad. The geographical
distribution of SO-MNEs changes significantly depending on their size and on the level
of participation held by the State. SO-MNEs from emerging economies are, on average,
predominantly majority owned and large. The nine SO-MNEs in the top 100 with a minority
State participation are all from developed countries. In Europe, many relatively small utility,
transportation or bank SOEs – often owned at the subnational level – maintain a few affiliates
in neighbouring countries due to the integrated nature of the region’s economies and small
national territories. These SOEs account for almost half of majority-owned SO-MNEs with
assets under $5 billion. In developed countries, many large SO-MNEs were (partially or fully)
privatized in the 1990s. As a result, SO-MNEs in developed economies are split among
small but majority-held SO-MNEs and a few large but minority-controlled SO-MNEs.

SO-MNEs’ M&A activity is slowing down. Until 2012, the growth in cross-border deals
was in line with the growth in the number of SO-MNEs, with increasing numbers of emerging-
markets SO-MNEs internationalizing their operations (figure I.20). In the last five years,
however, cross-border acquisitions from emerging markets have been on a downward
trend, mostly due to increasing concerns about competition and foreign State ownership of

Cross-border acquisitions by majority-owned SO-MNEs, number and share of total value

Figure I.20.
by home region, 1995–2018 (Number and per cent)

Developed economies Developing and transition economies Combined share of value of total cross-border deals
Number Share
250 25

200 20

150 15

100 10

50 5

0 0
1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Source: UNCTAD, based on information from Refinitiv Eikon.

26 World Investment Report 2019 Special Economic Zones

domestic assets and mounting scrutiny of acquisitions, especially in the United States and
Europe. (See also chapter III on investment measures related to national security.)

The number of SO-MNEs’ cross-border acquisitions has never accounted for more than 2
per cent of the total number of deals, but such deals are typically larger than the average
value of international deals. The value of SO-MNEs’ cross-border acquisitions accounted
for less than 7 per cent of the total in the last five years, down from almost 10 per cent
between 2009 and 2013. The spike in 2009 was due to a general decline in all cross-
border deals, but the spikes recorded in 2002, 2013 and 2017 are all explained by very
large single transactions. In 2002, Swedish majority State-owned Telia AB merged with
Finnish majority State-owned Sonera Corp to create a single telecommunication group
worth $6.3 billion. In 2013, Russian oil company Rosneft purchased TNK-BP Ltd for $55
billion. And in 2017, Chinese chemical giant ChemChina purchased Swiss group Syngenta
for almost $42 billion.

Over the 2010–2018 period, the highest numbers of acquisitions by SO-MNEs

occurred in utilities, followed by the hydrocarbon and mining industries. These
three industries together attracted almost half of all deals (figure I.21). Other attractive
industries were financial services and real estate. High-technology industries, including
both hardware providers and software and IT services, accounted for 5 per cent of the
acquisitions. This industry breakdown largely holds across SO-MNE home regions, except
in the mining industry, where SO-MNEs from emerging markets target foreign mining
companies more often than SO-MNEs from developed economies do.

Figure I.21. SO-MNEs’ cross-border acquisitions by industry, cumulative 2010–2018

(Per cent of all SO-MNE deals)

Primary
Oil and gas 13
Mining 11

Secondary
High technology 5
Food and beverages 4
Chemicals 3

Machinery 2
Pharmaceutical and health 2
Automotive 2
Aerospace and defence 1
Other secondary 5

Tertiary
Utilities 20
Financial services 11
Real estate 9
Telecommunication 4
Construction 2
Other services 7

Source: UNCTAD, based on information from Refinitiv Eikon.

Chapter I Global Investment Trends and Prospects 27

 D. THE GLOBAL
FDI NETWORK

A new view on bilateral investment relationships. Bilateral FDI stock data from the
balance of payments focus on direct investment relationships among countries. They
provide a granular and detailed map of the relative positions of countries in the global
investment network, showing where financial claims and liabilities are created and where
they are held. (Bilateral FDI data are accessible at UNCTAD Stat.)

The direct investor perspective is significantly affected by financial centres and investment
hubs, which play a systemic role in global FDI. An alternative view by ultimate investor
reveals some key underlying patterns – where the investment decision is made, where the
capital is originated, who bears the risks and reaps the benefits of the investment – that
can be more relevant in the analysis of international production. In the special case of
round-tripping, the ultimate investor perspective unveils the underlying domestic nature of
a foreign direct investment.

UNCTAD has created a new database of bilateral investment positions by ultimate investors
for more than 100 recipient countries, covering about 95 per cent of total FDI stock and
including many developing countries (box I.1). In addition to its analytical value for mapping
international production, a comprehensive picture of the global FDI network by ultimate
investors can provide important policy insights. Such information can inform policy areas
such as the coverage of international investment treaties, national policies to attract and
facilitate foreign investment and ongoing efforts to reform the international tax system
(WIR15 and WIR16).

UNCTAD FDI estimates by ultimate investing country (UIC) highlight the leading role of
large industrial economies in global investment (table I.11). The rankings of bilateral FDI
links based on UIC versus direct investors are considerably different: only two of the top
10 FDI links based on UIC appeared in the top 10 ranking based on direct investors in
2017. This difference highlights the prominent role that investment hubs now play as a tool
for investors.

Comparing the current picture based on ultimate investors with the picture based on direct
investors as of 2005 shows that the difference then was not as pronounced. That indicates
that investors’ reliance on investment hubs to channel their FDI has become far more
significant over the past decade. The discrepancy between the two rankings – by direct
and by ultimate investor – could narrow over the next few years, however, as a result of
initiatives to tackle tax avoidance.

Table I.11 reveals that cross-border investment from the United States to China is far more
significant than direct investment data would suggest. Based on estimates by ultimate
investors, FDI by United States MNEs in China features among the 10 largest bilateral
investment stocks worldwide, accounting for some 10 per cent of total Chinese inward
FDI. Yet according to official FDI data, that share is only 3 per cent, as much of the FDI
from United States MNEs has been channeled through (mainly regional) investment hubs,
including Singapore and Hong Kong (China). FDI estimates based on UICs thus provide
a more accurate perspective on the bilateral investment relationship between the United
States and China, as well as intra-firm trade between United States MNEs and their Chinese
foreign affiliates.

28 World Investment Report 2019 Special Economic Zones

 Table I.11. Top 10 FDI links by ultimate investor, estimated bilateral inward stock, 2017
Bilateral FDI by ultimate investor (estimates) Bilateral FDI by direct investor (data)

Rank 2017 Investor Recipient Rank 2017 Rank 2005

1 United Kingdom United States 6 1

2 Hong Kong, China China 1 2
3 Japan United States 11 5
4 Canada United States 12 7
5 United States United Kingdom 15 3
6 Germany United States 20 6
7 United States Canada 18 4
8 Switzerland United States 21 12
9 France United States 27 13
10 United States China 30+ 30+
Source: Bilateral FDI by ultimate investing countries: UNCTAD estimates. Bilateral FDI by direct investing countries: UNCTAD bilateral FDI database (complemented by data on
investment from and to special purpose entities).

The ultimate investor perspective, when applied to FDI from the European Union to the United Kingdom (relevant
to the current discussion on Brexit), results in the opposite effect. The share of EU firms as ultimate investors in
the United Kingdom remains sizeable at 33 per cent, but it is nonetheless lower than the 47 per cent measured
by standard bilateral FDI data. Official data are affected by major investments hubs located within the EU, which
channel FDI from UICs located elsewhere.

Regional integration. According to standard Intraregional investment, bilateral

Figure I.22.
bilateral FDI data, cross-border investment within inward stock, 2017
(Per cent of regional FDI in total FDI)
the same geographic region accounts for about
half of total FDI stock (figure I.22). This share has
46
been stable since 2005 (46 per cent of total stock Total 38
in 2017, compared with 49 per cent in 2005). Such
53
intraregional investment is particularly high in Europe Developed economies 46
and Asia, accounting for 81 and 47 per cent of these
regions’ total inward FDI, respectively. In Africa, Europe 81
67
this share is only 10 per cent, similar to the ratio in
Latin America and the Caribbean (11 per cent). The North America 17
20
regional proportion of total GVC flows is also low in
Africa, as well as in Latin America and the Caribbean Developing economies 36
24
(see WIR13, figure IV.10).4 Modest intraregional
FDI and GVC flows in these regions suggest that Asia 47
33
regional economic cooperation initiatives still have
significant potential to promote regional trade and Africa 10
7
investment links.

Yet the share of intraregional investment in global FDI Latin America and the Caribbean 11
8
decreases from 46 to 38 per cent when bilateral FDI
is based on UICs. This illustrates the outsized role Transition economies 6
8
that regional investment hubs play in intraregional
investment flow. For example, the Netherlands Direct investors Ultimate investors
and Luxembourg in Europe, as well as Hong Kong
(China) and Singapore in Asia, are often gateways Source: Bilateral FDI by ultimate investing countries: UNCTAD estimates. Bilateral
FDI by direct investing countries: UNCTAD bilateral FDI database
for investment in the region. In Africa, Mauritius
(complemented by data on investment from and to special purpose
entities).

Chapter I Global Investment Trends and Prospects 29

 Investment in developing economies, plays the same regional hub role, although on a
Figure I.23. smaller scale.
bilateral inward stock, 2017 (Per cent)
In developed economies, the correction is relatively
1 small (from 53 to 46 per cent), as the situations in
Developing economies
(South-South) North America and Europe balance each other. In
53
71 Developed economies Europe, regional hubs inflate intraregional FDI in
Transition economies official data. In North America, in contrast, FDI based
on UICs reveals a stronger regionalization than in
47
28 official data, highlighting that part of the investment
between the United States and Canada is channeled
By origin of By origin of
direct investor ultimate investor through investment hubs. In developing countries,
Source: Bilateral FDI by ultimate investing countries: UNCTAD estimates. Bilateral FDI
however, the correction is more significant: the share
by direct investing countries: UNCTAD bilateral FDI database (complemented of intraregional investment drops from 36 per cent
by data on investment from and to special purpose entities).
(based on official FDI data) to 24 per cent (according
to UNCTAD’s estimates by ultimate investors). This
is mostly driven by developing Asia (from 47 to 33 per cent). In relative terms, however,
the reduction is sizeable in Africa (from 10 to 7 per cent), as well as Latin America and the
Caribbean (from 11 to 8 per cent). Transition economies, by contrast, register a higher
proportion of intraregional investment when taking ultimate investors into account.

South–South FDI. Behind regional gateways to developing economies are often ultimate
investors based in the developed world. The share of South–South investment in total
investment to developing economies plummets from almost 50 per cent (when measured
based on standard FDI data) to 28 per cent when based on UICs (figure I.23). Although
the rise of investment in developing economies from other developing economies, such
as China or India, is an important trend in the global investment landscape, FDI estimates
by UICs reveal that it is nonetheless less significant than what official data indicate. As
a result, South–South FDI is likely to take longer than expected to reshape the global
production landscape. A thorough assessment of the investment links between developing
economies is especially important in the year of the Buenos Aires Conference on South–
South Cooperation.

The coverage of international investment agreements. The gap between immediate

and ultimate investors generated by indirect FDI has implications for the coverage of
international agreements and regional economic cooperation frameworks (see also
WIR16). The share of investment covered by an agreement in the total inward investment
to member countries may change significantly depending on the view (figure I.24). The
UIC perspective highlights the multilateralizing effect of indirect FDI. For some treaties and
economic groupings, such as the African Continental Free Trade Area (AfCTA) and the
Association of Southeast Asian Nations (ASEAN), in which regional hubs (Mauritius and
Singapore, respectively) have a relevant role, the share of direct investment covered by the

Figure I.24. Intraregional investment in selected economic groupings, share of inward stock, 2017 (Per cent)

By direct investing country By ultimate investing country

AfCFTA 9 6

ASEAN 24 19

CPTPP 16 17

USMCA 20 24

Source: Bilateral FDI by ultimate investing countries: UNCTAD estimates. Bilateral FDI by direct investing countries: UNCTAD bilateral FDI
database (complemented by data on investment from and to special purpose entities).

30 World Investment Report 2019 Special Economic Zones

treaties is higher than the share of investment by UICs. For others, the opposite is true:
the treaty’s weight is more relevant under the ultimate investor perspective. This occurs
when the agreement includes major industrial partners, as is the case of the United States–
Mexico–Canada Agreement (USMCA) and the Comprehensive and Progressive Agreement
for Trans-Pacific Partnership (CPTPP).

Box I.1. UNCTAD estimates of bilateral FDI by ultimate investing country

The large and growing divergence between bilateral FDI positions held by direct investors (as reported by standard bilateral FDI data)
and by ultimate investors is one of the main issues affecting FDI statistics. According to 2016 FDI statistics reported by Germany, for
example, Luxembourg and the Netherlands account for a combined 41 per cent of total bilateral inward FDI in Germany, and the United
States for only 8 per cent. FDI positions by ultimate investors (reported by Germany and few other developed countries) radically modify
this picture, however: the share of the United States rises to 21 per cent, and Luxembourg and the Netherlands combined make up only
14 per cent of German inward FDI stock. Similar differences apply to all other countries whose reported data allow direct comparison.
In this context, standard bilateral FDI data cannot properly uncover ultimate investor relations. The need for bilateral statistics by
ultimate investors to complement standard bilateral FDI is now largely acknowledged by the international community (OECD Benchmark
Definition of Foreign Direct Investment, edition 2008, page 110, item i). Nevertheless, progress in reporting FDI positions on the basis of
ultimate investors has been slow; currently only 14 developed countries provide statistics by ultimate investors. Statistical and analytical
efforts at the international level to bridge this gap are ongoing (Damgaard and Elkjaer, 2017; Borga and Caliandro, 2018).

Box figure I.1.1. UNCTAD approach to search ultimate investors illustrated

…. Non-conduit Conduit
Investors Ultimate investor

B2 B1
Step 3: A0 C0 D0 E0
1-COND (B)
B0
FDI (E, B)
Reversed investment process
Investment process

E2 E1
Step 2: A0 B0 C0 D0
COND (E)
E0
(D, E)
FDI

D2 D1
Step 1: A0 B0 C0 E0
COND (D)
D0
COND(X) = share
FDI (X,Y) = share of Y in total D) of direct investment
(C,
direct investment into X FDI from X made by
conduit entity

Step 0: A B C D E
Recipients Final recipient

Source: UNCTAD.

UNCTAD’s probabilistic approach to estimating investment positions held by ultimate investors combines standard bilateral FDI data,
available for a large set of countries, with appropriate assumptions on conduit FDI. This provides a transition rule to link final recipient
countries to ultimate investors, effectively looking through conduit jurisdictions. More specifically, the distribution of FDI based on
direct investing countries provides the overall exposure of recipient country X to direct investment from investor country Y; at the same
time, assumptions on conduit FDI define whether direct investor Y is an intermediate or an ultimate investor. If investor Y qualifies as
intermediate, the investment process iterates until an ultimate investor arises. Box figure I.1.1 illustrates the logic behind this approach.
Framing the dynamics represented in the figure within the probabilistic setting of absorbing Markov chains makes it possible to
analytically derive the distribution of ultimate investors. The final outcome of the UNCTAD approach is a novel bilateral matrix providing
inward positions by ultimate counterparts for more than 100 recipient countries, covering about 95 per cent of total FDI stock and
including many developing countries.
Source: UNCTAD.
Note: Full methodological details and an empirical validation can be found in the technical background paper on UNCTAD’s UIC data set, published as UNCTAD Insights
in Transnational Corporations (Casella, 2019).

Chapter I Global Investment Trends and Prospects 31

 NOTES
1
FDI data may differ from one WIR issue to another as data are continually revised, updated and corrected
by relevant national authorities, such as central banks and statistical offices, which provide FDI data to
UNCTAD.
2
Only about a third of the MNEs in the top 100 ranking from developing and transition economies reported
R&D expenditures, as most of the State-owned MNEs from extractive or industrial sectors are private and
do not report sufficient information in this context. These are, however, not top R&D investors.
3
State-owned MNEs are defined here as separate legal entities engaged in commercial activities, including
FDI operations through foreign affiliates. In addition, a governmental entity should either own at least
10 per cent of the capital, be the largest shareholder, or hold a “golden share” – a type of share that
gives the government special voting rights to block key strategic decisions, especially takeovers by other
shareholders. Subnational entities in countries with federal governments but significant functions at the
state level (e.g. German Länder, Republics as federal subjects in the Russian Federation, states in the
United States) as well as municipalities are considered State owners.
4
The updated UNCTAD-Eora GVC database can be found at http://worldmrio.com/unctadgvc.

32 World Investment Report 2019 Special Economic Zones