Measuring Carbon Emissions

of Foreign Direct Investment

in Host Economies
NOVEMBER 17,
2021

Maria Borga
Balance of Payments Division
Deputy Division Chief
IMF
IMF | Statistical
Statistics Forum 1
4
 Measuring Carbon Emissions of Foreign Direct
Investment in Host Economies

1. Introduction
2. Methodology and data
3. Limitations
4. Results
5. Conclusions and policy implications

IMF | Statistics 2
 Introduction (1)
▪ Foreign Direct Investment (FDI) effects on host economies are numerous and
complex but common themes in the literature point to
➢ rising wages;

➢ productivity and knowledge spillovers to domestic firms;

➢ exports diversification and introduction of new industries and

➢ increasing growth

▪ Environment and sustainability effects of FDI are unclear; but some have argued
that:
➢ If demand for environmental quality increases as incomes rise, then as FDI
increases incomes, environmental damage falls
➢ Countries with low incomes tend to set low pollution standards to attract
resource seeking as well as pollution intensive FDI leading to FDI that
increases emissions
➢ FDI deploys new technologies that are cleaner than domestic producers, thus
supporting improvements in the environment of the host country
IMF | Statistics 3
 Introduction (2)
▪ In this paper, we develop a framework for estimating possible effects of FDI on
emissions in host countries using industry level information on production, trade,
investment, and carbon emissions.

▪ Framework aims to provide estimates to address the following key questions:

➢ What is the effect of greenfield investment and capacity extension resulting
from FDI on emissions in host economies?
➢ What is the effect of the operations of foreign owned enterprises on emissions
in host economies?
➢ What effect does international trade activity of foreign owned enterprises have
on emissions in host economies?

IMF | Statistics 4
 Methodology and data (1)

… FDI benefits host economies by expanding capacity through greenfield investments or

new investments in existing operations, which results in carbon emissions in the
production units involved in the creation of the new capacity or expansion of existing
capacity.
Indicator Concept Methodology Data Source
Carbon The direct and indirect (1) In each industry take the Output multipliers for IEA production-
emissions amount of carbon each contributing industry from the World based emissions,
in supply to dioxide emitted into the Input-Output table (Output Multiplier) OECD Input
Gross Fixed atmosphere from the (2) Calculate a carbon emission intensity (Carbon Output Database
Capital production used in emissions / output) and FDI financial
Formation Gross Fixed Capital (3) Multiply the Carbon emission intensity by the flows database
of FDI Formation (GFCF) output multiplier by the final use of GFCF by the
funded by FDI. FDI ratio to GFCF

IMF | Statistics 5
 Methodology and data (2)

… Increased capacity increases the scale of economic activity, results in export

diversification, and leads to structural changes in the economy through the introduction
of new industries but also generates carbon emissions in the host economy.
Indicator Concept Methodology Data Source
Carbon The direct and (1) In each industry take the output multipliers for IEA production-
emissions indirect amount of each contributing industry by ownership from based emissions,
embodied in carbon dioxide the Activities of Multinational Enterprises Intercountry Input
Multinational emitted into the Inter Country Input-Output table (Output Output Tables from
Enterprises atmosphere from Multiplier) the OECD
(MNEs) output/exports of (2) For each industry by ownership calculate a Activities of
Output/Exports MNEs. carbon emission intensity (Carbon emissions / Multinational
(2 indicators) output) Enterprises
(3) Multiply for a given industry the Carbon database.
emission intensity * output multiplier *
output/exports

IMF | Statistics 6
 Limitations
▪ Data:
➢ Unavailability of estimates on FDI by use (i.e., Greenfield FDI);
➢ Limited geographic coverage;
➢ Source data used available for only 2005-2015 (no Input Output Tables (IOTs) for other
periods);
➢ Direct emission estimates are not disaggregated between MNEs and Domestic Owned
enterprises (DOEs);
➢ Direct emission estimates are available based on only Tier 1 method.

▪ Methodology:
➢ Conflict between the hypothesis of matrix calculation and the IOT balancing approach
resulted in aggregation of some sectors of MNEs or DOEs affecting reliability of
interrelations.
➢ Central equation system of input-output analysis fails to reflect dynamic interactions
between the respective variables
➢ IOTs pertain to lack of constraints on the factors of production and impose on the supply
side, a fixed input structure and fixed ratios for production for each industry

IMF | Statistics 7
 Carbon emissions associated with the investment
effect of FDI - Results (1)
…emissions associated with FDI effect on GFCF have continued to fall even as FDI recovers.

2,500 900,000

800,000
Metric tons per 1 million US$ of final

2,000
700,000

600,000
1,500
demand

500,000

400,000
1,000
300,000

200,000
500

100,000

- 0
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

CO2 emissions in GFCF of FDI Total inward FDI flows (US$ millions) - right axis

IMF | Statistics 8
 Carbon emissions associated with the investment
effect of FDI - Results (2)
…sector emission estimates are highest in electricity, …Country estimates are highest in Australia and Estonia
manufacturing, and construction and lowest in France and Switzerland

United States
Agriculture
United Kingdom
Construction
Switzerland
Elect. Sweden 2006
2006
IT 2007 Portugal 2007
2008 Poland, Rep. of 2008
Manufacturing 2009
2009 Mexico
2010 2010
Mining
Iceland
2011 2011
Other business Greece 2012
2012
2013 France 2013
Others
2014 Estonia, Rep. of 2014
Publishing 2015 2015
Czech Rep.
Transportation
Canada
Wholesale and retail trade Australia

- 1,000 2,000 3,000 4,000 5,000 - 500 1,000 1,500 2,000 2,500 3,000
IMF | Statistics 9
 Carbon emissions associated with the investment
effect of FDI - Results (3)
… electricity, construction and manufacturing sectors account for most emissions (in metric tons per million US$ of
output) at country level as well

Estonia
Australia
250
400
350 200
300
250 Construction 150 Construction
200 Elect., gas, & water
Elect., gas, & water 100
150
Manufacturing Manufacturing
100 50
50
0 0
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Iceland Poland
120 80
70
100
Agric., forestry and 60
80 fishing 50 Construction
60 Construction 40
Elect., gas, & water
40 30
Manufacturing Manufacturing
20
20
10
0 0
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
IMF | Statistics 10
 Carbon emissions from ongoing operations of MNEs –
Results : distribution of output emissions, 2005-2015
….shares of MNEs emissions were generally low with IT,
….. manufacturing, trade, and electricity had the highest contributions manufacturing, electricity, financial and insurance activities having
to MNE emissions the highest shares.
MNE emissions by industry/total MNE emissions MNE emissions by industry/total emissions by industry
Manufacturing IT
Wholesale and retail trade Manufacturing
Electricity Electricity
Financial activities Financial
Transportation Publishing
Other business Wholesale and retail trade
Mining Telecommunications
Construction Transportation
IT Other business
Telecommunications Mining
Publishing Construction
Real estate Accommodation
Accommodation Arts
Agriculture Human health
Human health Real estate
Arts Agriculture
Education Education
Public admin. Public admin
0% 10% 20% 30% 40% 50% 0% 2% 4% 6% 8% 10%

IMF | Statistics 11
 Carbon emissions from ongoing operations of MNEs -
Results - output emission intensities, 2005-2015
….Country estimates show cases of higher emission intensities in
….. Industry level intensities show lower emission intensities for MNEs compared to domestic enterprises (Malta, Luxemburg, Norway,
MNEs compared to domestic enterprises Hong Kong, Switzerland, and Ireland)
Russian Federation
Mining China (People's Republic of)
Electricity
Basic metals
South Africa
Other business Germany
Wholesale and retail trade Poland
Chemicals Bulgaria
Coke Estonia
Transportation Australia
Financial Singapore
Agriculture
Other non-metallic mineral
Mexico
Textiles Turkey
Machinery United Kingdom
Electrical equip Philippines
Computer Netherlands
Rubber and plastic Canada
Fabricated metal DOEs
Greece
Paper products MNEs
Motor vehicles
Belgium
Food products Latvia
Real estate Israel 1
Other manufacturing Malta
Other transport Chile
Telecommunications Luxembourg
Wood Colombia DOEs MNEs
Construction
Public admin.
Slovenia
Publishing Denmark
Accommodation Austria
IT Croatia
Human health Hong Kong, China
Education Switzerland
Arts Iceland
0 5 10 15 20 25 30 0 0.5 1 1.5 2 2.5

IMF | Statistics 12
 Carbon emissions from ongoing operations of
MNEs - Results - emissions in exports, 2005-2015
….sizeable share of the emissions in the low carbon intensity countries is driven by foreign demand - MNEs in
Belgium, Hungary, Luxemburg, Malaysia, Slovak, Slovenia, Switzerland and Philippines have relatively low
emission intensities, but more than half of their output is exported

75%
MNEs exports/MNEs output

70% Malta

65%
Belgium
60% Ireland
Hungary Malaysia
Luxembourg
55% Slovenia Slovak Republic Singapore
Thailand
Philippines
Israel
50% Viet Nam Estonia Saudi Arabia
Switzerland Netherlands
Cyprus Chinese Taipei
45%
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Carbon intensity of MNEs

IMF | Statistics 13
IMF | Statistics
30%
70%
80%

10%
20%
40%
50%
60%

0%
Iceland
Costa Rica

countries
Switzerland
Ireland
Sweden
Croatia
Finland
Austria
Hungary
Denmark
Slovenia
Portugal
Cyprus 2
Colombia
Lithuania
Slovak Republic

Carbon intensity - right axis

Norway
Chile
New Zealand
Morocco
Israel 1
Latvia
Malta
Luxembourg
Brazil
Hong Kong, China
Romania
Argentina

MNEs emisions in exports/MNEs emisions in gross output

Belgium
Greece
Canada
Spain
Philippines
Czech Republic
Chinese Taipei
Netherlands
Turkey
Italy
United Kingdom
Mexico
France
Korea
Viet Nam
Estonia
Thailand
Australia
Singapore
Malaysia
Bulgaria
Poland
Indonesia
Germany
Japan
India
DOEs emissions in exports/DOEs emissions in gross output

South Africa
Carbon emissions from ongoing operations of

United States
…. export related emissions shares are higher for MNEs compared to domestic enterprises in most

China (People's Republic…

Saudi Arabia
Russian Federation
2

0
1
MNEs - Results - emissions in exports, 2005-2015

2.5

0.5
1.5

14
 Policy implications

▪ Home economies:
➢ Incentivizing domestic investors to meet high environmental standards at home and abroad could be
important to reducing global emissions
➢ Could reduce emissions directly by inducing firms to use lower carbon production functions and technology
and also by inducing them to demand lower carbon infrastructure and transportation in the host economies
➢ If firms were also encouraged to reduce emissions along their supply chains, could induce them to demand
their suppliers reduce emissions

▪ Host economies:
➢ Remove barriers to investment in environmental goods and services sectors as well as in low carbon
technologies to promote positive spillovers and knowledge and technology transfer to the domestic economy
➢ Include an analysis of the impact on carbon emissions as part of their FDI attraction strategies ;

▪ Developing a standard for companies to disclose their carbon emissions will provide valuable information that can
help us better understand the role of all enterprises, both MNEs and DOEs, in carbon emissions

IMF | Statistics 15