Economics 4413
Keith Maskus
Fall, 2003
Policy Paper 2: Environmental Protection, Trade, and Competitiveness
The linkages between changes in economic activity and environmental
degradation are complex and not easily predicted. The same is true for the relationships
between openness to trade and FDI and environmental use. These impacts presumably
work in both directions and are dependent on circumstances.
How Does Openness Affect Environmental Protection and Use?
It is convenient to use the representation set out by Grossman and Krueger (1993)
to describe the various impacts of any changes in an economy’s fundamental variables –
endowments, prices, technologies, and policies – on aggregate use of the environment.1
For this purpose, I use the word “emissions” to proxy for all forms of environmental use,
including air and water pollution, deforestation, habitat destruction, and waste deposits.
Put in simplest terms, the amount of emissions an economy generates depends on three
essential factors: the size of the economy, the share of output that is produced by
emission-intensive (“dirty”) sectors, and the degree of emissions intensity in those
sectors.
This means that any change in pollution emissions can be decomposed into three
effects. First, the scale effect refers to an increase in emissions associated with a larger
GDP, holding constant the relative mix of outputs and pollution intensities across sectors.
That is, a ten-percent increase in all productive factors, everything else held constant,
should raise pollution by ten percent.2 Second, the composition effect refers to a change
in the share of dirty goods in GDP, which may come about because of a price change
favoring their production. With a constant scale of the economy and no change in
emissions intensities per industry, this effect would increase total pollution.
Third, the technique effect refers to a change in the amount of emissions per unit
of output across sectors (an “emissions intensity” change). A higher pollution tax would
induce firms to adopt cleaner technologies, for example. One important reason that a
government would increase the pollution tax (or generally increase environmental
regulation) is that voters demand a cleaner environment as incomes rise, assuming (with
justification) that environmental health is a normal good. Thus, in the applied literature
the term technique effect generally refers to the idea that anything raising per-capita
income in the economy generates an endogenous increase in environmental taxes, thereby
reducing the pollution intensity of production.
It is possible to characterize the impacts of exogenous changes in an economy
along these lines. Thus, for example, suppose that capital-intensive goods are also
pollution-intensive. Then capital accumulation over time will tend to increase pollution
through the composition effect because it pushes the economy toward specialization in
capital-intensive goods. It also tends to expand the scale of the economy, which
1
Copeland and Taylor (2001) provide a clear presentation of this framework.
This assumes that net output (total output minus resources devoted to pollution abatement) is produced
with constant returns to scale.
2
1
increases emissions. Note, however, that an increase in the labor force would tend to
reduce emissions through the composition effect by inducing the economy to shift
resources into labor-intensive goods, though the scale effect would remain positive.
Next, consider an exogenous improvement in production technology. If it were
neutral across sectors, there would be only a scale effect, raising pollution. However, if it
increased the productivity of capital and not labor, the change would be effectively
equivalent to an increase in the capital stock, thereby increasing emissions through both
the scale and composition impacts. It is possible that the technological improvement
would reduce the ratio of emissions to output across all sectors, in which case there
would be a pure technique effect tending to diminish emissions, other things equal.
Next, a change in commodity prices has important impacts on environmental use.
An increase in the price of clean goods, for example, would reduce emissions through a
composition effect. If this price increase tended to raise aggregate real incomes, the
technique effect would further reduce pollution intensities.
Consider the role of trade liberalization and FDI in this context. Trade
liberalization implies reducing impediments to imports and exports.3 This policy would
affect environmental use through its impacts on prices, which then filter through to
production and consumption. For example, suppose that an East Asian developing
economy is labor abundant and therefore has a comparative advantage in labor-intensive
goods. Freeing up trade would raise the price of labor-intensive goods in this economy.
The composition effect would arise from the shift of capital and labor into production of
these goods. If these goods are produced with cleaner technologies on average,
emissions will fall as output shifts away from dirtier goods, even if emissions intensities
do not change. Note that pollution could rise if the goods that a country exports are
dirtier, so comparative advantage is an important factor.
This shift in resources is efficient in the sense of raising aggregate productivity in
the economy (even if it worsens the environment), thereby raising real GDP. Thus, the
scale effect of trade liberalization is to increase pollution; in general a rise in economic
activity, other things equal, will raise demands to use the environment. Finally,
aggregate and per-capita incomes would also be higher as a result of trade liberalization,
tending to raise the demand for a cleaner environment.4 Thus, the technique effect of
tariff cuts is to improve the environment by reducing emissions per unit of output.
Three other comments are relevant here. First, the impacts of openness are less
clear when account is taken of input flows. Inputs themselves may be dirty or clean, so
there is an additional comparative advantage effect to consider. If trade liberalization
raises access to imported coal and petroleum, the net impact may be to expand output of
dirty final goods even if there is a direct comparative disadvantage in them. Moreover,
tariff cuts could expand consumption of cheaper imported fuels.
Second, many analysts suggest that as incomes rise there will be an incentive to
improve environmental technologies endogenously due to higher demand associated with
rising incomes. One direct means by which this could come about is that more open
economies have greater access to foreign technologies. In this context, trade openness
3
It could also mean domestic deregulation of services and changes in other policies, such as intellectual
property rights, the effects of which are left aside here.
4
Per-capita income would be higher despite changes in the distribution of income.
2
provides an endogenous technique effect that could be decisive in improving
environmental use.
Third, for the standard technique effect to operate, citizen demand for a cleaner
environment must be mediated effectively through an increase in the cost of polluting the
environment. In some cases this may be done through informal pressures, as analyzed by
Wang and Wheeler (1999). Generally, however, the government must be responsive to
citizen preferences and raise environmental charges as incomes grow. Again, therefore,
the openness and efficiency of the political process matter for environmental policy, as
they do for labor standards.5
In many dimensions the role of FDI may be analyzed in terms similar to those
applied to trade liberalization. If MNE’s exist disproportionately in dirtier industries,
their arrival in a country is like an increase in the capital stock that worsens the
environment through the composition effect. However, if they exist disproportionately in
cleaner industries, their arrival can improve the environment. This is clearly an empirical
question, as is discussed below. By expanding economic activity, MNE’s would generate
a scale effect that increases environmental use.6
Perhaps most fundamentally, the operations of MNE’s can generate beneficial
technique effects. They are likely to transfer cleaner technologies, developed in their
home economies, to recipient countries, for a variety of reasons (Moran, 2002). If they
pay higher than average wages without generating offsetting unemployment, the impact
would be higher per-capita incomes and an induced demand for stronger environmental
protection. Thus, even if MNE’s do not change the composition of output much, one
would anticipate their operations to improve environmental use. In this context,
economies that are more open to FDI, other things equal, would tend to have stronger
environmental standards and cleaner technologies. Again, this is an empirical question.
There is important empirical evidence available on both of these questions and I
provide here only a selective and brief review. The most prominent study is by Antweiler
and others (2001). Using data on sulfur dioxide concentrations in 293 sites in 44
countries from the Global Environment Monitoring Project over the period 1971-1996,
they decomposed emissions into scale (GDP), composition (capital-labor endowment
ratios), and technique (real income) effects, interacting each with a measure of openness
to trade. They found that international trade generates relatively minor changes in
concentrations of air pollution when it alters the composition of output, but that the
associated technique and scale effects reduce pollution. Overall they found that within
their sample, openness to trade actually reduces sulfur dioxide concentrations on average.
Beghin, et al (2000) constructed an empirical simulation model to consider
relationships between trade liberalization and pollution in Chile. Trade policies
examined included Chile’s accession to NAFTA, MERCOSUR, and unilateral free trade.
In their CGE model, unilateral liberalization generated substantial worsening of air
pollution by providing cheaper and dirtier energy sources. However, if this trade policy
were combined with an appropriate tax on emissions there would be significant welfare
gains for Chile. This insight underscores an important observation about the nexus
5
Barrett and Graddy (2000) find that increases in political freedoms significantly improve environmental
quality.
6
Not all would agree that FDI expands economic activity so much as it eliminates domestic competition.
3
between trade policy and environmental policy. In general, if environmental distortions
are internalized efficiently, open trade is a welfare-enhancing policy.
Dean (2002) analyzed the impacts of trade liberalization on water pollution in
Chinese provinces from 1987 to 1995, a period in which there were both an extensive
pollution levy system and significant opening to trade. She adapted the Copeland-Taylor
framework to a setting with endogenous environmental policy, estimating a two-equation
model in which trade openness both directly affects environmental use through a
composition effect and indirectly through an induced technique effect. She found that
China has a comparative advantage in pollution-intensive goods, so that trade
liberalization has aggravated environmental damage. However, openness also generated
higher per-capita incomes, mitigating the environmental costs through stronger
regulation. She ran counterfactual simulations and discovered that emissions per unit of
industrial output in China would have been much higher in the absence of trade reform,
so that China’s opening to trade was beneficial for the environment overall.
Arunanondchai (2001) developed a CGE model to assess the impacts of cuts in
tariffs and export taxes in logging and timber products from the Uruguay Round on
Indonesian and Malaysian exporters. She found that trade liberalization would not
necessarily generate more log production because it might not cause net producer prices
to rise. However, foreign tariff cuts in plywood and sawn lumber would generate
significant gains for Indonesia. On the other hand, she found that removal of export taxes
on logs would reduce world log prices, tending to worsen joint welfare for exporting
nations.
In sum, it is not possible to predict the implications of trade liberalization for
environmental use and economic welfare in general. However, the balance of
econometric and empirical evidence suggests that it can be beneficial.
Turning to FDI, studies of its impacts on environmental damage are surprisingly
7
scarce. Moran (2002) provides some anecdotal evidence that MNE’s are associated with
improvements in environmental use, though this depends on its industry of operation.
Environmental Standards and Competitiveness
Turning to the opposite flow of causation, what is the evidence that weak
environmental protection generates either an export advantage or attracts FDI? This
notion is referred to as the “pollution haven hypothesis” and has attracted much study.
Note first that the idea is more sensible than its counterpart in labor standards, for the
environment cannot complain or shirk when it is treated badly, suggesting that the ability
to exploit it without regulation could reduce costs, at least up to limits from congestion.
In contrast, workers who are treated poorly may simply react with inefficient efforts,
which could raise costs overall. In principle, then, one might expect a stronger
identification of weak regulation with trade performance and FDI in the environmental
area.
Prior Evidence
This prediction was borne out in the study of environmental standards and trade
by Wilson, et al (2002). They regressed net exports in five pollution-intensive industries
on measures of factor endowments and environmental laws in 24 countries between 1994
7
It is possible they exist but I could not locate any.
4
and 1998. There were six OECD countries (including South Korea) and 18 developing
countries, while they developed an index for the stringency of legislation. Their results
suggested that, controlling for various sources of heterogeneity, more stringent
environmental laws generated lower net exports of these goods. Thus, weak
environmental regulation is associated with higher trade performance. Moreover, a
hypothetical trade agreement to harmonize laws at levels higher than those in developing
countries would reduce trade by up to 11 percent per year.
The Wilson, et al paper is unique in this finding. Earlier studies by Tobey (1990),
Low and Yeats (1993), and Xu (1999) discovered no evidence that a country with stricter
environmental standards would have lower exports of pollution-intensive goods.
Grossman and Krueger (1993) found no contribution of U.S. pollution intensity to U.S.
imports from Mexico. Most observers have concluded that, because environmental
controls typically amount to a small percentage of total costs, they are relatively
unimportant in determining trade patterns.
In contrast, Levinson and Taylor (2001) point out that the use of pollution
abatement expenditures or particulate concentrations as an independent variable in such
regressions suffers significant biases from the fact that they are endogenous to trade
liberalization and FDI (the composition and technique effects). Incorporating more
appropriate measures of environmental control costs in the United States during the
1970s and 1980s, they found that those industries experiencing the largest increases in
costs had the largest relative increases in net imports. Thus, trade flows do react to such
regulation. Unfortunately, there are not comparable measures in developing countries to
see if this result holds more widely.
The role of weak environmental protection as a “pollution haven” for migrating
industry has been subject to more recent study. Again, the preponderance of evidence
suggests that little impact can be found that FDI is sensitive to international variations in
environmental costs. For example, Eskeland and Harrison (1997) found almost no
evidence that MNE’s investing in developing countries are attempting to escape higher
environmental costs in their home countries. They also discovered that foreign-owned
plants in developing countries are less polluting than comparable domestic plants.
Wheeler (2001) discussed why the “race to the bottom” idea in environmental standards
makes little sense in terms of attracting FDI. He found that indexes of air pollution
actually improved markedly in the major cities in China, Brazil, and Mexico during an
era of extensive investment inflows. Smarzynska and Wei (2001) examined firm-level
data in 24 transition economies, controlling for corruption levels, and found little support
for the pollution haven idea.8 Levinson (1996) discovered no evidence that differences in
environmental standards across states affected the location choices of manufacturing
plants.
Again, this view is not universal. Lucas, et al (1990) claimed that increasingly
strict environmental regulations in OECD countries led to displacement of pollutionintensive industries. List and Co (1999) found that FDI in the United States was
negatively related to regulatory expenditures per firm. Mani, et al (1997) indicated that
spending on environmental damage abatement was higher in more pollution-intensive
industries in India, which was a factor in plant location decisions. Again, however,
8
See also Jaffe, et al (1995).
5
Keller and Levinson (2002) point out problems with endogeneity in such measures. The
issues are inherently empirical.
APEC (1999), Study of Non-Tariff Measures in the Forest Products Sector, Committee
on Trade and Investment, APEC: Singapore.
APEC (2001), Survey of Environmental Markets in APEC, Committee on Trade and
Investment, APEC: Singapore.
Antweiler, Werner, Brian R. Copeland, and M. Scott Taylor (2001), “Is Free Trade Good
for the Environment?” American Economic Review 91: 877-908.
Arunanondchai, May (2001), “Trade Policy and the Welfare of Southeast Asian Timber
Exporters: Some Implications for Forest Resources,” Economy and Environment
Program for Southeast Asia, Research Report.
Barrett, Scott and Kathryn Graddy (2000), “Freedom, Growth, and the Environment,”
Environment and Development Economics 5: 433-456.
Beghin, John C., Bradley J. Bowland, Sebastien Dessus, David Roland-Holst, and
Dominqique van der Mensbrugghe (2000), “Trade Integration, Environmental
Degradation, and Public Health in Chile: Assessing the Linkages,” Environment and
Development Economics 7: 241-267.
Copeland, Brian R. and M. Scott Taylor (2001), “International Trade and the
Environment: A Framework for Analysis,” National Bureau of Economic Research
Working Paper 8540.
Dean, Judith M. (2002), “Does Trade Liberalization Harm the Environment? A New
Test.” Canadian Journal of Economics, forthcoming.
Eskeland, Gunnar S. and Ann E. Harrison (1997), “Moving to Greener Pastures?
Multinationals and the Pollution Haven Hypothesis,” The World Bank, manuscript.
Grossman, Gene M. and Alan B. Krueger (1993), “Environmental Impacts of a North
American Free Trade Agreement,” in Peter Garber, editor, The U.S.-Mexico Free Trade
Agreement, Cambridge: MIT Press.
Jaffe, A., S. Peterson, P. Portney, and R. Stavins (1995), “Environmental Regulation and
the Competitiveness of U.S. Manufacturing. What Does the Evidence Tell Us?” Journal
of Economic Literature 33: 132-163.
Keller, Wolfgang and Arik Levinson (2002), “Environmental Regulations and FDI to
U.S. States,” Review of Economics and Statistics, forthcoming.
6
Levinson, Arik (1996), “Environmental Regulation and Manufacturers’ Location
Choices: Evidence from the Census of Manufactures,” Journal of Public Economics 62:
5-29.
Levinson, Arik and M. Scott Taylor (2001), “Trade and the Environment: Unmasking the
Pollution Haven Effect,” Georgetown University, manuscript.
List, John. A. and Catherine Y. Co (1999), “The Effects of Environmental Regulations on
Foreign Direct Investment,” Journal of Environmental Economics and Management 40:
1-20.
Low, Patrick and Alexander Yeats (1992), “Do Dirty Industries Migrate?” in Patrick
Low, editor, International Trade and the Environment, World Bank Discussion Paper
159: 89-103.
Lucas, R. E. B., D. Wheeler, and H. Hettige (1990), “Economic Development,
Environmental Regulation, and the International Migration of Toxic Industrial Pollution,”
Paper presented at the Symposium on International Trade and the Environment,
Washington DC.
Mani, M., S. Pargal, and M. Huq (1997), “Does Environmental Regulation Matter?
Determinants of the Location of New Manufacturing Plants in India,” World Bank,
Policy Research Working Paper 1718.
Moran, Theodore H. (2002), Beyond Sweatshops: Foreign Direct Investment and
Globalization in Developing Countries, Washington DC: Brookings Institution.
Organization for Economic Cooperation and Development (1996), Trade, Employment,
and Labor Standards: A Study of Core Workers’ Rights and International Trade, Paris:
OECD.
Pargal, Sheoli, Hemamala Hettige, Manjula Singh, and David Wheeler (1997), “Formal
and Informal Regulation of Industrial Pollution: Comparative Evidence from Indonesia
and the United States,” World Bank Policy Research Working Paper no.1797.
Smarzynska, Beata and Shang-Jin Wei (2001), “Pollution Havens and Foreign Direct
Investment: Dirty Secret or Popular Myth?” Centre for Economic Policy Research
Discussion Paper 2966.
Tobey, J. A. (1990), “The Effects of Domestic Environmental Policies on Patterns of
World Trade: An Empirical Test,” Kyklos 43: 191-209.
Wang, Hua and David Wheeler (1999), “Endogenous Enforcement and Effectiveness of
China’s Pollution Levy System,” World Bank, Development Research Group,
manuscript.
7
Wang, Hua, Nlandu Mamingi, Benoit Laplante, and Susmita Dasgupta (2002),
“Incomplete Enforcement of Pollution Regulation: Bargaining Power of Chinese
Factories,” World Bank, Development Research Group, manuscript.
Wheeler, David S. (2001), “Racing to the Bottom? Foreign Investment and Air Pollution
in Developing Countries,” Journal of Environment and Development 10: 225-245.
Wilson, John S., Tsunehiro Otsuki, and Mirvat Sewadeh (2002), “Dirty Exports and
Environmental Regulation: Do Standards Matter to Trade?” World Bank, Development
Research Group, manuscript.
8