World Trade Review

http://journals.cambridge.org/WTR

Additional services for World Trade Review:

Email alerts: Click here

Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here

Environment and trade: the implications of imperfect

information and political economy

DANIEL STURM and ALISTAIR ULPH

World Trade Review / Volume 1 / Issue 03 / November 2002, pp 235 - 256

DOI: 10.1017/S1474745602001234, Published online: 14 February 2003

Link to this article: http://journals.cambridge.org/abstract_S1474745602001234

How to cite this article:

DANIEL STURM and ALISTAIR ULPH (2002). Environment and trade: the implications of imperfect
information and political economy. World Trade Review, 1, pp 235-256 doi:10.1017/
S1474745602001234

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/WTR, IP address: 128.218.248.200 on 14 Mar 2015

 World Trade Review (2002), 1 : 3, 235–256 Printed in the United Kingdom
f Daniel Sturm and Alistair Ulph DOI : 10.1017/S1474745602001234

Environment and trade: the implications of

imperfect information and political
economy
DANIEL STURM
University of Munich and CEPR

A L I S T A I R U L P H*
University of Southampton

Abstract : The last ten years have seen an upsurge in interest in the nexus of trade
and environmental policies. In part this reflects the need to deal with major global
pollution problems, and in part a concern that globalization may have adverse
impacts on the environment. Environmentalists worry that globalization may
trigger a race-to-the-bottom in environmental standards. While they would like to
see upward harmonization in environmental standards, they are sceptical about
the ability of supra-national agencies to achieve this. Industrialists also raise
concerns about the need for a ‘level playing field’ in environmental regulations
because of fears about the impact of environmental regulations on
competitiveness. However, developing countries question whether disputes over
differences in environmental regulations simply reflect a covert form of ‘ green
protectionism’. In this paper we review what light recent developments in
economic analysis (conceptual and empirical) can shed on these concerns. We
quickly summarize conventional trade models in which government bodies have
perfect information and are welfare maximizers, and show that this analysis does
not provide much support for the concerns or proposed policy recommendations.
We then turn to models of political economy and imperfect information to see
whether they provide a better explanation for the concerns and policy
recommendations.

1. Introduction
The links between environmental policy and international trade have been a very
prominent item in public policy debates over the last decade. This prominence was
fuelled by a number of concerns. The first was probably the increasing awareness of

* The first draft of this paper was written while the second author was visiting the Economics Program,
Research School of Social Sciences, Australian National University, Canberra, whose hospitality is
gratefully acknowledged. We are grateful to the editor and two anonymous referees for comments, which
helped to improve the exposition.

235

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 236 DANIEL STURM AND ALISTAIR ULPH

transboundary or even global environmental problems such as acid rain, ozone

depletion and climate change. To deal with such problems required international
environmental agreements (IEAs) and it was recognized that standard free-rider
problems might make it difficult to get effective agreements. However, in the
presence of increasing international trade these difficulties might be exacerbated by
the ‘ leakage ’ problem : action to regulate pollutants by any group of countries (or
individual country) might encourage production and associated pollution to switch
to unregulated countries. This led to the call to back IEAs with trade measures,
which raises the question of how such measures would be viewed by GATT/WTO.
A second concern was that, even with purely local pollution problems, liberal-
ization of trade and capital flows might pose a threat to the environment. The
increasingly vociferous campaign against globalization has argued that trade lib-
eralization would first expand production, consumption and transport of goods,
exacerbating damage to the environment and second make governments more
concerned about the loss of competitiveness if they set tougher environmental
policies than other countries. Governments might then set weaker environmen-
tal standards than warranted by environmental damage costs (‘ environmental
dumping ’), causing a ‘ regulatory chill ’ or even a ‘ race-to-the-bottom’ in en-
vironmental standards as governments competed to attract internationally mobile
capital (Esty (1994)). Environmentalists have argued that countries with tough
environmental standards should be able to impose countervailing trade measures
against countries with weaker standards elsewhere. However, at least if these in-
volved production and process methods (PPM), such measures would be outlawed
by GATT (as confirmed by the famous tuna–dolphin case). Fears about ‘ environ-
mental dumping ’ also led to calls for some supra-national agency to coordinate and
even harmonize environmental standards across countries, although many en-
vironmentalists were sceptical about a body such as the WTO playing this role,
fearing it had been captured by transnational companies, and would always
privilege free trade over the environment. The call for harmonization of environ-
mental standards was supported by industrialists arguing for a ‘ level playing field ’,
but strongly opposed by developing countries who saw this as a form of ‘ green
protectionism ’, denying them an ability to exploit one source of comparative ad-
vantage – a relatively undamaged environment.
Finally there have been a large and growing number of trade disputes over
national product standards. The typical pattern is that a country implements a new
product regulation, which it argues to be necessary for either consumer or en-
vironmental protection, while importers to that market challenge the regulation as a
‘disguised barrier to trade ’ or simply ‘ green protectionism ’. The EU and the United
States have, for example battled intensively over imports of genetically modified
food into the EU and have spent a decade in court over hormones found in US beef
exports to Europe. Supra-National Institutions, such as the WTO and the European
Court of Justice, were forced to rule on such trade disputes and have frequently
struck down the contested national regulations as unnecessarily trade restrictive.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 237

This has left environmentalists with the conviction that in our increasingly global-
izes world national environmental standards are at the mercy of politically un-
accountable supra-national institutions, which they suspect to be willing to promote
international trade at the expense of sound environmental protection.
Alongside this public debate there has been a substantial growth in the economics
literature on trade and the environment. In part this reflected the natural wish for
researchers to address issues of current concern, but also the development of new
tools with which to address the issues.1 There are already a number of surveys of this
literature including some written by ourselves (Dean (1992), Wilson (1996), Ulph
(1997a, b), Rauscher (2001), Schulze and Ursprung (2001), Sturm (2002)). How-
ever in this survey we shall give more emphasis to recent work, including some of our
own, which stresses the importance of political economy aspects and informational
problems, and the difficulties these might pose for resolving environmentally driven
trade disputes. In the next section we give an overview of the conventional literature,
which assumes welfare-maximizing governments and perfect information. In sec-
tion 3 we introduce informational problems, in section 4 political economy issues,
and in section 5 we argue that it is the interaction between informational problems
and political economy that poses the real challenge.

2. Environmental and trade policies with welfare-maximizing governments

and perfect information
In this section we summarize2 what economic analysis has to say about first-best and
second-best trade and environmental policies when governments are welfare
maximizing and there is perfect information. In considering optimal policies we
distinguish between what might be optimal from the perspective of an individual
country acting in its own self-interest from what might be globally optimal. For ease
of exposition we concentrate on pollution linked to production (PPM). Initially we
ignore transboundary pollution.
We begin with the textbook model of competitive markets and a small open
economy. With no transboundary pollution, optimal policies for each country
would be to have free trade and Pigouvian environmental policies (policies which
ensure that marginal abatement costs are equalized across all polluters and equal to
marginal damage costs). Comparative advantage applies : amongst countries with
the same relative endowment of other factors, those which are well-endowed with
environmental resources (in a general sense – see Rauscher (2001)) will export
goods which are on average ‘ dirty ’ goods, while those which have relatively scarce
environments will export ‘clean ’ goods. In general there is no reason to believe that
countries will set the same environmental standards.

1 There is now a new JEL-code for this field – F18.

2 A fuller version of this section can be found in Sturm and Ulph (2002).

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 238 DANIEL STURM AND ALISTAIR ULPH

Turning to second-best policies, if a country had an irremovable tariff in place

which encouraged too much production of a dirty good, then environmental policy
should be made tougher than Pigouvian to offset this trade distortion. Similarly if
there was inadequate environmental policy it might be desirable to adjust trade
polices ; if there was inadequate internalization of environmental damages on a
good that was exported it might be better to have a tariff on exports. But trade
policies will rarely be even second-best policies for dealing with environmental
damage.
If markets are competitive but a country is large enough to affect its terms-of-
trade, then the optimal policies for a country acting independently are to use trade
policies to exploit market power and Pigouvian policies to deal with environmental
problems (the policy targeting principle). Of course this outcome is not globally
efficient : efficiency requires free trade and Pigouvian environmental policies.
Turning to second-best policies, suppose, because of trade liberalization, a country
with market power cannot impose tariffs. Then, if it is an exporter, it will set
environmental policies which are tougher than Pigouvian (to restrict domestic
supply and hence drive up the world price it receives), while, if it is an importer, it
will set policies weaker than Pigouvian (to expand domestic supply and hence drive
down the world price it pays). There is no reason to believe there will be a systematic
tendency for all countries to engage in either environmental dumping or environ-
mental protectionism. However, if a country, for some reason, is unable to set
Pigouvian environmental policies, there may be a case for adjusting trade policies to
reflect this, in the way analysed in the small open economy case, but again it is
unlikely that trade policies are even second-best policies to address environmental
problems.
Provided environmental policies properly internalize environmental damages,
trade liberalization benefits all countries. If there are inadequate environmental
policies, trade liberalization may not be beneficial to some countries, but, as argued,
trade instruments are unlikely to be second-best ways of dealing with environmental
policy failures. What about the impact on the environment ? Following Grossman
and Krueger (1993) and Copeland and Taylor (1994) the impact of a trade liber-
alization on environmental quality can be decomposed into three effects : the
composition effect (changing the mix of goods produced) ; the technique effect
(changing how goods are produced) ; and the scale effect (higher income expands
demand for all normal goods, including the environment). The net impact of all
three effects is ambiguous, and so it is an empirical matter how trade liberalization
affects the environment. With a strong enough income effect on demand for a
cleaner environment, it is possible that trade liberalization will not only raise
welfare in all countries, but also lead to a cleaner environment in all countries.
Suppose now that markets are imperfectly competitive. Optimal policies from the
perspective of each government are again to use trade policies to achieve trade
objectives and Pigouvian environmental policy to deal with externalities. Suppose
again that because of trade liberalization, trade policies cannot be used. Then it is

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 239

possible to construct models, for example, extensions of the well-known ‘ rent-

shifting’ model of Brander and Spencer (1985) to include pollution (Conrad (1993),
Barrett (1994), Kennedy (1994), Ulph (1996a)) in which it will pay all governments
to set weaker environmental policies than Pigouvian as a proxy for an export
subsidy. While this might explain ‘ environmental dumping ’ by all governments, it is
important to keep in mind that this is a second-best result. As several authors,
including Walz and Wellisch (1997) and Bagwell and Staiger (2001), have stressed,
the incentive to undertake strategic environmental policy only arises if trade policies
are restricted.
Furthermore the conclusions of the model are by no means robust to changes in
assumption from those based on the Brander–Spencer model. Allowing for different
forms of market competition, for several firms in an industry to be located in the
same country, or strategic behaviour by producers (e.g. investment in R&D) can
imply that the second-best policies for governments acting independently would be
to set policies which are tougher than Pigouvian policies. We also note that all the
conclusions we have drawn for models of imperfect competition also apply when
plants are internationally mobile.
However even if ‘environmental dumping ’ was the result of imperfectly com-
petitive markets and free trade, the appropriate policy response is to coordinate
environmental policies so that all countries are better off than with environmental
dumping (this will sometimes imply policies tougher than Pigouvian). However
such coordination does not imply harmonization. Indeed, as noted by Ulph (1999),
harmonization may be worse, for some countries, than just allowing environmental
dumping to proceed. It is sometimes thought that a better approach than harmo-
nization is to set minimum environmental standards, so that some states can set
tougher standards if they wish, and so one can ‘ ratchet up ’ environmental standards
in all states. But Kanbur, Keen, and van Wijnbergern (1995), and Ulph (1999)
showed that minimum standards may not produce a ratchet effect : as some
countries raise their environmental standards others may relax them, and the
outcome, for some countries, may be worse than no policy at all.3

3 For all countries to benefit from minimum standards it is important that all countries toughen their
environmental policies. But whether this happens depends crucially on the policy instruments used. Suppose
countries use emission taxes. When one country raises its emission tax, and hence reduces domestic output,
that raises profitability in other countries and encourages them to expand output. If the governments leave
emission taxes unchanged, that leaves abatement unchanged and so all the extra output will just increase
pollution. That will cause marginal damage costs to rise while marginal abatement costs are unchanged. So it
will be optimal to also increase abatement, which requires that those countries respond by also raising their
emission taxes. So with emission taxes there is a ratchet effect. But suppose governments set emission limits.
If one government tightens its limits, that again reduces domestic output and increases profits in other
countries, encouraging them to expand output. But now if emission limits are kept unchanged, all the
associated extra pollution must be abated. So now marginal abatement costs rise, while marginal damage
costs are unchanged. So it will be optimal to allow some of the associated extra pollution to be emitted. But
that means other countries respond by relaxing their emission limits. So we do not get a ratchet effect. In
technical terms emission taxes are strategic complements, while emission limits are strategic substitutes.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 240 DANIEL STURM AND ALISTAIR ULPH

The analysis summarized so far has ignored transboundary pollution. There are
two implications for policy. First there is the well-known free-rider problem that
each state will only consider the impact of domestically generated pollution on itself,
and ignore the costs it imposes on others. Dealing with this requires International
Environmental Agreements, which goes beyond the scope of this survey. Second,
even if a state ignores the environmental damage it imposes on other states, it will
be concerned about the damage that other states impose on it, and specifically
about ‘ leakage effects ’ – if it toughens its environmental standards this may
simply cause production, and hence pollution, to expand elsewhere, and damage
its own environment. In the small country case, there is nothing a country can
do about leakage. But in the large country case or imperfectly competitive case,
then there are steps a country could take. Essentially transboundary pollution
introduces a ‘ pollution-shifting ’ incentive for a country to induce a bit more
domestic pollution and reduce foreign pollution. It can do this through trade policies
(to encourage exports or reduce imports) or by relaxing domestic environmental
policies.
The analysis of the previous sections shows that (i) there is no general conclusion
that liberalization of trade in goods and capital is good for welfare or the en-
vironment, although provided Pigouvian environmental policies are in place and
other distortions addressed by appropriate policies, then the usual arguments for
trade liberalization go through ; (ii) there is no robust conclusion that competition
between states will lead to a race-to-the-bottom in environmental policies. These are
ultimately empirical matters. We cannot give details here, but a broad summary of
the empirical literature is that there is little impact of environmental policies on
trade, that trade liberalization has not had a damaging effect on the environment
(and may even have improved it), and there is little evidence that states engage in
environmental dumping.4
Our review of conventional economic analysis and empirical evidence leads to the
conclusion that there are neither strong conceptual nor empirical arguments to
support many of the concerns or policy recommendations that we outlined in the
introduction. In particular there seems little basis for the concern that globalization
will damage the environment by inducing a ‘race-to-the-bottom ’ in environmental
standards, and even if such a concern was warranted, harmonization of environ-
mental standards is not the appropriate policy response. However this analysis has
been based on the assumptions of welfare-maximizing governments and perfect
information, for example about environmental damage costs in different countries.
In the rest of this paper we see how far these conclusions need to be modified
when we allow for asymmetric information and governments which are influenced
by special interest groups. We address these issues first in isolation and then
together.

4 For more detailed reviews of the empirical literature see Jeppesen, List, and Folmer (2000), Rauscher
(2001) and WTO (1999).

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 241

3. Information problems
To illustrate one of the information issues we are going to be concerned with,
suppose we take as given that there is trade liberalization, so we know that in both
the large country case and the imperfect competition case welfare-maximizing
governments acting non-cooperatively will seek to set environmental policies which
differ from Pigouvian in order to gain some trade advantage. So there is a case for
some supra-national agency to coordinate environmental policies to overcome the
inefficiencies caused by non-cooperation, due to both strategic trade considerations
and transboundary pollution. We have emphasized in the last section that com-
parative advantage means that it will be efficient for countries with different en-
vironmental endowments, broadly defined, to set different environmental stan-
dards. So simply observing different environmental standards tells us nothing about
whether countries are distorting their environmental policies. A supra-national
agency would need to know damage costs in different countries both to decide
whether there are any distortions and to compute appropriate efficient cooperative
environmental standards.
But standard subsidiarity arguments suggest that countries may have better in-
formation about their local damage costs than a supra-national agency might have.
If this is correct,5 does this informational asymmetry mean either that it would be
better to just leave environmental policy-making at the national level, or that, if
policy is set at a supra-national level the supra-national agency should just set
uniform standards, which has the advantage of simplicity and not having to justify
why it sets different standards in different countries? A negative response to these
questions was given in Ulph (2000). The context was a federal system where im-
perfect competition meant that states acting non-cooperatively would engage in
environmental dumping.
Ulph (2000) considered only strategic trade aspects, but Bigano (2002) shows this
analysis can be extended to include transboundary pollution. States know their own
damage costs, but not others, and the federal government knows only the distri-
bution of possible damage costs, which is the same for each state. There are three
possible ways environmental standards could be set. They could be set at the state
level, which has the advantage that this exploits the good information held by states,
but does not deal with either environmental dumping or transboundary pollution.
They could be set by the federal government based just on their best guess about
damage costs in each state – which in this case would be expected damage costs and
would be the same for each state, so standards would be harmonized. Or they could

5 Of course this might not be true. Economies of scale or scope in collecting information may mean that a
supra-national agency might be better placed to collect good damage cost data than individual national
agencies. But this would just reinforce the argument for having a supra-national agency set policy. For a
discussion of informational problems and the optimal level of decentralisation of environmental policies for
watershed management in developing countries see Coxhead (2002).

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 242 DANIEL STURM AND ALISTAIR ULPH

be set by the federal government in such a way that states have incentives to
truthfully reveal their information.
As Bigano (2002) has shown, the incentives for states to misrepresent their
damage costs depend on the degree of transboundary pollution. If there is very little
transboundary pollution, then the incentive will be for states with high damage costs
to report they have low damage costs in order to be allowed to set lax environmental
standards and hence get a bigger market share.6 But if there is significant trans-
boundary pollution, then the incentive will be for a state which suffers from
transboundary pollution and has low damage costs to claim it has high damage costs
to get the federal government to set tougher environmental standards on those states
which pollute it. To overcome these incentives to misreport their damage costs the
federal government sets appropriate environmental standards. In the case with low
transboundary pollution these emission standards will differ less between states
with different damage costs than they would with full information, but this does not
mean harmonization. It turns out that having the federal government set standards
in this sophisticated way is always the best policy, so the benefits from overcoming
environmental dumping outweigh the losses caused by asymmetric information.
Moreover, if the federal government uses harmonization, then, if damage costs
have more than a moderate variance, harmonization is worse than setting policy
at the national level. So information problems cannot justify harmonization as a
means of overcoming environmental dumping. We shall return to this problem in
section 5.

4. Political economy
So far we have assumed that governments seek to maximize the overall welfare of all
its citizens. In this section we consider the implications of assuming that govern-
ments are prey to capture by special interest groups. Why might we be interested in
this approach ? As Anderson and Blackhurst (1992) note, the nexus of trade and
environmental policies ‘ have an above average risk of being exploited by special
interest groups’ so this might be a more realistic description of how policies get set.
But the more interesting questions are how would using this approach change the
conclusions we have reached so far, and are there issues we can address with this
approach that we cannot answer assuming welfare-maximizing governments ?
There are five sets of questions we are going to be interested in.
First, we know from the work of Buchanan and Tullock (1975) that there may be
good reasons why various interest groups would lobby to have inefficient en-
vironmental policies in a closed economy. In brief, existing producers would prefer
to have a given level of emissions reduction implemented through quantity con-
straints in which existing producers get grandfathered entitlements to emit pol-
lution. The reason is that this effectively cartelises the industry and protects it from

6 This is the result originally derived by Ulph (2000).

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 243

new entrants. One interesting question is how these arguments carry over to an
open-economy setting.
Second, it would be important to know to what extent and under which cir-
cumstances the interests of environmental and industrial lobby groups converge. It
is certainly easy to find examples for situations in which the interests of environ-
mental and industrial lobby groups are opposed. However, as noted in the intro-
duction it is also frequently the case that both industrialists and environmentalists
support a policy of harmonization of environmental policies. To what extent can
political economy models illuminate the mechanisms behind these observations ?
Third, our analysis of welfare-maximizing governments suggested that, as trade is
liberalized, there are no robust predictions that countries will systematically weaken
environmental policies to compensate. Yet there are strongly held views that this is a
significant risk. Moreover recent empirical work by Barrett and Graddy (2000)
suggests there is a negative correlation between environmental standards and an
index of corruption. Would political economy models give a more robust prediction
that trade liberalization will result in environmental dumping ?
Fourth, there is a large gap between the analytical and empirical findings of
section 2 that environmental policy has little impact on competitiveness and the
public perception. WTO (1999) quotes the example of a Wall Street Journal poll in
1990 in which a third of respondents believed that their jobs were at risk from
environmental regulation, when data showed that between 1987 and 1990 only
about 0.1 % of US layoffs could be attributed to environmental regulations. There
are two ways political economy models might account for this. One is that the
potential impacts on competitiveness are indeed small, and that this gap in per-
ception just reflects the success of lobby groups in creating a public fear to influence
policy. A more subtle explanation is that impacts on competitiveness are potentially
larger than the data suggests, but that industrialists have been successful in ensuring
that where relatively tough environmental regulations are introduced these are
accompanied by subsidies or protection which mitigate most of the effects.
Finally, we have already argued that the introduction of a supra-national organ-
ization that coordinates national policies could be an obvious way to overcome the
inefficiencies that are created, if national policy making results in environmental
dumping. However, such a new institution would also be the subject of lobbying
efforts of both environmental and industrial groups. Environmentalists, for ex-
ample, are opposed to the idea that the WTO could play this role in the area of
environmental policy, fearing that the WTO is prone to be captured by multi-
national companies. An important challenge is therefore to determine how lobbying
will shape the policies of a supra-national agency and how these policy outcomes
compare to national policy making.
There is now a small literature which applies political economy models to the
study of trade and environment issues. Before we review this literature we briefly
describe the ways in which political considerations have been integrated into en-
vironment and trade models. The most frequently used approach in the literature

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 244 DANIEL STURM AND ALISTAIR ULPH

have been lobbying models, which portray the political process as a strategic in-
teraction between the government and various special interest groups, while elec-
tions do not play an explicit role. In a number of contributions the influence of lobby
groups is captured through the political support function approach, which assumes
that an incumbent has an objective function which is a weighted sum of welfare and
the contributions of different interest groups. A recent rigorous reformulation of this
approach was given by Grossman and Helpman (1994) in the context of trade
protection and has been used extensively in the trade and environment literature.
Alongside the lobbying approach there are also a number of political economy
approaches, which explicitly consider the role of elections. One such approach is the
familiar median voter model, which is really a way of aggregating diverse individual
preferences, but without explicit behaviour by special interest groups. Another
approach is the electoral competition model in which candidates for political parties
first select what platform to stand on, and then campaign contributions influence the
probabilities of different candidates being elected. Finally there is the political
agency model, which views the political process as a principal–agent relationship in
which the voters as principals have to provide incentives for their political agent
through elections. We now address the five questions set out above.

4.1 Efficiency of environmental policies

In an early, and relatively informal, analysis, Hoekman and Leidy (1992) and Leidy
and Hoekman (1994) argued that the normal incentives for producers to favour
inefficient forms of environmental regulation (quantity-based with grandfathering
of pollution permits for existing producers as opposed to an emission tax) as a
means of cartelising an industry, would be reinforced in an open-economy
setting. They point out that getting the industry to cooperate in this cartel manner
would strengthen the industry’s ability to press for trade protection to offset the
‘injury ’ caused by increased imports resulting from the tougher environmental
regulations.
Using more rigorous models of lobbying behaviour based on the Grossman and
Helpman (1994) model of the political support function, Aidt (1998) and Schleich
(1999) reach different conclusions. Schleich (1999) considers a model in which the
government can use either production or consumption taxes/subsidies to deal effic-
iently with environmental damages linked directly to either production or con-
sumption, or they can use trade policies, which are inefficient. Furthermore the
government is lobbied by a number of industry lobby groups. As noted above, the
Grossman and Helpman (1994) approach is that the government effectively maxi-
mizes a weighted sum of lobby contributions and social welfare. An important
implication of this assumption is that the government has an interest in using
instruments efficiently. This mechanism drives the finding of Schleich (1999), that
governments only use their efficient production or consumption taxes to internalize
the environmental damages, rather than inefficient trade taxes. The trade policy
instruments are only used to redistribute income between the different lobby groups.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 245

Aidt (1998) considers a very similar model in which the government faces the choice
between a tax on a polluting input and output taxes and subsidies. He also shows
that only the efficient tax on the polluting input will be used to correct the en-
vironmental damages in the political equilibrium.

4.2 Commonality of interests

In general, the interests of environmentalists and industrialists are likely to be
opposed, simply because reducing emissions reduces profitability, particularly of
industry-specific capital. However this need not always be the case. Hillman and
Ursprung (1992, 1994) consider commonality of interests in an electoral compe-
tition model in which the only policy instrument is trade policy and the choice is
between a free-trade candidate and a protectionist candidate (who would impose a
prohibitive tariff on imports). Governments can be lobbied by domestic and foreign
firms and by domestic environmentalists, and the question asked by Hillman and
Ursprung is how environmentalists will vote. If pollution is caused by production,
and environmentalists care only about domestic pollution, then environmentalists
vote for free trade, since that reduces domestic production and hence pollution. If
pollution is related to consumption, or environmentalists care about pollution in
both countries (Hillman and Ursprung refer to these groups as ‘ supergreens ’ ) or the
environmentalists can coordinate their lobbying, then they will vote for protection.
Rauscher (1997) using a political support function approach shows that if only one
policy instrument is available then the interests of environmentalists and in-
dustrialists generally diverge, but they will converge if policies (either trade or
environmental) are targeted at foreign polluters. He also allows for the possibility
that there may be some kinds of specific capital, say pollution abatement capital,
whose returns are increased by tougher environmental policies.

4.3 Trade liberalization and environmental dumping

We now turn to the question whether political economy models suggest that govern-
ments will respond to a trade liberalization by relaxing environmental policies
and how the predictions of political economy models differ from the policy choices
of a welfare-maximizing government. Explicit consideration of the impact of trade
liberalization is provided by Frederiksson (1999) also using a Grossman–Helpman
political support function approach. In his model there is a numeraire sector and a
single import-competing polluting industry, and two lobby groups – an environ-
mental group and a group representing an industry-specific factor whose return is
increasing in the level of emissions, so the interests of industrialists and en-
vironmentalists are opposed. There is an exogenously given import tariff, so, as we
saw in section 2, a welfare-maximizing government would impose environmental
policy tougher than the Pigouvian tax to correct for the excess domestic pollution
resulting from the tariff. His main finding is that in the political equilibrium the
environmental policy could be higher or lower than the welfare-maximizing policy

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 246 DANIEL STURM AND ALISTAIR ULPH

and that trade liberalization has an ambiguous effect on the environmental policy
that emerges in the political equilibrium. The intuition for this result is that the
reduction in domestic output in response to the trade liberalization reduces the
incentives for both groups to lobby. So, depending on parameters, environmental
policy could get weaker or tougher and environmental quality could get worse or
better in the wake of a trade liberalization.
Bommer and Schulze (1999) argue that Frederiksson’s results depend on his
assumption that lobbying is concentrated on the importing sector, while the ex-
porting sector remains unorganized. In their model there are two sectors, which
differ in their pollution intensity. Both sectors use a specific factor, which organizes
as a lobby group. Trade liberalization is assumed to increase the relative price of the
dirty good, which expands output of the dirty sector and, with constant emission
tax, increases pollution and also increases the return on the dirty specific factor and
reduces the return on the clean specific factor. They present some evidence that this
assumption is a good characterization of most OECD countries. We saw in section 2
that in addition to this composition effect of a trade liberalization there would be
technique and scale effects in which a welfare-maximizing government would
tighten environmental policy in response to both the increase in emissions and an
increase in income. Bommer and Schulze, using a political support function, but
without the Grossman–Helpman micro-foundations, argue that there will be a
further reason why a government may tighten environmental policy as trade is
liberalized – to offset the distributional effects of trade liberalization on the return to
the specific factor in the clean industry.
Both Frederiksson and Bommer and Schulze assume competitive industries. Johal
and Ulph (2001a, 2002) use a model of imperfect competition in which, with trade
liberalization, welfare-maximizing governments will engage in environmental
dumping. They introduce special interest groups in an electoral competition model
in which there are ‘ green’ and ‘ brown ’ parties who give, respectively, too high or
too low a weight to environmental damages relative to social welfare, and so, ceteris
paribus, would set too tough (lax) environmental policies relative to welfare-
maximizing governments. There are environmental and industrial special interest
groups who can give campaign contributions to parties of their own type at home or
abroad. Johal and Ulph show that if countries act non-cooperatively then, in a
political equilibrium, lobbying increases the probability of electing green govern-
ments. There are two reasons. First, although environmentalists are assumed to care
only about domestic pollution, they share with industrialists an interest in having
the foreign government set tougher environmental standards, and this gives in-
centives for environmentalists to lobby abroad as well as at home. Second, having
weak environmental policies in all countries reduces industry profits by expanding
industry output, and this reduces the incentive of industrialists to lobby for brown
governments.
None of these models suggest that introducing political factors leads to a strong
prediction that trade liberalization will lead to environmental dumping.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 247

4.4 Environmental policies and competitiveness

We have already seen that Hoekman and Leidy (1994) argued informally that, in an
open economy, industrialists were more likely to press for inefficient forms of
environmental regulation, since this would strengthen their ability to press for
compensating trade protection. They conjecture that this could be one reason why
there has been so little evidence of the impact of environmental policy on trade. They
provide some evidence for this conjecture by pointing out that in the US the in-
dustries that have the highest abatement costs as a proportion of total costs account
for a very high proportion of anti-dumping cases, and claim that the same is true of
the EU and Australia.
The idea that political economy forces might ensure that sectors which have to
bear high pollution abatement costs would be compensated through other policy
instruments is pursued further in Eliste and Frederiksson (2002). They use a
Grossman–Helpman type model to show that stricter environmental policy for a
sector can result in an endogenous increase in transfers to this sector. They test this
prediction on a cross-country dataset, which provides information on the stringency
of pollution control in the agricultural sector and transfers to farmers. They find that
stringent environmental policy is positively correlated with larger transfers to
farmers.
4.5 Policy coordination
In section 2 we saw that in the case of either large-country models or models of
imperfect competition, welfare-maximizing governments acting independently may
have incentives to set environmental policies which differ from Pigouvian to secure
some trade advantage, but, even without transboundary pollution, these policies
impose externalities on other countries. Indeed all countries can end up worse off by
these beggar-thy-neighbour policies. So there is a case for coordination of such
policies. This is reinforced if there is transboundary pollution. But how is this
argument affected if agencies in either states or some supra-national coordinating
agency are influenced by special interest groups?
We know that with transboundary pollution individual states acting non-
cooperatively will set too weak environmental policies, both because of free-riding
problems and, in the large-country case, to reduce leakage. It is sometimes argued
that the activities of green lobby groups may help to overcome this failure of
coordination, by inducing governments to set tougher environmental policies than if
they were welfare maximizing. Conconi (2000), using a Grossman–Helpman type
model of political lobbying within large countries, shows that this presumption is
only true if countries can use trade instruments to deal with leakage effects. Then
indeed green lobby groups press for higher emission taxes, and uncoordinated
environmental policies may give a better outcome than coordinated policies.
However, if countries are unable to use trade taxes to deal with leakage effects
(because of trade liberalization), then, if leakage effects are strong enough, green
lobby groups may press for lower emission taxes than when non-cooperative

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 248 DANIEL STURM AND ALISTAIR ULPH

governments are welfare maximizing, and so this strengthens the case for coordi-
nation of environmental policies.
Conconi assumes there are only green lobbyists. Similar results are found by
Schleich and Orden (2000), who extend the analysis of Schleich (1999) to consider
two large countries, but with lobby groups who also press for support for industries.
When governments act non-cooperatively, then the outcome is as in Schleich:
governments act efficiently, using environmental policies to deal with externalities
and trade policies to exploit terms of trade ; but they will give discounts to sectors
with organized lobby groups at the expense of sectors with no organized lobby
groups. Because there is transboundary pollution, countries impose two kinds of
externalities on each other : through transboundary pollution and through terms-of-
trade effects. If the governments set policies cooperatively, this will affect the
political equilibrium. Surprisingly they show that with cooperation there may be
more environmental damage than without. The fact that governments no longer try
to exploit terms-of-trade effects has ambiguous effects on environmental policies,
while the fact that they internalize the damage caused in other countries unambigu-
ously toughens policies. But these effects would apply with welfare-maximizing
governments. The additional effect that comes through political models is that be-
cause cooperation brings about efficiency gains this means governments can satisfy
lobby groups at a lower cost to other lobby groups. They increase output and hence
specific factor returns in organized sectors, and this effect on environmental damage
may outweigh other effects.
Johal and Ulph (2001a, 2002) also consider what happens when governments
coordinate their policies, to overcome environmental dumping and, in the first
paper, transboundary pollution. They model various asymmetries in lobbying
behaviour designed to capture some of the concerns of environmentalists; there are
differences in ability to influence electoral outcomes between lobby groups from
different countries (e.g. northern countries have more influence than southern
countries), between lobby groups of different types (e.g. industrialists have more
influence than environmentalists), or between different levels of government
(national and supra-national – to reflect ‘democratic deficit ’ at the supra-national
level). They show that despite these asymmetries, it is always better to have policies
coordinated at the supra-national level. So the benefits of dealing with policy ex-
ternalities outweigh the costs of political distortions.
None of these papers supports a general conclusion that the introduction of pol-
itical influence either makes international coordination of environmental policies
to deal with both environmental dumping and transboundary pollution unnecess-
ary or so prey to political distortion as to be less desirable than leaving policy making
uncoordinated.

4.6 Summary
To conclude this section we summarize what our review of this small literature tells
us about the five questions we posed at the start of this section. Firstly, in an open

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 249

economy the influence of special interest groups does not lead to governments using
inefficient (trade) instruments to deal with environmental problems when more
efficient instruments are available. Secondly, while political economy models can
provide explanations of when interests of environmentalists and industrialists co-
incide, there are no robust generalizations about when this occurs. Thirdly, allowing
for the influence of special interest groups does not provide any stronger or more
robust support for the concern that trade liberalization will lead to environmental
dumping. Fourthly, there is theoretical and empirical support for the view that
traded goods sectors which might be adversely affected by stringent environmental
policies are successful in getting other forms of support to offset the impact of the
environmental policies. Finally, there is no general support for the view that the
influence of special interest groups at either the national or supra-national level
makes the coordination of environmental policies either unnecessary or so distorted
as to be undesirable. Such general conclusions as we have been able to draw have
mostly been of a negative kind – denying the existence of strong general results. This
reflects the current state of the literature : there are relatively few papers, the models
are rather different and special so the conclusions they reach are model-dependent.

5. Information and political economy

We now put together the concerns of the last two sections. To see why this might be
necessary, we could ask the simple question – if members of the public know that
politicians are likely to be influenced by special interest groups, why do they not take
steps to prevent this ? One way of doing this is through the election process. Poli-
ticians who blatantly stood for policies that favoured narrow sectional interests
should not get elected. Politicians who are subsequently found to have swayed their
policies in the interests of certain groups may not get re-elected. If politicians value
a career in politics and if elections are relatively frequent, this could act as a
reasonable discipline. But of course this depends crucially on voters being able to tell
when politicians have acted to favour special interests. We argued in the last section
that even in political economy models, there are incentives for efficient choices of
policies. But that assumes perfect information. It is frequently argued informally
that one reason why governments may choose inefficient policies is because it is
easier to hide the fact that they are acting to favour certain interests. Crudely, it may
be easier to give certain industries trade protection on the grounds that it is offsetting
potentially serious competitive effects from environmental policies than to give
them protection outright.
Another way voters may try to limit the scope for special interest groups to gain
favours is to limit the discretion of politicians, for example by pressing for inde-
pendent agencies with tightly specified objectives which are close to welfare-
maximizing policies (for example independent central banks with narrow inflation
targets). But that again requires voters to be able to work out what welfare-max-
imizing policies would be and to know whether or not agencies have implemented

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 250 DANIEL STURM AND ALISTAIR ULPH

them. If politicians or their officials get access to information that is important in

designing welfare-maximizing policies, then restricting their scope for manoeuvre in
advance will mean that policies may not be based on the best available information.
Thus, it can be argued that information problems play an important role in
understanding how the political process operates and institutions might be designed
to deal with these problems (see Laffont (2000) for a general discussion).
Sturm (2001) gives an example of the first approach to limiting political influ-
ence – the threat of not being re-elected. He considers the case of trade disputes over
products whose consumption causes potential damage to health or the environment.
One example for such a dispute would be EU/US dispute over growth hormones in
beef. The paper develops a simple two-country model. Suppose that one of the goods
that the home country imports from the foreign country can cause some environ-
mental or health damages, if it is consumed. For simplicity it is assumed that the risk
of damages can either be high or low and that only units of the good which have been
produced in the foreign country can cause damages. Furthermore observing the ex
post damages does not reveal perfectly what the ex ante level of the risk was. A
crucial assumption is that only politicians in power know the true value of the risk,
and that this is common knowledge to politicians in both countries, so there is no
underlying scientific dispute between the countries.
The governments face a decision whether or not to impose a product standard on
the foreign product. Imposing the standard in the home country reduces consumer
surplus, raises domestic profits and eliminates the risk of damage ; in the foreign
country it has similar effects but lowers foreign profits. The parameters of the model
are chosen in such a way that if governments were welfare maximizers, they would
both agree to implement the standard if the risk was high, and not implement it if the
risk was low. This ensures that any dispute is purely political. More generally there
will be a range of risk values for which, in terms of expected welfare, the home
country would want to impose the standard but the foreign country would not, so
there could be genuine welfare differences.
In the political context, the median voter cares only about social welfare. Poli-
ticians can be either ‘ good ’ or ‘ bad ’. Good politicians will just pursue welfare-
maximizing policies – imposing the standard only if the risk is high. Bad politicians
want to favour producers of the good. The efficient way to do this would be by a
lump-sum transfer. But voters could detect that, and would not re-elect them, which
politicians care about, and so will not do this. The other, inefficient, way of
favouring producers is, in the home country to implement the product standard even
in the low-risk state (‘ green protectionism ’), and in the foreign country not to
implement the standard even in the high-risk state (‘ environmental dumping ’).
Voters know that bad politicians will want to act in the manner just described.
The problem is that voters do not get information on what the true risk is, which
would allow them to tell when politicians were acting badly. All they can observe is
whether or not politicians in the two countries have implemented the same policies.
But this is a very imperfect signal. If there is a trade dispute, all voters can infer is that

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 251

at least one of the politicians must be bad. They cannot tell whether it is a high-risk
state with a bad foreign politician not implementing the standard, while the home
politician, who could be good or bad, does implement the standard; or a low-risk
state with a bad politician at home implementing the standard while the foreign
politician, who could be good or bad, does not. Equally the absence of a trade
dispute does not mean both politicians are good – at least one must be good, but
voters cannot tell which. Voters will use their observations of whether or not a trade
dispute has taken place to decide whether or not to re-elect politicians. Sturm shows
that, given the fuzziness of the signal voters get, there can be a political equilibrium
in which the threat of not being re-elected is sufficiently low that bad politicians will
indeed choose to act in the way described above. So this paper gives a formal
demonstration of how environmental trade disputes can arise through the failure of
the political process, which allows politicians who want to favour certain groups to
use inefficient means of doing so because this is how they avoid detection and hence
punishment at the ballot box.
In a second step the paper goes on to discuss two possible mechanisms to over-
come or avoid trade disputes over product standards. One mechanism often sug-
gested is mutual recognition of standards. This means those goods, which meet the
standard of the exporting country must also be accepted by the importing country.
Implementing this rule in the model would eliminate ‘green protectionism ’ by bad
home politicians, but would not eliminate ‘ environmental dumping ’ by bad foreign
politicians and so may not increase welfare in the home country. Another possible
mechanism would be harmonization. Given the simplifying assumptions of the
model the same standard would be optimal in both countries and harmonization
would therefore not entail any efficiency losses. However, the potential problem
with harmonization is that there is no guarantee that the political process will only
implement the harmonized standard in the high-risk state. The paper shows that the
outcome of the political process could be both ‘ upward ’ harmonization (the stan-
dard is implemented in both countries even though risks are low) and ‘downward ’
harmonization (the standard is not implemented even though risks are high).
An example of the second approach to limiting the influence of special interest
groups – restricting the scope for policy discretion – is provided by Johal and Ulph
(2001b, c). These papers combine the analysis of Ulph (2000), summarized in
section 3, with the political economy models of Johal and Ulph (2001a, 2002)
referred to in section 4.5. In addition to the question of whether to have environ-
mental policy set at the national or supra-national level (setting policy at the supra-
national level overcomes environmental dumping but the supra-national agency has
less good information about national damage costs than national agencies), there is
the problem that governments or agencies at both the national and supra-national
levels can be captured by special interest groups. As in Sturm (2001), only national
governments in power get information that tells them their true damage costs. So
voters cannot tell whether lax environmental policies are a response to information
that damage costs are lower than expected or an attempt to buy influence from

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 252 DANIEL STURM AND ALISTAIR ULPH

industrial special interests (similarly with tough environmental policies and green
special interests).
Suppose now that voters try to limit the influence of special interest groups, say by
requiring the creation of independent environmental agencies with narrowly drawn
policy limits. Since the information available to voters at the outset is just expected
damage costs, and since Johal and Ulph assume this is the same for all states, limiting
politicians’ discretion will imply harmonization. Note that harmonization is playing
a very different role than in Ulph (2000) – it is a consequence of attempts to limit
political influence, not the consequence of informational asymmetries between
national and supra-national agencies. Note also that attempts to limit the influence
of special interests can be made whether policy is set at the national or supra-
national level.
Now there is a trade-off. Having an independent agency with narrow policy limits
restricts the influence of special interest groups, but it also restricts the ability of
the agency to adjust policy in the light of better information about damage costs.
Not surprisingly, it will pay to limit policy discretion if the potential gain in in-
formation is smaller than the potential distortion in policies by special interest
groups. But Johal and Ulph (2001b, c) show that it never pays to restrict political
influence if policy is set at the supra-national level if it does not pay to restrict it
when policy is set at the national level. So one cannot use the argument that because
policy is being moved to the supra-national level to overcome environmental
dumping as the rationale for harmonization to limit political influence if there
was no such attempt to limit political influence when policy was set at the national
level.
One important implication of the above analysis is that when one considers
problems of imperfect information in a political setting (and we argued that the two
are intimately linked), one finds not just that politicians may choose inefficient
policies because that is a means of exploiting poorer information held by voters, but
that overcoming the attempt by politicians to serve special interests may also involve
the use of policies, such as harmonization, which are criticized by economists as
being inefficient.
The above discussion has been entirely theoretical. Is there any empirical support
for these models of environmental policy setting with political economy and im-
perfect information ? List and Sturm (2002) use a model, which is similar to Sturm
(2001), in which politicians may have private preferences for the environment
(which may reflect their desire to benefit certain interest groups), which differ from
those of potential swing voters. Faced with the prospect of re-election, politicians
trim their policies to suit those of the swing voters. But if a politician faces a binding
term limit (and therefore is a ‘lame duck ’), she is free to set policies which reflect her
own preferences. This suggests that environmental policies should change when
politicians enter a lame duck phase. In the US term limit legislation constrains
governors in a substantial number of states to serving no more than two terms in
office. List and Sturm (2002) tests their model on US data on state environmental

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 253

policies and find substantial support for the view that environmental policy changes
during years in which a governor is a lame duck.
The papers in this section have provided formal analysis of the notion that
asymmetries of information between politicians and voters limit the extent to which
policy making can be shielded from the influence of special interest groups, but are
ambivalent on the question whether apparently inefficient policies like harmon-
ization might be supported as an indirect means of limiting political influence.
However, the literature is too small and the results too dependent on rather special
models to draw strong general conclusions.

6. Conclusions
In this paper we have outlined a number of concerns that have been expressed about
the nexus of policies linking trade and the environment and reviewed some of the
recent literature by economists to see what light it sheds on these concerns. In section
2 we reviewed the ‘ conventional ’ literature, mainly theoretical but also empirical,
and showed that it did not lend strong support either to the concerns or some of
the policy recommendations. In particular, while there is danger that if further trade
liberalization rounds succeed in preventing governments using conventional trade
instruments they may distort their environmental policies for trade purposes,
there are no robust arguments for saying that this will imply environmental policies
which are too lax – they may be too strong. The empirical evidence suggests that
trade liberalization has not had a major damaging effect on the environment, nor
have environmental policies had a marked impact on trade. One strong conclusion
from this literature was that harmonization was unlikely to be desirable.
We then reviewed a much smaller literature on what we believe to be key issues in
sorting out trade and environment issues – information problems and the possible
capture of politicians by special interest groups. The models are too disparate at this
stage to allow us to draw firm conclusions, but we would argue that they do not
provide any stronger basis for the concerns outlined in the introduction. In par-
ticular they give no more reason for believing that in a world of trade liberalization
there will be a systematic race to the bottom in environmental standards. There is at
best ambivalent and model-specific support for the view that harmonization might
be justified as a means of limiting the influence of special interest groups.
What does this imply for the discussions that will be taking place over the next few
years about whether or how environmental considerations should be built into the
next round of trade liberalization ? Bierman (2001) argues that what is called for is
an Authoritative Interpretation of Article XX which clarifies the role of trade
measures in support of multilateral environmental agreements, allows countries to
protect themselves against imports of goods whose consumption causes environ-
mental damage, provided this is done in a non-discriminatory fashion, and rules out
unilateral use of trade measures against countries who have different environmental
standards for production and process methods. Unless multilateral environmental

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 254 DANIEL STURM AND ALISTAIR ULPH

agreements are interpreted very widely, it is not clear that this addresses the issue of
countries distorting their environmental policies for trade or political reasons.
The fundamental challenge, as we have set out in the last half of this survey, is to
discriminate between differences in environmental standards that arise through
comparative advantage and differences that arise from attempts to exploit market
power or favour special interests. In facing this challenge it is important to recognize
that local governments may have better information about local environmental
damages than supra-national agencies, and that they might use this informational
advantage to favour special interest groups. Using expert scientific panels may
resolve some aspects of environmental disputes, though as Sturm (2001) notes
bodies of scientific experts may not be sufficiently isolated from political pressure,
and in any case the disputes may be much more about how different countries value
a given environmental impact. Similarly, looking for inconsistencies in the way
governments apply environmental standards in areas exposed to trade and areas not
exposed to trade will be a useful reality check. But governments can apply en-
vironmental policies consistently and in a non-discriminatory manner and still be
engaging in either environmental dumping or green protectionism. The challenge is
to design mechanisms that both get local politicians to reveal information truthfully
and limit their scope for exploiting it to favour special interest groups. In some cases
this may lead to adopting what normally look like inefficient policies, and we gave
the example of harmonization, but stress that the literature is at too early a stage for
this to be a robust conclusion. What the best intervention will look like needs a lot
more careful analysis than the research to date has provided.

References
Aidt, T. (1998), ‘Political Internalization of Economic Externalities and Environmental Policy’, Journal of
Public Economics, 69 (1): 1–16.
Anderson, K. and R. Blackhurst (eds.) (1992), The Greening of World Trade Issues, Ann Arbor: University
of Michigan Press.
Bagwell, K. and R. Staiger (2001), ‘Domestic Policies, National Sovereignty and International Economic
Institutions’, Quarterly Journal of Economics, 116: 519–562.
Barrett, S. (1994), ‘ Strategic Environmental Policy and International Trade’, Journal of Public Economics,
54 (3): 325–338.
Barrett, S. and K. Graddy (2000), ‘ Freedom, Growth and the Environment ’, Environment and Develop-
ment Economics, 5: 433–456.
Bierman, F. (2001), ‘ The Rising Tide of Green Unilateralism in World Trade Law – Options for Reconciling
the Emerging North-South Conflict’, Journal of World Trade Law, 53 (3): 421– 448.
Bigano, A. (2002), ‘ Environmental Dumping, Transboundary Pollution and Asymmetric Information’,
Chapter 4, Ph.D. thesis ‘ Environmental Regulation of the Electricity Sector in a European Market
Framework’, Department of Economics, Katholieke Universiteit Leuven, Belgium.
Bommer, R. and G. Schulze (1999), ‘Environmental Improvement with Trade Liberalization’, European
Journal of Political Economy, 15 (4): 639–661.
Brander, J. and B. Spencer (1985), ‘Export Subsidies and International Market Share Rivalry’, Journal of
International Economics, 18: 83–100.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 Environment and trade 255

Buchanan, J. and G. Tullock (1975), ‘Polluters’ Profits and Political Response: Direct Controls vs. Taxes’,
American Economic Review, 65, 139–147.
Conconi, P. (2000), ‘Can Green Lobbies Replace a World Environment Organization ?’, Nota di Lavoro
56.2000, FEEM, Milan.
Copeland, B. and T. Scott (1994), ‘North-South Trade and the Environment ’, Quarterly Journal of
Economics, 109: 755–787.
Conrad, K. (1993), ‘Taxes and Subsidies for Pollution-Intensive Industries as Trade Policy’, Journal of
Environmental Economics and Management, 25 (2): 121–135.
Coxhead, I. (2002), ‘It Takes a Village to Raise a Pigouvian Tax _ or Does It Take More? ’, Paper presented
at conference on Sustaining Food Security and Managing Natural Resources in Southeast Asia :
Challenges for the 21st Century, Chaingmai, Thailand.
Dean, J. (1992), ‘ Trade and the Environment: a Survey of the Literature’, in P. Low (ed.), International
Trade and the Environment, World Bank Discussion Paper 159.
Eliste, P. and P. Frederiksson (2002), ‘Environmental Regulations, Transfers and Trade: Theory and
Evidence’, Journal of Environmental Economics and Management, 43: 234–250.
Esty, D. (1994), ‘ Greening the GATT : Trade, Environment and the Future’, Institute of International
Economics, Washington, DC.
Frederiksson, P. (1999), ‘The Political Economy of Trade Liberalization and Environmental Policy’,
Southern Economic Journal, 65 (3): 513–525.
Grossman, G. and E. Helpman (1994), ‘ Protection for Sale’, American Economic Review, 84: 833–850.
Grossman, G. and A. Krueger (1993), ‘ Environmental Impacts of a North American Free Trade Agree-
ment’, in P. Garber (ed.), The Mexico US Free Trade Agreement, Cambridge, MA: MIT Press.
Hillman, A. and H. Ursprung (1992), ‘ The Influence of Environmental Concerns on the Political Deter-
mination of Trade Policy’, in K. Anderson and R. Blackhurst (eds.), The Greening of World Trade
Issues, Ann Arbor: University of Michigan Press.
Hillman, A. and H. Ursprung (1994), ‘ Greens, Supergreens and International Trade Policy: Environmental
Concerns and Protectionism’, in C. Carraro (ed.), Trade, Innovation, Environment, Dordrecht:
Kluwer, pp. 75–108.
Hoekman, B. and M. Leidy (1992), ‘Environmental Policy Formation in a Trading Economy: A Public
Choice Perspective’, in K. Anderson and R. Blackhurst (eds.), The Greening of World Trade Issues,
Ann Arbor: University of Michigan Press, pp. 221–246.
Jeppesen, T., J. List, and H. Folmer (2000), ‘ Environmental Regulations and New Plant Location Decisions:
Evidence from a Meta-analysis’, mimeo, University of Maryland.
Johal, S. and A. Ulph (2001a) ‘ Global Environmental Governance, Political Lobbying and Transboundary
Pollution’, in J. List and A. de Zeeuw (eds.), Recent Advances in Environmental Economics,
Cheltenham: Edward Elgar.
Johal, S. and A. Ulph (2001b), ‘International Environmental Policy Coordination, Harmonization, and
Limiting Political Discretion’, mimeo, University of Southampton.
Johal, S. and A. Ulph (2001c), ‘ Limiting Political Discretion and International Environmental Policy
Coordination with Active Lobbying ’, in M. Rauscher and C. Withagen (eds.), The International
Dimension of Environmental Policy, Dordrecht: Kluwer (forthcoming).
Johal, S. and A. Ulph (2002), ‘ Globalization, Political Lobbying and the Design of International En-
vironmental Governance ’, Review of International Economics, 10 (3): 387–403.
Kanbur, R., M. Keen, and S. van Wijnbergen (1995), ‘Industrial Competitiveness, Environmental Regu-
lation and Direct Foreign Investment’, in I. Goldin and A. Winters (eds.), The Economics of Sus-
tainable Development, Cambridge: Cambridge University Press, pp. 289–302.
Kennedy, P. (1994), ‘Equilibrium Pollution Taxes in Open Economies with Imperfect Competition’,
Journal of Environmental Economics and Management, 27 (1): 49–63.
Laffont, J-J. (2000), Incentives and Political Economy, Oxford: Oxford University Press.
Leidy, M. and B. Hoekman (1994), ‘‘‘Cleaning Up’’ While Cleaning Up? Pollution Abatement, Interest
Groups and Contingent Trade Policies’, Public Choice, 78 (3–4): 241–258.
List, J. and D. Sturm (2002), ‘Politics and Environmental Policy: Theory and Evidence from US States’,
mimeo, University of Munich and University of Maryland.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200

 256 DANIEL STURM AND ALISTAIR ULPH

Rauscher, M. (1997), International Trade, Factor Movements and the Environment, Oxford: Oxford
University Press.
Rauscher, M. (2001), ‘ International Trade, Foreign Investment and the Environment ’, Working Paper
No. 39, University of Rostock.
Schleich, J. (1999), ‘Environmental Quality with Endogenous Domestic and Trade Policies’, European
Journal of Political Economy, 15 (1): 53–71.
Schleich, J. and D. Orden (2000), ‘ Environmental Quality and Industry Protection with Noncooperative
versus Cooperative Domestic and Trade Policies’, Review of International Economics, 8 (4):
681–697.
Schulze, G. and H. Ursprung (eds.) (2001), International Environmental Economics: A Survey of the Issues,
Oxford: Oxford University Press.
Sturm, D. (2001), ‘ Product Standards, Trade Disputes and Protectionism’, CEP Discussion Paper No. 486.
Sturm, D. (2002), ‘ Trade and the Environment : A Survey of the Literature’, in M. Rauscher and C.
Withagen (eds.), The International Dimension of Environmental Policy, Dordrecht : Kluwer
(forthcoming).
Sturm, D. and A. Ulph (2002), ‘Environment, Trade, Political Economy and Imperfect Information – a
Survey’, Discussion Paper 0204, Department of Economics, University of Southampton.
Ulph, A. (1996a), ‘Environmental Policy Instruments and Imperfectly Competitive International Trade’,
Environmental and Resource Economics, 7 (4): 333–355.
Ulph, A. (1997a), ‘International Trade and the Environment ’, in H. Folmer and T. Tietenberg (eds.),
International Yearbook of Environmental and Resource Economics 1997/8, Cheltenham: Edward
Elgar, pp. 205–242.
Ulph, A. (1997b), ‘Environmental Policy and International Trade – A Survey of Recent Economic Analysis’,
in C. Carraro and D. Siniscalco (eds.), New Directions in the Economic Theory of the Environment,
Cambridge: Cambridge University Press, pp. 147–192.
Ulph, A. (1999), ‘ Ecological Dumping, Harmonization and Minimum Standards ’, in E. Petrakis, E.
Sartzetakis and A. Xepapadeas (eds.), Environmental Regulations and Market Structures, Chel-
tenham: Edward Elgar, pp. 233–250.
Ulph, A. (2000), ‘ Harmonization and Optimal Environmental Policy in a federal System with Asymmetric
Information’, Journal of Environmental Economics and Management, 39: 224–241.
Walz, U. and D. Wellisch (1997), ‘ Is Free Trade in the Interest of Exporting Countries When There Is
Ecological Dumping? ’, Journal of Public Economics, 66: 275–291.
Wilson, J. (1996), ‘Capital Mobility and Environmental Standards: Is There a Theoretical Basis for a Race
to the Bottom?’, in J. Bhagwati and R. Hudec (eds.), Fair Trade and Harmonization: Prerequisites
for Free Trade?, Cambridge, MA: MIT Press.
WTO (1999), ‘ Trade and Environment ’, Special Study No. 4, WTO, Geneva.

http://journals.cambridge.org Downloaded: 14 Mar 2015 IP address: 128.218.248.200