INTERNATIONAL TRADE: COMMERCIAL POLICY*

J. Peter Neary
University College Dublin and CEPR
July 19, 2001

Abstract
Following a brief historical introduction and a discussion of different types of
commercial policy, this paper reviews the arguments for and against trade protection. In the
bench-mark case of a competitive, small, open economy, free trade maximizes aggregate
national welfare, although some individual groups will lose unless compensation is actually
paid. Guidelines for policy include the uniform reduction and "concertina" rules for tariff
cuts, and the principle of targeting: corrective measures should be applied as close to the
source of the "distortion" as possible. Relaxing the bench-mark assumptions allows
exceptions to the case for free trade: "optimal" tariffs to manipulate world prices; "strategic"
tariffs or export subsidies when home firms engage in oligopolistic competition with foreign
rivals; and infant industry protection to allow home firms benefit from learning by doing.
Protection can also raise the growth rate, though it is less likely to raise welfare in a growing
economy. Overall, with due allowance for some ambiguity, both theoretical arguments and
empirical evidence suggest a pragmatic case for free trade. Finally, the paper notes the
political pressures for and against protection, and the role of international institutions such as
the GATT in underpinning moves towards freer trade.

Address for Correspondence: Department of Economics, University College Dublin,

Belfield, Dublin 4, Ireland; tel.: (+353) 1-716 8344; fax: (+353) 1-283 0068; e-mail: peter.
[email protected]; http://www.ucd.ie/~economic/staff/pneary/neary.htm.

* Prepared for the International Encyclopedia of the Social and Behavioral Sciences, Elsevier
Science, Amsterdam (forthcoming 2001).

1
20851A3/4/045 International Trade: Commercial Policy

Commercial policy' describes any form of government intervention towards international

trade. The study of commercial policy is a branch of international trade theory, itself a subfield of microeconomics. None of this sounds likely to arouse passions, but in practice trade
policy has often prompted bitterly divisive political debates and has been a central concern
of domestic and foreign policy.

In the early modern period, "Mercantilist" writers rationalized the use of restrictive trade
policy by expansionist monarchs to foster exports and ensure trade surpluses. David Ricardo
in 1817 provided the intellectual case against this approach. His theory of trade patterns,
based on specialisation according to comparative advantage (see International Trade: Models
of Trade), predicted that no free-trading country would lose relative to autarky. This laid the
foundations of the modern enthusiasm for free trade on the part of most mainstream academic
economists. However, Ricardos arguments were probably less important than the increasing
ascendancy of commercial over landed interests in leading to the repeal of the U.K. Corn
Laws in 1848, which ushered in an era of (mostly) falling barriers and expanding trade.
Twentieth-century world wars and depression reversed this trend and were themselves
influenced by trade policy: commercial and imperial rivalries contributed to the onset of the
First World War, and the 1930 Smoot-Hawley tariff worsened the depression in the U.S. and
hastened the collapse of world trade. Post-war attempts to restore the multilateral trading
system led in 1947 to the General Agreement on Tariffs and Trade (GATT), reconstituted in
1995 as the World Trade Organisation (WTO). Under their auspices, successive rounds of
trade negotiations have yielded progressive reductions in trade barriers. However, the chaotic

2
events inside and outside the abortive Seattle meeting in 1999, which it had been hoped
would launch another trade round, showed that progress towards further liberalization will
face opposition from critics of "globalization".

As this brief background makes clear, a full account of the theory and practice of trade policy
would require an extensive discussion of general intellectual currents, of economic history,
and of contemporary international relations. This article has the more limited objective of
summarising what economic theory has to say on the topic. It begins with some necessary
taxonomy and then reviews the principal theoretical arguments for and against trade
restrictions.

1. Varieties of Commercial Policy

The most obvious form of commercial policy, and historically often the most important, is
a tariff, a tax on imports which raises their domestic price above the world price, and so
"protects" domestic producers, at the expense of home consumers. Confusingly, an export
subsidy has a similar effect, raising the price of an export good to domestic producers and
consumers above its world price. The two measures have opposite effects on the relative
price of imports to exports, which is the basis of the Lerner symmetry theorem: a uniform
tariff on all imports has exactly the same effects on relative prices as a uniform tax on all
exports. Both raise the relative price of imports at home and thus discourage trade. A
corollary is that trade can be liberalized either by reducing tariffs or by leaving them in place
and subsidising exports: politically a more expedient route and one followed successfully by

3
some of the newly industrializing countries of East Asia.

Tariffs have declined in importance since the Second World War relative to non-tariff
barriers, such as import quotas, "voluntary export restraints" (i.e., quotas imposed by
exporting countries) and government procurement rules. Such policies are qualitatively
similar to tariffs in their protective effects, though the conditions for exact equivalence rarely
hold. Finally, most domestic policies (taxes, subsidies, health and safety regulations, etc.),
even if not explicitly discriminatory, have external repercussions. Though not strictly forms
of commercial policy, their effects are increasingly recognized in trade negotiations.

Constructing a true measure of trade policy is an index-number problem: how to aggregate

all these different types of trade restriction into a single measure which is comparable across
countries and across time. Solutions in principle to it have been devised, but implementing
them in practice is extremely difficult. In applied work, levels of protection are usually
measured by trade-weighted average tariffs and, even less satisfactorily, by "coverage ratios",
the percentage of traded commodities which are subject to non-tariff barriers.

This article considers only the case where commercial policy applies indiscriminately to
imports from whatever source. The desirability of this is enshrined in Article I of the GATT,
which requires that all trading partners be treated as favourably as the "most-favoured nation".
However, Article XXIV allows exceptions for regional trade agreements, which have grown
in importance in recent years with the widening and deepening of the European Union and
the signing of the North American Free Trade Agreement.

For further details, see

International Trade: Economic Integration. The GATT also tolerates tariffs imposed on

4
exporters found guilty of "dumping" - selling below cost or below the price charged in their
home market.

Such "anti-dumping" tariffs are an important form of protection in the

contemporary world economy: even the threat of imposing them can deter foreign exporters.

2. Trade Policy in a Competitive Small Open Economy

Devising criteria for trade policy which will hold universally is a daunting task, and it makes
sense to begin with a simple bench-mark case. The Classical starting point is an economy
which is competitive - individual consumers and firms cannot affect domestic prices - and
small - the economy as a whole cannot affect world prices. Free trade must then maximize
real national income, since it removes the constraint requiring an exact match between
domestic production and consumption patterns. Specialisation in production increases the
value of aggregate output at world prices, while consumers benefit by being able to buy from
the cheapest supplier worldwide.

However, individuals are both consumers and income recipients, and aggregate gains can
mask big shifts in internal income distribution. The near-certainty that there will be some
losers is implied by the Stolper-Samuelson theorem. This was originally formulated for a
special model, where it predicts that protection will raise real wages (so trade liberalisation
will lower them) if imports use labor relatively intensively. More generally, the logic of the
theorem implies that there are almost always some factor-owners who will lose from a
reduction in trade barriers. Most obviously, this will be true of factors which are specific
(even if only in the short run) to import-competing sectors.

5
Losers notwithstanding, the existence of national gains from trade ensures that "aggregate
welfare" must rise, meaning that it would be possible to tax some of the winners gains,
compensate the losers, and still leave no one worse off. Free trade is thus the archetype of
a situation which is potentially Pareto-efficient or simply "efficient" (confusingly, the term
has a more precise sense than in common parlance). The same result holds even if the
government has limited taxing and spending powers, and can only redistribute income through
changes in commodity (or "indirect") taxes. Of course, all this is poor consolation for the
losers if the compensation is not actually carried out. Nonetheless, trade theorists tend to
emphasise the efficiency gains; and prefer to try and devise programs of adjustment assistance
to help those adversely affected rather than to recommend foregoing the national gains. In
this they are motivated by professional division of labour (losses to particular groups mandate
changes in the tax and social welfare system, not protection), and a belief that the poor rarely
gain from highly restricted trade, rather than heartlessness.

For the same reasons, the

remainder of this survey concentrates on the effects of trade liberalization on aggregate

welfare, and will not repeat these essential qualifications about its distributional consequences.

Even though the case for free trade is clear, the best way to move towards it may not be
(except in the trivial case where there is only a single tariff). Abolishing all tariffs at once
is unlikely to be politically feasible. Two rules of piecemeal trade liberalization are then
available. The first is the uniform reduction rule: reduce all tariffs by an equiproportionate
amount. Heuristically, this kind of reform leaves relative tariff rates unchanged, so it is "as
if" there is only a single tariff rate, which is steadily reduced. Hence it is not surprising that
(pathological cases apart) it guarantees a welfare improvement. The second is the concertina
rule: reduce the highest tariff rate. A sufficient condition for this to raise welfare is that the

6
good in question is a substitute for all other goods subject to tariffs. Substitutability is not
necessary, however. For example, if all goods subject to tariffs are complements for each
other, then a reduction in any tariff (not just the highest) raises imports of all tariffconstrained goods and a welfare gain is again assured. Finally, the concertina rule does not
justify increasing the lowest tariff, unless all exports are subsidised at higher rates: only
raising the lowest distortion guarantees a welfare gain.

This discussion illustrates the distinction between "first-best" and "second-best" welfare
economics. Policy recommendations are more complicated in the latter case, when some preexisting distortions cannot be abolished. Nevertheless, a general principle applies: activities
which from a welfare perspective are under-supplied in the absence of intervention should be
encouraged, and vice versa. A related rule of thumb with many useful applications in
practical policy-making is the principle of targeting: intervention should be applied as closely
as possible to the desired target, whether this is to offset an irremovable distortion or to attain
a "non-economic" objective (such as restricting imports of certain types of goods, or
protecting industries deemed essential to cultural independence or national security). From
this perspective, trade policy is rarely a first-best instrument. For example, if there is a
minimum wage in the import-competing sector, protection may raise welfare because it
partially offsets the minimum wage. But other forms of intervention, such as employment
or production subsidies, would have the same effect at lower welfare cost.

3. Trade Policy in a Large Open Economy

7
Relaxing the assumption that the economy is "small" admits a specifically economic argument
for protection which was clearly stated by Bickerdike in 1906. Reducing home demand for
imports now lowers their world price, improving the home countrys terms of trade (i.e.,
reducing the price of imports relative to exports) and yielding a welfare gain. The optimal
tariff is the tariff which just balances this gain from manipulating the world price against the
loss from trading at a different price from the rest of the world. A corollary is that, if a
country is a major supplier of a good, and if local producers are competitive, an export tax
can improve national welfare. In effect, the home government acts as a monopolist: a role
which uncoordinated private producers cannot adopt by themselves. It should be stressed that
these policies are only optimal from a national point of view (which explains why some
authors prefer the term "exploitative" to "optimal"). World welfare definitely falls, though
once again transfers, this time international, would be needed to compensate losers from
universal free trade.

In practice, very few if any individual countries have a major influence on world prices.
Cases where a group of countries acting together would have such power are more common,
especially producers of primary commodities with few close substitutes. Hence the tendency
for cartels and commodity price agreements in such markets. However, such groupings are
typically unstable, since each member has an incentive to "free-ride" on its partners output
restrictions. The Organization of Petroleum Exporting Countries (OPEC) is the outstanding
counter-example and is probably best explained by a dominant firm model, where one large
producer with very low marginal production costs (Saudi Arabia) in effect sets the price, and
a "competitive fringe" of other oil-producing countries (whether OPEC members or not)
adjusts its outputs accordingly.

4. Trade Policy with Economies of Scale and Imperfectly Competitive Markets

The arguments reviewed so far assume that firms are perfectly competitive and produce in
equilibrium with constant returns to scale. A great deal of research in recent years has
relaxed these assumptions and explored the implications for trade policy.

A useful starting point is the case of Marshallian "external economies": individual firms lack
market power but expansion of the industry as a whole lowers costs for all. This isolates the
implications of increasing returns, while retaining the assumptions of perfect competition. In
such markets, industries which enjoy significant economies of scale are likely to concentrate
in large countries under free trade. However, this need not justify protection by small
countries, since this would condemn them to high-cost local production and to foregoing the
gains from participating in the international division of labour. It is only medium-size
countries which are likely to lose from free trade by specialising in the "wrong" commodities.
Even for them the principle of targeting continues to apply: production subsidies rather than
tariffs are the optimal policy.

Relaxing the assumption that firms are perfectly competitive complicates matters considerably,
since there are many varieties of imperfect competition. It is convenient to distinguish
between two: monopolistic competition and oligopoly. To begin with the former, it resembles
perfect competition in two important respects: individual firms are too small to influence their
rivals; and they enter or leave the industry in response to profit opportunities, so that in
equilibrium profits are zero. The key distinguishing features are that firms enjoy economies

9
of scale and that each produces a distinct variety. This in turn reflects consumers tastes,
which exhibit a preference for diversity. In this kind of economy, restricting trade has an
additional harmful effect: it reduces the range of choice available to consumers. Policy choice
is complicated because the free market may lead to more or less varieties being produced than
the social optimum. However, the principle of targeting continues to apply: provided antitrust measures are used if necessary to ensure an optimal number of domestic firms, free trade
remains desirable.

A different set of issues arises in the case of oligopoly, where there are barriers to entry and
a relatively small number of firms. Since firms perceive that they are interdependent, they
behave "strategically", taking into account the reactions of their rivals. In such markets, there
may be scope for governments to intervene in favour of home firms, an idea which has come
to be known as the theory of strategic trade policy.

The key insight is that, if a home and a foreign firm make their decisions simultaneously, the
home firm cannot credibly commit to a level of output which would maximise home welfare
conditional only on the behaviour of the foreign firm. By contrast, the home government is
assumed to be able to credibly commit to policies before both firms take their decisions.
Hence there is scope for the home government to make the commitment on behalf of the
home firm. In the simplest example, where a single home firm competes against a single
foreign firm in a third market, the implications are dramatic. The optimal policy for the home
government is to provide a positive export subsidy. This allows the home firm to credibly
commit to more aggressive behaviour, raising its output, market share and profits at the
expense of its foreign rival. This "profit shifting" result seems to provide a rationale for the

10
support of "national champions". Similar results apply to import-competing firms: tariffs may
serve to raise welfare by shifting profits from foreign to home firms.

However, the profit-shifting argument turns out to be subject to many qualifications. The
underlying model assumes that firms compete in the market-place by choosing their outputs,
taking the output choices of their rivals as given (i.e., that firms engage in Cournot
competition). This sort of behaviour is plausible when technology requires firms to commit
in advance to their capacity output levels. If instead output can be varied with little change
in marginal cost, then firms are more plausibly modeled as price-setters (engaging in Bertrand
competition). In that case, firms behave more aggressively in the absence of intervention, and
the optimal policy is an export tax rather than a subsidy. The rationale for intervention is the
same in the two cases: the home government uses its superior commitment power to achieve
an outcome which the domestic firm cannot achieve on its own. However, the practical
relevance of the theory is reduced by the sensitivity of the actual policy prescription to
assumptions about how firms behave. One slight defense of intervention comes from recent
research, which suggests that the ambiguity is reduced when subsidies are given to preproduction variables such as R&D or marketing expenditures rather than directly to exports.

Other criticisms of strategic trade policy are that with many home firms there is a
countervailing incentive to tax them (just as in the competitive large open economy case of
Section 3); that the gains from intervention are more than offset if foreign governments also
subsidise their own firms (see Section 6); and that general-equilibrium interactions with the
rest of the economy are ignored. For example, if applied to a number of sectors, all of whom
draw on a limited supply of some factor of production (such as skilled labour), an export

11
subsidisation policy merely raises the wages of that factor with little or no effect on the
pattern of output or the level of national welfare. For these and other reasons, attempts to
quantify the likely gains from strategic trade policy suggest that they are very small at best.
However, the gains could be much larger if subsidies made it possible for a home firm to
compete in the first place, especially if the alternative was domination of the world market
by a foreign monopoly. (The huge subsidies to Airbus by European governments are often
justified on these grounds.)

A final issue which arises when markets are imperfectly competitive is that trade policy can
itself affect the degree of competition. With price in excess of marginal cost, firms are
producing below their optimal scale. Hence any policy, including protection, which raises
domestic output may increase welfare. However, as noted in Sect. 2, protection is not the
best form of intervention. Exposing home firms to foreign competition is likely to be more
effective, reducing prices to consumers and allowing any surviving home firms to produce at
a more efficient scale.

5. Trade Policy and Growth

To assess the effects of trade policy in a growing economy, all the issues discussed in
previous sections remain relevant, and some new ones arise. Until recent years, the standard
approach to modelling economic growth was the neoclassical model of exogenous growth due
to Solow. In that framework, the long-run or steady-state rate of growth is determined by
exogenous rates of population growth and technological progress. Hence, trade policy cannot

12
affect the steady-state growth rate, though it may affect the rate at which the steady state is
approached.

More recently, attempts have been made to provide endogenous explanations for economic
growth. These stress the importance of resources devoted to research and development
(R&D), both in encouraging technological innovations and in facilitating the introduction of
new and higher-quality products. They also emphasise the importance of externalities, as the
benefits of R&D typically cannot be fully appropriated. This has immediate implications for
trade policy. For example, if a sector of the economy is disproportionately engaged in R&D,
protecting that sector will raise the long-run growth rate. These arguments are related to an
older, "infant-industry" argument, which defends transitional protection to enable a new firm
to benefit from learning-by-doing and scale economies.

For such firms, a tariff, by

guaranteeing higher home sales, may allow a firm to compete in export markets.

Once again, the principle of targeting must be mentioned. At best, these arguments justify
production or R&D subsidies: since the industries involved are likely to be oligopolistic, they
provide a case for strategic industrial rather than trade policy. Many of the qualifications
noted in Sect. 4 continue to apply. Success is more likely to come from general policies
which foster a culture of innovation and enterprise, rather than specific interventions which
seek to "pick winners".

A final caution is that, even when income distribution and

externalities are ignored, GNP per head is not the same as welfare in a growing economy.
Policies which raise the growth rate in the short or long term may do so at the expense of
current consumption.

13
Given the theoretical ambiguities surrounding the effects of trade policy on growth, it is
hardly surprising that empirical studies have failed to find a conclusive link between trade and
growth. Some authors have shown that various measures of "openness" can explain the
relative growth performance of different countries, but such measures are not directly related
to trade policy. The case against modest restrictions on trade is not proven. Nevertheless
there are strong theoretical and empirical reasons for believing that countries, especially those
with small home markets, which close themselves off from international movements of goods,
factors and ideas, are likely to have lower levels of welfare and growth.

6. Institutions, Politics and Trade Policy

The discussion so far has confined attention to the case of trade policy which is set
unilaterally by a single country. It has also adopted an exclusively "welfarist" perspective in
evaluating trade policy: the practical relevance of this is questionable unless decisions on
trade policy are devolved to public-spirited bureaucrats. A variety of approaches have been
taken to relaxing these assumptions. While much remains to be done, recent research has
thrown light on the political pressures for and against protection, and on the role of
international institutions such as the GATT in underpinning moves towards freer trade.

Many of the potential gains from unilateral intervention identified earlier must be qualified
when it is recognised that foreign governments face similar incentives.

Though often

described as "retaliation", this phenomenon is usually modelled by viewing governments as

players in a simultaneous-move game, often assuming only two countries for simplicity. The

14
effects on welfare in the cases of optimal tariffs (from Sect. 3) and optimal export subsidies
to Cournot firms (from Sect. 4) are similar: at least one country must lose relative to free
trade and both may lose. The latter outcome (which must ensue if the countries are relatively
similar) illustrates a "prisoners dilemma": unless the countries explicitly cooperate on free
trade, each has an incentive to adopt an interventionist policy and the result is lower welfare
for both. The case of optimal export taxes on Bertrand firms is different: now, both countries
may gain relative to free trade (though only at the expense of consumers). Nevertheless, it
exhibits a feature that holds in all cases: cooperation between governments leads to higher
welfare than non-cooperative choice of trade policy.

With or without international cooperation, all the theories of trade policy discussed so far are
normative, assuming that national welfare is the primary policy objective. Hence they fail
to explain why governments so frequently restrict trade by more than the welfare-maximising
extent. In response, the burgeoning field of political economy has proposed a variety of
positive explanations for the pattern and prevalence of trade policy. It may be determined
by spending on "rent-seeking" by factors of production for protection of the sectors which use
them intensively (or, in the limiting case of immobile factors, exclusively). It may be
determined by direct voting on tariff rates, in which case the rate chosen will reflect the factor
ownership of the median voter. Finally, it may be the outcome of campaign contributions to
politicians by lobby groups, whether to increase the probability of their preferred party being
elected, or to influence the policies of an incumbent government.

All these theories rely on special assumptions about the nature of the political system (often
assuming that politicians seek to maximise some weighted average of welfare and political

15
support) and the manner in which individual preferences influence political decisions. They
also assume that trade policy is used for redistributive purposes, and for the most part they
allow no role for other forms of public policy. They are therefore vulnerable to the criticism
that they fail to explain satisfactorily why trade policy is the preferred instrument, when more
efficient methods of redistributing income are available.

Finally, just as the widespread use of trade policy is a puzzle, so also is the steady trend
towards greater liberalisation in the period since the Second World War. It can be explained
by shifts in prevailing ideology, combined with the superior economic and political
performance of those countries which were first to move towards open markets.

But

economic explanations have also been proposed. To the extent that trade is intra- rather than
inter-industry (which appears to be the case for much trade between developed countries),
trade liberalisation imposes lower costs of adjustment and so has less distributional impact.
The GATT can be rationalised as a mechanism for implementing the international transfers
needed to compensate countries for foregoing nationally optimal tariffs which would lower
world welfare. And free-trade agreements of all kinds may serve as a commitment device:
a government has an incentive to join them because they provide a way to credibly distance
itself from the domestic pressure groups which lobby for protection.

7. Conclusions

In the bench-mark case of a competitive, small, open economy, free trade must raise
aggregate national welfare, although some individual groups will lose unless compensation

16
is actually paid. Relaxing the bench-mark assumptions allows exceptions to the case for free
trade: "optimal" tariffs to manipulate world prices; "strategic" tariffs or export subsidies when
home firms engage in oligopolistic competition with foreign rivals; and infant industry
protection to allow home firms benefit from learning by doing. Protection can also raise the
growth rate, though it is less likely to raise welfare in a growing economy. All these possible
arguments for protection are subject to many qualifications. Moreover, on closer examination,
most economic arguments for protection turn out instead to be arguments against laisser faire,
and so must be qualified by the principle of targeting: corrective measures should be applied
as close to the source of the "distortion" as possible, suggesting that other forms of
intervention (such as R&D or production subsidies) are preferable to trade protection in most
cases. Overall, with due allowance for some ambiguity, both theoretical arguments and
empirical evidence suggest a pragmatic case for free trade.

Suggestions for Further Reading

Bhagwati (1988) and Irwin (1996) provide contemporary and historical background.
Bhagwati (1971), Corden (1974) and Dixit (1985) give overviews of the theory of trade and
welfare, using mainly prose, diagrams and algebra respectively. More recent updates are
given in the contributions to Grossman and Rogoff (1995): see especially the chapters by
Brander on strategic trade policy, Feenstra on estimating the effects of trade policy, Grossman
and Helpman on technology and trade, Rodrik on political economy and Staiger on rules and
institutions for international trade. For more details and further references on particular
topics, see Anderson (1992) on dumping and anti-dumping; Anderson and Neary (1996) on

17
index numbers of trade restrictiveness; Bagwell and Staiger (1999) on the GATT; Dixit and
Norman (1980), especially Sects. 3.2 (on redistribution through commodity taxation), 9.1 (on
trade and competition), and 9.3 (on product differentiation and intra-industry trade); Ethier
(1982) on trade policy under increasing returns; Grossman and Helpman (1991) on trade
policy and growth; Krugman (1984) on import protection as export promotion; Neary (1995)
on tariffs and quotas; Neary and Leahy (2000) on strategic trade and industrial policy; and
Rodriguez and Rodrik (1999) on empirical studies of trade policy and growth. Among the
many important topics not covered are effective protection (see Ethier (1977)) and the
interaction of trade and environmental policy (see Neary (2001)).

Bibliography

Anderson J E 1992 Domino dumping I: Competitive exporters. American Economic

Review. 82: 65-83

Anderson J E, Neary J P 1996 A new approach to evaluating trade policy. Review of

Economic Studies. 63: 107-25

Bagwell K, Staiger R 1999 An economic theory of GATT. American Economic Review.

89: 215-48

Bhagwati J N 1971 The generalized theory of distortions and welfare. In Bhagwati J N,

Jones R W, Mundell R A, Vanek J (eds.) Trade, Balance of Payments and Growth:
Essays in Honor of C.P. Kindleberger. North-Holland, Amsterdam

18
Bhagwati J N 1988 Protectionism. MIT Press, Cambridge Mass.

Corden W M 1974 Trade Policy and Economic Welfare. Oxford University Press, Oxford

Dixit A K 1985 Tax policy in open economies. In Auerbach A J and Feldstein M (eds.)
Handbook of Public Economics, Volume 1. North-Holland, Amsterdam

Dixit A K, Norman V 1980 Theory of International Trade: A Dual, General Equilibrium

Approach. Cambridge University Press, London

Ethier W J 1977 The theory of effective protection in general equilibrium: Effective-rate

analogues of nominal rates. Canadian Journal of Economics. 10: 233-45

Ethier W J 1982 Decreasing costs in international trade and Frank Grahams argument for
protection. Econometrica. 50: 1243-68

Grossman G, Helpman E 1991 Innovation and Growth in the World Economy. MIT Press,
Cambridge Mass.

Grossman G, Rogoff K (eds.) 1995 Handbook of International Economics, Vol. III. NorthHolland, Amsterdam

Krugman P 1984 Import protection as export promotion: International competition in the

presence of oligopoly and economies of scale. In Kierzkowski H (ed.) Monopolistic

19
Competition and International Trade. Oxford University Press, Oxford 180-93

Irwin D A 1996 Against the Tide: An Intellectual History of Free Trade. Princeton
University Press, Princeton

Neary J P 1995 Trade liberalisation and shadow prices in the presence of tariffs and quotas.
International Economic Review. 36: 531-54

Neary J P 2001 International trade and the environment: Theoretical and policy linkages.
In Cararro C, Siniscalco D (eds.) Advances in Environmental Economics: Theory and
Policy. Cambridge University Press, Cambridge, forthcoming

Neary J P, Leahy D 2000 Strategic trade and industrial policy towards dynamic oligopolies.
Economic Journal. 110: 484-508

Rodriguez F, Rodrik D 1999 Trade policy and economic growth: A skeptics guide to the
cross-national evidence. Discussion Paper No. 2143, CEPR, London

J. Peter Neary
University College Dublin and CEPR