American Economic Association
Is Free Trade Passé?
Author(s): Paul R. Krugman
Source: The Journal of Economic Perspectives, Vol. 1, No. 2 (Autumn, 1987), pp. 131-144
Published by: American Economic Association
Stable URL: http://www.jstor.org/stable/1942985
Accessed: 31-12-2017 12:04 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
American Economic Association is collaborating with JSTOR to digitize, preserve and extend
access to The Journal of Economic Perspectives
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Economic Perspectives- Volume 1, Number 2-Fall 1987-Pages 131-144
Is Free Trade Passe?
Paul R. Krugman
I f there were an Economist's Creed, it would surely contain the affirmations "I
understand the Principle of Comparative Advantage" and "I advocate Free
Trade." For one hundred seventy years, the appreciation that international trade
benefits a country whether it is "fair" or not has been one of the touchstones of
professionalism in economics. Comparative advantage is not just an idea both simple
and profound; it is an idea that conflicts directly with both stubborn popular
prejudices and powerful interests. This combination makes the defense of free trade as
close to a sacred tenet as any idea in economics.
Yet the case for free trade is currently more in doubt than at any time since the
1817 publication of Ricardo's Principles of Political Economy. This is not because of the
political pressures for protection, which have triumphed in the past without shaking
the intellectual foundations of comparative advantage theory. Rather, it is because of
the changes that have recently taken place in the theory of international trade itself.
While new developments in international trade theory may not yet be familiar to the
profession at large, they have been substantial and radical. In the last ten years the
traditional constant returns, perfect competition models of international trade have
been supplemented and to some extent supplanted by a new breed of models that
emphasizes increasing returns and imperfect competition. These new models call into
doubt the extent to which actual trade can be explained by comparative advantage;
* Paul R. Krugman is Professor of Economics, Massachusetts Institute of Technology, Cambridge,
Massachusetts.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
132 Economic Perspectives
they also open the possibility that government intervention in trade via import
restrictions, export subsidies, and so on may under some circumstances be in the
national interest after all.
To preview this paper's conclusion: free trade is not passe, but it is an idea that
has irretrievably lost its innocence. Its status has shifted from optimum to reasonable
rule of thumb. There is still a case for free trade as a good policy, and as a useful
target in the practical world of politics, but it can never again be asserted as the policy
that economic theory tells us is always right.
Rethinking International Trade Theory
From the early nineteenth century until the late 1970s, international trade theory
was dominated almost entirely by the concept of comparative advantage, which we
can define loosely as the view that countries trade to take advantage of their
differences. In formal models, economies were assumed to be characterized by
constant returns to scale and perfect competition. Given these assumptions, trade can
arise only to the extent that countries differ in tastes, technology, or factor endowments.
The traditional Ricardian model emphasizes technological differences as the
cause of trade; the Heckscher-Ohlin-Samuelson model emphasizes differences in
factor endowments. Additional models can be generated by varying assumptions
about the number of goods and factors, by placing restrictions on the technology, and
so on. These alternative models have different implications in important respects; for
example, income distribution effects are absent in the Ricardian model, extremely
strong in the Heckscher-Ohlin-Samuelson model. Nonetheless, the underlying commonality
among conventional trade models is such that until a few years ago
international trade theory was one of the most unified fields in economics.
Thoughtful international economists have long known that comparative advantage
need not be the whole story, that increasing returns can be an independent
cause of international specialization and trade.' Ohlin himself repeatedly emphasized
this point. Furthermore, at least since the late 1950s empirical workers and informal
observers have been dissatisfied with formal trade theory, so that there has been a sort
of "counter-culture" in international trade research, a set of informal arguments
stressing sources of trade other than those represented in the formal models. Authors
such as Steffan Burenstam-Linder (1961) and Raymond Vernon (1966) emphasized
endogenous technological change, while many authors have discussed the possible role
of economies of scale as a cause of trade separate from comparative advantage. A few
papers attempted formal models of trade under increasing returns. However, all such
efforts were plagued by the problem of modeling market structure. Except under the
There have been many surveys of the new developments in international trade theory. For a synthetic
presentation of much of the positive side of this work, see Elhanan Helpman and Paul Krugman (1985); for
an informal presentation of arguments for and against new forms of trade intervention, see the volume
edited by Paul Krugman (1986); for a survey that also covers related topics in the border area between
trade and industrial organization, see Paul Krugman (forthcoming).
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Krugman 133
implausible hypothesis that economies of scale are completely external to firms,
increasing returns must lead to imperfect competition. Yet until the late 1970s, there
was no generally accepted way to model imperfect competition in general equilibrium.
Since mainstream trade theory derived its power and unity from being stated in
formal general equilibrium terms, alternative views were relegated to the footnotes. As
recently as 1980, many textbooks-and even survey articles on the theory of international
trade-failed even to mention the possibility that trade might arise for reasons
other than exogenous differences in tastes, technology, and factor endowments.
During the 1970s researchers in industrial organization began to develop models
of imperfect competition that, while admittedly lacking generality, were easy to use
and apply. In particular, Chamberlinian large-group competition was given a grounding
in utility maximization and placed in a general equilibrium framework by such
authors as A. Michael Spence (1976) and Avinash Dixit and Joseph Stiglitz (1977). It
quickly became clear to trade theorists that these new models supplied the necessary
framework for formal modeling of the role of increasing returns as a cause of
international trade. Simultaneously and independently, Avinash Dixit and Victor
Norman (1980), Kelvin Lancaster (1980), and this author (1979) published papers in
which economies of scale led to arbitrary specialization by nations on products within
monopolistically competitive industries. These models immediately established the
idea that countries specialize and trade, not only because of underlying differences,
but also because increasing returns are an independent force leading to geographical
concentration of production of each good. Indeed, at a logical level, increasing returns
are as fundamental a cause of international trade as comparative advantage.2
The role of increasing returns in trade was not, as already noted, a new idea,
although the new models gave it more clarity and precision than in the past. The
main new insight from these models was that to the extent that trade driven by
economies of scale is important in the world economy, imperfect competition is
important as well. International trade theory thus becomes inextricably intertwined
with industrial organization. In retrospect this conclusion is obvious. After all, most
trade is in the products of industries that economists classify without hesitation as
oligopolies when viewing them in their domestic aspect. For international economics,
however, this was a radical reorientation.
Although the new models of trade challenged the traditional view that all trade
represents exploitation of comparative advantage, the new trade theory did not at first
challenge the proposition that trade is of mutual benefit to the trading nations.
Indeed, if anything, the introduction of increasing returns and imperfect competition
into trade theory strengthens the case that there are gains from trade. In addition to
2One need not, of course, abandon comparative advantage entirely. From early in the development of new
trade theory most models have represented trade as arising from both increasing returns and some form of
comparative advantage, such as differences in factor endowments. This behavior of such models depends on
the underlying parameters. For example, if scale economies are large and countries are similar in their
factor endowments, the model will behave very differently from when scale economies are weak and
countries differ greatly; in the latter case traditional trade theory yields the right predictions, in the former
it does not.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
134 Economic Perspectives
benefitting from complementary differences in resources and technology, trading
countries can specialize in the production of different goods, achieving increased scale
of production while maintaining or increasing the diversity of goods available.
Admittedly, a second-best world of imperfect competition offers no guarantee that
potential benefits from trade will necessarily be realized. In most formal models,
however, it turns out that the presence of increasing returns increases rather than
reduces the gains from international trade. Furthermore, by creating larger, more
competitive markets, trade may reduce the distortions that would have been associated
with imperfect competition in a closed economy. Thus the initial implication of new
trade theory seemed, if anything, to reinforce the traditional view that trade is a good
thing, and thus to strengthen the case for free trade.
However, showing that free trade is better than no trade is not the same thing as
showing that free trade is better than sophisticated government intervention. The view
that free trade is the best of all possible policies is part of the general case for
laissez-faire in a market economy, and rests on the proposition that markets are
efficient. If increasing returns and imperfect competition are necessary parts of the
explanation of international trade, however, we are living in a second-best world
where government intervention can in principle improve on market outcomes. Thus as
soon as the respectability of non-comparative-advantage models in international
trade was established, international trade theorists began to ask whether the new view
of the causes of trade implied new views about appropriate trade policy. Does
acknowledging economies of scale and imperfect competition create new arguments
against free trade?
New Arguments Against Free Trade
The new view of international trade holds that trade is to an important degree
driven by economies of scale rather than comparative advantage, and that international
markets are typically imperfectly competitive. This new view has suggested two
arguments against free trade, one of which is a wholly new idea, the other of which is
an old idea given new force. The new idea is the strategic trade policy argument, which
holds that government policy can tilt the terms of oligopolistic competition to shift
'excess returns from foreign to domestic firms. The old idea is that government policy
should favor industries that yield externalities, especially generation of knowledge that
firms cannot fully appropriate.
Strategic Trade Policy
The strategic trade policy argument begins with the observation that in a world
of increasing returns and imperfect competition, lucky firms in some industries may be
able to earn returns higher than the opportunity costs of the resources they employ.
For example, suppose that economies of scale are sufficiently large in some industry
that there is only room for one profitable entrant in the world market as a whole; that
is, if two firms were to enter they would both incur losses. Then whichever firm
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Is Free Trade Passe? 135
manages to establish itself in the industry will earn super-normal returns that will not
be competed away.
A country can raise its national income at other countries' expense if it can
somehow ensure that the lucky firm that gets to earn excess returns is domestic rather
than foreign. In two influential papers, James Brander and Barbara Spencer
(1983, 1985) showed that government policies such as export subsidies and import
restrictions can, under the right circumstances, deter foreign firms from competing for
lucrative markets. Government policy here serves much the same role that "strategic"
moves such as investment in excess capacity or research and development (R & D)
serve in many models of oligopolistic competition-hence the term "strategic trade
policy."
The original Brander-Spencer analysis and the literature that followed it uses the
machinery of duopoly analysis: firms choose levels of R & D and/or output conditional
on other firns' choices, and an equilibrium occurs where the reaction functions
of firms intersect. The essence of the strategic trade policy concept, however, is so
simple that it can be conveyed with a numerical example. Indeed, focussing on such
an example may convey the essentials more clearly than a more formal treatment.
Suppose, then, that two countries are capable of producing a good. For concreteness,
let the good be a 150-seat passenger aircraft, and call the "countries" America
and Europe. Also, let there be one firm in each country that could produce the good:
Boeing and Airbus, respectively.
To focus attention on the competition for excess returns, assume that neither
America nor Europe has any domestic demand for the good, so that the good is
intended solely for export; this allows us to identify producer surplus with the national
interest. Also, assume that each firm faces only a binary choice, to produce or not to
produce. Finally, assume that the market is profitable for either firm if it enters alone,
unprofitable for both if both enter.
Given these assumptions, the game between Boeing and Airbus may be represented
by a matrix like that shown in Table 1. Boeing's choices to produce (P) or not
to produce (N) are represented by upper case letters, Airbus's corresponding choices
by lower case letters. In each cell of the matrix, the lower left number represents
Boeing's profit (over and above the normal return on capital), the upper right number
represents Airbus's profit.
As the game is set up here, it does not have a unique outcome. To give it one, let
us assume that Boeing has some kind of head start that allows it to commit itself to
produce before Airbus's decision. Then in the absence of government intervention, the
outcome will be Pn, in the upper right cell: Boeing will earn large profits, while
deterring entry by Airbus.
Clearly Europe's government would like to change this outcome. The strategic
trade policy point is that it can change the outcome if it is able to commit itself to
subsidize Airbus, at a point before Boeing is committed to produce. Suppose that
Europe's government can commit itself in advance to pay a subsidy of 10 to Airbus if
it produces the plane, regardless of what Boeing does. Then the payoff matrix is
shifted to that represented by Table 2. The result is to reverse the game's outcome.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
136 Economic Perspectives
Table 1
Hypothetical payoff matrix
Airbus
p n
-5 0
P -5 100
Boeing 100 0
N 0 0
Table 2
Hypothetical payoff matrix after European subsidy
Airbus
p n
5 0
P -5 100
Boeing
110 0
N 0 0
Boeing now kno
well, and it will make losses. Thus Boeing will be induced not to produce, and the
outcome will be Np instead of Pn. The surprising result will be that a subsidy of only
10 raises Airbus's profits from 0 to 110! Of this, 100 represents a transfer of excess
returns from America to Europe, a gain in Europe's national income at America's
expense.
The strategic trade policy argument thus shows that at least under some
circumstances a government, by supporting its firms in international competition, can
raise national welfare at another country's expense. The example just presented
showed this goal being achieved via a subsidy, but other policies might also serve this
purpose. In particular, when there is a significant domestic market for a good,
protection of this market raises the profits of the domestic firm and lowers the profits
of the foreign firm in the case where both enter; like an export subsidy, this can deter
foreign entry and allow the domestic firm to capture the excess returns. As businessmen
have always said, and as economists have usually denied, a protected domestic
market can-under some circumstances!-promote rather than discourage exports,
and possibly raise national income.
The strategic trade policy argument is immensely attractive to noneconomists,
since it seems to say that views always condemned by international trade theorists as
fallacious make sense after all. In defense of free trade, a number of analysts have
quickly acted to point out the weakness of strategic trade as a basis for actual
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Krugman 137
intervention. Before considering these arguments, however, I turn to the other
justification for government intervention in trade suggested by the new theory.
External Economies
There is nothing new about the idea that it may be desirable to deviate from free
trade to encourage activities that yield positive external economies. The proposition
that protection can be beneficial when an industry generates external economies is
part of the conventional theory of trade policy.3 However, the rethinking of international
trade theory has given at least the appearance of greater concreteness to the
theoretical case for government intervention to promote external benefits.
It is possible to imagine bees-and-flowers examples in which externalities arise
from some physical spillover between firms, but empirically the most plausible source
of positive externalities is the inability of innovative firms to appropriate fully the
knowledge they create. The presence of problems of appropriability is unmistakable in
industries experiencing rapid technological progress, where firms routinely take each
others' products apart to see how they work and how they were made. In traditional
international trade models with their reliance on perfect competition, however,
externalities resulting from incomplete appropriability could not be explicitly recognized,4
because the knowledge investment by firms that is the source of the spillover
could not be fitted in. Investment in knowledge inevitably has a fixed-cost aspect;
once a firm has improved its product or technique, the unit cost of that improvement
falls as more is produced. The result of these dynamic economies of scale must be a
breakdown of perfect competition. As a result, perfectly competitive models could not
explicitly recognize the most plausible reason for the existence of external economies.
This did not prevent trade theorists from analyzing the trade policy implications of
externalities, and in fact this is a well-understood topic. Since investment in knowledge
was not explicitly in their models, however, external economies seemed abstract,
without an obvious real-world counterpart. In traditional trade models, one industry
seems as likely as another to generate important external economies-so that the
theory seems remote from operational usefulness.
Once increasing returns and imperfect competition are seen as the norm, this
problem of abstractness is reduced. The dynamic scale economies associated with
investment in knowledge are just another reason for the imperfection of competition
that has already been accepted as the norm. External economies can now be identified
with incomplete appropriability of the results of R & D, which immediately suggests
that they are most likely to be found in industries where R & D is an especially large
part of firms' costs. So by making tractable the modeling of a specific mechanism
3See, for example, W. M. Corden (1974). As the conventional literature points out, however, protection is
only a second-best policy; direct correction of the domestic market failure is preferable.
4The one exception is the case of zero appropriability. In this case technological progress will occur only
through learning-by-doing, because there will be no incentive for firms to invest deliberately in knowledge
creation; since the fixed costs associated with knowledge investment are absent, perfect competition may be
preserved.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
138 Economic Perspectives
generating externalities, the new trade theory also seems to offer guidance on where
these externalities are likely to be important.
The emphasis on external economies suggested by new trade theory is similar to
the strategic trade policy argument in offering a reason for government targeting of
particular sectors. However, the external economies argument differs in one important
respect; policies to promote sectors yielding external economies need not affect other
countries adversely. Whether the effect of one country's targeting of high-externality
sectors on other countries is positive or negative depends on whether the scope of the
externalities is national or international. There is a conflict of interest if knowledge
spills over within a country but not between countries. Suppose that the research of
each computer firm generates knowledge that benefits other computer firms. This is
only a case for sponsoring production of computers in the United States as opposed to
Japan if U.S. firms cannot benefit from Japanese research.5 In many cases it seems
unlikely that spillovers respect national boundaries; a firm can "reverse engineer" a
product made abroad as well as one made at home. The best candidates for nationally
limited externalities are where knowledge spreads largely by personal contact and
word of mouth. This is a much more restricted set of activities than R & D in general,
although it is presumably the force behind such spectacular agglomerations of
high-technology industry as Silicon Valley and Route 128.
Despite the restriction that only externalities at the national level make industrial
policy a source of international conflict of interest, it is clear that the changes in trade
theory have strengthened the view that nations are competing over who gets to realize
these externalities. This reinforces the new strategic trade policy argument in offering
a more respectable rationale for deviating from free trade than has been available
until now.
Critique of the New Interventionism
The positive economics of the new trade theory, with its conclusion that much
trade reflects increasing returns and that many international markets are imperfectly
competitive, has met with remarkably quick acceptance in the profession. The
normative conclusion that this justifies a greater degree of government intervention in
trade, however, has met with sharp criticism and opposition-not least from some of
5Even if externalities are national in scope, one might argue that no conflict need be involved; simply let
each country provide the optimal subsidy. There are three answers to this. First, in practice countries often
pursue industrial policy objectives with second-best trade policy tools. It is interesting to ask why they do
this, but as long as they do the attempt to promote sectors is an attempt to promote them at other countries'
expense. Second, much of the practical argument over industrial policy in the U.S. concerns the urgency of
action when we are reluctant to intervene: if foreign countries' policies crowd out externality-generating
domestic sectors, then the costs of not subsidizing become larger. Finally, if economies of scale (internal or
external) are large enough, conflict of interest becomes unavoidable. Suppose there is room for only one
Silicon Valley in the world, yet the agglomeration will yield valuable external economies to the country that
gets it. Then the conflict cannot be avoided except through side payments.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Is Free Trade Passi? 139
the creators of the new theory themselves. The critique of the new interventionism
partly reflects judgements about the politics of trade policy, to which we turn below.
There are also, however, three economic criticisms. First, critics suggest that it is
impossible to formulate useful interventionist policies given the empirical difficulties
involved in modeling imperfect markets. Second, they argue that any gains from
intervention will be dissipated by entry of rent-seeking firms. Third, it is argued that
general equilibrium considerations radically increase the empirical difficulty of formulating
interventionist trade policies and make it even more unlikely that these
policies will do more good than harm.
Empirical Difficulties
The previous numerical example assumed that the European government knew
the payoff matrix and knew how Boeing would respond to its policy. In reality, of
course, even the best informed of governments will not know this much. Uncertainty is
a feature of all economic policy, of course, but it is even greater when the key issue is
how a policy will affect oligopolistic competition. The simple fact is that economists do
not have reliable models of how oligopolists behave. Yet the effects of trade policy in
imperfectly competitive industries can depend crucially on whether firms behave
cooperatively or noncooperatively, or whether they compete by setting prices or
outputs.6 Furthermore, in many oligopolistic industries firms play a multistage game
whose rules and objectives are complex and obscure even to the players themselves.
The externality argument for intervention runs up against the empirical problem
of measuring external economies. By their nature, spillovers of knowledge are elusive
and difficult to calculate; because they represent non-market linkages between firms,
they do not leave a "paper trail" by which their spread can be traced. A combination
of careful case study work and econometrics on the history of an industry may be able
to identify significant external economies, but what we need for trade policy is an
estimate of the future rather than the past. Will a dollar of R & D in the semiconductor
industry convey ten cents worth of external benefits, or ten dollars? Nobody really
knows.7
By itself, the argument that making policy based on the new trade theory is an
uncertain enterprise would only dictate caution and hard study, not inaction. When it
is linked with the political economy concerns described below, however, it raises the
question of whether the political risks associated with action outweigh any likely gains.
6In a widely cited paper, Jonathan Eaton and Gene Grossman (1986) showed that in a duopoly model
where the optimal strategic policy was an export subsidy with quantity competition, it was an export tax
with price competition.
70f course, one response is to try to find out. The central direction of current research in the new trade
analysis is the effort to produce quantitative models of competition in imperfectly competitive industries.
These efforts are currently primitive, and even the authors are skeptical about the robustness of the results,
but it is encouraging to see an effort to make the theory operational. See in particular Avinash Dixit
(forthcoming), Richard Baldwin and Paul Krugman (forthcoming), Anthony Venables and M. Alasdair
Smith (1986), Dani Rodrik (1987), and Richard Harris and David Cox (1984).
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
140 Economic Perspectives
Entry
Suppose that a government is somehow able to overcome the empirical difficulties
in formulating an interventionist trade policy. It may still not be able to raise national
income if the benefits of its intervention are dissipated by entry of additional firms.
Consider first the case of a strategic trade policy aimed at securing excess returns.
Our example was one in which there was room for only one profitable firm. Suppose,
however, that the market can actually support four or five firms, a sufficient number
so that the integer constraint does not matter too much and free entry will virtually
eliminate monopoly profits. In this case, as Ignatius Horstmann and James Markusen
(1986) have emphasized, a subsidy, even if it succeeds in deterring foreign competition,
will be passed on to foreign consumers rather than securing excess returns for
domestic producers. Or as Avinash Dixit has put it, when there is a possibility of new
entry we need to ask, "Where's the rent?"
A similar issue arises with policies aimed at promoting external economies.
Suppose that external economies are associated with the manufacture of semiconductor
chips, seemingly justifying a subsidy to chips production. If additional resources of
labor and capital are supplied elastically to the industry, the external benefits of larger
production will not be confined to the promoting country. Instead, they will be passed
on to consumers around the world in the form of cheaper chips. The national
advantage can come only to the extent that some factors are supplied inelastically to
the industry-Santa Clara valley real estate?-or external benefits conveyed by the
semiconductor industry to other industries. The point is that entry of new factors and
new firms further reduces, though it does not eliminate, the extent to which competition
for external economies represents a valid source of international conflict.
General Equilibrium
Even in a world characterized by increasing returns and imperfect competition,
budget constraints still hold. A country cannot protect everything and subsidize
everything. Thus interventionist policies to promote particular sectors, whether for
strategic or externality reasons, must draw resources away from other sectors. This
substantially raises the knowledge that a government must have to formulate interventions
that do more harm than good.
Consider first the case of strategic trade policy. When a particular sector receives
a subsidy, this gives firms in that sector a strategic advantage against foreign
competitors. However, the resulting expansion of that sector will bid up the price of
domestic resources to other sectors, putting home firms in these other sectors at a
strategic disadvantage. Excess returns gained in the favored sector will thus be offset to
at least some extent by returns lost elsewhere. If the government supports the wrong
sector, the gain there will conceal a loss in overall national income.
The implication of this general equilibrium point is that to pursue a strategic
trade policy successfully, a government must not only understand the effects of its
policy on the targeted industry, which is difficult enough, but must also understand all
the industries in the economy well enough that it can judge that an advantage gained
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Krugman 141
here is worth advantage lost elsewhere. Therefore, the information burden is increased
even further.
A similar point applies to externalities. Promoting one sector believed to yield
valuable spillovers means drawing resources out of other sectors. Suppose that
glamorous high-technology sectors yield less external benefit than the government
thinks, and boring sectors more. Then a policy aimed at encouraging external
economies may actually prove counterproductive. Again, the government needs to
understand not only the targeted sector but the rest of the economy to know if a policy
is justified.
The general equilibrium point should perhaps not be emphasized too much.
Sectors of the economy differ radically and visibly in both the extent to which they are
imperfectly competitive and in the resources they devote to the generation of knowledge.
There may not be a one-to-one correspondence between small numbers of
competitors and excess returns, or between high R & D expenditure and technological
spillovers, but there is surely a correlation. Governments may not know for sure where
intervention is justified, but they are not completely without information. However,
the general equilibrium critique reinforces the caution suggested by the other cri-
tiques.
To say that it is difficult to formulate the correct interventionist policy is not a
defense of free trade, however. Thus the economic critique of the new interventionism
is only part of the post-new-trade-theory case for free trade. The other indispensable
part rests on considerations of political economy.
The Political Economy Case for Free Trade
Like most microeconomic interventions, the interventionist policies suggested by
new trade theory would affect the distribution of income as well as its level. The
well-justified concern of economists is that when policies affect income distribution, the
politics of policy formation come to be dominated by distribution rather than
efficiency. In the case of trade interventions, this concern is at two levels. First, to the
extent that the policies work, they will have a beggar-thy-neighbor component that
can lead to retaliation and mutually harmful trade war. Second, at the domestic level
an effort to pursue efficiency through intervention could be captured by special
interests and turned into an inefficient redistributionist program.
Retaliation and Trade War
Strategic trade policy aimed at securing excess returns for domestic firms and
support for industries that are believed to yield national benefits are both beggar-thyneighbor
policies that raise income at the expense of other countries. A country that
attempts to use such policies will probably provoke retaliation. In many (though not
all) cases, a trade war between two interventionist governments will leave both
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
142 Economic Perspectives
countries worse off than if a hands-off approach were adopted by both. For example,
consider the case of the European telecommunications equipment industry. This
industry is a likely candidate for targeting on both oligopoly and external economy
grounds. It is also a sector where nationalistic procurement by government-owned
firms allows countries to pursue protectionist policies without violating agreements on
international trade. The result of such protectionist policies, however, is by most
accounts harmful to all concerned. Each country tries to be largely self-sufficient in
equipment, and no country is able to realize the scale economies that would come
from supplying the European market as a whole. Arguably, the structure of the game
between countries in telecommunications equipment, and probably in other sectors as
well, is that of a prisoners' dilemma where each country is better off intervening than
being the only country not to intervene, but everyone would be better off if nobody
intervened.
The way to avoid the trap of such a prisoners' dilemma is to establish rules of the
game for policy that keep mutually harmful actions to a minimum. If such rules are
to work, however, they must be simple enough to be clearly defined. Free trade is such
a simple rule; it is easy enough to determine whether a country imposes tariffs or
import quotas. New trade theory suggests that this is unlikely to be the best of all
conceivable rules. It is very difficult to come up with any simple set of rules of the
game that would be better, however. If the gains from sophisticated interventionism
are small, which is the import of the economic critique of the last section, then there is
a reasonable case for continuing to use free trade as a focal point for international
agreement to prevent trade war.
Domestic Politics
Governments do not necessarily act in the national interest, especially when
making detailed microeconomic interventions. Instead, they are influenced by interest
group pressures. The kinds of interventions that new trade theory suggests can raise
national income will typically raise the welfare of small, fortunate groups by large
amounts, while imposing costs on larger, more diffuse groups. The result, as with any
microeconomic policy, can easily be that excessive or misguided intervention takes
place because the beneficiaries have more knowledge and influence than the losers.
Nobody who has followed U.S. trade policy in sugar or lumber can be very sanguine
about the ability of the government to be objective in applying a policy based on the
Brander-Spencer model.
How do we resolve the problem of interest group influence on decision-making in
the real world? As in the case of the problem of international conflict, one answer is to
establish rules of the game that are not too inefficient and are simple enough to be
enforceable. To ask the Commerce Department to ignore special-interest politics while
formulating detailed policy for many industries is not realistic; to establish a blanket
policy of free trade, with exceptions granted only under extreme pressure, may not be
the optimal policy according to the theory but may be the best policy that the country
is likely to get.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
Is Free Trade Passe? 143
The Status of Free Trade
The economic cautions about the difficulty of formulating useful interventions
and the political economy concerns that interventionism may go astray combine into a
new case for free trade. This is not the old argument that free trade is optimal because
markets are efficient. Instead, it is a sadder but wiser argument for free trade as a rule
of thumb in a world whose politics are as imperfect as its markets.
The economic cautions are crucial to this argument. If the potential gains from
interventionist trade policies were large, it would be hard to argue against making
some effort to realize these gains. The thrust of the critique offered above, however, is
that the gains from intervention are limited by uncertainty about appropriate policies,
by entry that dissipates the gains, and by the general equilibrium effects that insure
that promoting one sector diverts resources from others. The combination of these
factors limits the potential benefits of sophisticated interventionism.
Once the expected gains from intervention have been whittled down sufficiently,
political economy can be invoked as a reason to forego intervention altogether. Free
trade can serve as a focal point on which countries can agree to avoid trade wars. It
can also serve as a simple principle with which to resist pressures of special-interest
politics. To abandon the free trade principle in pursuit of the gains from sophisticated
intervention could therefore open the door to adverse political consequences that
would outweigh the potential gains.
It is possible, then, both to believe that comparative advantage is an incomplete
model of trade and to believe that free trade is nevertheless the right policy. In fact,
this is the position taken by most of the new trade theorists themselves. So free trade is
not passe-but it is not what it once was.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms
144 Economic Perspectives
References
Baldwin, Richard, and Paul Krugman, "Market
Access and Competition: A Simulation Study
of 16K Random Access Memories." In Feenstra,
Robert, ed., Empirical Research in International Trade.
Cambridge: MIT Press.
Brander, James A., and Barbara J. Spencer,
"Export Subsidies and International Market Share
Rivalry," Journal of International Economics, 1985,
18, 83-100.
Brander, James A., and Barbara J. Spencer,
"International R & D Rivalry and Industrial
Strategy", Review of Economic Studies, 1983, 50,
707-722.
Burenstam-Linder, Steffan, An Essay on Trade
and Transformation. New York: Praeger, 1961.
Corden, W. Max, Trade Policy and Economic
Welfare. Oxford: Clarendon, 1974.
Dixit, Avinash K., "Optimal Trade and Industrial
Policy for the U.S. Automobile Industry."
In Feenstra, Robert, ed., Empirical Research in International
Trade. Cambridge: MIT Press, forth-
coming.
Dixit, Avinash K., and Victor Norman, Theory
of International Trade. Cambridge: Nisbet, 1980.
Dixit, Avinash K., and Joseph E. Stiglitz,
"Monopolistic Competition and Optimum Product
Diversity," American Economic Review, 1977, 67,
297-308.
Eaton, Jonathan, and Gene M. Grossman,
"Optimal Trade and Industrial Policy Under
Oligopoly," Quarterly Journal of Economics, May
1986, Cl.2, 383-406.
Harris, Richard, and David Cox, Trade, Industrial
Policy and Canadian Manufacturing. Toronto:
University of Toronto Press, 1984.
Helpman, Elhanan, and Paul Krugman, Market
Structure and Foreign Trade: Increasing Returns, Imperfect
Competition, and the International Economy. Cambridge:
MIT Press, 1985.
Horstmann, Ignatius, and James R. Markusen,
"Up the Average Cost Curve: Inefficient Entry
and the New Protectionism," Journal of International
Economics, May 1986, 20.3/4, 225-248.
Krugman, Paul R., "Increasing Returns, Monopolistic
Competition, and International Trade,"
Journal of International Economics, 1979, 9, 469-479.
Krugman, Paul R., ed., Strategic Trade Policy and
the New International Economics. Cambridge: MIT
Press, 1986.
Krugman, Paul R., "Industrial Organization
and International Trade." In Schmalensee,
Richard, and Robert Willig, eds., Handbook of
Industrial Organization. Amsterdam: North-Holland,
forthcoming.
Lancaster, Kelvin, "Intra-industry Trade Under
Perfect Monopolistic Competition," Journal of
International Economics, 1980, 10, 151-175.
Rodrik, Dani, "Imperfect Competition and
Trade Policy in Developing Countries," Harvard
University mimeo, 1987.
Spence, A. Michael, "Product Selection, Fixed
Costs, and Monopolistic Competition," Review of
Economic Studies, 1976, 43, 217-235.
Venables, Anthony J., and M. Alasdair Smith,
"Trade and Industrial Policy Under Imperfect
Competition," Economic Policy, 1986, 3, 621-672.
Vernon, Raymond, "International Investment
and International Trade in the Product Cycle,"
Quarterly Joumal of Economics, May 1966, 80.2,
190-207.
This content downloaded from 131.211.208.19 on Sun, 31 Dec 2017 12:04:57 UTC
All use subject to http://about.jstor.org/terms