Paul Krugman is the winner of the 1991 John Bates Clark Medal, given every two years to "that American economist under forty who is adjudged to have made a significant contribution to economic thought and knowledge." He obtained his B.A. from Yale in 19 74, and his doctorate in 1977 from M. I. T. He returned to Yale as Assistant Professor from 1977 to 1980, only to double back to M.I.T., first as Associate Professor, then as Professor since 1984. He was a Senior Economist at the Council of Economic Advisers in 1982-83. His career has been thus confined geographically to the northeast corridor, but the imagination and power of his ideas have ranged far more widely.1 The range of his writing is equally wide. Although he is being honored by his fellow professionals for his creative contributions to the frontiers of economic knowledge, he is best known outside the profession for a book that disseminates ideas from frontier economics to the general public.

Here is Paul's typical modus operandi. He spots an important economic issue coming down the pike months or years before anyone else. Then he constructs a little model of it, which offers some new and unexpected insight. Soon the issue reaches general attention, and Krugman's model is waiting for other economists to catch up. Their reaction is generally a mixture of admiration and irritation. The model is wonderfully clear and simple. But it leaves out so much, and relies on so many special assumptions including specific functional forms, that they don't think it could possibly do justice to the complexity of the issue. Armies of well-trained economists go to work on it, and extend and generalize it to the point where it would get some respect from rigorous theorists. In this process they do contribute some new ideas and find some new results. But, as a rule, they find something else. Krugman's special structure was so well chosen that most of its essential insights survive all the extension and generalization. His special assumptions go to the heart of the problem, like a narrow and sharp stiletto. By contrast the followers' work often resembles thoracic surgery, involving much clumsy breaking of ribs; sometimes it proves to be no more than an autopsy of the issue.

Over the years Paul has earned the credit for many such kills: intraindustry trade, strategic trade policy, behavior of exchange rates under target zones, speculative attacks on currencies, and most recently, the renaissance of economic geography.

If Krugman is rare among economists in the originality and elegance of his theoretical thinking, he is almost unique in the simplicity and clarity of his expository writing. His two books with Elhanan Helpman, Market Structure and Foreign Trade and Trade Policy and Market Structure,2 have been admired by professional economists and avidly read by graduate students. Now he has reached the general public with The Age of Diminished Expectations (6]. His clear and simple explanations of quite complex economic issues and ideas in a few sentences have to be read to be believed.

How does Krugman maintain such consistently high quality both in his frontier research and in his popular writing? Inherent talent is surely involved, but equally importantly, he uses theoretical modeling and observation of the real world to discipline and reinforce each other. In the introduction to Currency and Crises [10] , his second collection of articles, he includes a "personal manifesto" of his approach: "[T]he essential purpose of this kind of economic analysis is to build intuition ... (and get] greater insight into an economic problem, an insight whose major content is often that of providing a language to discuss that problem. ... This methodology brings with it a natural attitude toward modeling style: a very strong preference for maximum simplicity. ... The point ... is to wear one's technique lightly." But modeling and technique are essential components. When the world presents us with a genuinely new problem, "guesswork is all that we have to go on, and those who discipline their guesses with models are more reliable than those who fly by the seat of their pants, no matter how well tailored.,3

I shall attempt to convey a flavor of Paul's research contributions grouped into some major themes. I will not try to be exhaustive; for reference I have listed his major publications in Table 1. Nor will I try to cover all the links of Paul's work with the small numbers of his precursors and contemporaries, or with the large masses of his followers. To do so would need half -a-dozen separate review articles.

Product Diversity and Monopolistically Competitive Trade

The traditional theory of international trade began with David Ricardo, and reached its peak in the mid-1960s. This theory explained trade in terms of comparative advantage: each country would export the good that it could produce at lower relative cost in autarky. Comparative advantage in turn was explained in terms of differences among countries. The Heckscher-Ohlin model, based on relative differences of primary factor endowments, came to dominate textbooks as well as research papers. Here each country had comparative advantage in the good that used relatively more intensively its relatively more abundant factor.

This theory had several further implications. First, we should see the largest volume of trade between countries that are most different in their endowments, for example industrialized and less developed countries. Second, the opening or liberalization of trade should lead to conflict between the owners of different factors of production. Since exporting a capital-intensive good to import a labor-intensive good is like exporting capital-and importing labor by proxy, trade indirectly faces domestic labor with greater market competition and laborers end up losers. Finally, a group of countries stand to gain most in the aggregate by forming a bloc with more liberal trade within it (such as a free trade area or a customs union) if they are complementary in their factor endowments, so they can produce different commodities when trade expanded.

Just as the dominance of the Heckscher-Ohlin model became complete, disquieting observations contrary to all of these implications began to accumulate. Since the second world war, the fastest growing component of trade was between industrial countries with very similar factor endowments. The European Common Market brought together countries that were not complementary in their factor endowments. Much of this trade expansion seemed to occur with relatively little distributive conflict within each country. Finally, in many emerging industries, one could not point to a clear comparative advantage for any country. Many patterns of production and trade seemed matters of chance; in fact there was a lot of two-way trade in very similar products such as automobiles.

Many different explanations for these facts were offered, and new ones are still being attempted. But the approach that Krugman helped pioneer in a pathbreaking series of papers (16], [17] and (18] was the most drastic departure from Ricardian tradition. The new view in fact went back to an even older tradition, namely Adam Smith's idea that division of labor lowers unit costs. Scale economies internal to firms are incompatible with the perfect competition that was assumed in all traditional models. Many economists throughout the history of the subject had mentioned scale economies as a cause of trade, but they did not have, and could not develop, the tools that would implement this view in models that could yield its logical implications. Krugman found the necessary techniques, and wielded them with such skill and finesse that led not just to a new paradigm, but to a synthesis of the old and the new views of trade.

The scale economies were internal to firms, but sufficiently moderate to ensure the survival of a large number of firms in the free-entry equilibrium of a group producing close but not perfect substitute products. Thus the market structure was that of Chamberlinian monopolistic competition. When such a sector expands, it does so through some combination of an increase in the number of firms (greater product variety) and the size of each firm (greater scale economies). Suppose all the products in the group require the same factor proportions. Let there be another sector, operating under constant returns to scale and perfect competition. When two such countries start to trade, their inter-industry trade (exports of the competitive sector against net imports of the Chamberlinian sector) are still governed by the factor endowment differences as in Heckscher-Ohlin. But when we examine the Chamberlinian sect or more closely, we see that the two countries produce disjoint sets of varieties; the choice of which ones are produced in which country is arbitrary. Each supplies the whole world's demand for the ones it produces, so we get two-way intra-industry trade. If the countries have identical factor endowments, there is no inter-industry trade (each produces an amount of the competitive good equal to its own consumption of it), but lots of intra-industry trade. All these results fit very well with the observations on the growing pattern of trade among industrial European countries cited above.

Even more remarkable is the implication for gains from trade. The availability of greater variety of goods in the Chamberlinian sector at lower unit costs is a benefit to all income-earners. This can be enough to outweigh the conflict over incomes (factor prices) themselves. Then trade liberalization can command general consent. This is more likely the more similar the economies. This again squares with the observation that the formation of the European Economic Community in its initial stage, when the members were very similar economies, generated much less distributional conflict. A similar more recent observation is that the US-Canada free trade agreement produced only minor local complaints of a distributive nature, whereas the expansion of that agreement to include Mexico is proving more controversial.

In all, Krugman's contribution to the development of the monopolistic competition model of intra-industry trade was a remarkable achievement for one so young. The only thing more remarkable is his continued career at the same high level.

Oligopoly and Strategic Trade Policy

Writers of detective stories tell us that the first murder is the hardest to commit; the next ones become easier or even inevitable. It took a lot of thought and tenacity to displace from its pedestal the traditional paradigm of trade based on differences among countries, and to establish as a viable alternative or complement the monopolistic competition theory of intra-industry trade among similar countries. But after that, other market structures became easier targets for analysis. Obviously oligopoly was next.

Jointly with James Brander, and building on an earlier attempt of Brander (1980), Krugman developed in [20] a model of international duopoly. Suppose initially there are two identical countries. There are two goods, one produced competitively under constant returns, and the other monopolized by one firm in each country. Now let trade open up, with some transport cost or other barrier that gives each firm a cost disadvantage in the other's home market. Under perfect competition there would be no trade. But if the firm's strategies are quantities, there can be a Cournot equilibrium where each firm has the majority market share in its own market. Since a Cournot competitor perceives a demand elasticity equal to the industry elasticity divided by its own market share, each perceives a more elastic demand in its export market, and is therefore willing to accept a lower markup there. Thus a common price in each market can be compatible with different costs.

This simple model has some dramatically new implications. (1) There is no comparative advantage but still trade occurs, in fact trade runs "uphill" overcoming the force of the transport cost. (2) There is two-way trade in absolutely identical products. In this respect oligopoly can be even more dramatically different from perfect competition than could monopolistic competition; there at least the two-way intra-industry trade was in varieties that were close but not perfect substitutes. (3) Such seemingly pointless trade can be mutually beneficial to the two countries, because the shift from monopoly to duopoly in each country brings prices closer to marginal costs. This benefit remains even though some resources are being lost in costly transport, so long as the transport costs are not too large.

This model has several special features that contribute to the results. The assumption of Cournot competition, and the assumption that each firm sets its quantity in each country with no resale or third-party arbitrage that would impose a constraint on the price differential, are particularly important. Later work has re-examined the conclusions in other settings; for example Venables (1990), Ben-Zvi and Helpman (1992). But the Brander-Krugman model with its simple structure and dramatic results served a very valuable function; it was a kind of wake-up call to the profession.

Once we were alerted to the potentialities of imperfectly competitive market structures for trade theory, the obvious next question was the role of trade policy in such conditions. Traditional theory based on the perfect competition paradigm had very little good to say about any measures to interfere with free trade. Tariffs could benefit one country if it was large enough to improve its terms of trade, and even this resulted in a net economic loss to the world as a whole. For any other market failures, for example domestic externalities, trade policies were at most inferior substitutes for other policies more precisely targeted at the problems. Export subsidies were particularly useless; they just worsened your terms of trade. Might things be different with imperfect competition?

Many researchers in different settings soon found new possibilities. The role of policy under monopolistic competition was examined by Flam and Helpman (1987) and others; the large-group model of Venables (1990) has some similar features. But the greatest impact and stimulus to research came from the idea of Brander and Spencer (1985) that export subsidies can give our firms a strategic commitment advantage over their foreign rivals, and thus shift some of the pure profit of the international oligopoly toward our firms and contribute to our national income. This is surveyed, discussed and evaluated in [4, Chapters 5, 8], [31, Sections 4, 5], and in Dixit (1987). Here I shall focus on three lines of contributions that Krugman made to this development.

The first was another of his dramatic wake-up calls. Traditional theory had taught us that in general equilibrium, import tariffs discourage exports. This is just the Lerner symmetry theorem. Krugman showed that within oligopolistic industries, things can be dramatically different.

In [21] he developed the first such model. Consider two firms, home and foreign, competing in any number of markets, including, those in the home and foreign countries. The markets are segmented in the sense explained above, and a Cournot duopoly occurs in each. Each firm's marginal cost is a decreasing function of its total output.

To facilitate the analysis, Krugman conducts a thought experiment where each firm is supposed to have constant marginal costs, the level being chosen equal to the actual marginal cost at the equilibrium output. Figure I shows the initial equilibrium E in two such markets by means of the solid reaction curves. Now let the home country impose an import tariff. This raises the delivered cost in the home market for the foreign firm, and shifts its reaction curve downward as shown by the dotted line. The intersection point shifts to El. But that is not the final equilibrium. Since the home firm's output rises, its marginal cost falls; similarly the foreign firm's marginal cost rises. This has further repercussions on their reaction curves in both markets: the home firm' s reaction function shifts to the right, and that of the foreign firm downward. These shifts in turn increase the home firms's sales and decrease the foreign firm's sales in both markets. This further lowers the home firms's marginal cost and raises the foreign firm's, that causes further output changes, and so on. Since all these changes work in the same direction, the qualitative prediction for the final outcome is unambiguous. The new reaction functions are shown by dashed lines, and the equilibrium is marked E*. The home firm has not only reinforced its advantage in the home market, but also gained in the export markets. Import protection has acted as export promotion; defense has proved the best form of attack.

Businessmen find in this a long overdue confirmation of their belief that a secure home base improves their export prospect. Indeed, in his paper Krugman explicitly set out "to show that there is a class of models in which the businessman's view ... makes sense." But therein lies a danger, and although I come to praise Krugman, I must attempt to bury any attempt to draw unwarranted inferences from his model. Makes sense for whom? The businessman, or society as a whole? Krugman constructed a model that showed the former, but did not even investigate the latter. When one firm or industry in the country expands, it draws resources from others, which must therefore contract. Now some of them may have excellent export prospects; and in an economy-wide sense import protection does hurt exports. Some of those industries might also be oligopolistic, and their strategic position will be worsened. Krugman's focus on one industry falls far short of an overall economic welfare analysis and does not constitute a policy prescription. But his clever and provocative title is just inviting to be hijacked by special interests in one industry, who do not care if the rest of the economy is hurt. I think that in his excitement over the new results, Krugman was insufficiently vigilant in deterring such hijackers.

Back to praise. In almost anyone else’s hands, this model would have got bogged down in a mathematical morass of matrix inverses and fixed points. It needed Krugman's deeper understanding of the problem to cut it down to its essentials and express the argument in simple diagrams. As he says in (10, p. xii], "often the truest sophistication is finding a way to express novel ideas with no more than a diagram or a numerical example."

Krugman's second paper in this vein (23] is even more dramatic. Now we have a whole sequence of industries, each with dynamic external economies. The government can protect each for a while, until its costs are reduced and it has a stand-alone competitive advantage in world trade. Then the policy moves on to protect the next industry in line. Again a very clever model, again without any overall normative analysis, and again, alas, a standing invitation for Japanbashers and advocates of a US industrial policy that would protect their own favorite industry.

Krugman is much more careful when he comes to discuss trade and industrial policies in the real world; he did this brilliantly in two scholarly articles (22], (23] and other subsequent popular writings. These articles carefully and evenhandedly assess the successes and failures of Japan's industrial policies in various industries especially steel and semiconductors, the effects of such policies on the US economy as a whole, and the case for and against attempting targeted industrial policies in the US. His recent position favors a small amount of government expenditure, say $10billion or so a year, to identify and promote industries that may have special roles (export profit potential or external effects to other sectors of the US economy).s

All these analyses of imperfectly competitive trade and strategic trade policy were armchair theorizing, informed at most by a casual reference to reality. But with [28], Krugman and Baldwin helped launch new line of 'industry case studies." These adapted the theory to the specifics of the chosen industry, fitted the resulting model to the available data, and then quantified the effects of various trade and industrial policies. They chose an important high-tech industry -- 16K DRAM semiconductor chips. They found that Japan's protection of its market helped it increase its market share dramatically. But the resulting profit was small, and the harm to downstream users of the chips, who had to pay higher prices for a time, was quite substantial. Thus they conclude that the policy was not a net benefit to the Japanese economy. Other studies for other industries and countries have similarly found that the gains from strategic trade policy are of modest magnitude at best. Thus the studies weaken the case of special interests for strategic policies to favor their own industries.

Such models have rapidly progressed from simple functional forms to more realistic ones, and from calibration (getting just enough data to solve for the parameters, or econometrics with zero degrees of freedom) to genuine empirical estimation; observe the difference between two studies of the US automobile industry by Dixit (1987) and Koujianou (1992). But Krugman's pioneering role remains memorable.

Dynamics of Trade and Exchange Rates

In this age of specialization, micro and macro rarely meet; international trade theorists rarely worry about exchange rates, and specialists in international finance pay little attention to the microfoundations of the underlying trade flows. To this rule, as to so many others, Paul Krugman constitutes an exception. Some of his earliest articles [12], [13) are about exchange rates. But he made his most striking contribution to the theory of exchange rate target zones.

The demise of the Bretton Woods system of fixed but adjustable exchange rates in the early 1970s, and the drift to a non-system of floating with some interventions, initially raised hopes that floating exchange rates would insulate each economy from disturbances originating elsewhere, and allow more stable conduct of monetary and fiscal policies. The reality proved otherwise. Nominal as well as real exchange rate fluctuations were much more dramatic than expected, and had large effects on domestic economies. For example, the monetary-fiscal policy mix followed by the UK in 1980, and imitated by the US soon thereafter, led to large real appreciation of the currencies of these countries, causing major recessions and lasting harm to their international competitiveness. Economists who had not long ago welcomed the move toward floating now began to advocate going back to some fixity or management of. exchange rates. John Williamson's (1983) advocacy of 'target zones," where rates would be allowed freedom within the zone but prevented from crossing outside, drew particular attention.

What was lacking was any good conceptual understanding of why exchange rates were so volatile, and any good theory of how they would behave in under an alternative system of target zones. Here were practical men flying by the seats of their well-tailored pants. Along came Krugman with his customary insightful contributions, in the form of a short monograph [3] based on his Lionel Robbins Lectures at the London School of Economics, and some research articles, (24], [32] (joint with Baldwin), and most importantly, [35].

The full story has two components. First, why are exchange rates so volatile? The answer is twofold. First, capital markets are not as efficient as some economists would have them [3, Chapter 3]. But, more interestingly to the theorist, trade flows respond sluggishly to exchange rate swings in a floating regime. Importers and exporters have to incur some sunk costs to move in or out of country markets. Their future revenues are denominated in one currency and their costs in another. When future movements of exchange rates are uncertain, so is their future profitability. Therefore they find an option value in the status quo -- in waiting to observe a little more of the evolution of the rate. This idea was developed in [32]; see also my own contribution (Dixit, 1989). Krugman put this together into an overall view of exchange rate volatility in general equilibrium in (3, Chapter 2]. Exchange rates have little effect on trade flows in the short run because traders wait in the face of large volatility. But the rates move to clear the markets, and they fluctuate so much precisely because the response of trade flows to them is so inelastic. Thus any initial disturbance has a magnified effect on exchange rate volatility.

The second component is how exchange rates would behave in a target zone. Some early commentators feared that they would offer speculators a one-way bet just as the fixed rates under the Bretton Woods system did, and therefore destabilize the regime. In [351, Krugman pointed out that credible target zones have exactly the opposite effect. When the exchange rate nears the top of the zone, it has only one way to go. The asymmetric expectation of its future decline leads investors to hold less of the currency, thus lowering the rate immediately. The opposite happens at the lower end. Thus the policy stabilizes the rate. In fact the function that relates the exchange rate to the fundamentals (like money supply or velocity) is S-shaped. Its slope is everywhere less than unity; this is the stabilization property. At the extreme edges of the band, the slope falls to zero. This remarkable fact follows from a subtle mathematical property of stochastic control and option pricing known as "smooth pasting. 11 Krugman discovered the property independently from purely economic arbitrage reasoning, and helped launch a line of research that Rudi Dornbusch has called "smooth pastry," a veritable bakery of models.

In fact some of this research was spurred by a counterfactual implication of Krugman's model. Since the exchange rate responds but little to the fundamentals near the edges of the bands, and the fundamentals are assumed to follow a random walk which has a uniform long run distribution over its whole range, the long run distribution of the exchange rate must have peaks at the edges. In fact, rates in systems like the European Monetary Union are found to spend much more time near the middle of the band. To resolve this discrepancy, people have modified Krugman's model in various ways, allowing inframarginal intervention, realignments of the bands, etc. For reviews of this burgeoning literature, see Bertola (1992) and Svensson (1992). Krugman himself has helped edit a volume [8], and contributed to it jointly with Rotemberg [381. If a target zone is not fully credible, the possibility of speculative attack on the currency in anticipation of the collapse of the regime should be considered. Krugman and Rotemberg do so, in the process returning to an early paper of Krugman's [151. The general insight is that "to understand how an exchange rate regime works, one must also understand how it ends" [38, p. 1311. In the process the paper contributes a nice new twist to " smooth pastry, " and launches a new mini - industry of dynamic analyses of the gold standard.

Krugman's analysis of exchange rate volatility led him to a cautious endorsement of a move toward greater fixity [3, pp. 99-102). I have diminished expectations, and remain skeptical. The markets are not managing exchange rates well, but to me the track record of experts, central bankers and governments, and of their regularly changing pronouncements, suggests that they will do even worse.

Trade, Debt and Development

I have spent too much space on my favorite topics and must be very brief with the rest. Krugman's contribution to the LDC debt issue was typically incisive. Practical men in well-tailored pants split into two groups, ones with harder heads advocating refinancing of the debt, and those with softer hearts favoring some forgiveness. Krugman pinpointed the essential trade-off. Refinancing gives the creditor an option, because if the country gets a lucky break the creditor stands to recoup more. But the debt overhang reduces the debtor's incentives to take actions that will promote more favorable outcomes, because part of the gain will go to the creditor. Thus refinancing reduces the likelihood of the lucky breaks. In extreme cases forgiveness may be in the aggregate interest of the creditors, too; the country may be on the wrong side of "the debt Laffer curve." This deeper understanding of the issue should have led to better policies; alas, practical men seem to prefer continuing to fly by the seats of their pants.

Early in his career, in [14] and [19] Krugman examined some implications of technology transfer and dynamic externalities for the relative growth of industrial and developing countries. The models were typically simple and incisive, but nowadays we would criticize them for accepting as exogenous some features that should be explained within the model. The papers are nonetheless seminal; in particular they were an early inspiration for the research of Gene Grossman and Elhanan Helpman that led to their magnum opus (1991).

Economic Geography and History

Krugman has recently argued that international trade-is but one aspect of the more general theory of location of production, and brought to economic geography the same new insights based on scale economies and industrial organization that he so effectively used in launching a new paradigm in trade theory. His articles [36], [37] and the monograph [7] have found some of the same implications in this new context -- an essential arbitrariness of industrial location, and intriguing implications for policy intervention. This work, and a related but more explicitly dynamic model [33], also emphasize the importance of history in explaining observed patterns of industry location and growth, supporting the well-known work of David (1985) and others. This work seems sure to launch another new wave of followers of Krugman.

Diminished Expectations

Paul's command of the research frontier in economics, his previous scholarly participation in policy debates, and his facility in writing, all came together in his brilliant popular book [6]. Its main purpose is to give the practical reader an understanding of current economic issues, of how and why we have come to "live in an 'age of diminished expectations,' an era in which our economy has not delivered very much but in which there is little political demand that it do better.' The most important issues - slow productivity growth, increasingly skewed distribution of income and wealth, and chronic unemployment that exceeds previous decades' experience - have become too difficult to do anything about. In the meantime the process tries to grapple with symptoms and side-issues, such as inflation, the twin deficits in US trade and the Federal budget, and financial crashes, with little effect on the main problems. "My happiness depends almost entirely on a few important things, like work, love, and health, and everything else is not really worth worrying about -- except that I usually can't or won't do anything to change the basic structure of my life, and so I worry about small things, like the state of my basement.

For me, the best thing about the book is not its vivid portrayal of the sorry but realistic picture of drift in the US economy and policy, but the amount of economics it teaches in the process of doing so. Even very difficult and subtle points, which were made by various original authors using very complicated mathematical models, are conveyed in a few clear lines of prose. The discussion of Grossman and Hart's theory of takeovers (pp. 163-4) is a particularly good example.6 Would that the original authors could express their own insights so simply and clearly. If Krugman were not too valuable to the profession for his own work, we should appoint him to a permanent position as the translator of economics journals into English.

Raised Expectations

Our expectations about the US economy and the US government’s future economic policies may have been diminished as a result of our observation of recent economic performance and policies. But our expectations about Paul Krugman's future economic research and writing have been greatly raised by our observation of his record so far. I am sure the Clark Medal is but one milestone of many to come in his career.

Books

[1] Market Structure and Foreign Trade (with Elhanan Helpman), Cambridge, MA: MIT Press, 1985.

[2] Strategic Trade Policy and the New International Economics (editor),

Cambridge, MA: MIT Press, 1986.

[31 Exchange Rate Instability, Cambridge, MA: MIT Press, 1988.

(4] Trade Policy and Market Structure (with Elhanan Helpman) Cambridge, MA: MIT Press, 1989.

[5] Foreign Direct Investment in the United States (with E. Graham), Washington, DC: Institute for International Economics, 1989.

[61 The Age of Diminished Expectations, Washington Post Briefing Books, 1990.

[7] Geography and Trade, Cambridge, MA: MIT Press, 1991.

[8] Exchange Rate Targets and Currency Bands, (editor, with Marcus Miller), Cambridge, UK: Cambridge University Press, 1992.

Collected Articles

[91 Rethinking International Trade, Cambridge, MA: MIT Press, 1990.

[10] Currencies and Crises, Cambridge, MA: MIT Press, 1992.

Articles

(12] "Purchasing power parity and exchange rates," Journal of International Economics, 8, 1978, 397-407.

[13] "Contractionary effects of a devaluation," (with Lance Taylor) Journal of International Economics, 8, 1978, 445-456.

[14] 'A model of innovation, technology transfer, and the world distribution of income," Journal of Political Economy, 87, 1979, 253-266.

(16] "Increasing returns, monopolistic competition, and international trade," Journal of International Economics, 9, 1979, 469-479.

[17] "Scale economies, product differentiation, and the pattern of trade," American Economic Review, 70, 1980, 950-959.

(18] "Intraindustry specialization and the gains from trade,' Journal of Political Economy, 91, 1981, 959-973.

[19] "Trade, accumulation, and uneven development,' Journal of Development Economics, 8, 1981, 149-161.

[20] 'A reciprocal dumping model of international trade," (with James Brander) Journal of International Economics, 15, 1983, 313-321.

[21] "Import protection as export promotion: International competition in the presence of oligopoly and economies of scale, 11 in Monopolistic Competition in International Trade, ed. Henryk Kierzkowski, Oxford, UK: Oxford University Press, 1984, 180-193.

[22] "Targeted industrial policies: Theory and evidence," in Industrial Change and Public Policy, Kansas City: Federal Reserve Bank, 1983, 123-155.

[23] "The US response to foreign industrial targeting and the US economy,'Brookings Papers on Economic Activity, 1, 1984, 77-121.

[24] The persistence of the US trade deficit," (With Richard Baldwin) Brookings Papers on Economic Activity, 1, 1987, 1-43.

(25] "Is free trade passé?" Journal of Economic Perspectives, 1, 1987, 131-144.

[26] "The narrow moving band, the Dutch disease, and the economic consequences of Mrs. Thatcher: Notes on trade in the presence of dynamic economies of scale,' Journal of Development Economics, 27, 1987, 41-56.

(28] "Market access and competition: A simulation study of 16K random access memories," (with Richard Baldwin) in Empirical Methods in International Trade, Cambridge, MA: MIT Press, 1988, 171-197.

[29] "Financing v. forgiving a debt overhang,' Journal of Development Economics, 29, 1988, 253-268.

Industrial Organization, eds. Richard Schmalensee and Robert Willig, Amsterdam: NorthHolland, 1989, 1179-1223.

[32] "Persistent trade effects of large exchange rate shocks,' (with Richard Baldwin) Quarterly Journal of Economics, 104, 1989, 635-654.

[33] "History versus expectations,' Quarterly Journal of Economics, 106,

1991, 651-667.

[34] "Is bilateralism bad?" in International Trade Theory and Policy, eds. Elhanan Helpman and Assaf Razin, Cambridge, MA: MIT Press, 1991, 9-23.

economics, 106, 1991, 669-682.

[36] "Increasing returns and economic geography,' Journal of Political Economy, 99, 199 1, 483-499.

[37] "History and industrial location: The case of the manufacturing belt," American Economic Review, Papers and Proceedings, 81, 1991, 80-83.

[38] "Speculative attacks on target zones," (with Julio Rotemberg) in Exchange Rate Target and Currency Bands, eds. Paul Krugman and Marcus Miller, Cambridge, UK: Cambridge University Press, 1992, 117-132.

Brander, James (1981) "Intra-industry trade in identical commodities,' Journal of International Economics, 11, 1-14.

David, Paul (1985) "Clio and the economics of QWERTY," American Economic Review, 75, Papers and Proceedings,

Dixit, Avinash (1988) "Optimal trade and industrial policies for the U.S. automobile industry,' in Empirical Methods in International Trade, ed. Robert Feenstra, Cambridge, MA: MIT Press, 1988, pp. 141-165..

Dixit, Avinash (1989) "Hysteresis, import penetration, and exchange rate

Flam, Harry and Helpman, Elhanan (1987) "Industrial policy under monopolistic competition,' Journal of International Economics, 22, 79-102.

Grossman, Gene and Helpman, Elhanan (1991) Innovation and Growth fn the Global Economy, Cambridge, MA: MIT Press.

Svensson, Lars E. 0. (1992) "Recent research on exchange rate target zones: An interpretation," Journal of Economic Perspectives, forthcoming.

Venables, Anthony (1990) "International capacity choice and national market games," Journal of International Economics, 29, 23-42.

Invited contribution for the Journal of Economic Perspectives. Preliminary draft; please do not quote or cite. Comments and suggestions welcome.

**. Avinash Dixit is John J. F. Sherrerd 152 University Professor of Economics at Princeton University.

1. Indeed, early in his career, Paul wrote a clever paper on interstellar trade, where goods are transported from one stellar system to another at speeds close to that of light, and therefore time and interest rates differ in different frames of reference.

2. These are items [1] and [4] respectively in Table 1, which lists Krugman's major publications.

3. My own favorite analogy is the army. Economic theorists are like the general staff; they are busy devising ways to win the last war. Practical economists are like infantry; they are superbly trained and equipped to win the war before the last one. When an actual new war erupts, the two groups must somehow combine their different talents and muddle through.

4. See also his later surveys and evaluations of this literature: [1, Chapters 6-9], [27, Section 2], and (31, Section 1].

5. I am somewhat more pessimistic, not -because I believe in any magic of the markets, but because I believe that there is no market failure so bad that the US government and political process could not do even worse.

6. At a slightly more technical level, equally remarkable is Krugman's geometric exposition [31, Section 21 of Krishna's (1989) difficult and subtle model of the effects of quotas in facilitating international duopolistic collusion.