InternationalTrade[221]

The Nobel Prize in Economics has been awarded to scholars in international trade twice, in 1977 to James Meade and Bertil Ohlin jointly, and in 2008 to Paul Krugman. Tony's work on the subject starts from where Ohlin and Meade ended and has been instrumental to Krugman's success.

What determines the pattern of international trade? What are the associ­ated welfare gains and losses? What are the sources of these gains and losses? Can individual incentives to trade diverge from societal objectives? Should governments intervene to promote or restrict international trade? Questions like these have always been at the core of the debate on trade down the years (indeed, over many centuries; see Irwin 1996). In the last few decades, they have gained new salience, in Europe and elsewhere. The world first embraced globalisation after the Second World War but then with the new century it caved into renewed protectionist pressures and resurgent nationalistic tendencies due to disenchantment with the globalised economy, especially after the financial crisis of 2008.

The ideas underpinning Krugman's Nobel Prize were the products of this roller coaster period, just like those motivating Meade and Ohlin were the product of an earlier period. To understand why, it is useful to divide the recent history of international trade relations into two waves of globalisation (see Baldwin and Martin 1999). The first started in the mid-nineteenth cen­tury and ran up the eve of the First World War. It roughly coincided with the Second Industrial Revolution, during which new manufacturing, transporta­tion and communication technologies diffused from Great Britain to Continental Europe and a small set of other countries worldwide. The result was the emergence of an industrialised “North”, exporting manufactures to a less developed and often colonised “South” in exchange for raw materials and primary products.

During this first wave of globalisation, due to North-South international specialisation in production, international trade was characterised by the exchange of different goods between structurally different countries. In eco­nomics, such an intersectoral pattern of trade soon found two robust theoreti­cal explanations at the core of what later came to be known as traditional trade theory (TTT). Both explanations highlighted the role of relative cost differences (comparative advantages) between countries, predicting that a country would export the goods that it is able to produce at relatively lower costs. The two explanations differed, however, in terms of the sources of cost differences, which were to be found in the uneven international distribution of either technologies (Ricardian model) or relative factor endowments (Heckscher-Ohlin model). It was for their contributions to the theory of comparative advantage that Ohlin and Meade were (belatedly) awarded the Nobel Prize in 1977.

The second wave of globalisation began to gain momentum just after the Second World War, and it is still going on despite changing moods about its merit, especially in places with fading manufacturing traditions, like Newport. In this period, further technological improvements in production, transporta­tion and communication technologies, and their steady diffusion to a growing number of countries, brought a substantial change in international trade pat­terns. These started to be dominated by the exchange of similar goods between structurally similar “northern” countries, sharing roughly the same technolo­gies and relative factor endowments. The rise of this type of intra-industry trade between rich countries created a conundrum for the traditional theories based on comparative advantage as these explained bilateral trade flows in terms of differences between trading partners (see Linder 1961; Grubel and Lloyd 1975; Greenaway and Milner 1986).

The counterfactual predictions of the Ricardian and Heckscher-Ohlin models were derived from two specific simplifying assumptions: constant returns to scale at the firm level and perfect competition in all markets. These assumptions anchored those models to the standard Arrow-Debreu paradigm of general equilibrium theory in which incentives to trade arise only when traders have different individual assessments of the relative values of the trans­acted goods. The larger the difference in those assessments, the higher their incentives to transact and thus the volumes of trade. Vice versa, individuals sharing the same assessments have no incentive to trade. This is indeed the case for countries sharing the same technologies and relative factor endow­ments, as their autarky relative prices are identical.

While it was clear that the Arrow-Debreu assumptions were putting a straitjacket on the ability to explain the structure of world trade, for a long time the lack of tractable general equilibrium models with increasing returns to scale and imperfect competition hampered progress in international trade theory. This state of affairs started to change in the late 1970s when new par­tial equilibrium models of oligopoly and monopolistic competition were bor­rowed from industrial organisation and transplanted to the general equilibrium framework of international trade theory (see Helpman 1984a). By the end of the 1970s and the beginning of the 1980s the so-called new trade theory (NTT), which would transform the field, had been born (see, for example, Krugman 1979, 1980; Dixit and Norman 1980; Markusen 1981; Brander and Krugman 1983; Helpman 1984b).

With the benefit of hindsight, it is now clear that what had held interna­tional trade theory back had been its “obsession” with general equilibrium. This obsession is easily explained and justified by the fact that the assessment of the effects of trade liberalisation on a national economy necessarily requires an understanding of what happens to factor incomes and prices. In other words, ‘you want a general-equilibrium story, in which it is clear where the money comes from and where it goes' (Fujita and Krugman 2004: 141).

At the same time, even armchair evidence makes it clear that a theoretical account of the structure of world trade cannot fly without a model of firm behaviour. By assumption, however, in the perfectly competitive Arrow- Debreu paradigm, the boundaries of the firm are undetermined. A firm, whatever that may be in an Arrow-Debreu world, is just a production func­tion and, as such, has no “behaviour” whatsoever. Yet, as John Cashmore would testify, firm behaviour is important in many respects: firms decide whether to launch new products and dispense with old ones, where to pro­duce and where to sell their goods, whether to compete in prices or quantities, how to organise their operations and so on.

The key that eventually unlocked the door of general equilibrium with imperfect competition was the monopolistically competitive model by Krugman (1980), heralded in earlier unpublished work by Norman (1976). The idea of monopolistic competition is a rather old one, dating back at least to the early 1930s. Chamberlin (1933) introduced the idea of “large group competition”, where firms retain some monopoly power thanks to product differentiation, yet are small in the aggregate economy. The idea that firms are small in the economy can be made precise by assuming that there is a “con­tinuum” of firms. In such a setting, firms are aware that they are price makers as they face finitely elastic demand for their products while their behaviour has no impact on market aggregates like gross domestic product (GDP), the number of firms, consumer income and price indices.

Such “non-strategic” behaviour allows one to sidestep a myriad of thorny technical problems that arise once we seriously think about oligopoly in gen­eral equilibrium, such as the existence of equilibria or diverging conclusions depending on whether firms maximise profits or the welfare of their share­holders by choosing prices or quantities. Though one may argue that the properties of monopolistic competition are rather special and may limit the generality of the analysis, they offer the advantage of laying out a clear frame­work within which macroeconomic issues can be parsimoniously examined. Thanks to the theoretical and empirical success of monopolistically competi­tive models in accounting for the exchange of similar goods between structur­ally similar “northern” countries, comparative advantage is today usually viewed as driving specialisation at the industry level, whereas product differ­entiation and economies of scale are utilised to explain what drives specialisa­tion at the product level.

Trade and gains from trade then arise not only because international exchange allows countries to specialise according to comparative advantage, but also because it increases the variety of products available and reduces the market power of domestic firms, thus leading to smaller markups, lower prices as well as larger quantities consumed. With increasing returns to scale, larger quantities serve to reduce prices not only because markups fall but because average production costs fall as well.

TTT and NTT are clearly complementary and in the 1980s their synthesis provided a unified view of international trade, changing the way economists understand the patterns, the gains and the sources of international trade (see Dixit and Norman 1980; Helpman and Krugman 1985). It also enriched the way that economists understand the effects of trade policies (see Helpman and Krugman 1989) and how trade barriers affect economic growth (see Grossman and Helpman 1993), economic geography (see Fujita et al.

For his contributions to NTT (and to the so-called new economic geogra­phy (NEG)), Krugman was awarded the Nobel Prize in 2008. Tony's contri­bution to the body of work that led to Krugman's Nobel is substantial, starting with key studies on the TTT-NTT synthesis, such as Venables (1987) and later Markusen and Venables (2000). Of particular relevance is the trilogy of articles published in the Journal of International Economics in the 1980s (see Venables 1982, 1984, 1985), which became reference points for subsequent generations of trade scholars. For instance, one can read Melitz and Ottaviano (2008) who point out that almost all the rich welfare effects of trade featured in the “new” NTT literature with heterogeneous firms, developed in the wake of Melitz (2003), had been already identified in the “old” NTT literature with representative firms. In particular, the welfare gains from additional product variety as well as the asymmetric welfare gains of trade induced by differences in country size and trade costs had been highlighted by Krugman (1980). Krugman (1979) had also shown how trade can induce pro-competitive effects in a model with monopolistic competition and endogenous markups, while Markusen (1981) had formalised and highlighted the pro-competitive effects from trade due to the reduction in market power of a domestic monop­olist. Horstmann and Markusen (1986) and Venables (1985) had extended this modelling framework to the case of oligopoly with free entry (while maintaining the assumption of a homogeneous traded good). These papers had also emphasised, among other things, how free entry could generate wel­fare losses for a country unilaterally liberalising imports by “reallocating” firms towards the country's trading partners. Venables (1987), a paper not in the trilogy but on which Tony had already been working in parallel when visiting the University of British Columbia in 1982-1983, had shown how this effect can be generated in a model with monopolistic competition and product differentiation with exogenous markups. The new NTT models additionally captured the welfare effects stemming from changes in average productivity based on the selection of heterogeneous firms into domestic and export markets. More recently, Haaland and Venables (2016) have discussed some of those models' policy implications in another paper in the Journal of International Economics that echoes Venables (1982) after more than 30 years.

The third paper of the 1980s trilogy hints at what would arguably become Tony's most important contribution to economics and his ongoing passion: the study of the interactions between geography, development and trade. Specifically, Venables (1984) developed a model of trade with monopolistic competition in which small perturbations in the parameters radically change the number and type of equilibria. For certain parameter values, the model exhibits multiple stable equilibria. In some of them, sectoral specialisation across countries is not complete and intra-industry trade takes place. In oth­ers, specialisation is complete and there is no intra-industry trade. This sug­gested that, differently from TTT, NTT models could lead to path dependency, lock-in effects and suboptimal outcomes in patterns of international trade and development, an insight at the core of the NEG.

3