Preface*

Almost twenty years have passed since the publication of Volume 3 of the Handbook of International Economics in 1995. Much has changed since then, both in international trade and international macroeconomics. The changes are fourfold: (a) new questions have arisen as the world trade and payment system has evolved; (b) new data sets have become available; (c) new theoretical models have been designed to address new issues, but they have also enabled sharper and deeper analyses of older issues; and (d) new empirical studies have greatly enriched our understanding of the global economy.

The chapters in this handbook review, illuminate, and interpret these developments in a systematic way, making this material—which is technical at times—accessible to professional economists and graduate students alike. Trade is covered in the first six chapters, and international macroeconomics is covered in the subsequent six chapters.

1 International Trade

Neoclassical analysis of foreign trade focused on comparative advantage at the sectoral level, be it due to variation in productivity or factor endowments. Firms as suppliers of unique brands of differentiated products and monopolistic competition were integrated into trade theory in the 1980s (see Krugman 1995 for a review). Yet as much as these improvements have been important, they focused on sectoral outcomes by treating firms within industries as symmetric entities. Given the aim of that research, which was to expand the neoclassical framework to accommodate intraindustry trade and large volumes of trade between similar countries, the symmetry assumption was a reasonable simplification. Except that it proved to be inadequate for interpreting evidence that emerged in the 1990s concerning the participation of firms in foreign trade, as new firm-level data sets became available. In these data, exporting firms differ systematically from nonexporters. Scholars responded with the development of new models in which firms are heterogeneous, and these models guided empirical studies with the new rich data sets.

Melitz and Redding review these developments in Chapter 1. After describing the patterns of firm heterogeneity in the data, they develop an integrated multisector analytical framework for discussing many issues that have been analyzed in the recent literature. In the data, exporters differ from nonexporters in a number of dimensions; e.g., exporters are bigger and more productive than nonexporters and they pay higher wages. Studies of trade liberalization show that it leads to substantial reallocation within industries; low-productivity firms exit while market shares are reallocated to more productive firms, and especially to exporters. In addition, firms and product margins are important determinants of trade flows, because variation in the number of products explains a large fraction of the variation in trade volumes. These are some of the findings that motivated the original theoretical analysis that Melitz and Redding review in this chapter.

The analytical framework consists of sectors that have the features developed in Melitz (2003), as well as a homogeneous good sector (although the latter is shut down in some applications). Firms enter an industry in anticipation of a productivity draw. After the entry cost is sunk and a firm’s productivity revealed, the firm has to decide whether to stay in the industry or exit. If it stays, it has to decide whether to serve only the domestic market or also export. There is a fixed cost of operation and a fixed cost of exporting, as well as variable trade costs. In the now familiar manner, firms choose among these strategies based on productivity: the least productive firms exit, the most productive become exporters, and intermediate-productivity firms serve only the domestic market.

Although these results are not new, the exposition is new and intuitive, making the analysis accessible to many readers. This useful feature characterizes the entire chapter. It is especially helpful in the discussion of within-sectoral reallocations in response to declining trade costs and the home market effect.

Melitz and Redding also explain how these models have guided estimation of trade flows and quantitative analysis. Some of these issues are discussed in more detail in Chapter 3 on gravity equations and in Chapter 4 on trade theory with numbers. They also discuss the integration of factor proportions into the multisector framework and some of its consequences, such as the relationship between factor endowments and endogenous sectoral productivity levels.

Up to this point preferences for variety were assumed to be of the constant-elasticity-of-substitution type, resulting in constant markups in percentage terms. In Section 8, Melitz and Redding replace those preferences with a generalized quadratic system that yields linear demand functions with an intercept that depends on sectoral conditions. Under these circumstances markups vary across firms and they respond to demand and supply shocks. Since trade is costly, international markets are not fully integrated and market size impacts markups and average sectoral productivity levels. As a result, productivity is higher and markups are lower in larger economies.

In this case multilateral trade liberalization leads to exit of low-productivity firms and market share reallocation toward more productive firms, including exporters, but in addition it reduces markups. The result is that prices are lower because of both higher productivity and lower markups.

Section 9 then examines endogenous firm-specific productivity levels, by allowing firms to engage in innovation or technology adoption that augments their productivity draws. Choice of product scope for multiproduct firms is also considered, which provides another endogenous source of productivity variation across firms. In a dynamic context this generates a complementarity between exports and productivity-enhancing investments, because the payoff to productivity improvements is higher for exporting firms.

A number of empirical studies have shown that key predictions of these models are consistent with the data. As a rule, exporters invest more than nonexporters in new technologies. And in the face of trade liberalization, multiproduct firms shed their worst products while nonexporters who switch to exporting invest the most in technology upgrading.

Finally, Melitz and Redding review the recent literature on the impact of trade on wage inequity in the presence of labor market frictions. Unlike the neoclassical writings that discussed wage inequality across workers with different characteristics, such as differences in human capital, the new literature emphasizes residual wage inequality—wage inequality between workers with similar characteristics—that has become a large contributor to the inequality of earnings. Firm heterogeneity plays an important role in this relationship, both theoretically and empirically, because firms with varying attributes within sectors pay different wages to workers with similar characteristics.

In Chapter 2, Antràs and Yeaple review the evolution of ideas about the role of multinational firms in the world’s trading system. Much has happened on this front since the publication of the previous volume of the Handbook of International Economics. On the theoretical side the internalization decision has been afforded new foundations and within-industry firm heterogeneity has been extended to cover multinationals. On the empirical side new hypotheses emanating from the novel theory have been examined with rich new data sets.

Antràs and Yeaple first review some stylized facts about the operation of multinational corporations. Some of these are similar to stylized facts about trade flows; e.g., foreign direct investment (FDI) is concentrated within the group of rich countries and it exhibits large two-way flows (intraindustry FDI). There are, however, substantial differences across sectors; e.g., multinationals operate primarily in capital-intensive and R&D-intensive sectors. Within sectors, both the parents of multinationals and their affiliates tend to be larger, more productive, more R&D intensive, and more export oriented than nonmultinational corporations. And moreover, parents tend to specialize in R&D while affiliates tend to sell goods primarily in the host countries.

After presenting this evidence, Antràs and Yeaple develop a simple analytical framework of trade with monopolistic competition that includes multinational corporations (MNCs), but no within-industry heterogeneity. They apply it to horizontal FDI, driven by the proximity-concentration tradeoff. In this view a firm can serve a foreign market via export or subsidiary sales. The former entails variable trade costs—such as tariffs or freight—that are saved by the latter mode of operation, while FDI entails fixed costs of acquiring and operating a production unit in the foreign country, which are saved by exporting. The evidence, based on U.S. data, supports this tradeoff: export relative to subsidiary sales is lower in sectors with high variable trade costs and higher in sectors with high plant-level fixed costs.

Following this discussion, the theory is extended to include heterogeneous firms. Now the most productive firms become multinational corporations while the least productive (that remain in the industry) serve only the domestic market. Firms with intermediate productivity levels serve foreign markets via export sales. This sorting pattern holds in several data sets. The model also predicts a proximity-concentration...