Introduction

Outsourcing and offshoring of services has attracted a lot of interest recently, perhaps in large part for the simple reason that it is a relatively new phenomenon. But much of the interest is not just academic curiosity. Some concern is generated by the offshoring of white-collar services to relatively low-income countries such as India. It was one thing when low to moderately skilled manufacturing jobs were gradually lost to lower-income countries over the last two decades, but somehow the potential competition for and loss of white-collar jobs seems just as threatening.

There has been little new theory to guide us in understanding this expansion of trade and investment into new activities which were previously classified as non-traded. Possibly, no new theory is needed, and the new trade in services is just a particular case of our more general models. But at the very least, a more detailed development of this ‘particular case’ seems warranted given the empirical and policy attention it has received. The purpose of this paper is thus to inquire how theory might be adapted and developed to shed light on the new offshoring of white-collar services. Particular attention will be paid to small open high-skilled economies, and how they might be affected by increased trade in skilled services.

Defining services has always proved difficult and ambiguous, and I will instead simply indicate a range of activities that I have in mind.

Introduction

Regionalism has been an important phenomenon in the world of international trade in the last decade. Groups of countries, typically geographically concentrated, are banding together to liberalise trade and investment among themselves. The EU is surely the furthest along, with relatively liberal provisions for labour migration added to trade and investment liberalisation.

One interesting and relatively novel feature of some of the new regional trade agreements (RTAs) is that they combine partners of very different levels of development. Typically this was not the case during previous decades, when such agreements tended to be among countries of similar per capita income levels. The NAFTA was pioneering in this respect, and may be expanded to include other Latin American countries in the next few decades. Similarly, the EU will surely consider substantial liberalisations in the future with countries from Eastern Europe and the former Soviet Union.

Factors which motivate and encourage these new ‘North–South’ or ‘East–West’ agreements may also differ from the older agreements among highly developed countries. The latter were in large part motivated by the objective of creating large internal markets in order to capture scale economies and other production efficiencies. But the newer agreements have a somewhat different focus. First, the developed partner(s) may be seeking a low-wage partner who can provide low-cost labour for labourintensive tasks of the developed country's firms. The less-developed partner(s) may be seeking access to inward investment and newer technologies. A somewhat more subtle motive for the less-developed country (LDC) is to obtain ‘insurance’ against capricious policy changes by the developed countries.

Introduction

International trade has a long tradition of explaining trade flows and international differences in sectoral production levels by differences in relative factor endowments among countries. Dual results relate real factor rewards to international prices and trade barriers. But factor-proportions trade theory, at least in its traditional competitive formulation, is not well suited to discussions about the role of trade in technologies and knowledge capital in determining real wages and national standards of living. Because of problems relating to the public goods nature of knowledge or to the firm-specific character of knowledge and skills, the services of these assets are often exploited internally within multinational firms in serving foreign markets.

Many theoretical and empirical developments have improved our understanding of which firm-level characteristics lead to industries dominated by multinationals. More recently, we have begun to incorporate these firm-based models into the general equilibrium theory of international trade so that we understand, for example, which country characteristics lead to international activity dominated by direct investment rather than trade.

The purpose of this paper is to exploit these recent developments in order to improve our understanding of how multinationals in turn influence certain variables in equilibrium, outputs and factor prices in particular. While previous work has given us a basis for understanding how country characteristics such as size, relative endowments, and trade costs lead to multinational firms, we now turn to the question of how the introduction of multinational production alters the inter-country distribution of production and the intra-country distribution of income.

We adapt our two-country, two-sector, two-factor static model developed in Markusen and Venables (1995a, 1996).

Introduction

In December 1992, the heads of state of the United States, Canada, and Mexico signed a trade agreement that could significantly liberalize trade between these neighboring countries in North America. This chapter provides an analysis of the effects of the new agreement on one industry – the automotive sector. We focus on production, employment, and consumer welfare effects of the agreement as simulated in a calibrated general equilibrium model that accounts for production and trade in automotive parts, engines, and finished automobiles. The model distinguishes between the effects of the agrement on the “Big Three” North American firms and on foreign firms producing in North America. This distinction is quite important because the new agreement will introduce significant nontariff barriers (NTBs) in the form of content rules for firms selling in the expanded North American market. The model we have developed provides a framework in which we can evaluate the effects of these important new nontariff barriers that may become permanent features of the North American economic landscape.

The analytical framework employed in this chapter is based on two earlier papers [Hunter, Markusen, and Rutherford (1990) and López-de-Silanes, Markusen, and Rutherford, 1992)]. In this chapter, we have extended our previous modeling work in several areas. First, we now distinguish two primary factor inputs to production: labor, and capital. Second, our new model accounts for more aspects of intrafirm competition in the international auto market.

Introduction

For those of us who received our classroom instruction in international trade theory prior to 1971, the 2 × 2 Heckscher-Ohlin model of trade constituted our “basic training.” The two-good, two-factor model with both factors perfectly mobile between the two sectors was the simple general-equilibrium model of factor proportions trade (simple one-factor Ricardian models constituted a technology-based explanation of trade). It is not clear what the justification for this exclusive focus was, but it is certainly true that the four theorems derived from that model constitute an elegant statement of theory: the Heckscher–Ohlin theorem, the factorprice- equalization theorem, the Rybczynski theorem, and the Stolper- Samuelson theorem.

In 1971, Ronald Jones published his paper, “A Three Factor Model in Theory, Trade, and History,” in which two goods are produced with three factors, two of which are sector specific and one of which is mobile between sectors. Jones expressed several motivations for this model. First, it removed the “straightjacket” of factor-price equalization, in that factor prices now depend on factor supplies as well as on commodity prices. Second, Jones found the model consistent with recent articles (at the time) in economic history and the theory of capital and international trade.

Gradually, the Jones specific-factors model has become more and more popular, both for pedagogic and for research purposes. The model in my opinion is now used at least as often as the Heckscher–Ohlin model to teach the simple general-equilibrium theory of trade. Several reasons for this popularity are found in later papers by Mayer (1974), Mussa (1974), and Neary (1978).