Migration, Unemployment and Trade

The pure theory of international trade was originally based on the mobility of goods and international immobility of factors with full employment. The Ricardian and Heckscher-Ohlin (HO) models of comparative advantage explain the pattern of trade using two sector models of general equilibrium in which goods are perfectly mobile while factors are mobile only within a country. Building on the HO model, Stolper and Samuelson (1941) derived an important theorem regarding tariffs and return to factors and later Rybczynski (1957) proved its dual relating factor endowments and output levels. In two pioneering papers, Samuelson (1947,1948), established that under certain conditions, free goods mobility resulted in factor price equalization across countries. The real theory of trade consists of these five core theorems: the Ricardian; the HeckscherOhlin, the Stolper-Samuelson; the Rybczynski and the factor price equalization theorem.

This chapter provides a simple and systematic treatment of a twocommodity (two-sector), two-factor general equilibrium model of a closed economy which is widely used in several real models of trade.1 This model is the cornerstone of the Heckscher-Ohlin, the Ricardo-Samuelson-Viner, the Harris-Todaro and many other models of trade.² This framework can be extended to analyse many interesting problems in economics. More importantly it is utilised in the trade and distortion literature for deriving optimal policies and other important results. This chapter is motivated by the desire to make this book as self contained as possible so that the reader does not have to search for background material in various other books or journals. The model is presented in both the primal and dual forms which are widely used in this work and trade literature.

The object of this chapter is to provide a comprehensive treatment of the generalised Harris-Todaro model (henceforth called GHT) which underlies several chapters in this book. The original Harris-Todaro model. (hereforth called HT) was generalised to include non-traded goods by Hazari. and Sgro (1991). The HT model was developed to explain internal migration and urban unemployment in the context of Third World Countries. This model consisted of two regions: urban and rural; two internationally traded goods, manufacturing and agricultural; a minimum urban wage and a migration function¹. This general equilibrium structure resulted in urban unemployment due to the presence of a minimum wage and via migration of rural workers to urban areas from the expected wage mechanism. We do not offer a view of migration based on the work of Stark (1991). His explanation of internal migration is based on variables such as insurance provided by having children so that one can be looked after when old, fertility, family connections and other variables. While these ideas are interesting they are not amenable to general equilibrium modeling as practiced by trade theorists. Moreover, too much neo-classical rationality is attached to the migration decision of poor and illiterate people who have negligible knowledge of the functioning of contingency markets, discounting and various other variables mentioned in Stark’s work.

In the previous chapter we presented a simple treatment of the Generalized Harris-Todaro (GHT) model. However, we did not provide a treatment for policy making in that framework. Recall that in the important papers of Bhagwati and Srinivasan ((1974)(1975)) on the HT model optimal policies were developed to reduce or eliminate unemployment and also to rank these policies in terms of welfare.¹ They established that the first best policy was a uniform wage subsidy in both sectors financed by lump-sum taxes. This policy removed the distortions and the economy arrived at a first best solution where DRS = DRT = FRT. The second best policies were a wage subsidy to manufacturing and a production subsidy to agriculture which resulted in the following condition: DRT ≠ FRT = DRS. These policies could not be ranked in terms of welfare due to the well known theories of the second best.² In all this analysis there is only one representative consumer whose marginal rate of substitution is given by DRS.

On the basis of the model presented in chapter two we now analyse the impact of temporary skilled out-migration on urban unemployment, the relative price of urban and rural non-traded goods, the real rewards of the factors of production, and welfare in each region in the source country. These migrant workers can be a large proportion of the labour force both in the source and the host country. For example, foreign workers from neighbouring Arab states and from Asian countries as a percentage of the total labour force constituted 44 per cent in Saudi Arabia to 89 per cent in the United Arab Emirates. Given the significance of this phenomenon it is important to study its economic effects. We shall confine ourselves to analysing the impact of emigration in the source country only.

So far it has been assumed that capital is not internationally mobile. In this chapter we relax this assumption. Recall that in the GHT model, the factors were assumed to be mobile in the following manner: no international mobility of factors; urban capital was assumed to be mobile within the urban region; similarly rural capital was only mobile in the rural region and labour was assumed to be totally mobile among all regions and sectors.

The GHT model can be adapted for a wide variety of applications and this chapter reinforces its flexibility by extending it to include a Free Trade Zone (FTZ) in the urban region. The FTZ sector produces a good that uses domestic, foreign capital, labour and an intermediate good. The urban region now consists of three sectors: an urban manufactured good, X _U ; an urban non-traded good X _NU ; and the FTZ sector X _E . For welfare purposes the disaggregation of national income into urban and rural is maintained. In this extended framework we examine the consequences of changes in final and intermediate good tariffs on the following important variables: regional welfare, structural adjustment, urban unemployment and the real exchange rate. Unlike previous contributions to this literature we do not argue a case for setting up a FTZ. The analysis concentrates on the effects of parametric changes on the above variables in the presence of a FTZ. The pioneering work of Hamada (1974) was concerned with the analysis of the various implications of establishing duty free zones in developing countries which were seen as an attempt at trade liberalization as they did not impose tariffs on final goods¹. Our emphasis is different from that of Hamada and others who followed the same tradition.

This chapter sets up a model of multilateral trade that involves both factor and commodity flows across countries’. This model is built on the basis of a customs union framework, which involves three countries and departs from the usual HO model which deals with the case of two-countries, two-commodities and two-factors. Our model has been inspired by the following scenario. Consider, for example, the three countries: the USA (home), Europe (foreign and developed) and India (third world). The USA and Europe produce three commodities, X, Y and Z while India produces only two goods X and Y reflecting the fact that India is less diversified than the other two economies. In view of globalization we shall assume that capital and skilled labour (of the USA and European origin) are fully mobile between these countries, but unskilled labour is immobile. Capital and labour do not generally flow freely from the USA and Europe to India although capital flows are now changing due to India’s economic reforms. However, for analytical convenience they will be ignored. The USA also receives skilled migrants on the basis of a quota from India (a typical immigration pattern between India and the USA). However, due to language, legal and cultural barriers such labour, in general, does not migrate to Europe.

In the previous chapter we examined the consequences of legal migration and its spillover effects in a model of multilateral trade. It was established that in models of multilateral trade with both factor and goods mobility, migration policies of one country affect important economic variables in other countries. Hence, in a globalized world, policy decisions cannot be made in isolation. This chapter changes the above perspective from two points of view. First, both legal and illegal migration are considered. Illegal migrants are defined as those who do not possess a valid visa in the country in which they are working or seeking work. Second, both types of migration are examined from the point of view of national welfare with no spillover effects.

In several chapters we have analysed the consequences of both internal and international migration of labour in the context of models of international trade. However, in all chapters, except parts of chapter eight, international migration has been treated as legal migration. As noted earlier, illegal migration is a worldwide phenomenon; present both in third world and advanced economies. For example, India receives illegal migrants from Nepal and Bangladesh. Americans receive both legal andlor illegal migrants from its neighbouring countries. Both types of migrants are of great concern to politicians and policy makers, since such migration has an impact on resident welfare¹.