tion is linked to trade, both because the volume of imports in many countries depends on the volume of remittances, and because trade in services often involves migration. A significant body of research suggests that trade can be a substitute for migration, which leads to the conclusion that free trade will reduce migration pressures. However, there is too little appreciation for the fact that trade and migration can also be complements, e.g., when technicians accompany specialized goods; when migrant workers create a demand for transportation, banking, and goods links to their countries of origin; or when construction firms include imported workers in their bids.
Trade, investment, and aid policies that seek to stimulate economic growth and reduce emigration pressures remind one why economics is sometimes called the dismal science: it is very hard both to help emigration countries and to leave no emigration- increasing side effects. Freer trade may increase imports before exports rise, producing a currency crisis, devaluation, recession, and emigration, as occurred in Mexico in 1995. Multinationals tend to use more imported components, so breaking up local monopolies and attracting direct foreign investment can increase imports, the use of capital-intensive production techniques, and exports without increasing the number of jobs. Finally, aid in the form of infrastructure can have the perverse effect of stimulating emigration; better roads meant to help farmers market their crops also permit cheap imported food to reach even remote areas of countries, thereby destroying jobs and stimulating emigration.
This sobering picture should not deter immigration countries from adopting, and persuading emigration countries to adopt, the economic policies that have the proven ability to speed up economic growth. Immigration countries should be comforted by how little--not how much--wage and job gaps must be narrowed to deter migration. Experience suggests that economically motivated migration practically ceases once wage gaps are narrowed to four or five and once there are signs of increased economic and job growth in the emigration area and widespread expectations of a continued narrowing of economic differences.