MEETING LABOR SHORTAGE WITH EXPATRIATE WORKERS: THE EFFECTS OF IMMIGRATION ON DOMESTIC WAGE AND EMPLOYMENT LEVELS IN THE PRIVATE NON-AGRICULTURAL SECTOR IN SAUDI ARABIA

Faced with an expected general labor shortage, Saudi Arabia allowed foreign workers to enter its labor markets or to be recruited by employers on a large scale since 1973. This led to a continuous inflow of labor migration, so much so that as early as 1980 there were more employed immigrants than natives. In this dissertation the Saudi policy to meet labor shortages with expatriate workers is evaluated using empirical analysis. Specifically, this study consists of: (1) a comparative study of other countries' policies on labor shortages and immigration and the implications for the Saudi case; (2) a critical assessment of the effects of immigration on domestic wages supplemented by an empirical analysis of the Saudi case; and (3) an empirical analysis of the elasticities of substitution between native and expatriate workers in Saudi production, in which we also consider the effects of immigration on the level of wages. First, our study of other countries' experiences with the use of foreign workers on a temporary basis clearly indicates that this can be a misleading strategy to counter labor shortages. The main reason is that several factors usually intervene and make the presence of foreign workers more of a permanent one. Most noticeable is the possibility that immigration is a cumulative and self-reinforcing phenomenon. The evidence from the Saudi case shows that it is not immune from this pitfall. Second, when we considered the effects of immigration on domestic wages, the evidence shows that, under almost any assumptions immigrants will reduce wage rates. The empirical analysis of the Saudi case shows that expatriate workers did reduce wage growth in the private sector. This effect seems to be, however, rather small and uneven among different labor groups. Our results show that had there been no immigration, wage growth could have been 3-10 percent higher than it was between 1968 and 1973 and 1-12 percent higher than it was betwen 1978 and 1982. Third, our study includes estimates of the elasticities of ...