Did Public Wage Premiums Fuel Agglomeration in Ldcs?
Univ. of Southampton Dept. of Economics Working Paper No. 20
46 Pages Posted: 3 Nov 2000
Date Written: October 2000
We build and test a model of how the growth of public jobs with wage premiums may help to explain the high and potentially inefficient level of urbanization in LDCs. Public jobs comprise about 40% of non-agricultural employment in LDCs, and have frequently offered substantial wage premiums. The Harris-Todaro model - and its extensions - suggest that wage premiums induce inefficient agglomeration, but that model critically assumes that wage premium jobs are allocated to favor local residents. This is inapplicable to public appointments in various LDCs. In the two-region general equilibrium model discussed here, the existence of spatial mobility costs are shown to be sufficient for wage premiums to result in inefficient agglomeration in regions that are allocated wage premium jobs. This weakens the assumptions under which wage premiums promote agglomeration, and extends the idea to LDCs such as Egypt, Ethiopia, and Kenya, where public jobs have, until recent reforms, offered substantial wage premiums, but are not allocated so as to favor local residents. The policy implications of this model also differ from Harris-Todaro. For example, if wage premiums are later reduced, the agglomeration persists: with mobility costs, the history of the location of jobs with wage premiums matters. We explore our hypothesis using Egyptian data. Between 1960 and 1986 the share of public jobs increased from 10% to 34% of the labor force, public jobs were centrally allocated, and offered a high total compensation premium. We find that public jobs' growth has substantially altered the pattern of regional mobility and population shares, in a way that is consistent with this theory of agglomeration due to wage premiums and mobility friction.
Keywords: Public Sector, Agglomeration, Migration, Developing Country
JEL Classification: J61, J68, J60, J45, H11, H40
Suggested Citation: Suggested Citation