ELSEVIER
Economics Letters 54 (1997) 277-282
economics
letters
Has the wage curve nullified the Harris-Todaro model?
Further US evidence
Mark D. Partridge a'*, Dan S. Rickman b
"Department of Economics, St. Cloud State University, St. Cloud, MN 56301-4498, USA
t'Oklahoma State Universi~.. OK, USA
Received 28 June 1996; received in revised tbrm 19 December 1996; accepted 22 January 1997
Abstract
Economists traditionally believed that regional unemployment rates and wages were positively related. This notion has
recently been attacked by proponents of the wage curve, which suggests the exact opposite. This paper attempts to replicate
the wage curve using different US data. © 1997 Elsevier Science S.A. © 1997 Elsevier Science S.A.
Keywords: Wage levels; Regional unemployment; Migration
JEL classification: J31" J64; R23
I. Introduction
Traditionally, regional and labor economists generally believed that a positive relationship existed
between regional wages and regional unemployment. This idea dales to Adam Smith (1776) and was
mathematically formalized by Harris and Todaro (1970). Supporting empirical evidence |br this
relationship was found in many studies including Hall (1972), Marston (1985), and Topel (1986). The
intuition for the Harris-Todaro (HT) result is simple. In a multi-regional labor market, equilibrium is
attained when expected utility is equalized across all regions and net migration between regions equals
zero. Hence, regions that persistently have higher unemployment rates require higher wages to act as a
compensating differential. Among other things, the HT model has become an important element in
regional and labor economic analysis of migration flows (e.g. Greenwood et al., 1991).
Recently, the HT compensating differential model has come under attack in several studies, most
notably those by Blanchflower and Oswald (BO). These studies are clearly summarized in their book
(Blanchflower and Oswald, 1994). Instead of a positive relationship, B© argue that wages are in fact
negatively related to (contemporaneous) unemployment. B© generally prefer an efficiency wage
interpretation for this result.
B© econometrically estimated their wage curve for numerous countries. They claim that the
*Corresponding author. Tel.: (320) 255-2072; fax: (320) 255-2228; e-mail: mpanridge@tigger.stcloud.msus,edu.
0165-1765/97/$17.00 © 1997 Elsevier Science S.A. All rights reserved
Pll S0165-1765(97)00028- i