RRpage6-dp102

“Job Stability in Developing and Developed Countries: Evidence from Colombia and the United States”

Julie Anderson Schaffner

IED Discussion Paper 102,November1999

While a large empirical literature describes the determination of wages in developing country labor markets, very little is known about other dimensions of employment contracts, especially in urban labor markets. Richer information about employment contracts in these markets would seem especially valuable, given the likely importance of contract design for facilitating training and other modern employment practices, and for improving work incentives and labor productivity more generally. This paper begins to fill this void. It presents a systematic comparison of job stability for a developed and a developing country, first documenting that jobs are much shorter in the developing country, and then studying why long-lasting jobs are less prevalent in the developing country.

Schaffner compares both cross-section distributions of current job tenure and estimates of job retention probabilities between Colombia and the United States, and within Colombia before and after the major liberalization of job security legislation in 1991. She demonstrates that the typical male worker in Colombia has been on the job for a substantially shorter period than his American counterpart,

and the typical job is much shorter in Colombia. A number of possible alternate explanations for these phenomena are shown unlikely to be relevant. For instance, current job tenures may appear shorter in Colombia owing to higher employment growth rates rather then differential job retention probabilities. Using two alternate methods to control for employment growth rates turns out to not change the basic conclusion. Similarly, cross-country differences in job length remain significant even after considering the effect of possibly counter-productive job security legislation prevalent in Colombia in the eighties, and/or differences in industrial structure (e.g., the possibility that Colombia specializes more in production activities where long-lasting employment relationships are less important).

The evidence in the paper is consistent with the hypothesis that employers face greater costs of implementing long-term employment contracts or the work organization practices that such contracts necessitate. This may explain why developing countries often exhibit higher rates of self-employment or employment in small enterprises, and lower labor productivity. As the author emphasizes, the evidence presented in this paper may motivate enrichment of theoretical models of cross-country growth rate differences by incorporating microeconomic and institutional obstacles to long-term growth and development.