| The Institute for Economic Development at Boston University -------- ---------------------------Research Review Spring 2001 |
|
“Turnover
and Job Training in Developing and Developed Countries: Evidence from
Colombia and the United States” Recent research documents that job turnover rates are often much higher in developing countries than in developed countries. This observation suggests the hypothesis that long-lasting employment contracts, which may be required for the economical adoption of high productivity production and personnel practices, are more costly to implement in developing countries. Higher costs of long-term employment contracting may in turn help to explain why labor productivity (and GNP per capita) are so much lower in developing countries than in developed countries, even after controlling for differences across countries in physical capital per worker and in levels of formal education. But empirical evidence of a link between high turnover pressures and low productivity in developing countries relative to developed countries is thus far lacking. This paper takes a step in the direction of examining the link between turnover pressures and productivity by examining the relationship between turnover and the incidence and cost of an observable practice thought to increase labor productivity: job training. It employs high quality household survey data from Colombia and the United States to shed light on two questions. First, do the economic circumstances underlying higher job turnover rates also result in lower incidence of job training in developing countries? |
Second, do higher job turnover rates imply higher costs of keeping trained positions filled? Paying careful attention to matters of measurement, weighting, and conditioning, the paper documents that the incidence of post-school formal job training acquired for the current main job is in fact higher in Colombia than in the United States. The paper argues that if high turnover pressures tend to reduce the incidence of training in developing countries relative to developed countries generally, then evidence of the problem should have emerged in this two-country comparison. Thus, the results appear to rule out the simplest route through which higher turnover and training costs might reduce productivity. The paper goes on to document, however, that while male private sector wage employees in Colombia accumulate training experiences at a higher rate over the course of their working lives than their counterparts in the United States, their wages tend to rise more slowly as they age. This result appears robust to attempts to mitigate biases associated with omitted cohort effects and endogenous selection into private sector wage employment. The most compelling interpretation of these patterns is that higher turnover rates cause the value of past job training investments to be eroded more rapidly in Colombia than in the United States. If this is the case, then higher turnover increases the cost of keeping trained positions filled. Because the increased costs take the form of more labor hours (of trainees and trainers) diverted away from directly productive activities toward training activities, they represent reductions in average labor productivity. The paper thus provides empirical evidence of a link between higher turnover pressures and lower labor productivity in developing countries. |
Page 3
Table of Contents
IED Home page
Updated: 7/11/01