On-the-Job Screening and Investments in General and Specific Training

Abstract

In recent years there has been wide interest in the role of education as a screening device. Spence [4], Stiglitz [6] and others have argued that education requirements may play two types of roles in the screening process. First, education may screen workers directly by ability. Second, even if education has no direct screening effect, education requirements may lead workers to screen themselves if they believe that they will be unable to recoup investments in education unless they meet some set of on-the-job screening standards.' In effect, in this case, investments in education may serve as a bond which workers foreit if they fail to meet screening criteria after they are hired. In the second type of model, the actual role of education in the screening process will depend heavily on the degree to which training can be used in other jobs and who bears the cost of this training. Nevertheless, there has been little attempt in screening models to introduce the usual distinction in the human capital models between general and specific training. In the human capital literature, general training is defined as training which is transferable between firms. Specific training is defined as traning which can only be used in a job in a specific firm and is non-transferable. Becker [1] and others argue that workers will always bear the costs of general training because employers cannot prevent workers from being pirated by other firms once training is completed. But in the case of specific training, they suggest that workers and firms may share training costs. Workers will be hesitant to invest in specific training because, being non-transferable, it cannot be used elsewhere and gives firms monopsony power over them. But at the same time, even though firms need not worry about piracy in the case of specific training, workers may still leave firms because of normal turnover. This may create an incentive for firms to encourage workers to invest in at least part of the specific training required in their jobs; once workers have made this type of investment, they are less likely to move. This paper seeks to link discussions of general and specific training in the human capital literature with models of on-the-job screening in which education may serve as an entry re-