On Employer-Specific Information and Internal Labor Markets

Abstract

The human capital literature, of which Becker's [2] paper is seminal work, stresses productivity enhancing effects of schooling and on-the-job training. Later work by Spence [21] and others emphasizes fact that it is difficult for employers to ascertain productivity of potential employees and suggests that activities such as schooling may provide employers with signals as to workers' productivities. In signalling literature, it is assumed that information generated concerning an individual's abilities is general in that it is known to all potential employers. Yet information that an employer gathers concerning his workers' abilities is not likely to be completely transferable to other employers. As a consequence, as Rosen [18,249] points out, there are limitations on ability of price system ... to achieve efficient assignments. If price system is incomplete there is a role for entrepreneurial activity of assembling an optimal work force. This paper analyzes employment relations that arise when employers' information concerning new workers' abilities is incomplete until after a period of employment. The acquisition of such employer-specific information is shown to lead to what has been referred to in literature as labor Internal labor markets, as identified by Dunlop [5,32], are the complex of rules which determine movement of workers among classifications within administrative units ... these movements may be transfers, promotions, demotions, or layoffs to exterior labor Doeringer and Piore [3] emphasize skill specificity and on-the-job training as key factors in emergence of internal labor market. Along similar lines, Williamson, Wachter, and Harris [22] point out that job idiosyncracy can result in employment relations with attributes like those of internal labor markets. By way of contrast, this paper emphasizes role of employer-specific information in generating internal labor markets. The theory provides an explanation as to why employers dismiss current employees at same time that new employees are hired. (Evidence that dismissal rates are not insignificant is presented in section IV.) The theory also provides a new rationale for long-term em-