Modern labour markets and the nature of the employment relationship
In fact, developments in theoretical labour market analysis since the mid- 1970s have provided several arguments that can be used to answer both the above questions and, at the same time, yield criteria for formulating hypotheses about the efficiency of different employment security regimes.
One major strand of theory production has been inspired by the empirical finding of prevailing long-term job attachments and delayed or incomplete employment adjustment to demand fluctuations and has proposed several arguments why both firms and workers may, indeed, share an (‘endogenous’) economic interest in open-ended, stable employment relationships extending over longer time periods. Central to most of these arguments has been the notion of market imperfections due to asset specificity and idiosyncratic exchange: that is, exchanges between workers and firms (as opposed to transactions in classical commodity markets) involve specific irreversible investments in relationship-specific capital (sunk costs) that generate economic rents for both sides (dual monopoly) as long as the exchange between the two parties continues, thereby generating a mutual interest in long-term relationships.6More specifically, most theoretical models in this tradition have focused on fixed employment costs (turnover costs) which have to be incurred at the beginning of the employment relationship (costs for screening job candidates, hiring, specific training and matching workers to jobs). Other models have emphasized workers’ higher risk aversion as compared to firms as a rationale for implicit contracts distributing income streams over longer periods. And even other theories, known as efficiency wage or effort regulation models, have focused on information asymmetries and on the costs involved in monitoring work effort in employment relationships and have provided arguments that seniority-based pay and deferred compensation schemes may be efficiency enhancing.
Internal labour market theories, finally, have combined several of the above explanations in order to account for the observed dualism of workforce adjustment inertia in some parts of the economy (primarily larger firms) and more rapid adjustments involving external labour turnover in others (for example, occupational labour markets, as prevailing in the traditional crafts sector, or secondary labour markets employing primarily casual labour). In this view,[Internal labour markets] develop as technological considerations require firmspecific investments and asymmetric information necessitates the monitoring of work effort: workers make sunk investments in training and monitoring by accepting deferred compensation. At the same time the firm makes sunk investments in shared training costs and in intangibles such as reputation in the labour market. (Cohen and Wachter, 1989, p. 245)
Taken together, these theoretical approaches provide strong arguments to explain why, even from a strictly microeconomic perspective, stable, longterm employment relationships are by no means incompatible with, but rather in many instances conducive to (if not even a necessary prerequisite for), economic efficiency and societal welfare maximization. Moreover, theory also shows that long-term employment relationships between workers and firms necessarily involve a high degree of uncertainty for both sides. First, the existence of information asymmetries implies the risk of opportunistic behaviour by either party: thus firms, in order to reduce the risk of ‘shirking’ by workers, will be inclined to design special compensation schemes rewarding work effort and, at the same time, will want to retain the right to dismiss workers whose work effort fails to match expected standards. Workers, on the other hand, in the absence of any binding rules or fairness standards, cannot, for instance, rely on their employers not dismissing them before they reach seniority and get compensated for past work effort, thus reducing workers’ willingness to accept efficiency-enhancing deferred compensation schemes in the first place.
Second, the parties share a lack of information and face uncertainty with regard to the future into which their relationship must continue if the returns to their initial shared investments are to be reaped: for example, the firm may be forced to reduce its workforce and to relinquish its promise of future compensation for past work effort in the event of an unforeseen major decline in the demand for its products. In the presence of information asymmetries and future uncertainty, the willingness of both parties to engage in long-term employment relationships involving sunk investments in relationship-specific capital, therefore, depends on whether they can devise and implement between them a governance structure that addresses the dual question of mutual trust and risk sharing.Mutual trust requires safeguards against opportunistic behaviour by the other side. This involves firms not being able to fire workers (or unilaterally alter contract terms) ‘at will’ or arbitrarily, thereby depriving workers of their share in the rents from mutual sunk investments, and workers not being able to lower the firm’s returns to such investments by withholding work effort (‘shirking’) without facing dismissal.
Risk sharing, on the other hand, requires enforceable rules that the parties, each within their relative capacity, equally share the potential losses resulting from unpredictable future events. This implies that firms cannot be forced to maintain the employment relationship if, owing to altered economic circumstances, this would involve a serious jeopardizing of the medium-term survival of the firm, but that they would have to compensate workers for unrewarded effort in the past.