FIGURE 10.5 The effects of a fall in labor demand in the Shapiro-Stiglitz model

With regard to short-run fluctuations, consider the impact of a fall in labor demand, shown in Figure 10.5. w and L move down along the no-shirking locus. Since labor supply is perfectly inelastic, employment necessarily responds more than it would without imperfect monitoring. Thus the model suggests one possible reason that wages may respond less to demand-driven output fluctuations than they would if workers were always on their labor supply curves. Again, however, the model is sufficiently stylized that it is difficult to gauge its quantitative implications,

Finally, the model imphes that the decentrahzed equilibrium is inefficient. To see this, note that since the marginal product of labor at full employment, eF(eL/N), exceeds the cost to workers of supplying effort, ¸, the first-best allocation is for everyone to be employed and exert effort. Of course, the government cannot bring this about simply by dictating that firms move down the labor demand curve until full employment is reached: this pohcy causes workers to shirk, and thus results in zero output. But Shapiro and Stiglitz note that wage subsidies financed by lump-stmi taxes or profits taxes improve welfare. Such a pohcy shifts the labor demand curve up, and thus increases the wage and employment along the no-shirking locus. Since the value of the additional output exceeds the opportunity cost of

This discussion compares steady states with different levels of labor demand rather than analyzmg the dynamic effects of a temporary or permanent change in labor demand. See Kimball (1994) for an analysis of the dynamics of the Shapiro-Stiglitz model.

Extensions

The basic model can be extended in many ways. Here we discuss three.

First, an important question about the labor market is why, given that unemployment appears so harmful to workers, employers use layoffs rather than work-sharing arrangements when they reduce the amount of labor they use. One might expect workers to place sufficient value on reducing the risk of imemployment that they would accept a lower wage to work at a firm that used work-sharing rather than layoffs. Shapiro and Stiglitzs model (modified so that the number of hours employees work can vary) suggests a possible explanation for the puzzling infrequency of work-sharing. A reduction in hours of work lowers the surplus that employees are getting from their jobs. As a result, the wage that the firm has to pay to prevent shirking rises. If the firm lays off some workers, on the other hand, the remaining workers" surplus is unchanged, and so no increase in the wage is needed. Thus the firm may find layoffs preferable to work-sharing even though it subjects its workers to greater risk.

Second, Bulow and Summers (1986) extend the model to include a second type of job where effort can be monitored perfectly. These jobs could be piece-rate jobs where output is observable, for example. Since there is no asymmetric information in this sector, the jobs provide no surplus and are not rationed. Under plausible assumptions, the absence of surplus results in high turnover. The jobs with imperfect monitoring continue to pay more than the market-clearing wage. Thus marginal products in these jobs are higher, and workers, once they obtain such jobs, are reluctant to leave them. If the model is extended further to include groups of workers with different job attachments (different bs), a higher wage is needed to induce effort from workers with less job attachment. As a result, firms with jobs that require monitoring are reluctant to hire workers with low job attachment, and so these workers are disproportionately employed in the low-wage, high-turnover sector. These predictions concerning wage levels, turnover, and occupational segregation fit the stylized facts about primary and secondare jobs identified by Doeringer and Piore (1971) in their theory of dual labor markets.

The third extension is more problematic for the theory. So far, we ha\ e assumed that compensation takes the form of conventional wage payments But, as suggested in the general discussion of potential sources of efficieno wages, more comphcated compensation pohcies can dramatically change the effects of imperfect monitoring. Two examples of such compensatior policies are bonding and job selling. Bonding occurs when firms requir-each new worker to post a bond that must be forfeited if he or she is caugh

producing it, overall welfare rises. How the gain is divided between workers and hrms depends on how the wage subsidies are financed.

10.5 Implicit Contracts

The second departure from Walrasian assumptions that we consider in this chapter is the existence of long-term relationships between firms and workers. Firms do not hire workers afresh each period. Instead, many jobs involve long-term attachments and considerable firm-specific skills on the part of workers. Akerlof and Main (1981) and Hall (1982), for example, find that the average worker in the United States is in a job that will last about ten years.

The possibility of long-term relationships implies that the wage does not have to adjust to clear the labor market each period. Workers are content to stay in their current jobs as long as the income streams they expect to obtain are preferable to their outside opportunities; because of their long-term relationships with their employers, their current wages may be relatively unimportant to this comparison. This section and the next two explore the consequences of this observation. This section considers the case where the pool of workers dealing with the firm is fixed; Sections 10.6 and 10.7 investigate the effects of relaxing this assumption.

iSee Shapiro and Stiglitz (1985) and Akerlof and Katz (1989) for further discussion of these issues.

shirking. By requiring sufficiently large bonds, the firm can induce workers not to shirk even at the market-clearing wage; that is, they can shift the no-shirking locus down until iX coincides with the labor supply curve. If firms are able to require bonds they wiU do so, and unemployment will be eliminated from the model. Job selling occurs when firms require employees to pay a fee when they are hired. If firms are obtaining payments from new workers, their labor demand is higher for a given wage; thus the wage and employment rise as the economy moves up the no-shirking curve. Again, if firms are able to sell their jobs, they will do so.

Bonding, job selling, and the like maybe limited by an absence of perfect capital markets (so that it is difficult for workers to post large bonds, or to pay large fees when they are hired); they may also be limited by workers fears that the firm may falsely accuse them of shirking and claim the bonds, or dismiss them and keep the job fee. But, as Carmichael (1985) emphasizes, considerations like these will not eliminate these schemes entirely, if workers strictly prefer employment to unemployment, firms can raise their profits by, for example, charging marginally more for jobs. In such situations, jobs are not rationed, but go to those who are willing to pay the most for them; thus even if these schemes are limited by such factors as imperfect capital markets, they still eliminate unemployment. In short, the absence of job fees and performance bonds is a puzzle for the theory.