تعهد، تبلیغات و بهره وری سرمایه گذاری دو طرفه در تعادل جستجوی رقابتی

Competitive search entails both commitment to and advertising of pay-off relevant aspects of market participants. This paper considers incrementally the implications of each in a labor market where both workers and firms invest prior to market entry. A wide range of institutional arrangements are addressed within the same general framework. When the characteristics of jobs or workers are advertised the efficient outcome pertains. Commitment without advertising typically leads to market unravelling: the Diamond paradox. But, whenever wages and human capital are advertised, firms become residual claimants; the private and social returns to investment coincide. Absent wage commitment, the Hosios condition implies efficiency when investments are advertised.

This paper looks at how labor market matching frictions affect incentives to invest in the quality of jobs and in human capital. It has two main goals. The first is to examine the extent to which the established efficiency results from competitive search extend to an environment with investment decisions on both sides of the market. The second follows from the recognition that competitive search requires that market participants can both commit to characteristics of the employment relationship, and make that commitment public (through advertising)—the paper examines incrementally the roles of commitment and advertising in shaping outcomes in this extended environment. For context, consider a labor market with matching frictions and endogenous vacancy creation, but no investment decisions. Moen (1997) showed that competitive search equilibrium generates efficient vacancy creation. Now, if the workers' ability to observe individual firms' wages (which I will call advertising) is removed, Diamond (1971) implies that the equilibrium terms of trade assign the entire match surplus to firms. Workers facing any positive cost of entry will choose autarky. Beyond this, we can also remove the ability to commit to the terms of trade. If the wage is then determined ex post by generalized Nash bargaining, the environment becomes that of Pissarides (2000). As long as both sides get some share of the match surplus, for sufficiently small cost of entry, equilibria other than autarky exist. And, we know that efficient vacancy creation transpires under the Hosios (1990) condition (that equates the elasticity of the matching function with respect to vacancies to the firm's share of the match surplus). The objective in the current paper, therefore, is to examine the extent to which these results pass through to a more general environment with two-sided investment.

This paper provides an assessment of the efficiency properties of a general class of search models with two-sided investment. When characteristics of the job and worker are knowable (i.e. either observed directly or, given the environment, can be ascertained from what is directly observable) we find that the efficiency result of Moen (1997) passes through to the more general environment. When neither side can commit to a wage so that the terms of trade are determined by Nash bargaining the Hosios condition ensures efficient vacancy formation but its implications for investment depend on the extent to which firms and workers advertise their investments. When investments are private information to the investor (only revealed in one-on-one meetings) the classical hold-up problem leads to under investment. However, if investments become public knowledge, then competitive search disciplines investors into making the right choices. A general theme that emerges from this analysis is that while competitive search leads to desirable outcomes it requires strong informational assumptions on the environment. Specifically, for the equilibrium condition associated with some job or worker characteristic to coincide with that of the optimality condition of Social Planner requires both commitment and advertising. When it is reasonable to make these assumptions depends on the specific context. Furthermore, we have seen that if advertising does not convey sufficient information, very bad outcomes are possible. For instance, when investments are hidden but wage commitments are advertised, autarky is the only possible outcome. In this case, ex post bargaining mitigates the severity of the moral hazard problem by providing a fixed share of the match surplus to investors. Thus, a job posting agency that requires pay-off relevant aspects of vacancies be made public and accurate would achieve efficient matching and investment. But in the absence of any enforcement of truth-in-advertising we might be better off with random matching.