Monthly Archives: July 2014
The closest paper in the literature is the recent one by Cukierman and Lippi (1998), which was written simultaneously with this one. They also consider a monetary policy game with many unions, and study the interaction among labor market centralization, and economic performance. The main difference is that they work with an ad-hoc model with the crucial assumption that the elasticity of substitution among the labor supplied by different unions is always increasing in the number of unions, and goes to infinity as the number of unions goes to infinity.
By contrast, we work with a micro-founded model which yields very different implications for the relationship between the number of unions and the elasticity of labor demand. These modelling differences matter a great deal in terms of results. Cukierman and Lippi (1998) reproduce the conventional wisdom that economic performance is U-shaped in the number of unions, and therefore an intermediate degree labor market centralization is worst. We find that, depending on parameter values, the opposite results obtain: economic performance is either always decreasing or U-shaped in the number of unions; in the latter case, an intermediate degree labor market centralization is best.
More specifically, the model below has the following implications for the relationship among CBC, CWS and macroeconomic performance.
The conventional wisdom that discretionary policymaking yields an inflation bias while leaving employment and output at suboptimal levels, relies on two special assumptions:
a) unions are myopic (they do not internalize the consequences of their actions);
b) unions suffer no costs from inflation.
For a flavor of the basic arguments (more detailed results follow), consider a standard monetary policy game in the spirit of Barro and Gordon (1983), but with two twists:
a) Wage setters are not atomistic; instead, they are organized in unions. These unions set nominal wages on behalf of their membership.
b) Unions dislike inflation, unlike the standard story in which they care only about the expected real wage.
In this context, and for an exogenously fixed price level, it should be clear that the outcome of the game among the unions should be Pareto inefficient, with a real wage that is too high and employment and output that are too low relative to the first best.
Two conjectures have become part of the conventional wisdom.
The first is that greater central bank conservatism (CBC, defined as a greater weight placed by the central bank on an inflation as opposed to an employment objective) reduces average inflation rates while leaving average real activity unaffected.1 In particular, greater CBC enables countries to overcome the inflation bias first stressed by Kydland and Prescott (1977) and Barro and Gordon (1983). Relevant work is also provided by Schaling (1995). Evidence for the effect of CBC on inflation is presented by Grilli, Masciandaro and Tabellini (1991), among others. Alesina and Summers (1993) provide some evidence that CBC has no impact on real activity; Hall and Franzese (1996) are less certain. The issue awaits more detailed statistical analysis.
Also, an assessment of an offer-of-settlement rule should take into account also the rule’s effect on the level of litigation costs in the event of trial. In our model, we took this level to be given exogenously. As is true for any fee-shifting rule, however, an offer-of-settlement rule would increase the litigation costs in trial, because, in the event of trial, each party would know that, with some probability, it would not have to bear fully its litigation costs.
Finally, even focusing only on the terms of settlement, it should be noted that our model has focused on only one reason — namely, asymmetric litigation costs — for settlement terms to deviate from the expected judgement. In particular, we have assumed that all litigants are risk-neutral, cannot credibly commit to a strategy that insists on a disproportionate share of the gains from settlement, and have an equal ability to make settlement offers. Relaxing these assumptions would introduce other sources for divergence between the terms of settlement and the expected judgment — such as differences between the parties in their ability to bear risk or to commit to a certain bargaining position. The presence of these other sources of divergence suggests that much analysis must be done before we know which rules and institutions would best align settlement outcomes with the expected judgment. We hope that future work will pursue this agenda for research.
As we have shown, one way to get settlements equal to the expected judgment is to require one of the parties to make a special offer with two-sided cost-shifting. As long as the cost-shifting includes all litigation costs, and is two-sided, this rule would produce an expected settlement equal to the expected judgment, whether the rule requires the plaintiff or the defendant to mate the offer. An alternative that yields the same outcome is to give each party an option to mate a special offer. Note that Rule 68 as it stands currently is quite different from the any of the rules that, under the identified circumstances, would ensure settlements that mimic the expected judgment. Under the existing Rule 68, (i) only the defendant may mate a special offer, yet this offer is optional, (ii) the cost-shifting is one-sided, and (iii) the cost-shifting is only partial and does not include attorneys’ fees.
While the potential use of offer-of-settlement rules to move settlement terms closer to the expected judgment is an interesting possibility, we wish to emphasize that we are at this stage far from being in a position to mate any recommendations concerning such use of these rules. To start with, thus far we have shown how to ensure settlement terms equal to the expected judgment only in the situation in which the judgment is equally likely to be higher or lower than its expected value. Disputes over damages alone may come close to this situation, but cases that include a dispute over liability are unlikely to do so. To produce settlements equal to expected trial outcomes in cases of disputed liability, offer-of-settlement rules would have to be more complex; we leave this subject to future research.
Our analysis has normative implications. The main effect of the law on outcomes is through its influence on settlement terms. Therefore, in designing an offer-of-settlement rule we should take into account the rule’s potential effect on the terms of settlement. The model developed in this paper enables us to identify this effect.
One interesting implication of the analysis is the possibility of designing rules to eliminate or reduce the divergence between settlement terms and the expected judgment. While settlement terms are always chosen in anticipation of the expected judgment at trial, they might well diverge from it rather than mimic it. To be sure, the legal system, recognizing that settlement outcomes might diverge from the expected judgment, might set the expected judgment not at the level of the desired outcome but rather at the level that would result in settlement terms close to the desired outcome (see Bebchuk (1997)). In many contexts, however, it is reasonable to assume that the legal system has set expected judgments at the level that is equal to the desired outcomes. In such cases, it would be desirable to have settlement terms mimic the expected outcome of the trial — that is, to eliminate the divergence between these terms and the expected judgment.
Equation (8) reveals why Rule 68 currently has little practical significance and why defendants in many cases do not take advantage of the opportunity to make special offers. The only difference that the existing Rule 68 mates is the introduction of the cost-shifting term, 0CdPr(O < D < S*), in the equation for the equilibrium settlement amount. If 0 is small, as it is likely to be under the existing rule, this term will be small even if the optimal special offer implies that Pr(0 < D < S*) is positive.