In the Lucas-Rapping (1969) model of the labour market, fluctuations in unemployment represent individuals optimally adjusting their labour supply behaviour in response to fluctuations in wage rates over the business cycle. In this paper I propose and implement a misspecification test of the Lucas-Rapping treatment of unemployment as labour supply behaviour using panel data. This test extends previous such work with micro data by simultaneously allowing for intertemporal substitution, uncertainty and endogenous unemployment. Using the standard specification of intertemporal labour supply behaviour, I find strong evidence against this interpretation of unemployment. There are two possible interpretations of the test results. The first is that it is necessary to turn to alternative models of the labour market in which unemployed workers are off a supply function. The second is that the test results indicate the necessity of moving to more complex models of intertemporal substitution. However, given current econometric techniques and data sets, these alternative models of intertemporal substitution will be extremely difficult to test.

This paper argues that adverse selection in the labour market, when viewed as part of a three-way interaction among workers, their current employers and a universe of alternative employers, may seriously impair a worker's freedom to change jobs. When current employers are better informed about the abilities of their workers than potential alternative employers, they will presumably concentrate their efforts to prevent turnover on their better workers. If these efforts lead to fewer quits among better workers, the stream of job changers should be composed disproportionately of less able ones. This will inhibit turnover in two ways. First, firms should be unwilling to hire from the job-changing pool except at low wages. Second, workers who change jobs are marked by being part of an inferior group, which lowers their future bargaining power and wages. Models of these phenomena can be made to account for many aspects of observed labour market behaviour.

In this note we show that the conditions that guarantee uniqueness and optimality of Walrasian equilibrium under uncertainty are not sufficient to guarantee these results for non-noisy rational expectations equilibrium under asymmetric information.

This paper develops a general asymptotic theory of regression for processes which are integrated of order one. The theory includes vector autoregressions and multivariate regressions amongst integrated processes that are driven by innovation sequences which allow for a wide class of weak dependence and heterogeneity. The models studied cover cointegrated systems such as those advanced recently by Granger and Engle and quite general linear simultaneous equations systems with contemporaneous regressor error correlation and serially correlated errors. Problems of statistical testing in vector autoregressions and multivariate regressions with integrated processes are also studied. It is shown that the asympotic theory for conventional tests involves major departures from classical theory and raises new and important issues of the presence of nuisance parameters in the limiting distribution theory.

Methods for exact Bayesian inference under a uniform diffuse prior are set forth for the continuous time homogeneous Markov chain model. It is shown how the exact posterior distribution of any function of interest may be computed using Monte Carlo integration. The solution handles the problems of embeddability in a very natural way, and provides (to our knowledge) the only solution that systematically takes this problem into account. The methods are illustrated using several sets of data.