Knowledge that Transforms

This paper studies the asymptotic properties of instrumental variable (IV) estimates of multivariate cointegrating regressions and allows for deterministic and stochastic regressors as well as quite general deterministic processes in the data-generating mechanism. It is found that IV regressions are consistent even when the instruments are stochastically independent of the regressors. This phenomenon, which contrasts with traditional theory for stationary time series, is a beneficial artifact of spurious regression theory whereby stochastic trends in the instruments ensure their relevance asymptotically. Problems of inference are also addressed and some promising new theoretical results are reported. These involve a class of Wald tests which are modified by semiparametric corrections for serial correlation and for endogeneity. The resulting test statistics which we term fully-modified Wald tests have limiting X2 distributions, thereby removing the obstacles to inference in cointegrated systems that were presented by the nuisance parameter dependencies in earlier work. Some simulation results are reported which seek to explore the sampling behaviour of our suggested procedures. These simulations compare our fully modified (semiparametric) methods with the parametric error-correction methodology that has been extensively used in recent empirical research and with conventional least squares regression. Both the fully-modified and errorcorrection methods work well in finite samples and the sampling performance of each procedure confirms the relevance of asymptotic distribution theory, as distinct from super-consistency results, in discriminating between statistical methods.

This paper proposes a model of firms' optimal employment policies under linear adjustment costs. We find that firing costs have a larger effect on firms' propensity to fire than to hire, and (slightly) increase average long-run employment. Calibrating the model with realistic parameter values, we argue that high firing costs, slower and more uncertain growth, and lower attrition rates after the first oil shock can explain some features of employment's dynamic behaviour in the largest European countries.

A given set of agents jointly own and operate a decreasing returns to scale technology (with a single input and a single output). They all contribute some input and receive some of the output. What first best allocation is equitable? We discuss three allocation mechanisms. The Equal Benefits solution gives every agent the same benefit, computed at the supporting price. The Equal Returns solution equalizes returns (output shares/input contributions) across agents. The Constant Returns Equivalent solution gives every agent his indirect utility level at some common price of output relative to input. A lower and an upper bound on individual welfares play a key role in these axiomatic characterization results.

It is useful to have bounds on expected utility in situations in which the relevant distribution is skewed. This paper assumes that the third moment of the distribution is known and uses knowledge of the first three moments to derive upper and lower Tchebycheff bounds.

This paper discusses why redundant skilled workers may be reluctant to accept interim unskilled jobs. If skilled work is more satisfying or less arduous for highly productive workers, then such workers invest more in moving quickly between skilled jobs. Thus, high productivity workers tend to search on-the-job, and if unemployed will specialise in job search, rather than take an interim position. If individual differences in productivity are known to the worker but not the potential employer, then search strategy may be used as a productivity signal, with more than the efficient proportion of workers searching on-the-job and too few accepting interim unskilled jobs. Optimal policy requires a subsidy on interim unskilled jobs.

This paper describes a simple framework for evaluating the allocative performance of economies characterized by trading frictions and unemployment. This framework integrates the normative results of earlier Diamond-Mortensen-Pissarides bilateral matching-bargaining models of trade coordination and price-setting, and consists of a set of general conditions for constrained Pareto efficient resource allocation that are applicable to conventional natural rate models. To illustrate, several conventional models of the labour market are reformulated as matching-bargaining problems and analyzed using this framework.

The failure of the asset market to be complete causes serial dependence in output and prices, which is suboptimal. We consider an economy with white noise shocks. When the asset market is complete, an optimal, competitive allocation inherits this strong stationarity. When the asset market is only sequentially complete, prices and output necessarily display serial dependence at equilibrium. The further incompleteness of a monetary economy explains co-movements in real and nominal variables.

This paper investigates the relationship between uncertainty and delay of agreement in the one-sided offer bargaining model with two-sided uncertainty where the seller makes offers. We construct a weak stationary equilibrium in which different types of the seller charge different prices in every period. We completely characterize the separating equilibrium by three regularity conditions, and show that the time interval between offers converges to zero, the seller's initial price offer in a separating equilibrium converges with probability 1 to the lowest valuation of the buyer if and only if the gain from trading is common knowledge.

In the mean-variance CAPM without a riskless asset, the possibility of satiation sometimes leads to non-existence of general equilibrium. Moreover, because portfolio preferences are not necessarily monotone, equilibrium asset prices, when they exist, may be negative or zero. To demonstrate the possibility of non-existence, and to develop an intuitive understanding of when and why equilibrium does or does not exist, this paper fully investigates the special case of utility functions linear in mean and variance and partially extends the results to the general case.

This article offers a new method of solution for linear difference equations with Rational Expectations. We provide a description of the complete set of solutions which is shown to depend on arbitrary martingales. We thus avoid the use of "differences of martingales" as introduced by Gourieroux, Broze, Szafarz, while describing the general solution as a dynamic equation with lower order. At the same time we provide a simple algorithm, based on polynomial divisions, to calculate some basic solutions, namely all ARMA solutions.