Over the years there have been frequent changes in unemployment insurance (UI) benefits. Such changes could influence spells in progress as well as future spells. If individuals anticipate changes, they are likely to adjust their behavior. The effect of changes in benefits on unemployment duration, therefore, will be difficult to predict accurately without accounting for individual expectations. This article investigates the extent to which individuals anticipate changes in UI entitlement. The results suggest that individuals have significant, although not perfect, foresight about changes in UI provisions. Assuming perfect foresight might represent actual expectations more closely than assuming no foresight.

This paper attempts to provide microfoundations for the knowledge production function in an idea-based growth model. Production of new ideas is made a function of newly reconfigured old ideas in the spirit of the way an agricultural research station develops improved plant varieties by cross-pollinating existing plant varieties. The model shows how knowledge can build upon itself in a combinatoric feedback process that may have significant implications for economic growth. The paper's main theme is that the ultimate limits to growth lie not so much in our ability to generate new ideas as in our ability to process an abundance of potentially new ideas into usable form.

Research using U. S. manufacturing data finds that job destruction fluctuates more over time than job creation, but some new data indicate that this behavior is not shared in growing sectors, where job creation varies more. An explanation for this finding based on the interaction between (S,s)-type adjustment and trend employment growth delivers some tight predictions for the relationship between trend growth and the volatility of creation relative to destruction. Although it scores some notable successes, the simple (S,s)-based model augmented with a low-frequency trend cannot fully account for the strength of the empirical relationship between relative gross-flow volatility and trend growth across one-digit industries.

This paper develops a general approach to characterizing optimal income tax and enforcement schemes. Our analysis clarifies the nature of the interplay between tax rates, audit probabilities and penalties for misreporting. In particular, it is shown that for a variety of objective functions for the principal the optimal tax schedule is in general concave (at least weakly) and monotonic; the marginal tax rates determine the audit probabilities; and less harsh penalties lead to higher enforcement costs. Our results imply that there exists a tradeoff between equity and efficiency considerations in the enforcement context which is similar to that in the moral hazard context for tax policy.

This paper is about the economic theory of biodiversity preservation. A cost-effectiveness methodology is constructed, which results in a ranking criterion sufficiently operational to be useful in suggesting what to look at when determining actual conservation priorities. The formula is firmly rooted in a mathematically rigorous optimization framework, so that its theoretical underpinnings are clear. The underlying model, called the 'Noah's Ark Problem, ' is intended to be a kind of canonical form that hones down to its analytical essence the problem of best preserving diversity under a limited budget constraint.

The least-absolute-deviations (LAD) estimator for a median-regression model does not satisfy the standard conditions for obtaining asymptotic refinements through use of the bootstrap because the LAD objective function is not smooth. This paper overcomes this problem by smoothing the objective function. The smoothed estimator is asymptotically equivalent to the standard LAD estimator. With bootstrap critical values, the rejection probabilities of symmetrical t and X 2 tests based on the smoothed estimator are correct through O(n -γ ) under the null hypothesis, where γ<1 but can be arbitrarily close to 1. In contrast, first-order asymptotic approximations make errors of size O(n -γ ). These results also hold for symmetrical t and X 2 tests for censored median regression models.