This article examines the impact of unemployment insurance (UI) on the allocation of labor across industries. An overlooked aspect of UI is the effect of imperfect experience rating on hiring. Firms in more stable industries generally pay more into the UI system than their workers ever receive in benefits, thus subsidizing more volatile industries. The results indicate that industry employment shares are significantly affected by UI and that there is a net shift of resources from the service industry to the construction industry. The estimates also imply that layoff unemployment is increased by about 5% because of UI-induced employment shifts.

This paper analyzes the implications for turnover of costly job-specific training. The presence of such costs in a search model implies that turnover decisions reduce the value of potential trades that are available to other market participants. There is too much turnover because of this external effect, and, therefore, too much retraining. When the investment in job training is endogenous, inefficient turnover again occurs, and the investment in specific skills is inefficiently high. The interactions between skill acquisition and turnover imply that it is essentially impossible for a brokerage institution to achieve efficiency.

A market composed of pairwise trading under incomplete information is modeled in order to analyze how resources are allocated among competing uses when information about trade gains is incomplete. Contrary to the results from studying a single such trade, sufficient homogeneity across potential trades guarantees that efficiency obtains. This is analogous to simple first-price auctions with homogeneous bidders, where bidders have a common bid function and, as a result, the high bidder also places the highest value on the auctioned object. With enough symmetry, the decentralized bilateral trades in the present model occur as if they were made in a first-price auction that occurs through time. The robustness of the efficiency result to heterogeneities among agents and to nontrivial search intensity decisions is then considered.

[A market composed of pairwise trading under incomplete information is modeled in order to analyze how resources are allocated among competing uses when information about trade gains is incomplete. Contrary to the results from studying a single such trade, sufficient homogeneity across potential trades guarantees that efficiency obtains. This is analogous to simple first-price auctions with homogeneous bidders, where bidders have a common bid function and, as a result, the high bidder also places the highest value on the auctioned object. With enough symmetry, the decentralized bilateral trades in the present model occur as if they were made in a first-price auction that occurs through time. The robustness of the efficiency result to heterogeneities among agents and to nontrivial search intensity decisions is then considered.]

This paper analyzes the relationship between changes in unionization and firm growth. Average growth is significantly lower in manufacturing firms that experience successful union elections, but these strong "effects" are largely illusory. We find no evidence of a significant relationship between unionization and firm growth, despite a strong cyclical pattern in election activity. Our results suggest that the significant negative effect of organizing activity on a firm's market value is not accompanied by any growth changes. We therefore cannot reject the hypothesis that the equity losses from union election activity represent a simple transfer of wealth from shareholders to workers.

This paper analyzes the relationship between changes in unionization and firm growth. Average growth is significantly low er in manufacturing firms that experience successful union elections bu t these strong "effects" are largely illusory. The authors find no evidence of a significant relationship between unionization and firm growth, despite a strong cyclical pattern in election activity. Thei r results suggest that the significant negative effect of organizing activity on a firm's market value is not accompanied by any growth changes. The authors, therefore, cannot reject the hypothesis that t he equity losses from union election activity represent a simple transf er of wealth from shareholders to workers.