Villani, Samuel Wagner, and Frank Husic, The Prohctiontiii d Application of'Ne(it Industrial Technology (New York: W. W. Norton, 1977). Mansfield, Edwin, Anthony Romeo, Mark Schwartz, David Teece, Samuel Wagner, and Peter Brach, Technology Tr naiis fel, Productivity, aind Economic Policy (New York: W. W. Norton, forthcoming). National Science Foundation, Methodology of Statistics on Research a1nzd Development (Washington, D.C.: Government Printing Office, 1959). Resarchili(hid D)evelopment in Idiidstry, 1974 (Washington, D.C.: Government Printing Office, 1976). Scherer, F. M., Industrial Market Structure antd Economic Pe orrniance, 2nd ed. (Chicago: Rand McNally, 1980). Shepherd, William, The Ecoiomics of Industl-ial Orgonizlation (Englewood Cliffs: Prentice-Hall, 1979). Teece, David, and Henry Armour, Innovation and Divestiture in the U.S. Oil Industry,' in D. Teece (ed.), R (ind D in Eniergv (Stanford: Stanford University, 1977).

consumers. In this paper, we begin to remedy this omission by developing a formal model that shows how risk affects the decisions of riskaverse consumers regarding the amount of search and whether or not to buy advertised products. We incorporate risk into the model by assuming that consumers are concerned not only with expected product quality, but also with the amount of variation that exists across the qualities available. The risk arises not because the quality is itself variable, but rather because there is variation in quality across products, and consumers cannot evaluate the quality of a particular brand before purchase. We consider two sources of information about quality, both of which are costly to consumers: firms supply consumers with implicit quality signals through advertisements that convey only that a product is advertised but not direct information about the quality of the product, and consumers supply themselves with direct quality information through search.' These methods affect both the expected value of quality received and its riskiness. We analyze search and advertising simultaneously in order to model the consumers' substitution of one information source for the other. We then use the model to show how firms may use advertising as an implicit signal that advertised products are less risky than unadvertised products and to explain why information is primarily provided by firms in some markets, whereas in other markets, consumers acquire their information through search. In addition to explicitly considering risk, our model has several other attributes that distinguish it from the extant search literature. In order to obtain a meaningful market equilibrium, we construct a full partial-equilibrium model in which the behavior of both buyers and sellers is examined and the distribution of qualities in the market is endogenously determined. This approach is used

Two fundamentally different subcontracting systems arise as distinct solutions to the quality control problem facing an input buyer. The "American" system involves competitive bidding on each contract, large orders, and inspections. The "Japanese" system involves repeat purchases from a supplier who earns a premium, small orders, and no inspections. Both systems may coexist as local solutions, but the global optimum is determined by the ratio of set-up to inspection costs. This suggests that the adoption of flexible manufacturing equipment and rising product complexity may be responsible for the shift from the American to the Japanese system observed in many industries.

An empirical analysis of the impact of transaction costs on contracting shows that imperfect in formation can seriously limit the effectiveness of private contracting. The case under study is the widespread failure of private crude oil producing firms to use US oil fields to reduce rent dissipation. Rent dissipation follows as multiple firms compete for migratory oil in common oil pools. Unitization is the obvious private contractual solution to rent dissipation. The authors argue that, despite large net gains from unitization, imperfect information and information asymmetries among the negotiating parties regarding lease values prevents consensus on unit shares. As a result, contracts are often not completed or only fragmented units are completed. 9 references, 1 figure, 4 tables.

This paper examines bargaining among firms to mitigate rent dissipation following the major oil discoveries of 1926-35. Because of high bargaining costs, firms chose prorationing instead of consolidation and unitization, and success varied. The analysis also shows that prorationing took the form it did because concession, such as per well quotas, were required to draw in small operations and the quotas led to predictable responses regarding rent dissipation. Prorationing, despite its costs, controlled total field production and costs, conserved natural reservoir energies, and lengthened field life. When private agreements failed, the parties successfully appealed for state enforcement. Since similar heterogeneities influnce regulations elsewhere in the economy, detailed analysis of bargaining among firms is essential for insight into the emergence of various institutional forms. 33 references, 3 tables.

This paper analyzes the application of liability to large-scale, long-term hazards. The key features distinguishing such hazards are the long temporal separation between exposure to a hazard and disease and the large damages when injuries finally emerge. The large scale of damages creates a strong incentive to avoid liability payments, and the long temporal separation creates numerous avenues through which parties can avoid paying possible damage awards. The analysis focuses on the incentive to avoid paying damages by vertically divesting production tasks associated with serious occupational risks. Such divestiture can lower liability costs if the small firm operating the risky stage goes out of business before latent injuries emerge or has insufficient assets to pay damages and declares bankruptcy when suits are filed. The paper then presents an empirical regression analysis of small-firm entry into the U.S. economy between 1967 and 1980, the period in which liability laws were changing. The point estimate is that, ceteris paribus, liability changes appear to have led to a large increase in small corporations in hazardous sectors. Hence the empirical analysis shows widespread attempts to avoid liability by shielding assets through divestiture.

This paper analyzes the application of liability to large-scale, long-term hazards. The key features distinguishing such hazards are the long temporal separation between exposure to a hazard and disease and the large damages when injuries finally emerge. The large scale of damages creates a strong incentive to avoid liability payments, and the long temporal separation creates numerous avenues through which parties can avoid paying possible damage awards. The analysis focuses on the incentive to avoid paying damages by vertically divesting production tasks associated with serious occupational risks. Such divestiture can lower liability costs if the small firm operating the risky stage goes out of business before latent injuries emerge or has insufficient assets to pay damages and declares bankruptcy when suits are filed. The paper then presents an empirical regression analysis of small-firm entry into the U.S. economy between 1967 and 1980, the period in which liability laws were changing. The point estimate is that, ceteris paribus, liability changes appear to have led to a large increase in small corporations in hazardous sectors. Hence the empirical analysis shows widespread attempts to avoid liability by shielding assets through divestiture.