When buyers' private information about the value of an item for sale is correlated, the seller can increase expected revenue in a sealed bid auction by making the winner's payment a function of information available after the end of the auction. Specifically, revenue can be increased by making the payment a function of all the losing bids. In addition there are gains to making the payment contingent upon some signal of the object's value which becomes public at a later date. That is, there are gains to introducing positive royalty rates.

In this paper we show that, under a variety of alternative assumptions about the private information of loan applicants, a competitive market for loans is characterized by screening. Banks separate out loan risks by offering higher loans at higher interest rates. Depending on the nature of the informational asymmetry, it may be that applicants with less risky projects select larger rather than smaller loans. Comparative statics implications are also examined. In particular, we explore the effects of an increase in banks' cost of funds on average loan quality.