【正文】
ough a broker. But our results also imply that it is not because these buyers have used a broker. Rather, this group of buyers would have paid a higher price regardless of the means by which the purchase was consummated. These buyers have higher ines, are more likely to be from out of town, are more likely to have employer assistance in the purchaseall factors that lead them to pay more for a house, but also to make them more likely to use a broker in purchasing their home. When the decision to use a broker is accounted for, these buyers do not wind up paying more for their homes, and there is some evidence that they actually pay less than a parable buyer who buys without an agent’s assistance. The layout of this study is as follows. The relevant literature is reviewed in section section 3 the data, variable selection, and model are described. Section 4 presents the empirical results, and the last section contains the conclusions of the study. Literature review A broad range of research has focused on the determinants of housing prices. These include traditional estimates of housing demand hedonic modeling of housing prices, the determinants of the tradeoff between price and time on the market, and models of the search process. Other literature has examined the role of real estate brokers and has focused on the impact of brokers on buyer search, or has examined the welfare and/or moralhazard implications of broker intermediation. Few studies have tried to measure the direct effects of real estate brokers on the housing market, and fewer still on home selling prices. The work that has been produced to date has often generated conflicting results. In part, these disparate results may reflect data availability problems, as virtually all of these earlier studies have been based on local data, and samples tended to be small, making generalizations difficult. 3 In one of the first empirical studies of broker choice, Janssen and Jobson ( 1980) find that real estate agents do have an impact on price. Using data from the Canadian housing market, Janssen and Jobson results indicate that with real estate firms of parable size, brokers who list parable properties for higher prices than peting brokers tend to realize significantly higher selling prices. The higher selling prices tend to be associated with transactions involving executive transfers and brokerarranged secondary financing. These results may in turn indicate that brokers obtain higher prices when dealing with buyers who are both less knowledgeable about local market conditions and less sensitive to price. Yavas and Colwell (1994) suggest that selling price may also be, at least to some degree, a function of the type of broker listing arrangement used by the seller. In a study of the residential market Jud (1983) estimates the demand for real estate brokerage services. Using housing transactions data from three urban areas in North Carolina, Jud finds that brokers do not affect the prices of the houses which they sell, although they do appear to influence the level of housing consumed by buyers. In a subsequent study Jud and Frew (1986), using different data, find that brokers do obtain higher prices for the homes they sell. Evidence is also presented that brokerassisted buyers have a greater demand for houses than their nonbrokerassisted counterparts. Their results lead them to conclude