This award funds work that develops a new data analysis technique for an important class of markets. The new technique will be used to analyze very detailed data about consumer mortgages.

In many markets prices are determined through a negotiation process between buyers and sellers. Sellers post or announce a price, but consumers may be able to negotiate a discount. The extent to which sellers negotiate may depend on customer characteristics. Despite the prevalence of discounting, existing empirical techniques for analyzing market data ignore discounting in differentiated products markets. This research project develops and applies a new empirical technique that allows for discounting.

The technique is used to analyze very detailed data on a large set of approved mortgages issued in Canada between 1992 and 2004. These data provide extremely valuable information on features of the mortgage as well as household and market characteristics. Mortgages are a good case study for the new method because many consumers negotiate a discount from the price/interest rate 'posted' by a neighborhood financial institution.

The project gives new insight into how the interplay of market concentration, differentiation, and information frictions determine the market power of banks. The first goal is a descriptive analysis of discounting and price discrimination in this market. The second goal is to understand how mergers in the Canadian banking industry affected mortgage interest rates. The third goal is to develop and estimate a full supply and demand model for this industry and use it to answer several different questions about the likely effect of credit regulations on demand for mortgages and housing.

Understanding how consumers shop and negotiate for mortgages will give us new insight into how this market functions to support housing markets.

Project Report

In many product markets prices are determined through a negotiation process between buyers and sellers. Sellers post a price, but consumers may be able to negotiate a discount. The extent to which sellers negotiate may depend on consumer characteristics; ignoring the actual pricing mechanism can lead to an incomplete and biased analysis. Moreover, discounting is an important form of price discrimination. Despite its prevalence, this form of pricing has been largely ignored by researchers studying market power. For example, Berry, Levinsohn and Pakes (JPE, 2004) study the demand for new automobiles abstracting away from the price setting mechanism actually used in the market. Adams, Einav and Levin (AER, 2009), on the other hand assume monopoly pricing in their analysis of the sub-prime used car loans. Similar examples can be found in insurance, housing, and consumer loans markets. Our case study is the market for mortgages. In this market most consumers negotiate a discount from the price posted by a financial institution located in their neighborhood. Therefore, the size of the discount a consumer can negotiate should depend on the number and characteristics of lenders present in his/her local market. In practice, however, predicting discount sizes is complicated by the fact that financial institutions are selling services that are differentiated. For instance, the location of retail branches determines the cost of shopping for mortgages (horizontal differentiation), while the quality of complementary services (e.g. convenience of teller/ATM network) affects the value of contracting with each institution (vertical differentiation). Consumers also differ in terms of their preferences for these amenities, as well as their ability to negotiate the best deal. To shed light on these issues, we analyze detailed transaction-level data on a large set of approved mortgages in Canada between 1992 and 2004 and administered by either the Canadian Mortgage and Housing Corporation or Genworth Financial. These data provide information on features of the mortgage, household characteristics (including place of residence), and market-level characteristics. The richness of these consumer data in combination with lender-level location data (MicroMedia ProQuest) allow us to empirically examine the functioning of this important market. Ultimately our goal is to understand how the interplay of concentration, differentiation and information frictions determines the ability of banks to exert market power and potentially distort housing and credit markets. We have so far completed three research papers. All three are co-authored with Jason Allen (Bank of Canada), and Robert Clark (HEC-Montreal). The first paper is titled ``Price dispersion in mortgage markets''. It is descriptive analysis of discounting and price discrimination in the mortgage market. Our goal is to document a series of facts about the distribution of transaction prices for new home buyers, and test for the presence of price discrimination motives on the part of lenders. We decompose discounts into factors related to consumer locations, demographics, bank characteristics, market structure, and the arrival of brokers. The paper was published in the Journal of Industrial Economics (2013). The second paper is ``The effect of mergers in search markets: Evidence from the Canadian mortgage industry''. In this paper, we perform retrospective merger analysis to study the relationship between local market structure and transaction prices. In particular, our analysis uses sharp local-market structure changes generated by several important national mergers in the Canadian banking industry. Between 1992 and 2000, the six largest Canadian banks acquired almost all of the trust companies. These six banks came to control approximately 70% per cent of the mortgage market following the merger wave -- almost double their 1980s market share. The focus of our analysis is to highlight the impact of mergers on the ability of lenders to discriminate between consumers with heterogenous search and negotiation abilities. We show in particular that the loss of a competitor raise interest rates by 10 basis points on consumers at the bottom of the distribution, and as a result decrease the dispersion of transaction rates. The paper was published in the American Economic Review (2014). The third paper is titled ``Search frictions and market power in negotiated price markets''. This paper develops and estimates a search and bargaining model designed to measure the welfare loss associated with frictions in oligopoly markets with negotiated prices. We use the model to quantify the consumer surplus loss induced by the presence of search frictions in the Canadian mortgage market, and evaluate the relative importance of market power, inefficient allocation, and direct search costs. Our results suggest that search frictions reduce consumer surplus by $20 per month per consumer, and that 17% of this reduction can be associated with discrimination, $30%$ with inefficient matching, and the remainder with the search cost. We are currently revising the paper for the Journal of Political Economy (revised-and-resubmit).

National Science Foundation (NSF)
Division of Social and Economic Sciences (SES)
Standard Grant (Standard)
Application #
Program Officer
Nancy A. Lutz
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
National Bureau of Economic Research Inc
United States
Zip Code