Discussion

Abstract

Previous articleNext article FreeDiscussionPDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreRobert Hall opened the discussion by presenting evidence from Hall and Woodward (2012) supporting the hypothesis that the retail mortgage market in the United States features large markups. He used this evidence to question the authors’ assumption that mortgage contracts in the United Kingdom are written in a completely competitive environment.Jonathan Parker pointed out that over the three different periods considered in the paper (1975 to 1985, 1986 to 1995, and 1996 to 2005), risk premia have declined in markets other than the mortgage market. Some of these markets, moreover, did not undergo the same institutional changes—such as increased competition and the growth of securitized lending—that the mortgage market experienced over the sample. He wondered whether the authors had any evidence to explain why these different markets exhibited such similar price behavior despite their structural differences. He also suggested supplementing figures 15 and 16 with information about loan sizes to document how that variable has changed across different sample periods.Raj Chetty made two points. First, he indicated that the cross-sectional relationship between the size of mortgage loans and their interest rates is particularly interesting. He suggested showing other similar cross-sectional relationships, possibly with different measures of creditworthiness such as credit scores or levels of unsecured debt. These relationships could be helpful in determining whether lenders were becoming less sensitive to borrower credit risks, or whether they simply expected house price increases to provide greater downside protection against such risks. Mark Gertler also wondered about the role of the housing price bubble in reducing the apparent riskiness of large mortgages during the 1995 to 2005 period.Second, Chetty expanded on concerns expressed by Bernard Salanié about the validity of the stamp duty rate instrument. In his view, the “Baseline” first-stage specification reported in the first column of table 31 does not adequately isolate exogenous variation in stamp duty rates. More appropriate is the “Collapsed” specification from the second column of table 3, which focuses on variation in stamp duty rates across regions and years, but not across individuals. This specification, however, suffers from lower power: the t-statistic on the stamp duty rate falls from 229 to 2.83. More importantly, he expressed doubts over whether the IVQR model remains identified after eliminating this variation across individuals, and recommended that the authors instead place greater emphasis on their OLS estimates.2Timothy Besley provided three responses to the issues raised by the discussants and other participants. First, following up on comments made by José-Luis Peydró, he explained that the authors may be able to match loans to lender characteristics, but that so far they have not been granted access to that data. Second, in response to Chetty, he explained that unlike how it is in the US mortgage market, credit scoring is not an established practice in the United Kingdom. It is therefore not possible for them to include those measures of creditworthiness in their anaylsis. Third, he agreed with Jonathan Parker and José-Luis Peydró that he and his coauthors should more carefully investigate possible sources of time variation in their data. He suggested that the rise in mortgage brokers over the later half of the sample might be an important phenomenon driving this variation. The data set used in the paper, however, does not contain information about whether a particular mortgage was brokered or not. So it would be difficult to tie this institutional trend directly to the data.Endnotes1Now table A3 in the appendix.2This part of the discussion refers to the conference draft of the paper. The final draft published in this volume does indeed primarily emphasize the OLS estimates, as suggested. The discussion here relates to an approach that is now mainly discussed in the appendix.ReferenceHall, Robert E., and Susan E. Woodward. 2012. “Diagnosing Consumer Confusion and Sub-Optimal Shopping Effort: Theory and Mortgage-Market Evidence.” Forthcoming in American Economic Review.First citation in articleGoogle Scholar Previous articleNext article DetailsFiguresReferencesCited by NBER Macroeconomics Annual Volume 272012 Sponsored by the National Bureau of Economic Research (NBER) Article DOIhttps://doi.org/10.1086/669200 Views: 76 © 2013 by The National Bureau of Economic Research. All rights reserved.PDF download Crossref reports no articles citing this article.