The residential mortgage crisis that began in 2007 is more severe, in terms of rates of foreclosure and decreases in home prices and residential wealth, than any since the 1930s. The building and mortgage boom of the 1920s and bust of the 1930s, as a result, provide striking historical benchmarks against which to measure our own experience. History can play a much more important role, however, because it offers the opportunity to investigate three issues that are of great importance and relevance as we struggle through our own mortgage crisis:
1. We need to examine why the collapse in residential housing markets during the 1930s was so severe, and the recovery from it so uneven, in order to better understand how disruptions to the mortgage market contributed to the depth and duration of the Great Depression. These issues are relevant today because the volume of mortgage lending grew rapidly during the 1920s just as it did before the current crisis; because rapid innovation transformed mortgage loan contracts and funding channels in the lead-up to both episodes; and because both crises were marked by widespread disruption in mortgage lending channels.
2. We need to assess whether federal emergency responses were effective in ameliorating the impacts of the 1930s mortgage crisis. These New Deal interventions were extraordinary in size and scope, but similar to recent policies in that they encouraged widespread mortgage loan modification. Examining the emergency policies of the 1930s, as a result, can help us better understand how responses during the current crisis have succeeded and fallen short.
3. We need to examine the performance of the mortgage market before, during and after the institutional structure of the nation?s mortgage market was permanently reshaped between 1932 and 1938 by the creation of the Federal Home Loan Bank System, the modern Saving & Loans industry, the FHA home mortgage loan insurance program, the Federal National Mortgage Association (Fannie Mae) and the introduction of the long-term, amortized, low-down payment mortgage loan. These institutions continue to form the basic framework of the modern mortgage market and understanding their origins will provide important guidance as we reshape them once again.
All three issues deserve close investigation, but discussion of financial crises during the Great Depression has for decades been focused almost exclusively on the banking system and the stock market. This project is designed to rectify that oversight by stimulating for the first time a serious examination of the issues enumerated above. It does so in two steps. First, the core of the proposal is to create a publicly available, high-quality data base of institution-level mortgage lending activity that will allow scholars to investigate the operation of local and regional mortgage markets between 1920 and 1940. Second, the co-PIs will use the data base as it is being assembled to investigate several of the topics enumerated above in order to stimulate a broad discussion of this important historical episode.
The importance of the data collection component of the project is easy to understand?research in economics is drawn to issues that can be investigated with readily available, high-quality data. Such data exist for the residential mortgage market before 1940, but the information is spread out in thousands of printed volumes held that are held in public and university archives. The data sit in this form because the dominant institutional mortgage lenders before 1940 were non-bank, state-chartered institutions such as Building & Loan Associations, Mutual Savings Banks and Life Insurance companies. The activities of these institutions were well-documented in annual reports of state regulatory authorities, but this information needs to collected, digitized and organized into a reliable and widely available data set. Most of the funding for this project will support photographing and digitizing the state-level reports; preparing extensive documentation so that these data can be easily used; and combining the institutional balance sheet data with neighborhood, city, and state level information on housing markets, economic activity and a broad range of demographic information. The data collection and assembly will take three years to complete and the collaborative structure of this project has been driven by its size and complexity.
The co-PIs were drawn to this project after independently using federally-generated, county-level data to investigate one of the largest and most dramatic of the federal policy interventions of the 1930s?the Home Owners? Loan Corporation. The HOLC modified distressed mortgages on one-tenth of the nation?s owner-occupied homes between 1934 and 1936 and the work of the co-PIs have shown that this activity cushioned homeownership rates and home prices during the 1930s, but did not stimulate new home construction. In the course of these investigations the co-PIs became convinced that much more can be learned about HOLC?and FHA insurance, Fannie Mae discounting and FHLB supervision?if institution-level data were available to complement the county-level data they have already collected on mortgage and housing markets. The second phase of the project, therefore, will be a series of solo- and joint-authored empirical investigations by the co-PIs using the assembled data base.
Carmen Reinhart and Kenneth Rogoff (in This Time is Different) have recently shown that the historical perspective can inform us about severe economic crises and shape our responses to them. It is unfortunate, in contrast, that we have yet to closely examine the mortgage crisis of the 1930s or to fully appreciate its hard-earned lessons. The goal of this project is to facilitate that discussion so that history can assist as we struggle to move through and beyond our own mortgage crisis.
