There is much discussion today about the size of the government’s role in the mortgage market. At first glance, the data plotted in Exhibit 1 support the impression that the government is playing a far larger role in the market than it had done historically. The parts of the chart shaded in green represent government agencies and the Government Sponsored Enterprises (GSEs), which we can collectively call the agency sector. The blue represents the private sector (with “PLS” standing for private-label securities, representing what many would think as Wall Street’s involvement).
Exhibit 1: Market Shares of Residential Mortgages 1925-2010: By Nominal Holder1
At the end of 2010, the total stock of single-family mortgages outstanding was $10.5 trillion. The agency sector either held or guaranteed a total of $5.8 trillion, 55 percent of the total. Put in the historical context shown in Exhibit 1, it looks like the footprint of the agency sector reached a level in the last 20 years that is historically unprecedented.
Exhibit 1 is the “official scorecard” (as reflected in the Federal Reserve’s Flow of Funds Accounts, for instance). It is not wrong – data can be interpreted in different ways. However, I believe it paints an incomplete picture of the mortgage market – because of the way FHA and VA loans are counted, this chart downplays the government’s centrality in the mortgage market since the 1930s.
FHA and VA loans are explicitly backstopped by the government. Consequently, I think they should be counted as government-backed loans when determining government versus private-sector market shares. Instead, they are counted as holdings of the nominal owner, be it a bank, thrift, GSE, or other entity. To the extent that they are held by banks and thrifts, for instance, they are counted as private-sector rather than government-backed loans.
Exhibit 2 presents the data taking this credit exposure into account. FHA/VA loans held by the GSEs are taken out of the GSE book and reassigned to FHA/VA. Similarly, FHA/VA loans owned by banks, thrifts, and insurance companies are moved from their portfolios to FHA/VA. This reassignment based on credit exposure presents a different landscape. It shows that government has had a major involvement in mortgage finance dating back to the 1930s.
Exhibit 2: Market Shares of Residential Mortgages 1925-2010: By Holders of Credit Risk
Exhibit 2 is rich in detail – to the extent that it makes it difficult to see the deeper trends. The next two exhibits summarize the data in order to draw out what are some of the underlying patterns.
Exhibit 3 isolates the boundary between the green and the blue areas of Exhibit 2. It shows the total government and GSE share of single-family mortgages. At the end of 2010, the share was 55 percent. This was exactly the same share as in 2001, before the turmoil of the housing bubble and the subsequent housing market decline.
Exhibit 3: Market Share of Government and GSEs as Holders of The Credit Risk of Single-Family Mortgages 1930-2010
What I believe is even more surprising is that the 2010 agency market share was not that much higher than it had been over half a century earlier, in 1956 – 55 percent now versus 45 percent then. This was in the middle of the Eisenhower administration, remembered by many as an era of minimalist government (though it coincided with the Federal Aid Highway Act of 1956 that funded the interstate highway system). The situation in 1956 represented the culmination of the government’s massive response to the two big upheavals of the 20th century – dealing with the Great Depression and housing returning GIs and their families after World War II.
Exhibit 4 is the mirror image of Exhibit 3. It isolates the market share of the private sector. To me, what stands out in this chart is the close correlation between the private-sector market share and financial crises:
Exhibit 4: Market Share of Private Sector as Holders of The Credit Risk of Single-Family Mortgages 1930-2010
In summary, the government has played a large role in the housing market dating back to the Great Depression. During those 80 years, its role has expanded and contracted with ramp-ups coming in response to downturns in the housing market followed by a declining share of the market in more expansionary times.
Director, Economics and Strategic Research
1Sources: Federal Reserve Board, Flow of Funds Accounts; Leo Grebler, David Blank and Louis Winnick, Capital Formation in Residential Real Estate, 1956; Saul Klaman, Volume of Mortgage Debt in the Postwar Decade, 1958; Saul Klaman, The Postwar Rise of Mortgage Companies, 1959; Historical Statistics of the United States, Millennium Edition; Federal Housing Finance Agency 2010 Annual Report to Congress; Fannie Mae; Freddie Mac; U.S. League of Savings Institutions, Savings and Loan Fact Book, Various.
2See: Jesse Jones, Fifty Billion Dollars, My Thirteen Years with the RTC, 1951, page 151; Gertrude S. Fish, The Story of Housing, 1979, page 207.
3See, for instance: http://www.crestmontresearch.com/docs/i-rate-relationship.pdf; http://www.wsjprimerate.us/wall_street_journal_prime_rate_history.htm
4See: L. William Seidman, Full Faith and Credit, 2000, page 177; FDIC, History of the 1980s, 1997, Volume 1, pages 168-9.
5See: General Accountability Office, Financial Audit: Resolution Trust Corporation’s 1995 and 1994 Financial Statements, Tables 1 and 4, GAO/AIMD-96-123, July 1996 at http://www.gao.gov/archive/1996/ai96123.pdf.
6See, for instance: Financial Crisis Inquiry Commission Report, Figure 5.1, page 69 at http://fcic-static.law.stanford.edu/cdn_media/fcic-reports/fcic_final_report_full.pdf; Paul Krugman, New York Times, June 3, 2010, Tables 4 and 5, at http://krugman.blogs.nytimes.com/2010/06/03/things-everyone-in-chicago-knows/; Paul Moulo and Matthew Padilla, Chain of Blame, page 281.
November 2, 2011