The global financial crisis has its origins in the failings of the US housing market – and, some believe, those of the US government. Since the onset of the crisis, nominal house prices in the US have fallen over 40%, delinquent mortgage debt has risen to an astonishing $1.5 trillion, and some of the world’s leading financial institutions have completely disappeared. US government support for mortgage lending to low-credit-quality households in the years leading up to the subprime crisis is seen by many as a critical factor in its severity and – by extension – the subsequent global financial crisis (see Leonnig 2008, Barrett 2008, Calomiris and Wallison 2008, and Congleton 2009).
Yet government intervention in the financial sector is nothing new. As far back as 1792, the US government was instrumental in the establishment of the First Bank of the United States. Indeed, government intervention is often well intentioned and justified by economic theory. But once governments are involved in the financial sector, the incentives for individuals to try to manipulate government policy and politicians’ votes can be irresistible. The results can be disastrous. Charles Calomiris (2009) argues that such policy interference “has been at the centre of the explanation of the propensity of banking crises for the past two centuries.”
Testing the theory
A long-standing and influential body of research argues that politicians vote in their economic interests (see Stigler 1971, Kalt and Zupan 1984, and Peltzman 1985). In this view, politicians respond to both the interests of voters – constituent interests – and those of specific groups such as an industry lobby – special interests – in order to increase the likelihood of re-election.
An alternative view argues that politicians primarily vote according to their ideological preferences (see Kau and Rubin 1979, 1993, Bernstein 1989, Poole and Rosenthal 1996, and Lee et al. 2004). Determining which of these two views is dominant presents a constant challenge in empirical work. It is possible, for example, that legislators with a track record of voting conservatively also face constituent and special interests that are naturally aligned with the conservative agenda (Kalt and Zupan 1990, Levitt 1996).
In recent research, (Mian et al. 2010), we make use of a unique dataset to examine how special interests, measured by campaign contributions from the mortgage industry, and constituent interests, measured by the share of subprime borrowers in a congressional district, may have influenced US government policy towards the housing sector during the subprime mortgage credit expansion from 2002 to 2007.
The expansion of subprime lending coincided with key US government policies, often with cross-party backing, that reduced regulation of subprime lenders and increased mortgage support for low-income households.
Among the most prominent was the affordable housing mandate imposed by the Department of Housing and Urban Development. The mandate required that mortgage providers and guarantors Freddie Mac and Fannie Mae purchase a larger fraction of mortgages that serve low- and moderate-income borrowers.
At the same time as the bills, campaign contributions from the mortgage industry intensified (see Figure 1).
Figure 1. Campaign contributions by the mortgage industry
What votes were being bought with the money?
Due to the large number of bills – over 700 roll calls in the House alone – that are related to housing between 1993 and 2008, disentangling the many possible factors influencing voter behaviour is impossible.
In light of this, we adopt an alternative approach to examining voting patterns on each single roll call. We aggregate all roll call votes for any legislation related to “affordable housing,” “homeownership,” or “subprime”, and find that the predictive power of mortgage campaign contributions on a representative’s voting behaviour increases sharply during the subprime mortgage credit expansion. The fraction of votes for which mortgage campaign contributions have an effect on voting patterns increases from 3% in 1995 to 20% by 2003.
Buying votes with voting behaviour
But the story is more complex than just subprime lenders buying government support for subprime lending. Beginning in 2002, mortgage industry campaign contributions increasingly targeted US representatives from districts with a large fraction of subprime borrowers. To measure constituent interest we use zip code level data on consumer credit. Combining this with electoral data, we then compare this with campaign contributions, lobbying expenditure, and eventually congressional voting data.
During the expansion years, we find that both mortgage industry campaign contributions and the share of subprime borrowers in a congressional district increasingly predicted congressional voting behaviour on housing-related legislation. In 1997 and 1998, the fraction of subprime borrowers in a representative’s district significantly predicts the representative’s votes on only 30% of roll calls. By 2003 the fraction increases to 70%. =
As the solid line in Figure 2 shows, beginning with the 107th Congress in 2002, there is a sharp relative increase in mortgage campaign contributions to high subprime share districts, while there is no such pattern for non-mortgage contributions. Our results suggest that a one standard deviation increase in the subprime share as of 1998 – before the expansion – leads to a relative increase in the growth rate of mortgage industry campaign contributions of 81%.
Figure 2. Relative campaign contribution growth in subprime congressional districts
Taken together, our results suggest that constituent interests, measured with the fraction of subprime borrowers in a given Congressional district before the subprime mortgage expansion, and special interests, measured with campaign contributions from the mortgage industry, both helped to shape government policies that encouraged the rapid growth of subprime mortgage credit.
Votes on specific bills
In the final section of the study, we examine voting and co-sponsorship patterns on six bills for which competing interests are well defined, these are:
- The American Dream Downpayment Act of 2003, which aimed to increase homeownership among low-income communities by providing downpayment and closing cost assistance;
- the Ney-Kanjorski Responsible Lending Act of 2005, which would have preempted state regulations on predatory lending;
- the Prohibit Predatory Lending Act of 2005, which would have placed more stringent controls on subprime lenders;
- the Mortgage Reform and Predatory Lending Act of 2007, which was a revised version of the Prohibit Predatory Lending Act that eventually passed the House (but failed in the Senate);
- and the Federal Housing Finance Reform Acts of 2005 and 2007, which sought to tighten regulation of Freddie Mac and Fannie Mae.
While these bills are complex in their implications, there were instances of a clear alignment of special and constituent interests. One example is the vote on Amendment number 600 of the 2005 version of the Federal Housing Finance Reform Act. The amendment outline describes it as seeking “to authorise the regulator to require one or both of the government sponsored enterprises to dispose or acquire assets or liabilities if the regulator deems those assets or liabilities to be a potential systemic risk to the housing or capital markets, or the financial system.” We find that politicians with ties to both special interests and constituent interests display a statistically higher propensity to vote against this amendment.
While we cannot tell whether contributions are causing co-sponsorship or whether contributions go to allies that are more likely to co-sponsor regardless of the contributions, the evidence suggests an alliance between co-sponsors and the mortgage industry in line with our aggregate results.
We show that campaign contributions and lobbying expenditure by mortgage lenders increased sharply during the subprime mortgage expansion. While it may be tempting to argue that mortgage lenders are completely at fault for the subprime crisis, our findings suggest a more nuanced reality. Pressure on the US government to expand subprime credit came from both mortgage lenders and subprime borrowers.
Moreover, given the nature of political influence and the complexity of government decisions, our results should not be seen as a “smoking gun”. Instead we provide suggestive evidence of the influence of subprime borrowers and lenders on policy.
This raises the question: What do our results mean for future policy – and future crises? In Mian et al. (2009), we examine Congressional voting patterns on the American Housing Rescue and Foreclosure Prevention Act of 2008 and the Emergency Economic Stabilisation Act of 2008 – two of the most significant pieces of federal legislation in US economic history. Focusing on the trade-off between ideology and economic incentives, we find that conservative politicians are less responsive to both constituent and special interests. This suggests that politicians, through ideology, might be able to commit against intervention even during severe crises.
Our combined findings demonstrate how politicians respond to ideology, constituent interests, and special interests. But this is just a start. Understanding how politicians respond to vested interests is of first-order importance given the historic magnitude of the global financial crisis and the government intervention that has followed it. We hope that our research in this area can inspire others.
Barrett, Wayne (2008), "How Andrew Cuomo Gave Birth to the Crisis at Fannie Mae and Freddie Mac", Village Voice, 5 August.
Bernstein, RA (1989), Elections, Representation, and Congressional Voting Behavior: The Myth of Constituency Control, Englewood Cliffs, Prentice-Hall.
Calomiris, Charles (2009), “Banking Crises and the Rules of the Game”, NBER WP 15403.
Calomiris, Charles and Peter Wallison (2008), “Blame Fannie Mae and Congress for the Credit Mess,” Wall Street Journal, 23 September.
Congleton, Roger (2009), “On the Political Economy of the Financial Crisis and Bailout of 2008-2009”, Public Choice, 140:287-317.
Kalt, Joseph P and Mark A Zupan (1990), “The Apparent Ideological Behavior of Legislators: Testing for Principal-Agent Slack in Political Institutions”, Journal of Law and Economics, 33(1):103-131
Kau, JB, and Rubin, PH (1979), “Self-interest, ideology, and logrolling in congressional voting", Journal of Law and Economics, 22:365-384.
Kau, JB, and Rubin, PH (1984), “Capture and ideology in the economic theory of politics”, The American Economic Review, 74:279-300.
Kau, JB, and Rubin, PH (1993), “Ideology, voting and shirking”, Public Choice, 76:151-172.
Lee, D, E Moretti and M Butler (2004), “Do voters a¤ect or elect policies? Evidence from the US House”, Quarterly Journal of Economics, 119(3):807-859.
Leonnig, Carol (2008), "How HUD Mortgage Policy Fed the Crisis," Washington Post, June 10: A01=
Levitt, Steven D (1996), “How Do Senators Vote? Disentangling the Role of Voter Preferences, Party Affiliation, and Senator Ideology”, American Economic Review, 86:425-441.
Mian, Atif, Amir Sufi, and Francesco Trebbi (2009), “The Political Economy of the U.S. Mortgage Default Crisis”, American Economic Review, forthcoming.
Mian, Atif, Amir Sufi, and Francesco Trebbi (2010), “The political economy of the subprime mortgage credit expansion”, NBER WP 16107.
Peltzman, Sam (1985), “An Economic Interpretation of the History of Congressional Voting in the Twentieth Century”,.American Economic Review, 75:656-675.
Poole, Keith T and Howard Rosenthal (1996), “Are Legislators Ideologues or the Agents of Constituents?”, European Economic Review, 40:707-717.
Stigler, G (1971), “The theory of economic regulation”, Bell Journal of Economics, 2(3):21.