“It’s the mortgage market, stupid.” It may not be catchy, but new research shows that it is true.
It is often said that Americans vote with their wallets, and there is a large literature showing that voters react to ups and downs in the economy. But we have a very limited understanding of what particular economic phenomena matter most at the polls, and it is not clear why they matter.
Do voters punish political parties that generate poor growth and high unemployment for selfish reasons or because they love their country? It could be either.
{mosads}My research at Columbia Business School with Alexis Antoniades of Georgetown University looked at the 2008 election — a period when mortgage underwriting standards quickly morphed from their most lenient in history around 2004 to their tightest ever as debased underwriting standards on mortgages produced a housing collapse and America sunk into the Great Recession.
In our study, we not only show that voters respond to economic news, but we show that they respond more to economic news that affects them personally. Voters punish the incumbent party if a contraction in mortgage credit supply results in a greater rate of denial in mortgage applications within their county.
This discovery gives political forecasters a new and powerful metric to gauge whether a challenger to the White House is likely to defeat an incumbent. Our findings also suggest that voters respond most strongly to economic news that affects their own personal well being.
To gauge the importance of the mortgage supply effect, our county-by-county study tested a simple question: What would have happened had the housing credit crisis not happened in 2007-2008?
We find that the results of the 2008 election between then-Senator Obama (D-Ill.) and Senator McCain (R-Ariz.) would have changed dramatically. North Carolina and its 23 electoral votes would have actually flipped for Senator McCain.
Three key swing states that went for Obama — Florida, Indiana and Ohio — would have almost gone red. Even though the overall election outcome wouldn’t have shifted, McCain would have actually closed his gap with Obama in nine different states by a full 51 percent of the vote.
The mortgage-credit supply effect proved to be much greater than other economic factors that we considered in our study. In particular, we found that the credit-supply contraction in 2008 mattered for swaying voters in swing states five times as much as the rising unemployment rate, which is one of the most commonly used variables in prior analyses of economic effects on voting.
Why is mortgage credit supply so powerful an influence? The number of Americans without jobs, even when the unemployment rate is comparatively high, will always reflect a small slice of voters. For many, the unemployment rate is an abstract economic statistic.
But most Americans apply for mortgages, and there is nothing abstract about getting your mortgage application rejected. Mortgage credit supply changes create real-life consequences that touch many everyday people — and potentially a wide swath of the electorate.
To see whether our results for 2008 generalize to other elections, we applied the same methods to the elections of 2000, 2004 and 2012.
Interestingly, we found that voters punished President Obama in the 2012 election for the continuing decline in mortgage credit availability from 2008 to 2012, which implies symmetry of the mortgage-credit supply effect across partisan lines.
But voters did not reward either Al Gore in 2000 or George W. Bush in 2004 for the boom in credit supply that voters experienced from the mid-1990s through 2004. Not even a little. Why do voters punish presidential candidates for mortgage credit-supply contraction, yet fail to reward them for mortgage credit-supply expansion?
One possibility is that when people get a mortgage approved by a bank, they credit themselves. They may find it more pleasant to believe it’s their own success that landed them that mortgage. But when a person’s mortgage application is rejected, they look for fault elsewhere, including to the White House.
This finding has an important political implication: Government actions that expand the availability of mortgage credit (of which there have been many) don’t produce in any direct sense more votes for the incumbent party in the presidential election.
If that is so, then why were presidential candidates (including Clinton in the 1990s and both Bushes) so eager to promote mortgage risk subsidies?
For example, the GSE Act of 1992, which proved to be an important contributor to the relaxation of mortgage standards that produced the financial crisis, was signed by George H.W. Bush during a tough re-election campaign. Presidents Clinton and Bush 43 both expanded the relaxations of credit standards under the GSE Act.
One possibility is that the candidates were unaware of our research findings, and believed that they would be rewarded by voters. Another possibility is that presidential candidates may see perks other than voters’ reactions to local credit conditions.
In other words, they may be rewarded in the metaphorical “smoke-filled backroom” with campaign contributions or other favors from vested interests.
Our research, therefore, also raises new questions about the political process that generates mortgage credit-supply surges, which we hope will stimulate other research. We also hope that it will encourage pundits, pollsters and political observers to focus more attention on mortgage market conditions.
Mortgage credit supply has been expanding dramatically over the past five years, and it is quite likely that within the next decade a major contraction of credit will occur again, with significant likely consequences for the presidential election.
Charles W. Calomiris is the Henry Kaufman professor of Financial Institutions at Columbia Business School, director of the business school’s Program for Financial Studies and its Initiative on Finance and Growth in Emerging Markets, and a professor at Columbia’s School of International and Public Affairs.