Author: Ran Duchin (University of Michigan) and Denis Sosyura (University of Michigan)
Publisher: Ross School of Business Working Paper No. 1165
Date Published: September 2011
Banks that received federal bailouts during the financial crisis went on to issue riskier loans and to increase their stake in higher-yield, riskier financial securities compared with banks that did not get government assistance, the authors of this study find. The shift in risk occurred largely in the same asset class, they note, and thus had little impact on closely watched capitalization levels.
“Consequently, bailed banks appear safer according to the capitalization requirements, but show a significant increase in market-based measures of risk,” the authors write. “Overall, our evidence suggests that banks’ response to capital requirements may erode their efficacy in risk regulation.”
In an attempt to stabilize the banking sector during the financial crisis, the U.S. Treasury invested nearly US$205 billion between October 2008 and December 2009 in 707 financial institutions through the Capital Purchase Program (CPP), the first and largest of 13 initiatives under the Troubled Asset Relief Program (TARP).
Whereas other studies have used proxies to stand in for government guarantees, the authors of this paper collected data from the CPP and several agencies that allowed them to study “an unusually large cross section of firms subject to a simultaneous, unexpected liquidity shock in the world’s largest and most developed financial market.”
The authors began with a list of all public domestically controlled financial institutions that were eligible for the CPP and were active in the quarter before the program kicked in. They chose public firms because their regulatory filings are available and they account for the vast majority (92.7 percent) of all capital that was invested through the CPP. The 295 recipients of these funds obtained about $190.1 billion, according to the Treasury.
Of the 529 institutions in the final sample, 424 (or about 80 percent) submitted requests for aid. Among the 337 institutions that were approved, 286 accepted the help and 51 ultimately did not. The researchers culled quarterly reports on these banks from the first quarter of 2006 through the fourth quarter of 2010. The firms in the study had average book assets of $2.2 billion. As a proxy for an institution’s level of exposure to the financial crisis, the researchers used the ratio of foreclosed assets to the value of all loans and leases. (The authors did not include in the final sample the first wave of CPP recipients — the nine institutions deemed “too big to fail,” including Citigroup, JPMorgan, and Bank of America. Adopting what they called a conservative approach, the researchers decided to exclude the nine because of their special nature. But had the group been included, the results of the authors’ study would have been the same, they added.)
The authors gathered loan application data from the Home Mortgage Disclosure Act Loan Application Register, which covers about 90 percent of mortgage lending in the United States (only the smallest rural banks, with assets under $37 million, are excluded). Because this data contains information on both approved and denied mortgages, the researchers could examine bank lending decisions on nearly all individual applications in the U.S. from 2006 to 2009, including information on the demographics of the borrower and characteristics of the loan (amount, type, and property location).
The borrower and loan specifics allowed the researchers to study the changes in the banks’ allotments across riskier and safer loans, and the granular data on location enabled them to examine how a range of banks in the same region made different lending decisions. The researchers also controlled for peculiarities in the local housing market, combining data from the U.S. Postal Service on the dynamics of home vacancies; county-level statistics on per capita income, population, and unemployment from the Bureau of Economic Analysis; and fluctuations in housing prices from the Federal Housing Finance Agency.
Although banks don’t have to divulge applicants’ credit scores or the interest rate for every mortgage, they must report the interest rate spread on loans with an APR that is at least 300 basis points above the Treasury yield. Previous research has demonstrated that a rate spread this large serves as a close proxy for subprime mortgages. The authors used borrowers’ loan-to-income ratio as another proxy for mortgage risk.
The analysis looked at three aspects of bank operations: retail lending (mortgages), corporate lending (large syndicated loans), and investment in financial assets. To examine corporate lending, the authors used the DealScan database, which covers large corporate loans that stem from several banks in a syndicate. According to the database, between 2006 and 2009, 179 qualified banks in the study originated $3.5 trillion in corporate credit.
In the first step of their analysis, the authors focused on the mortgage market and found that after receiving federal funds, banks that had received bailouts shifted their credit issuance toward riskier mortgages, whereas those denied federal funds decreased their portfolio of subprime loans. The share of CPP recipients issuing risky loans increased from 91.2 percent before the bailout to 96.7 percent afterward. In addition, the approval rates among bailed-out banks for the least risky mortgage applications fell from 74 percent to 63 percent, while approval of the most risky loans rose from 48 percent to 53 percent.
The findings were similar for large corporate loans. After getting government funds, banks increased their share of credit issuance to the riskiest corporate borrowers, as measured by their credit ratings and bond yields, and cut back on their loans to safer companies.
“Altogether, our findings for both retail and corporate loans suggest that the bailout was associated with a shift in credit rationing rather than the volume of credit, leading to a marked increase in the riskiness of originated credit by government-supported institutions,” the authors write.
The risk-taking trend also extended to the investments made by government-supported banks. After receiving federal capital, banks boosted their investments in risky securities, such as equities scooped up in an attempt to profit from short-term price movements, mortgage-backed securities, and long-term corporate debt. In the average portfolio of a bailed-out bank, the combined value of these assets increased by 10 percent, in place of safer assets such as Treasury bonds, short-term paper investments, and cash.
After controlling for bank fundamentals and changes in security valuations, the authors found that non-recipients were much less likely to make risky investments. Because high-yield bonds have lower credit ratings and are widely regarded as more likely to default, the researchers used asset yield as a market measure of risk. For bailed-out banks, the average interest yield on investment portfolios shot up by 9.4 percent after getting government aid compared with banks that received no federal funds.
And there is a real economic consequence for bailed-out banks, say the authors, who measured the level of systemic risk through an analysis of firms’ earnings volatility, leverage, stock return volatility, and market beta — a measure of the risk contained in a security or portfolio vis-à-vis the overall market. Compared with applicants that did not receive federal funds, recipient firms increased their exposure to systemic risk after getting government money, the authors found, with their beta measure increasing substantially, from 0.8 to 1.01. In contrast, non-participating banks experienced no changes in systemic risk over the same period.
Overall, the authors write, their analysis “indicates a robust increase in risk taking in both lending and investment activities by bailed financial institutions, as compared to fundamentally similar banks, which were denied federal assistance.”
Financial institutions that received government assistance during the economic crisis of 2008–09 went on to issue riskier mortgage loans and invest in volatile, higher-yield portfolios than banks that received no federal aid, this paper finds. As a result, these financial firms exposed themselves to even more risk after receiving bailouts.