Comments on A Crisis in Student Loans? How Changes in the Characteristics of Borrowers and in the Institutions They Attended Contributed to Rising Loan Defaults Karen M. Pence Federal Reserve Board Note: The views expressed are mine alone and not those of the Board of Governors or its staff.
Hooray for this paper! Superb and timely paper Candidate for inaugural BPEA award for service to the economics and public policy professions Policy interest in student loans > data available to answer policymaker questions This paper + online data appendix contains the information needed to address many of these questions
Policy concern 1: Government-guaranteed student loan balances have tripled since 2001 $1,200 Balances on Government-Guaranteed Student Loans ($B) $1,000 $800 $600 $400 $200 $0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Year Source. Looney and Yannelis.
Policy concern 2: Default rates on these loans have almost doubled 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% 10% Three-Year Cohort Default Rate on Government- Guaranteed Student Loans 10% 10% 11% 12% 13% 15% 17% 18% 18% 19% 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Cohort Source. Looney and Yannelis.
This paper: increased defaults stem from a compositional change in borrowers and deteriorating outcomes Borrowers in Repayment (thousands) Three-Year Cohort Default Rates 1000 900 800 700 600 500 400 300 200 100 0 241 929 For-Profit 283 447 Selective 30% 25% 20% 15% 10% 5% 0% 19% 26% For-Profit 4% 7% Selective 2001 2011 2001 2011 Source. Looney and Yannelis.
Is the increase in debt entirely bad news? Broader context During the financial crisis and ensuing recession: Households suffered major shocks to their most widely held assets earnings and houses Access to private student loan, mortgage, auto, and credit card debt contracted sharply State budgets came under pressure: tuitions increased at state-supported educational institutions
In contrast Terms eased for government student loans Widely available underwriting only at the school level for most programs (exception: parent PLUS) Loan limits increased on unsubsidized Stafford loans (2008) Parent PLUS underwriting loosened (2008) Interest rates decreased on subsidized Stafford loans (2009-12) Opportunity cost of attending college decreased because of poor outside labor market options
In principle, the increase in debt could be a good thing The increases in enrollment and student loan balances may be sensible response to the changes in the labor market, credit market, and household balance sheets during the financial crisis and recession For-profit schools can expand more quickly than other sectors to meet increases in demand If students invest in valuable human capital, the student loan program might be considered an effective counter-cyclical policy tool
However, education does not appear to have paid off for many for-profit students 60% 50% Borrowers with Larger Balances 5 Years after Starting Repayment 55% 25% 20% Borrowers Unemployed 2 Years After Starting Repayment 21% 40% 30% 20% 33% 20% 22% 15% 10% 14% 7% 7% 10% 5% 0% For-Profit Selective 0% For-Profit Selective 2001 2009 2001 2011 Source. Looney and Yannelis.
How bad are these outcomes? As thought experiment, compare student loans originated for attending for-profit colleges with subprime mortgages
Default rates on for-profit student loans are as high as those on subprime mortgages 50% 40% 30% 20% 10% 9% 31% Five-Year Cumulative Default Rates by Year Borrower Entered Repayment 12% 34% 35% 35% 35% 23% 32% 39% 45% 45% 0% 2003 2004 2005 2006 2007 2008 2009 Subprime Mortgages Student Loans for For-Profit Schools Sources. Subprime mortgages, Palmer (2014). Student loans, Looney and Yannelis.
Borrowers with student loans from for-profit schools outnumber subprime mortgage borrowers 12,000,000 10,000,000 8,000,000 6,000,000 4,000,000 2,000,000 Borrowers with Outstanding... - 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Subprime Mortgages Student Loans for For-Profit Schools Sources. Subprime mortgages, Federal Reserve staff estimates. Student loans, Looney and Yannelis. Note. Subprime mortgage estimates assume number of subprime mortgages = number of borrowers.
Outstanding balances on subprime mortgages are much higher than for-profit student loans Outstanding Balances (billions) $1,400 $1,200 $1,000 $800 $600 $400 $200 Average loan size: Subprime mortgage: $175,000 Student loan, for-profit school: $20,000 $- 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Subprime Mortgages Student Loans for For-Profit Schools Sources. Subprime mortgages, Federal Reserve staff estimates. Student loans, Looney and Yannelis.
Incidence of subprime mortgage losses Most subprime mortgage borrowers had small or no down payments In the event of default, the borrower loses the house, but the debt is (largely) extinguished Credit losses from subprime mortgages were widely dispersed throughout the financial system in opaque and complicated ways My guess: majority of direct losses from subprime mortgages were borne by the financial sector
Incidence of for-profit student loan losses Human capital is non-transferable: a lender cannot seize and resell knowledge if a borrower defaults Student loan debt is difficult to discharge in bankruptcy, and DoEd has extraordinary collection powers DoEd recovers 80-100% of defaulted principal and interest My guess: losses accrue primarily to the household sector, with the rest on the government s balance sheet. Schools lose very little.
Final caveat and question Hard to know the counterfactual for students who attended for-profit schools during the recession Because of their characteristics, they would have struggled at other types of schools, and if they didn t attend school at all But this paper suggests that the paramount policy question is not so much Why is student loan debt so high? but rather How can we improve the returns to and decrease the risks of investing in education for disadvantaged students?