«This version: January 2, 2016 Keywords: Adverse Selection, Loan Maturity, Consumer Credit. JEL codes: D82, D14. † Columbia University. Email: ...»
ADVERSE SELECTION ON MATURITY: EVIDENCE FROM ON-LINE CONSUMER
ANDREW HERTZBERG†, ANDRES LIBERMAN‡, AND DANIEL PARAVISINI††
ABSTRACT. We provide evidence of adverse selection on maturity in consumer credit. Our estimation
compares two groups of observationally equivalent borrowers that took identical 36-month loans, but
where only one of the groups is selected on maturity: borrowers chose the 36-month loan when a 60-month maturity option was also available. Borrowers who self-select into short maturity loans default less a year after origination, and have future credit ratings that are higher and less volatile.
Consistent with the insurance role of long term credit, the ﬁndings suggest borrowers self-select on their exposure to shocks to their future ability to repay.
This version: January 2, 2016 Keywords: Adverse Selection, Loan Maturity, Consumer Credit.
JEL codes: D82, D14.
† Columbia University. Email: firstname.lastname@example.org.
‡ New York University. Email: email@example.com.
†† London School of Economics. Email: D.Paravisini@lse.ac.uk.
We thank Emily Breza, Anthony DeFusco, Theresa Kuchler, Adair Morse, Holger Mueller, Christopher Palmer, Philipp Schnabl, Antoinette Schoar, Amit Seru, Felipe Severino, Johannes Stroebel, and participants at Bocconi University, Columbia University, Credit and Payments Markets Conference (Federal Reserve Bank of Philadelphia), Crowdfunding Symposium (Berkeley), CUHK, Dartmouth University (Tuck), HKU, HKUST, LSE (Economics and Finance departments), NBER Corporate Finance 2015 Fall meeting (Stanford), NBER Household Finance Summer Institute 2015, NYU (Stern), NYU-Columbia Junior Faculty Seminar, UBC (Sauder), and UNC (Kenan-Flagler). We thank Siddharth Vij for outstanding research assistance. All errors and omissions are ours only.
1 2 I. Introduction A household’s ability to repay debt is adversely affected by income and consumption shocks such as job loss, illness, or divorce. Since debt repayment capacity —creditworthiness— is priced in consumer credit markets, a household that borrows through a sequence of short term debt contracts will see the price of credit increase precisely in those states of the world where expected income is lower and marginal utility of consumption is higher. Financing with a long maturity loan, in contrast, protects the household from price increases due to repayment capacity shocks. In fact, when prepayment is costless, a long-term loan is identical to a contract that combines an equally priced short-term loan and an option to delay payment that the household can exercise when subject to an adverse shock. Thus, a fundamental function of long term debt contracts is to allow households to insure against the risk of becoming, or being re-classiﬁed as, a bad risk in the future.
The insurance role of long term contracts is not exclusive to consumer credit markets, as it is also played, for example, by long term corporate debt (Diamond (1991)), long-term employment contracts (Holmstrom (1983)), and long-term health and care insurance (Cochrane (1995), Finkelstein, McGarry, and Suﬁ (2005)). In theory, adverse selection, which results from households’ private information about their risk exposure, may limit the ability of competitive markets to provide insurance through long term contracts.1 Lenders facing adverse selection will offer a menu of contracts with varying maturities, such that households that are more exposed to shocks self-select into longer maturity contracts that offer more protection, and are priced accordingly. Although many common consumer loan products, such as mortgages, auto loans, and personal loans, offer borrowers a choice over loan maturity, the potential for adverse selection in this contract dimension and the role of maturity as a screening device remain, to date, largely theoretical propositions.2 The present paper provides the ﬁrst evidence consistent with the insurance and screening roles of debt maturity in consumer credit markets. We show that borrowers self-select into long maturity loans when they are privately informed about a high exposure to shocks to their future repayment capacity. The main challenge in measuring how ex ante selection on maturity affects ex post repayment performance stems from the fact that maturity itself affects borrower behavior. Suppose we compared the performance of two observationally equivalent groups of borrowers, the ﬁrst with a short term loan and the second with a long term loan. The default rate of the two groups of borrowers may differ due to adverse selection, but also due to the effect that a different maturity, installment amount, and interest rate have on the borrowers’ repayment behavior. For this reason, identifying empirically the consequences of selection on repayment requires comparing how selected and unselected borrower samples behave when facing the same contract.
To illustrate this point and provide a motivation for our empirical strategy, consider the idealized setting for identifying adverse selection on maturity, depicted in Figure 1. Suppose we observe two groups of prospective borrowers, A and B, before they take a loan. Group A is offered only a short ST maturity loan at an interest rate of rST. The default rate of these borrowers is γA. Group B is offered two options: the same short maturity loan as group A (at rate rST ), and a long maturity loan for the same amount at a rate of rLT. Group B borrowers that choose the short term (long term) loan default ST LT
at a rate γB (γB ). Borrowers from group B who take the short term loan are selected on maturity:
they could have taken a long term loan, but chose not to. Group A borrowers, in contrast, are an unselected group. Further, both group A and group B short term borrowers face identical loan terms (interest rate, amount, and maturity). Thus, any difference in the repayment of the short term loans ST ST between group B and group A borrowers, γB − γA, must be driven by the selection induced by the ST ST long maturity loan. In particular, γB −γA 0 would indicate that borrowers with a higher privately observed default risk select into the long maturity loan.
We exploit the staggered roll-out of long maturity loans by an on-line lending platform, Lending Club (hereafter, LC), as an empirical setting that closely resembles this idealized one. When a borrower applies for a loan at LC she is assigned to a narrow risk category based on FICO score and other observable characteristics. All the borrowers in a risk category are offered the same menu of loan choices, e.g. the same interest rate for every amount and maturity combination.
Loans of amounts between $1,000 and $35,000 are available in either short—36 months—or long maturities—60 months. In January 2013 the long maturity loan was available only for amounts above $16,000. During 2013 the available menu of long term loan options expanded twice: 1) to loans amounts between $12,000 and $16,000 in March 2013, and 2) to loan amounts between $10,000 and $12,000 in July 2013. Crucially for our analysis, the terms of all other previously available menu items were unchanged during this time (within each risk category), and the roll-out was not advertised on the LC website or accompanied by any additional marketing campaign.
Borrowers would only notice the new options once they began applying for a loan.3 Our empirical strategy compares the default rate of short term loans between $10,000 and $16,000 issued before and after the availability of the long maturity option at the corresponding amount, which approximate groups A and B of the idealized setting of Figure 1, respectively.4 Simple before-after comparisons are potentially confounded by time-of-origination shocks. To account for these shocks we estimate a difference-in-differences speciﬁcation that exploits the staggered roll-out of the long term loans, and that uses short term loans of amounts just above and just below the $10,000 to $16,000 to construct counterfactuals. And to perform comparisons between 3According to the information reported in the website Internet Archive, LC’s website at the time of the expansion advertised that 60 month loans were available for amounts above $16,000 until November 2013.
4For example, borrowers choosing a 36-month $10,000 loan before July 2013 resemble those in group A of Figure:
these borrowers did not have a long term option in the menu at the time of making the choice. Borrowers choosing a 36-month $10,000 loan after July 2013 resemble borrowers in group B: they chose the 36-month loan when a longer maturity loan was available, and are thus a sample selected on maturity.
4 observationally equivalent borrowers, our speciﬁcations include month-of-origination, risk category, 4-point FICO range, and state ﬁxed-effects.5 We validate the identiﬁcation assumptions behind the empirical strategy by documenting that the bulk of self-selection into long maturity loans occurs among borrowers that would have borrowed between $10,000 and $16,000. We ﬁnd that the number of short maturity loans between $10,000 and $16,000 drops by 14.5% after the long maturity loans become available, relative to loans issued at amounts just above and below this interval. Further, the decline was permanent and occurred the same month the 60-month loan appeared in the menu for the corresponding amount.
We ﬁnd that the average default rate of short maturity loans decreases by 0.8 percentage points and the average future FICO score increases by 2.7 points when a long maturity loan is available at the time of origination. This implies that borrowers with unobservably higher default risk (and risk of suffering a decline in their future credit rating) self-select out of short term loans and into long term ones. The increase in performance of 36-month loans is due to the 14.5% of borrowers that self-select into long maturity, which implies that these borrowers would have had a default rate that is 5.5 percentage points higher (0.8/14.5) than the default rate of the average 36-month borrower in our sample (9.2%). These results indicate that selection on maturity provides a powerful device for identifying, among a pool of observationally identical borrowers, those with the poorest repayment prospects.6 Having established that borrowers who select into long term loans are privately informed about their poor repayment prospects with a short term loan, we turn to understanding the mechanism that is driving this behavior. A natural question of interpretation is whether it is insurance provided through loan maturity or, instead, a borrower’s preference regarding minimum payment size that matters for screening. We note that, for an unconstrained borrower, the fundamental difference between any short and long-term amortizing loan where the rate is locked at origination is insurance.
An unconstrained borrower with a short-term loan can always effectively lower the minimum payment by undertaking additional borrowing (for example, using a credit card) each period. The only difference between this series of short-term loans and a long maturity loan is that the additional borrowing must be done at market interests rates at the time of the new lending. Thus, with the 5Our empirical setting has several additional advantages that underline the robustness of our estimates. First, loans offered on the LC platform are funded by investors at the terms set by LC’s pricing algorithm. These terms compare favorably to other investments of similar risk, thereby ensuring that all loans are funded. Moreover all terms of the 36-month loans including interest rate remain unchanged during our sample period. This rules out that selection is occurring based on supply side screening decisions or by reverse causality of loan terms impacting default. Second, LC charges an upfront origination fee between 1.1 and 5% of a borrower’s loan amount (subtracted from the amount borrowed). Thus, borrowers who took a short maturity loan prior to the expansion could not costlessly swap them for long maturity ones after the expansion. This ensures that the pool of borrowers who select the short maturity loan prior to the expansion is not impacted by the expansion itself.
6Two robustness tests ensure our results are not simply capturing time-of-origination varying differences in creditworthiness for loans of different amounts. First, our results are unchanged when we reduce our sample to a narrower bandwidth around the affected amounts. Second, we test for differences in creditworthiness in the same time period but at other amounts in the borders of the interval of affected amounts. These tests ﬁnd no evidence of time-varying differences in creditworthiness.
5 sequence of short term loans the borrower is left exposed to shocks to her observable creditworthiness.
In contrast, a long maturity loan insures the borrower against this reclassiﬁcation risk. A long-term loan with costless prepayment is equivalent to a short-term loan with an additional embedded option to borrow the difference in minimum payment size each month at terms that are ﬁxed at the time of origination.
This equivalence does not hold for a borrower that faces immediate and temporary liquidity constraints, as in Attanasio, Koujianou Goldberg, and Kyriazidou (2008). This suggests that borrowers who have private information about their exposure to temporary liquidity constraints may self-select into long term loans. We can empirically distinguish between the insurance and the liquidity constraints mechanisms in our setting by looking at how selection predicts default at different horizons from origination. If adverse selection is driven by temporary liquidity constraints, then the differential propensity to default will be large close to origination, when these constraints bind, and decline over time as the constraints relax.
We ﬁnd the opposite result in the data: the decline in the default probability of 36-month loans when the long term loan is available occurs only twelve months after origination, even though, unconditionally, close to half of the loans that default do so within twelve months of origination. In other words, borrowers who select into long term loans have an incremental propensity to default (relative to those that self-select into short-term loans) that increases with time since origination.