• DOI:10.2139/ssrn.4278917
• Corpus ID: 253581672

On the Convergence of Credit Risk in Current Consumer Automobile Loans

@article{Lautier2022OnTC,
  title={On the Convergence of Credit Risk in Current Consumer Automobile Loans},
  author={Jackson P Lautier and Vladimir Pozdnyakov and Jun Yan},
  journal={SSRN Electronic Journal},
  year={2022}
}

• Jackson P Lautier, V. Pozdnyakov, Jun Yan
• Published 16 November 2022
• Economics
• SSRN Electronic Journal

Risk-based pricing of loans is well-accepted. Left unstudied, however, is the conditional credit risk of a loan that remains current. Using large-sample statistics and asset-level consumer automobile asset-backed security data, we find that default risk conditional on survival converges for borrowers in disparate credit risk bands well before scheduled termination, a phenomenon we call credit risk convergence. We then use actuarial techniques to derive the conditional market-implied credit spread by credit risk band to estimate current deep subprime and subprime borrowers eventually overpay by annual percentage rates of 285-1,391 basis points. Our results are robust to various sensitivity tests. JEL Codes: C58, D11, D12, D18, D53, G14, G32, G53, P46 Keywords— Adjustable premium loans, asset-level disclosures, competing risks, Coronavirus, COVID-19, economic inequality, financial literacy, market inefficiency, Reg AB II ∗All authors are at the University of Connecticut in the Department of Statistics. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. DHE 1747453. Jackson P. Lautier was employed with Prudential Financial, Inc. in various actuarial and financial roles from 2010-2019, including its fixed-income asset management subsidiary PGIM Fixed Income from 2016-2018. While the material presented herein may be of interest to Prudential and its subsidiaries, the views, research, data, and dissemination of this work have no financial, intellectual, proprietary, or other connection to Prudential Financial, Inc. and should not be construed as such. The authors have no other conflicts to disclose. Correspondence: Jackson P. Lautier, University of Connecticut, Room 323, Philip E. Austin Building, 215 Glenbrook Road, Unit 4120, Storrs, Connecticut 06269-4120; e-mail: [e-mail] 1 ar X iv :2 21 1. 09 17 6v 1 [ qfi n. ST ] 1 6 N ov 2 02 2 For nearly one hundred years the American consumer has relied on installment credit to purchase automobiles and other expensive goods (e.g., Hughes and Cain, 2011). And for nearly as long, the discourse of political economy has contemplated the amount of protection required for individual consumers in financial transactions. After the economic calamities of the Great Depression, for example, Franklin D. Roosevelt famously pledged “a new deal for the American people” in his 1932 acceptance speech for the Democratic nomination (Roosevelt, 1932). Exactly thirty years later, the eventual Nobel laureate Milton Friedman wrote, “Underlying most arguments against the free market is a lack of belief in freedom itself.”(Friedman, 2002). Forty-eight years later in 2010, shortly following the Global Financial Crisis amid a historical cyclicality befitting poetry, then President Barack Obama remarked during the signing of the Dodd-Frank Wall Street Reform and Consumer Protection Act that, “We all win when consumers are protected against abuse. And we all win when folks are rewarded based on how well they perform, not how well they evade accountability”(Obama, 2010). The literature has naturally followed suit, with noteworthy contributions on payday loans (e.g. Melzer, 2011; Bertrand and Morse, 2011), credit cards (e.g. Gross and Souleles, 2002; Agarwal et al., 2014), and automobile loans (e.g. Adams et al., 2009; Grunewald et al., 2020) to name but a few. Much of the traditional focus in these studies is on the moments prior to and including contract signing or the aftereffects of such consumer contracts. We examine the well-accepted practice of risk-based pricing (Edelberg, 2006) but spanning the entire lifetime of a current loan. In other words, the risk-based cost of borrowing is traditionally a function of a singular point-in-time historical assessment of a consumer’s credit risk profile (Phillips, 2013). A borrower’s credit profile is dynamic, however, and this disconnect leads to a natural question: how many on-time payments does it take for a high-risk borrower to no longer be considered high-risk? We find that the answer is considerably less payments than the original term of the loan, which raises an inevitable second question: what are the financial and economic implications of loan contracts for borrowers originally charged a high-risk cost of borrowing that remain current? This paper is dedicated to providing the first empirical answers to both of these questions. Our findings may be split into two major categories. The first is a statistical argument that borrowers of secured consumer automobile loans in disparate credit risk bands who remain current eventually converge in default risk. We call such a phenomenon credit risk