Hidden Culprit in Financial Crisis
The Financial Crisis Inquiry Commission, created by the Fraud Enforcement and Recovery Act (May 20, 2009), held its first public meeting today, launching its mandate to examine causes of the financial crisis. The statute enumerates 22 possible culprits, all now usual suspects, like executive compensation, bad regulatory oversight, financial derivatives, complex securitization schemes and the like. Also on the list is the general notion of bad lending practices at banks, which two Commissioners today likewise noted in general terms.
Not on the list, and not mentioned today, or much discussed in the cacophonous litany, are the credit scoring systems lenders use to make first and sometimes final cuts on loan decisions. The credit scores lenders used before the crisis are still used today. But they do not measure credit-worthiness, or probability of loan repayment, as much as they measure whether applicants are good customers of banks, compared to those less reliant on banks. They value debt, and over-leverage is a recurring culprit in all financial crises.
To illustrate, take a pop quiz. Suppose the following two people apply for a mortgage loan on a home in suburban Maryland (say for $500,000). (1) Which is more credit worthy? (2) Which is more likely to be approved? (3) Would the answers differ if the decision were made today, after the financial crisis, or two years ago, before that?
Applicant A is a tenured Professor of Medicine at Johns Hopkins University, with a mid-6 figure base salary, doubled by significant practice income, revenue on patents she shares with the University, and rental income from resort properties she inherited from an aunt years ago that are owned free and clear of any liens or loans. The Doctor has no outstanding debt, a net worth in the low seven figures and has bought and sold two previous homes, paying off related mortgages before their maturity date.
Applicant B is a local real estate developer, with a low-6 figure income, ¼ of which is in contingent bonus compensation, and no other source of income. His net worth consists mainly of a modest retirement account, stock options and grants from his employer, and some cash in the bank, about enough to make the down-payment contemplated by the mortgage loan application. He has several lines of credit outstanding, all with meaningful balances, that he has increased regularly for years, though always paying monthly minimum balances when due.
(1) A, the Doctor, is by far the more creditworthy of the two.
(2) B, the real estate developer, will have a much higher credit score.
(3) No, under the so-called Desktop Underwriting software that lenders use, before crisis and now (hawked as enabling these decisions to be “made in less than 15 minutes”), B gets a far higher credit score than A. B passes lender underwriting cut-offs; A does not.
Why? Credit scores increase when people have large amounts of debt and decrease when they do not have large amounts of debt. Credit scores are unaffected by factors like employment security, equity net worth, or income. True, after credit score screening, banks may apply an additional test to borrowers with good credit scores, to assure requisite ratios of income to debt payment obligations. But the credit scores bias even that evaluation and still overlook significant factors. They essentially foreclose the possibility of making loans highly likely to be repaid.
In short, credit scores do not reflect people’s probability of repaying loans (whether they are creditworthy) but whether someone is a good customer of the banking system. Contributing to the crisis, banks made mortgage loans to millions of people with wonderful credit scores who lacked creditworthiness. They avoided lending to people with bad credit scores and strong credit quality. The practice continues. The Commission should put credit scoring on its list of possible causes of the financial crisis.