Salesmen without Scruples: My Observations as a Loan Officer -- Part III

Science Vs. Shoeleather: The Case Against Quants

Based on the prevailing financial models in 2006, sub-prime loans defaulted at a six standard deviation shock beyond expected calculations. A natural occurrence of once in 216 million years. I wouldn’t buy a one dollar lottery ticket with those odds. The banks bet billions.

The math couldn’t account for every variable. Why didn’t the big investment banks have detectives on the case? Everything was reduced to an excel spreadsheet; billions were bet on blind calculations. A little gumshoe work would have put a question mark next to the math, especially when they discovered that smooth-talking ex-cons were providing the numbers and earning the commissions.

Sometimes it was only a simple stroke of white-out that stood in the way of a loan’s approval and the commission that came with it.

At Global Home Loans, a $10,000 commission of Danny’s appeared in jeopardy. The loan required the borrower to have a bank balance showing sufficient cash reserves. But, their bank statements revealed that they had no cash flow. They spent everything they earned.

Danny knew what to do. He applied a little swab of white-out, a stroke of his pen, and bank statements showing a satisfactory balance were faxed to the lender. Loan approved.

The products themselves were as questionable as the people selling them.

The most popular product was a 2 & 28 Adjustable Rate Mortgage known as an ARM. Most traditional mortgages offered a 30-year payment plan where the interest rate and payments remained fixed for the life of the loan.

With an ARM, the interest was fixed for only the first two years, at a low teaser rate unavailable to sub-prime borrowers in a standard loan. After the first two years the interest rate jumped for the remainder of the loan. But, the 28-year piece was never intended to come into play. After two years the borrower’s credit was supposed to improve to prime territory allowing them to refinance at a better rate.

The rub? The same spending habits that created their credit issues remained. If they received cash back as part of the refinancing it compounded their problems.

Do you remember Mr. Perez? The landscaper from Sarasota? His home was worth $400,000. He owed $300,000 on the house; we gave him a loan for $340,000. The $40,000 difference was his to spend. If he was like most people he did, and stupidly.

Cash back became our calling card. It was one of the first questions we asked on cold-calls to hook in customers. A better interest rate bored them, but $40,000?

Let’s make a deal.

Instead of paying off bills with the cash back, they bought flat screens TV’s big enough to skate on, bunker-sized barbeque grills, and 8-cylinder SUV’s. Their credit never improved.

They used their homes like ATMs, refinancing three and four times as their property value increased. They collected cash back every time; so did the loan officer through his commissions. But, every cash back opportunity increased the loan amount. This posed no problem as long as the property held its value.

It didn’t. Mr. Perez’s $340,000 loan was issued when the property was worth $400,000.

Housing values in Florida have since fallen 25%, valuing the house at less than $300,000. He owed more than the house was worth—this is called negative equity.

Like most of our customers, Mr. Perez was sold a 2 & 28 mortgage. After two years, he couldn’t refinance; banks don’t offer loans for more than a house is worth. The teaser rate ended, the monthly payments jumped, and he couldn’t cover the added burden. This happened all over the country, creating waves of foreclosures that crested into what mariners have mythologized as a rogue wave. A six standard deviation shock to scientists.

This was sub-prime as I saw it, not as an abstraction on a spreadsheet. I spoke with hundreds of these customers; I heard their stories; I saw their credit report. I doubt the rating agencies spoke to even a single loan customer.

PhD’s in physics, math, and computer science became hot commodities on Wall Street. Why accept a professor’s pay when you could quadruple your salary on Wall Street? Their models became gospel.

Quantitative analysts and the spreadsheets they spawn will always have a place in business, but numbers are never so surefire to make intuition obsolete.

Clearly something went wrong with the models. The models measured what they could, but some of the most important factors were immeasurable. They couldn’t account for shadiness and stupidity on an Excel spreadsheet.

A bad business joke gets the gist:

The president of a car dealership stands in his showroom looking concerned. There’s a new dealership across the street and their American flag flies higher against the horizon. He won’t be upstaged, so he asks the janitor to measure the competitor’s flagpole.

The janitor returns the next morning and says 20 inches in circumference.

The rating agencies used statistical models to assess patterns of default. Statisticians want to know out of 1,000 mortgages, based on historical performance what percent of people will pay their loans? They assumed the past would remain relevant in a world where new products were unveiled every month.

A common product during the boom called a N.I.N.J.A. loan sums up the whole story. The acronym stands for No Income, No Job, or Assets. Though it was implied the borrowers had all these things, they weren’t documented. These loans appealed to cash businesses like landscaping where most income isn’t claimed on tax returns. People willingly paid a higher interest rate to avoid alerting the I.R.S. of their true earnings. And naturally loan officers loved them; there were few documents to deal with.

These new loans lacked both historical data and individual loan facts. Everything was based on credit scores and loan to value ratios. How can an equation be solved without knowing any of the variables?

You can’t make projections without historical precedent. The outcome itself, known as the sub-prime crisis was without precedent.

Next: The Closing


Popular posts from this blog

The Farthermost Seas of the South Pacific

Grand Central vs. Penn Station

Five Years After the Storm: A Celebration of the Crescent City