“Know Your Customer”

When I took “Management of Lending” at The Harvard Business School, I will never forget Professor Charlie Williams telling us that the most important rule in lending is to “Know your customer.” Before mortgage lending became so sophisticated and complicated, bankers religiously followed this very simple but effective principle.

However, the advent of securitization—the practice of bundling together a group of mortgages into “mortgage backed securities” (MBS) or “collateralized debt obligations” (CDOs), and then selling off these new securities to groups of investors separate from the banks that made the mortgage loans in the first place— severed the relationship between the lender and the borrower, fundamentally changing the way in which mortgage loans are financed. This change is at the heart of the current U.S. mortgage crisis.

To appreciate how different today’s practices are, it’s important to understand how it used to work. In the past, the local banker typically made mortgage loans to people in the same community where he or she lived. The banker often knew the borrower personally—perhaps they had even gone to school together. The banker was very familiar with the borrower’s family and background; knew where the borrower worked; and had a sense as to whether the borrower was a responsible member of the community. The banker knew whether the borrower could be trusted to live up to the obligations of the mortgage.

Before being considered for a mortgage, the borrower would fill out a lengthy application, detailing his or her education, employment history, income and assets. A letter from the employer, pay stubs and tax returns were often requested to support statements made in the application. After the bank was satisfied that the borrower was financially qualified, it would hire a real estate appraiser to conduct an objective, third-party appraisal of the property. In most cases, the amount of the loan was limited to 80 percent of the value. In other words, the borrower was required to have significant equity in the property, putting up 20 percent of the value in cash. Once the loan was made, the banker and the bank had to live with the loan. If the borrower made all of the payments on time, the bank had a solid asset on its books that generated regular and consistent interest income.

Under this system, mortgage defaults happened, but they were relatively rare. Because the banker and the borrower lived in the same community, the banker was often the first to become aware of any change in a borrower’s financial status. If the borrower’s income happened to change, the borrower and banker would sit down and work out a solution, which might include rescheduling mortgage payments. Since home and real estate values seldom declined, the value of the bank’s collateral and the borrower’s equity tended to increase every year. As a result, the banker had flexibility to work with a borrower through difficult financial times. Likewise, the borrower had a very strong incentive to make the mortgage payments and to stay on good terms with the banker to preserve the property’s equity value.

For all of these reasons, real estate mortgage lending has traditionally been considered one of the safest investments to make in the United States. Banks and savings and loan organizations that have concentrated on mortgage lending have typically been sound and stable financial institutions. With many U.S. families owning their own homes, and real estate prices rising with the growth of the U.S. economy, real estate mortgages have become one of the largest investment categories in the United States. At the beginning of this year, there were approximately $14.7 trillion of mortgages outstanding.

This system, which served the U.S. economy quite well in the 50 years since the end of World War II, began to change dramatically in the 1990s. The pressure on banks by government legislation and regulators to extend loans to less than creditworthy borrowers, combined with a flood of new money to make such loans, lead to a dramatic increase in subprime loans.

As the amount of subprime loans went up, and home prices increased, the country’s overall mortgage portfolio began to deteriorate and become inherently more risky. Bankers began making loans that were 90 percent, or even 100 percent of the value of the property. Appraisal firms were pressured to over-value properties, justifying a loan amount that would give the borrower the money he or she needed. Whether the borrower could afford the payments, or whether the real underlying value of the property justified the loan, were completely different matters. “Stated income” loans were made that merely required the borrower to tell the bank how much he or she made—no supporting documentation needed. Adjustable Rate Mortgages (ARMs) that initially required interest only payments enabled borrowers to qualify for bigger mortgages. Of course, when payments were adjusted upwards at a later date to provide for principal repayments or higher interest rates, they often became too high for the borrower, leading to payment defaults. And finally, so-called “Ninja” loans were made to borrowers with “no income, no jobs, and no assets.”

Because bankers could sell off loans to Fannie Mae, Freddie Mac or Wall Street underwriters anxious to underwrite MBS and CDOs, the bankers no longer had to live with the loans they had made. They could simply make the loans, collect the associated origination fees (the higher-risk loans typically carried the highest fees), and then pass on the responsibility for the future performance of the loans to others. By bundling a group of mortgages together, the risk of any one loan going bad was reduced, or so the theory went. If there was ever any doubt as to the quality of the loans in the mortgage package, AIG would happily provide credit insurance by issuing a “credit default swap.” (A credit-default swap has been called both “the most important instrument in finance” by former Federal Reserve Chairman Alan Greenspan and a financial “weapon of mass destruction” by Warren Buffett. In all, AIG wrote some $79 billion in insurance on CDOs backed mainly by subprime mortgages. The total value of AIG’s credit default swap portfolio was $527 billion, according to a regulatory filing.)

The increase in subprime loans was one important reason for increased home ownership and the housing bubble in the United States. The overall U.S. homeownership rate increased from 64 percent in 1994 (the same level it had been since 1980) to a peak in 2004 of 69.2 percent. Between 1997 and 2006, American home prices increased by 124 percent. Many homeowners used the increased property value to refinance their homes with lower interest rates and to take out second mortgages to get more cash for spending. U.S. household debt as a percentage of income rose to 130 percent during 2007, versus 100 percent earlier in the decade.

Despite the risks being created for the economy, this revolution in banking and lending was self perpetuating because it had something in it for everyone. The bank could generate more and more fees, without having to worry whether borrowers could actually make future payments. Fannie Mae and Freddie Mac, which back nearly one-half of the total mortgages outstanding, could make a great deal of money originating and selling off loan packages. Investment bankers and insurance companies were happy because this new system created an unending supply of underwriting and insurance fees. Consumers were happy because they could borrow more money and buy a bigger or more expensive house with the same amount of equity. And finally, the politicians were happy because this seemed to be a great way to help every American realize the dream of home ownership.

With so many risky new loans being made, though, it was only a matter of time before delinquency rates began to increase and real estate values to crack. Approximately 16 percent of subprime ARMs were 90-days delinquent or in foreclosure proceedings as of October 2007, roughly triple the rate of 2005. By January 2008, the delinquency rate had risen to 21 percent and by May 2008 it was 25 percent. By November 2007, average U.S. housing prices had fallen approximately 8 percent from their 2006 peak, and by May 2008 they had fallen 18.4 percent. As prices declined, more homeowners were at risk of default and foreclosure. Absent any other damage to the financial markets, the current crisis will not begin to be over until real estate prices in the United States at least stabilize.

If something ever seems too good to be true, it’s usually because it is, as everyone is learning once again. “Trees don’t grow to the sky” and prices don’t go up forever in the capital markets. Although increased sophistication is generally a good thing, there is no question that it has made the current crisis an even larger one than it would have been otherwise. This week, Treasury Secretary Paulson will propose that Congress approve a $700 billion rescue package, by far the largest in history.

Once the dust settles, a great deal of analysis will be done to determine the causes of Meltdown Monday and the measures that should be taken to prevent a re-occurrence. As China’s capital markets develop, Chinese government officials and economists will have a once in a lifetime opportunity to learn from the U.S experience and get it right by putting in place regulations that preserve the benefits of developed capital markets, but prevent their potentially costly side effects.