![]()
Part One
THE U.S. HOUSING SLUMP AND THE GLOBAL FINANCIAL CRISIS
The United States has experienced a few severe housing slumps since World War IIānotably, the sharp decline in housing prices from 1989-1993. But as Figure I.1 shows, declining housing prices from that slump were not reflected in falling prices of U.S. stocks. In fact, the Standard & Poorās 500 Index (S&P 500) rose by over 15 percent for the years 1989 through 1993, while the home price index fell by over 13 percent.1
Other countries have experienced even more severe housing slumps than the United Statesāfor example, the fall in Japanese housing prices during the 1990s. Although this Japanese housing slump was paralleled by a decline in the Japanese stock market during the 1990s, neither led to a global decline in stocks or bonds. Indeed, prior to 2008, no housing slump in any country has ever led to a global financial crisis.
Figure 1.1 Inflation-Adjusted Home Price Index vs. Real S&P 500 Stock Price Index
SOURCE: Robert Shiller irrationalexuberance.com.
So why did the U.S. housing slump in 2007 and 2008 trigger a global financial crisis? The answer lies in the excessive debt of American families and financial institutions, combined with the securitization process that spread mortgage-backed securities (MBS) across the world.
After being burned by stocks in the burst of the dot -com bubble, investors in 2001 were looking for other places to put their money. Most Americans felt that real estate was a safe bet because they believed that home prices always went up. With lots of mortgage financing available at low interest rates, many Americans bought housing and piled on the debt. By 2007, U.S. household debt reached a record high of over 130 percent of household income.
To increase their profits, many U.S. financial institutions also borrowed heavily to buy assets, supported by relatively small amounts of capital. In 2004, the Securities and Exchange Commission (SEC) allowed the five largest investment banks to double their ratio of assets to capital to over 30:1. The largest banks created separate shell companies to issue bonds, which could be found only in footnotes of their financial statements. Moreover, money poured into unregulated hedge funds, which often took out large loans to pursue aggressive trading strategies.
As long as the music was playing, everyone kept dancing. But when the music stopped in late 2006, all the dancers ran for the exits at the same time, crushing each other in the panic. As prices of housing and mortgage-backed securities plummeted, many investors tried to sell assets to raise cash and pay down their debts. These sales, in turn, drove asset prices lower, leading to more selling and more losses.
Huge losses were suffered not only by American investors but also by foreign financial institutions due to the global distribution of mortgage-backed securities. Traditionally, when lenders made home mortgages, they held on to them until they were paid off. But now lenders could sell most of their mortgages to specialized entities created by Wall Street banks or to either of two quasi-public corporations called the Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac). In turn, these entities and corporations sold MBS and bonds to investors across the world.
Figure I.2 provides a simple flow chart for the mortgage securitization process. A woman buying a house borrows money from a lender and signs a mortgage on the home, which secures her promise to repay the loan. She then makes monthly payments of principal and interest on the mortgage to the lender. This is the M in MBS. If she fails to repay her loan, the lender has the right to foreclose on the mortgage and take the house. The lender may sell the mortgage to Fannie Mae or Freddie Mac, which finances the purchase by selling bonds to investors. Alternatively, the lender may sell the mortgage to a special purpose entity (SPE), a shell corporation formed by a Wall Street firm to gather mortgages into a pool. An SPE raises money to buy these mortgages by selling to investors various types of bonds that are backed by (the B in MBS) the monthly payments from the mortgages in the pool. These bonds are securities (the S in MBS).
Figure 1.2 Mortgage Securitization Process
As the total residential mortgage debt in the United States more than doubled from 2000-2007, so did the total amount of MBS that were sold to investors throughout the world. The total residential mortgage debt in the United States skyrocketedāfrom approximately $5 trillion in 2000 to over $11 trillion in 2007āas U.S. home prices soared. In parallel, the total amount of MBS doubled from $3.6 trillion in 2000 to $7.3 trillion in 2007.2
Following the huge run-up in U.S. home prices from 2000-2006, these prices plummeted in 2007 through 2009 as the housing bubble burst and mortgage defaults rose to record high levels. In turn, these higher rates of mortgage defaults led to dramatic decreases in the prices of MBS based on these pools of troubled mortgages. Prices fell because the monthly mortgage payments backing the MBS were evaporating at an unforeseen rate. Because large amounts of MBS were held by financial institutions in the United States and abroad, this dramatic decline in MBS prices resulted in significant capital losses at institutions across the world. As a result, the securitization of loans has virtually halted since the end of 2008.
Although flawed, the securitization process still has significant benefits so it needs to be fundamentally reformed, rather than eliminated. Most importantly, securitization multiplies the volume of lending by increasing the liquidity of mortgages. Instead of holding mortgages to maturity, lenders can sell them to investors and use the proceeds to originate another round of loans. To take a simple example, compare two identical lendersāone that makes and holds one $400,000 mortgage that pays off after 15 years, while the second lender makes a $400,000 mortgage each year for 15 years and sells a $400,000 mortgage each year to investors. Over 15 years, the second lender will use the same monies to lend $5.6 million more than the first lender. Thus, the securitization process provides more loans to Americans and, with more money available to fund mortgages, lowers interest rates on mortgages across the country.
At the same time, the securitization of mortgages offers the benefits of risk diversification to both banks and investors. Before mortgage securitization, banks located in a particular region of the country were vulnerable to declining housing prices in that region. Now regional banks can sell mortgages on local homes and buy a portfolio of MBS based on pools of mortgages from across the country. Investors can choose the specific package of risks they want because one pool of mortgages often issues several separate types of MBS (called tranches) with different risk characteristics. For instance, a low risk tranche of an MBS with a relatively low yield might have the right to the first principal and interest payments from the pool. In contrast, a high-risk tranche of an MBS would have a much higher yield, but would incur the first loss if any mortgage in the pool defaults.
In short, the challenge is to retain the substantial benefits of the mortgage securitization without its negative aspects that led to investor losses around the world. Part I will rise to this challenge by analyzing the mortgage securitization process in four chapters. Each chapter will explain one important aspect of the securitization process, evaluate the reforms taken so far and offer further proposals to remedy the abuses in that aspect of the process.
ā¢ Chapter 1 will identify the driving forces behind the United States housing bubble and suggest what should be done to reduce the likelihood of another bubble.
ā¢ Chapter 2 will explain why Fannie Mae and Freddie Mac went bankrupt despite their government charters and assess what role, if any, they should play in the future.
ā¢ Chapter 3 will analyze why the private process of mortgage securitization came to an abrupt halt and delineate what reforms are needed to restart the process.
ā¢ Chapter 4 will propose a new regulatory framework for credit default swaps and, more generally, discuss the lessons learned from the misuse of mathematical models.
![]()
Chapter 1
The Rise and Fall of U.S. Housing Prices
Contrary to popular perceptions, residential housing prices in the United States rose by only 10 percent above the rate of inflation from 1949-1997āgoing from an index of 100 to an index of 110, as demonstrated by Figure 1.1. Next, housing prices rose sharply by 21 percent above inflation between 1997 and 2001 (from an index of 110 to an index of 133), and then suddenly took off like a rocket between 2001 and 2006ārising 53 percent higher than the inflation rate (from an index of 133 to 203). But this meteoric rise was unsustainable; at the end of 2008, U.S. residential housing prices had plunged by 33 percent from their 2006 high (from an index of 203 to 137), and have declined further during 2009.3
Figure 1.1 Inflation Adjusted Home Price Index: 1949-2008
SOURCE: Robert Shiller irrationalexuberance.com.
Many factors contributed to this rise and fall of housing prices. In this chapter, we will focus on three key factors: abnormally low interest rates, unscrupulous sales practices of certain mortgage lenders, and incentives for certain house purchasers to avoid personal responsibility. (We will discuss additional important factors in other chapters, for instance, Chapter 5 on short selling by hedge funds and Chapter 6 on excessive leverage of financial institutions.)
The Fed Kept Interest Rates Too Low
Low interest rates in the United States were a key factor driving domestic housing prices sky high between 2001 and 2006. Because mortgages were so cheap, some purchasers were willing to pay more for homes that they were going to buy and other purchasers were able to afford homes for the first time. United States interest rates were pushed lower during this period by a combination of the savings glut in the emerging markets and the Federal Reserveās extended response to the 2001-2002 recession.
Between 2000 and 2007, the foreign exchange reserves of central banks in emerging markets ballooned from less than $800 billion to over $4 trillion.4 In part, this sharp increase resulted from the rising prices of oil and gas in countries with natural resources, such as Saudi Arabia, Brazil, and Russia. In part, this sharp increase resulted from the rapid growth in trade surpluses of China and other Asian countries with the United States, where American consumers gobbled up imports.
In turn, the central banks in the commodity-producing countries and Asian exporters invested much of their rising foreign currency reserves in U.S. Treasuries. Such investments boosted the value of the U.S. dollar, which supported the price of oil (denominated in U.S. dollars) and encouraged Americans to buy relatively cheap imports from Asia. Between 2000 and 2007, U.S. Treasuries owned by foreign investors rose from $1 trillion to $2.4 trillion. China alone increased its holdings of U.S. Treasuries from $60 billion in 2000 to $478 billion in 2007.5
In other words, there was an implicit agreement on a global recycling process. By consuming massive amounts of imported goods and oil, the U.S. ran huge trade deficits, which resulted in large trade surpluses with oil producers and Asian exporters. These two groups of countries then recycled most of these surpluses back to the United States by investing in U.S. Treasury securities. This global recycling process kept the rates on long -term Treasury bonds approximately 1 percent lower than they otherwise would have been.6
The role of the Federal Reserve in elevating U.S. housing prices is more complex. In response to the 2001 recession resulting from the burst of the dot-com bubble and the September 11, 2001 terrorist attacks on the World Trade Center, the Fed aggressively lowered the interest rate on short-term U.S. Treasuries (e.g., one week to three months), which declined to almost 1 percent at the end of 2002. The Fed then held the short-term rate close to 1 percent until the middle of 2004. Concerned about the fragility of the economic recovery, the Fed held interest rates too low for too long. Only toward the very end of 2006 did the Fed bring the short-term interest rate back to normal levels.7 To see how far the Federal Reserve suppressed interest rates during this period, consider Figure 1.2.8 The chart compares the actual low level of interest rates set by the Federal Reserve to the level determined by the Taylor ruleāa well-recognized method of setting central bank rates developed by Stanford University professor and former Treasury official John Taylor. As the chart shows, actual rates were dramatically below those suggested by the Taylor rule from 2001 through 2005.
Low Interest Rates Stimulated Appetite for High-Yield Mortgages
The decline in interest rates on U.S. Treasury bonds stimulated the appetite among foreign investors for higher yields from other types of debt securities. Between 2001 and 2006, foreign ownership of MBS increased from 6 percent to over 18 percent.9 Si...
