In the first installment of a three-part series, economic historian Price Fishback showed just how different the basic macroeconomic facts are in the current financial situation versus during the Great Depression.
In the second of three blog posts, Fishback turns to a discussion of the recent financial meltdown compared to the one that accompanied the Great Depression. For everyone still scratching their heads about what happened to the financial sector this fall, Fishback offers one of the clearest descriptions I’ve seen yet. He then discusses the similarities and differences of the financial collapse that began in 1929.
The Financial Meltdown Now and Then
By Price Fishback
A Guest Post
The “greatest financial disaster since the Great Depression” has become the catchphrase for the current situation. In Old Testament fashion, the U.S. appears to be paying for the financial excesses associated with a real estate boom between 2000 and 2006. The reasons for the boom were many. As the stock market dropped at the beginning of the decade, investors sought a new haven for their assets in housing and real estate. The boom was fueled in part by low interest rates stemming from a loose monetary policy and a series of financial innovations that led mortgage lenders to make many more loans.
The Bush administration and many members of Congress pushed for an “ownership society,” in which a larger share of people owned their own homes. One way to achieve this was to find ways to loan to people who had long been considered bad credit risks because they lived in neighborhoods where housing values were at risk or because lenders considered them to have inadequate capacity to repay the loans. Some of the efforts were designed to correct the discrimination in lending to minorities that many once perceived to be commonplace.
Fannie Mae and Freddie Mac, the government-sponsored entities established to create markets for mortgage loans, were given more incentives to purchase subprime and alt-A mortgages. The subprimes are the highest risk loans; the alt-A mortgages carried less risk than the subprimes, but were risky nonetheless because they lent to borrowers who provided minimal or no down-payments and were not required to document their incomes and assets. Fannie and Freddie eventually bought 44 percent of the subprime loans in 2004, 33 percent in 2005, and 22 percent in 2006. The plan was for Fannie and Freddie to use their rigorous standards and procedures to monitor and ensure that these risky loans did not default. The goals of the ownership society were met to some degree as home-ownership rates rose from 67 percent in 2000 to an all-time high of 69 percent in 2004 and 2006.
The heads of Fannie and Freddie were not dragged kicking and screaming into these purchases. In fact, Fannie and Freddie have become infamous for their active lobbying of their regulator and Congress to allow them to make these purchases. Both entities were profit-seeking corporations selling stock on Wall Street and issuing bonds. Originally established to purchase the conventional mortgages guaranteed by the F.H.A. or V.A., Fannie and Freddie had slowly been expanding the range of mortgages they purchased through the 1980’s and 1990’s. Both enterprises profited greatly from the common perception that they were backed by the federal government.
The perceptions of backing held even though President Lyndon Johnson had removed the federal guarantee for Fannie in 1968 and Freddie had been established as Fannie’s competitor without government backing in 1970. This perceived guarantee, which proved true in September 2008, allowed them to issue bonds and borrow at interest rates only slightly higher than the no-risk federal treasury bill rates. They could then purchase mortgages that paid higher rates and either hold them or package them up in mortgage-backed securities and sell them to investors. Despite this advantage, Fannie and Freddie faced increasing competition from other financial institutions who provided the same services. To maintain their strong position in the mortgage markets, they lobbied strongly to expand their base.
The modern boom was fueled by far more than policy by government and government-sponsored agencies. The shift of assets out of stocks and into housing touched off a rise in housing demand that outstripped the existing housing supply in Arizona, California, Florida, and a number of the largest cities around the country. Even as the recession of 2001 led to increased mortgage foreclosures, the losses to lenders were dampened greatly because foreclosure sale prices were near or even above the value of the loans. Private lenders therefore began perceiving less risk from lending even to riskier borrowers. Meanwhile, the economic models used in the mortgage industry were showing that borrowers in high-price areas were able to meet their mortgage payments with much higher monthly payment to income ratios than the 36 percent maximum long recommended in the industry.
Seeing the rising housing prices, a higher share of the population saw home ownership as an opportunity to buy the houses as an investment opportunity. An increasing number of borrowers sought adjustable-rate mortgages with low initial rates with the expectation that they could easily “flip” the home for a higher price when the interest rate reset at a higher level. A number of mortgages, particularly alt-A and subprime loans, carried “teaser” rates that were incredibly low with the proviso of a substantial repayment penalty if the loan were refinanced. A borrower faced with this option still thought it was a good one if they expected to move and resell the house at the much higher price that seemed likely in the go-go atmosphere of the early 2000’s.
There were also opportunities for fraudulent activity on both sides of the transactions, although we have no way to really measure its extent. Some lenders misrepresented the terms of the loan, glossing over repayment penalties or the probability of an interest rate jump on the loan. Meanwhile, a number of borrowers misrepresented their income and assets. Some hid the fact that they had borrowed the down-payment on their homes.
The financial industry built an innovative series of financial instruments on top of these mortgages and other types of loans. A large number of lenders and financial institutions joined Freddie and Fannie in packaging up the mortgages into mortgage-backed securities (MBS’s) that they then sold to investors. The mortgage is just like a bond in that it represents a stream of payments, so the MBS is similar to a bond fund that fluctuates in price with fluctuations in interest rates and the likelihood of default or early repayment of the underlying asset. The MBS’s offered more liquidity to the housing markets because lenders like Countrywide had more options to sell the mortgage to others. They could then turn around and use the cash they received to make more loans. Investors saw the MBS’s as a new investment opportunity in an asset that was far more liquid than the real estate it is based on. Furthermore, a large portfolio of mortgages inside the MBS was a way to spread risk just as risk could be spread by buying stock and bond funds. Even better, the MBS’s offered a new form of investment that an investor could add to a portfolio of domestic stock funds, bond funds, exchange-rate funds, and foreign-investment funds. To the extent that movements in value were uncorrelated with the movements in value of the other funds, the purchase reduced the risk in the overall portfolio.
Unfortunately, the financial industry did not stop with the MBS’s. Many firms began packaging up large numbers of MBS’s and pieces of MBS’s into collateralized debt obligations (CDO’s) that could be sold to large institutional investors. Again, the goal was to offer new opportunities for investment and offer people a broader portfolio. The CDO’s, which took off in the 2000’s, contained “tranches” of MBS’s with different risk characteristics. Rather than having to buy a large number of MBS’s to diversify a portfolio, an investor could purchase a CDO that contained a range of risks. Over the course of the decade, CDO’s increasingly contained more tranches to fulfill this role.
As a means of reducing risk, financial institutions added another layer to this asset tower in the form of credit default swaps (CDS’s), which were essentially insurance contracts. Holders of CDO’s created contracts with firms like A.I.G. The CDO owner made payments to A.I.G., which in return promised to purchase the CDO at an agreed-upon price if the CDO declined in value. Given the extent of the tower built on the mortgages, one might think that it was hard for anyone to know what was in the CDS’s, the CDO’s, and the MBS’s, but that is not actually true. Credit-rating agencies were evaluating the quality of assets in the structure. The owners, typically large investment operations, were all using the analyses of a large number of smart people to evaluate the securities themselves. Meanwhile, other firms were evaluating the assets as well, as they considered purchasing them or wrote CDS contracts on CDO’s owned by other firms.
Problems arose as the rise in housing prices slowed. By 2007 housing prices were falling. The Case-Schiller housing-price index for 20 cities across the nation rose from 100 to 230 between 2000 and 2006. By October 2008 it had fallen back to around 170. It should be noted that the Case-Schiller index is dominated by the highest-flying housing markets in the country. An alternative index from the Office of Federal Housing Enterprise Oversight (OFHEO) shows that prices for homes with conventional and prime mortgages rose from 100 in 2000 to around 160 in 2007 before falling to 142 in November 2008.
Builders had responded to the rising housing prices with a surge in building, while the Federal Reserve had pushed the target federal funds rate from 1 percent in June 2003 to 5.25 percent in June 2006. Over the period the quality of the typical loan had deteriorated as the share of subprimes and alt-A loans expanded. As the housing-price rise slowed and prices began falling, the best-laid plans of many borrowers were interrupted. People facing the prospect of substantially higher monthly payments when the adjustable rate reset found it difficult to sell their houses at a profit. Subprime and alt-A borrowers began falling behind on their payments, and foreclosure rates began a steady rise from 1.7 percent in 2005 to 2.8 percent in 2007.
As each new cohort of adjustable-rate mortgages reset, the problems increased and housing prices in the markets where prices boomed the most began falling. As housing prices have fallen, the number of loans where the house value fell below the value of the loan has increased. Given that mortgages in most parts of the country are nonrecourse loans, some people have walked away from their loans, leaving the mortgage owner with the house. One of the remaining fears is that a large number of these borrowers who are “under water” will decide to follow suit.
The combination of rising foreclosures and falling house prices jeopardized a larger share of mortgages and the asset towers built upon them began to crumble. The forecasts of the downsides of foreclosure by ratings agencies and the owners of the CDS’s, CDO’s, and MBS’s turned out to be overly optimistic, and values of the assets began to slide. The stock price of Lehman Brothers and Bear Stearns, which was heavily invested in these financial instruments, started experiencing problems in the summer of 2007.
Meanwhile, A.I.G., which had sold CDS’s on CDO’s through 2005, began facing a substantial problem that it had not anticipated. The main risk it had anticipated was the default risk on the underlying mortgage loans, and its models seem to have done a good job of measuring that risk. What it had not anticipated was the possibility that it might have to post collateral to show that it could pay off the CDS’s if either the assets insured by the swaps declined or A.I.G.’s own corporate debt rating declined.
As the markets weakened between fall 2007 and fall 2008, A.I.G.’s CDS trading partners began demanding that A.I.G. post collateral. By August 2008, A.I.G. had posted $16.5 billion in collateral with more demands coming. Meanwhile, the Dow Jones Industrial Stock Index hit an all-time peak of 13,930 in October 2007 and then began a series of declines with temporary interruptions in April and the summer of 2008.
The Federal Reserve responded to these problems by purchasing bonds in open-market operations. To signal its efforts, it reduced the target for the federal funds rate from its peak of 5.25 percent in June 2006 to 4.25 percent in December 2007 to 2.25 percent in March 2008. Bear Stearns’s problems escalated to the point where in March 2008 Treasury Secretary Henry Paulson forced the sale of Bear Stearns to Chase for $2 per share (later adjusted to $10 per share) with an unusual guarantee of $29 billion to Chase against potential future losses on Bear Stearns assets in the sale. The Dow Jones responded with a temporary up-tick in April and then continued to slide, despite the Fed’s purchase cut in the federal funds rate target to 2 percent on April 30th.
The whole financial system seemed to be unraveling in September 2008. On September 9, Fannie and Freddie were put under conservatorship under federal regulators with up to $200 billion in liquidity backstops. Lehman Brothers was still failing and seeking buyers, but the buyers were seeking loan guarantees for the sale from the Treasury and the Federal Reserve. Secretary Paulson argued that the problems in Lehman’s balance sheet were too large and let Lehman Brothers go bankrupt instead on September 15. The next day he reversed field and the federal government took an 80 percent ownership stake in A.I.G. and offered $123 billion in credit to the struggling company.
In late September, Paulson, Federal Reserve Chair Benjamin Bernanke, the Congress, and the president negotiated a $700 billion bailout bill. The bill passed on October 3 and established the Temporary Asset Recovery Program (TARP). The same day, the Federal Deposit Insurance Corporation raised the limit on accounts eligible for deposit insurance from $100,000 to $250,000 through the end of 2009. On October 7, the Fed announced the unusual step that it would purchase commercial paper directly, and the next day the Fed led a global interest rate cut by lowering the target federal funds rate to 1.5 percent. Although the TARP’s announced goal was to purchase the “toxic” assets on the balance sheets of financial institutions, Paulson switched gears and used a large part of the first $350 billion to take ownership stakes in major banks. Another significant portion of the TARP funds went to shoring up A.I.G. with loans of up to $150 billion to meet collateral calls and to bail out Citigroup in November 2008. Meanwhile, the Fed has been pouring more money into the banking system by cutting the target federal funds rate to 0 to 0.25 percent in December 2008.
The Financial Meltdown from 1929 to 1933
The major similarity between the financial meltdown between 1929 and 1933 and today is the sharp drop in stock prices in the first year after the peak. The Dow Jones Index peaked around 380 in October 1929 and then declined 48 percent to 199 in October 1930. In the modern era, the Dow declined 37 percent between the October 2007 peak and October 2008. As the Great Depression continued, the Dow Jones Index fell to a low of 41.2 in the middle of 1932, a fate that seems unlikely today.
The similarities basically stop there. The housing boom of the 1920’s was more a quantity boom than a price boom. Rudimentary estimates of housing prices show a price rise on average of about 45 percent between 1920 and 1930, compared to the more than doubling in the current period. Claims of drops in lending standards during the housing boom are overstated because the typical borrower still could only borrow up to 40 to 60 percent of the value of the home — substantially less than the loan-to-value ratios on prime loans today. Home-ownership rates hovered around 48 percent, so nearly all the loans at the time of the loan would have been considered prime loans today. The typical loan required repayment of interest only over five years with a balloon payment of the principal at the end. In normal times, most borrowers rolled the loan over. As unemployment soared and financial institutions failed right and left, many borrowers struggled to find new loans. My sense is that the direction of causation in the 1930’s was from the economic troubles to the mortgage meltdown — the opposite of the current situation
The differences in the government reactions to the financial crises then and now are dramatic. Prior to 1933 there was no federal insurance of bank deposits and no savings and loan deposits. A number of states had experimented with their own deposit-insurance programs, but economic historians describe them as unsuccessful. Between 1920 and 1929, the Federal Reserve and state bank regulators allowed an average of 630 banks per year to suspend operations on the grounds that many were unsound and not worth saving. Most of the banks were small and thus on average less than 0.7 percent of bank deposits were located in the suspended banks in any one year. Between 1930 and 1933 the problems worsened dramatically as bank runs led to a reduction in the number of banks from 25 thousand to 17.8 thousand. The shares of deposits in suspended banks rose to 2 percent in 1930, 4.5 percent in 1931, 2.4 percent in 1932, and 11 percent in 1933.
Meanwhile, the Federal Reserve kept responding too little and too late in its role as a lender of last resort. The Fed did cut the discount rate, the interest rate at which it lent reserves to members of the Federal Reserve system. The discount rate was lowered in 11 steps from 6 percent in October 1929 to 1.5 percent in 1931. But the rate cuts failed to offset the problems created by rapid deflation. The Consumer Price Index stayed constant in 1929 and then fell 2.4 percent in 1930 and 9 percent in 1931. This meant that the real discount rate (the discount rate minus the inflation rate) rose slightly from 4.5 to 4.77 percent in 1930 and then jumped to 10.5 percent in 1931. The Fed raised the discount rate back to 3.5 percent in late 1931 before allowing it to fall to 2.5 percent for most of 1932. Yet the 10 percent deflation rate that year caused the real discount rate to rise to 12.5 percent. Basically, the deflation made the discount rate a nearly useless tool for the Federal Reserve.
This contrasts with the current period. The Fed has cut its target federal funds rate close to zero while the CPI in December 2008 was slightly higher than in December 2007.
In response to the current crisis, the Federal Reserve and the Treasury together have flooded the banking system with liquidity through open-market purchases of bonds, expressed a willingness to purchase commercial paper, and provided extensive guarantees of assets of various types of financial institutions. The Federal Reserve between 1929 and 1932 focused on maintaining the international gold standard and therefore did way too little to stave off the disasters in the banking system. Its boldest move was the purchase of one billion dollars in bonds in open-market operations in the spring of 1932. Yet the move, which could have done a great deal to stem the tide had it come in 1930, was far too little and too late in 1932 after three years of bank failures.
In February 1932, the Hoover administration established the Reconstruction Finance Corporation (RFC), which made a series of loans to troubled banks. In an effort to be transparent, the RFC announced which banks received the loans. The announcements, in turn, had the unfortunate effect of creating doubts about the soundness of the banks in the minds of depositors. Hence, the initial loan program met with little success. When the RFC began to take short-term ownership stakes in the banks, a precedent for the recent moves by Henry Paulson, the agency was more successful at preventing suspensions and failures.
State governments began trying to resolve problems with bank runs by declaring bank “holidays” that closed the banks until the depositors calmed down. Faced with a large number of runs on banks in February 1933, Roosevelt’s first move upon inauguration was to declare a federal bank holiday. Bank examiners then subjected each bank to a rigorous examination and allowed the strong banks to reopen with an implicit federal seal of approval. The government worked to merge weak banks into stronger banks and shore up their balance sheets. Current Treasury Secretary Timothy Geithner’s recently announced “stress test” for financial institutions seems to be a descendant of the bank holiday policy.
The difference in the timing of the government moves in the 1930’s and today is stark. During the debacles of the Great Depression, it took the Federal Reserve and the federal government three years, a rise in the unemployment rate to over 20 percent, and a decline in annual output to 70 percent of its 1929 level before these emergency moves were made. Before the crisis of 2008, the U.S. had far more institutional checks in place. Meanwhile, Federal Reserve Chair Benjamin Bernanke, a distinguished scholar of the Great Depression, and Treasury Secretaries Henry Paulson and Timothy Geithner had made numerous emergency moves even before the monthly unemployment rate reached 8 percent and year-to-year growth in real G.D.P. has turned negative.