How Psychological Pitfalls Generated the Global Financial Crisis
Click to Print This Page
The root cause of the financial crisis that erupted in 2008 is psychological. In the events which led up to the crisis, heuristics, biases, and framing effects strongly influenced the judgments and decisions of financial firms, rating agencies, elected officials, government regulators, and institutional investors. Examples involving UBS, Merrill Lynch, Citigroup, Standard & Poor’s, the SEC, and end investors illustrate this point. Among the many lessons to be learned from the crisis is the importance of focusing on the behavioral aspects of organizational process.
Acknowledgments: I thank Mark Lawrence for his insightful comments about UBS; Marc Heerkens from UBS; participants at seminars I gave at the University of Lugano and at the University of California, Los Angeles; and participants in the Executive Master of Science in Risk Management program at the Amsterdam Institute of Finance, a program cosponsored with New York University. I also express my appreciation to Rodney Sullivan and Larry Siegel for their comments on previous drafts.
In this article, I present evidence that psychological pitfalls played a crucial role in generating the global financial crisis that began in September 2008. The evidence indicates that specific psychological phenomena—reference point–induced risk seeking, excessive optimism, overconfidence, and categorization—were at work. I am not saying that fundamental factors, such as shifts in housing demand, changes in global net savings rates, and rises in oil prices, were not relevant. They most certainly were relevant. I suggest that specific psychological reactions to these fundamentals, however, rather than the fundamentals themselves, took the global financial system to the brink of collapse.
To what extent did analysts see the crisis coming? In late 2007, four analysts (among others) forecasted that the financial sector would experience severe difficulties. They were Meredith Whitney, then at Oppenheimer; Dick Bove, then at Punk Ziegel & Company; Michael Mayo, then at Deutsche Bank; and Charles Peabody at Portales Partners (see Berman 2009). For example, in October 2007, Mayo issued a sell recommendation on Citigroup stock. Two weeks later, Whitney issued a research report on Citigroup stating that its survival would require it to raise $30 billion, either by cutting its dividend or by selling assets. More than any other analyst, Whitney raised concerns about the risks posed by the subprime mortgage market—and by the attendant threat to overall economic activity.
How timely were analysts in raising the alarm? As it happens, public markets had begun to signal concerns early in 2007. At that time, the VIX was fluctuating in the range 9.5 to 20, having fallen from its 2001-2002 20 to 50 range. On 27 February 2007, an 8.8 percent decline in the Chinese stock market set off a cascade in the global financial markets. In the United States, the S&P 500 Index declined by 3.5 percent, which was unusual during a period of relatively low volatility. Among the explanations that surfaced in the financial press for the decline in U.S. stocks was concern about weakness in the market for subprime mortgages.
In a book published in 2008, I argued that psychological pitfalls have three impacts that analysts should be aware of (Shefrin 2008b): First is the impact on the pricing of assets, particularly the securities of firms followed by analysts. Second is the impact on decisions by corporate managers that are germane to companies’ operational risks. Third is the impact on the judgments of analysts themselves.
The financial crisis contains illustrations of all three impacts. I use five specific cases to explain how psychological pitfalls affected judgments and decisions at various points along the supply chain for financial products, particularly home mortgages, in the crisis. The cases involve (1) UBS, a bank; (2) Standard & Poor’s (S&P), a rating firm; (3) American International Group (AIG), an insurance company; (4) the investment committee for the town of Narvik, Norway, an institutional investor; and (5) the U.S. SEC, a regulatory agency.
I use these cases to make two points. First, common threads link the psychological pitfalls that affected the judgments and decisions of the various participants along the supply chain. In this respect, a relatively small set of psychological pitfalls were especially germane to the creation of the crisis. The key mistakes made were not the product of random stupidity but of specific phenomena lying at the heart of behavioral finance.
Second, the major psychological lessons to be learned from the financial crisis pertain to behavioral corporate finance (Shefrin 2005). Many readers think of behavioral finance as focusing on mistakes made by investors, but issuers (corporations, governments, and so on) are people too and are just as prone to mistakes; behavioral corporate finance focuses on their side of the equation. Specifically, behavioral corporate finance focuses on how psychology affects the financial decisions of corporate managers, especially those in markets that feature mispricing. The key decisions that precipitated the crisis need to be understood in the context of behavioral corporate finance. Moreover, behavioral corporate finance offers guidelines about what to do differently in the future. For analysts, the general lesson to be learned is the importance of including a behavioral corporate perspective in their toolbox.
The five cases are intended to be representative. For example, UBS is hardly unique among investment banks, as the fates of Lehman Brothers, Merrill Lynch, and Bear Stearns illustrate. As discussed later in the paper, Citigroup engaged in strategies similar to those pursued by the investment banks. Indeed, in April 2009, the Washington Post reported that banks relied on intuition instead of quantitative models to assess their exposure to a severe downturn in the economy. This statement was based on interviews with staff at the Federal Reserve Bank of New York and the U.S. Government Accountability Office.
The source material for the five cases is varied. For UBS, the main source is an internal document from the firm itself. For the SEC, the main source material is an audio transcript from an SEC meeting. For the other three cases, the main source material is press coverage. Material from press coverage features both strengths and weaknesses. One of the key strengths is that information comes from the level of the individual decision maker, as revealed in interviews with decision makers and their colleagues. From a behavioral perspective, this level of detail is invaluable. One of the key weaknesses is that press coverage is less than fully comprehensive and is prone to distortion. In this regard, I discuss an example illustrating a case of distorted coverage.
FUNDAMENTALS AND CONTROVERSY
Mohamed El-Erian (2008) described a broad set of fundamentals related to the financial crisis. He identified the following three structural factors associated with changes in the global economy during the current decade: (1) a realignment of global power and influence from developed economies to developing economies, (2) the accumulation of wealth by countries that in the past were borrowers and that have now become lenders, and (3) the proliferation of new financial instruments, such as collateralized debt obligations (CDOs) and credit default swaps (CDS).1
El-Erian described how these structural factors worked in combination. Developing countries’ external accounts, which had been in deficit before 2000, switched to being in surplus after 2000, with the current account surplus rising to more than $600 billion in 2007. In contrast, the United States ran an external deficit of almost $800 billion in 2007. El-Erian explained that these imbalances permitted U.S. consumers to sustain consumption in excess of their incomes. He pointed out that U.S. financial markets facilitated this pattern by providing a way for U.S. consumers to monetize their home equity. And, he added, emerging economies purchased U.S. Treasury instruments, mortgages, and corporate bonds as they converted their trade surpluses into long-term investment accounts.
El-Erian described how these structural elements have affected financial markets. For instance, in 2004, the U.S. Federal Reserve Board increased short-term interest rates with the expectation that long-term rates would also rise. Instead, long-term rates fell—to the point where, in November 2006, the yield curve inverted. This phenomenon puzzled many investors at the time. El-Erian suggested that the inversion might have been caused by emerging economies purchasing long-term T-bonds in an attempt to invest their growing trade surpluses at favorable (high) interest rates. Those purchases drove prices up and yields down. During the 2005–06 period, the yield spread of 10-year over 2-year T-bonds fell from +125 bps to more than –25 bps.
Typically, yield-curve inversions are precursors of recessions. The U.S. stock market was robust in 2005 and 2006, however, with the S&P 500 rising from 1,200 to 1,400, which hardly signaled recession. Moreover, perceptions of future volatility, as measured by the Chicago Board Options Exchange Volatility Index, were at very low levels. As a result, El-Erian concluded that the bond market, stock market, and options market were providing mixed signals during a period he characterized as exhibiting “large systemic uncertainty."
The biggest puzzle for El-Erian is what he called “the ability and willingness of the financial system to overconsume and overproduce risky products in the context of such large systemic uncertainty" (p. 20). He suggested that as risk premiums declined from 2004 on, investors used leverage in a determined effort “to squeeze out additional returns" (p. 21). This behavior created a feedback loop that further depressed risk premiums, which, in turn, induced additional leverage. He went on to say:
Think of it: At a time when the world’s economies seemed more difficult to understand . . . and multilateral financial regulation mechanisms were failing us, the marketplace ended up taking on greater risk exposures through the alchemy of new structured products, off-balance sheet conduits, and other vehicles that lie outside the purview of sophisticated oversight bodies . . . . More generally, the pressure to assume greater risk, especially through complex structured finance instruments and buyout loan commitments, combined with overconfidence in a “just in time" risk management paradigm led to the trio that would (and should) keep any trustee, shareholder, or policy maker awake at night: a set of institutions taking risk beyond what they can comfortably tolerate; another set of institutions taking risk beyond what they can understand and process; and a third set of institutions doing both! (pp. 51–53)
Is the institutional behavior that El-Erian described (1) rational risk taking in which the outcomes simply turned out to be unfavorable, (2) rational risk taking responding to problematic incentives, or (3) irrational risk taking? I argue that the phrases “beyond what they can comfortably tolerate" and “beyond what they can understand and process" suggest that the answer is irrational risk taking. In this regard, it seems to me that El-Erian laid out the market fundamentals that preceded the crisis and then described behavioral patterns that represent irrational responses to those fundamentals: Rather than responding to a riskier environment by cutting back on risk, institutions took more risk.
Akerlof and Shiller (2009) argued that irrational decisions associated with the subprime housing market were central to the financial crisis. In this respect, consider some history. From 1997 to 2006, U.S. home prices rose by about 85 percent, even after adjustment for inflation, making this period a time of the biggest national housing boom in U.S. history. The rate of increase was five times the historical rate of 1.4 percent a year. As a result, the authors suggested, the sentiment of many people at the time was that housing prices would continue to increase at well above their historical growth rates. This belief supported a dramatic increase in the volume of subprime mortgages, especially mortgages requiring no documentation and little or no down payment. Later in this article, I discuss the time-series properties of loan-to-value ratios (LTVs), limited documentation, and 100 percent financing in the mortgage market.
Housing prices peaked in December 2006, when the Federal Reserve was raising short-term interest rates, and then declined by 30 percent over the subsequent 26 months. During the decline, many new homeowners (and some old ones who had engaged in repeated cash-out refinancings) found that the values of their mortgages exceeded the values of their homes. Some in this situation chose to default on their mortgages. Some homeowners had taken out adjustable-rate mortgages with low initial rates that would reset after a period of time to rates that were much higher. These homeowners were planning on refinancing before rates reset. Once housing prices began to decline, however, they did not qualify for refinancing. Many were unable to afford the higher rates and had to default.2
The mortgage product supply chain began with mortgage initiation by financial institutions such as Indy Mac, Countrywide, and Washington Mutual. It continued with such firms as Fannie Mae and Freddie Mac, which purchased and "securitized" mortgages, thus creating mortgage-backed securities (MBS). Next in the chain were investment banks, such as Lehman Brothers, Merrill Lynch, Citigroup, and UBS, which created and sold CDOs backed by the MBS. The supply chain also included financial firms such as AIG, which insured against the risk of default by selling CDS. The risks of both the products and the financial firms were rated by rating agencies, such as Moody’s Investors Service and S&P. At the end of the supply chain were the end investors, such as foreign banks, pension funds, and municipal governments, who ultimately held the claims to cash flows generated by the mortgages.3 Along the way, the supply chain was subject to regulation by various bodies, such as the SEC, the Board of Governors of the Federal Reserve, the Federal Reserve Bank of New York, and the Office of Thrift Supervision.
Taken together, the viewpoints expressed by El-Erian and Akerlof–Shiller suggest that financial institutions exhibited behavior inconsistent with the predictions of the Akerlof adverse-selection, “lemons" model, in which all agents use the information at their disposal to make rational decisions. The lemons model predicts the collapse of trade, resulting in, for example, a credit freeze when rational agents who perceive themselves to be at an information disadvantage assume the worst (e.g., all cars are lemons) when forming their expectations. In contrast to this model, despite the opaqueness of securitized asset pools, CDOs, and CDS—with their attendant information asymmetries—the subprime mortgage market did not collapse; it proceeded as if no cars could be lemons thrived.
Whether financial institutions behaved irrationally and whether the associated market movements reflected market inefficiency are the subject of controversy. Posner (2009a) maintained that institutions behaved rationally in light of the incentives they faced. He wrote, “At no stage need irrationality be posited to explain" the collapse of financial markets in 2008 and the deep recession in 2009. In an interview, Eugene Fama contended that past market movements are consistent with the notion of market efficiency.4
In his critique of Akerlof and Shiller’s 2009 book, Posner (2009b) stated, “But mistakes and ignorance are not symptoms of irrationality. They usually are the result of limited information". This line of reasoning leads him to conclude that the stock market increases of the late 1920s and late 1990s did not reflect mispricing and that in 2005 and 2006, people did not overpay for their houses in an ex ante sense.
Of Akerlof and Shiller’s (2009) contention that irrational decisions in the subprime housing market were central to the financial crisis, Posner (2009b) wrote, “They think that mortgage fraud was a major cause of the present crisis. How all this relates to animal spirits is unclear, but in any event they are wrong about the causality." Posner then provided his own list of what caused the crisis:
The underlying causes were the deregulation of financial services; lax enforcement of the remaining regulations; unsound decisions on interest rates by the Federal Reserve; huge budget deficits; the globalization of the finance industry; the financial rewards of risky lending, and competitive pressures to engage in it, in the absence of effective regulation; the overconfidence of economists inside and outside government; and the government’s erratic, confidence-destroying improvisational responses to the banking collapse. Some of these mistakes of commission and omission had emotional components. The overconfidence of economists might even be thought a manifestation of animal spirits. But the career and reward structures, and the ideological preconceptions, of macroeconomists are likelier explanations than emotion for the economics profession’s failure to foresee or respond effectively to the crisis. (Posner, 2009b)
Posner might well be correct in identifying problematic decisions in the regulatory process. Whether he is correct in his view that institutions acted rationally is another matter. One way of dealing with this issue is to examine decision making on a case-by-case basis, as I will do in this article, to identify the nature of the decision processes within financial institutions.5
This discussion needs to be based on a well-defined notion of rationality. In financial economics, rationality is typically understood in the neoclassical sense. Neoclassical rationality has two parts: rationality of judgments and rationality of choice. People make rational judgmentswhen they make efficient use of the information at their disposal and form beliefs that are free from bias. People make rational choices when they have well-defined preferences that express the trade-offs they are willing to make and choose the best means to meet their objectives. In financial economics, rationality is typically said to prevail when decision makers act as Bayesian expected-utility maximizers who are averse to risk.
1 For example, in an 18-month period beginning in January 2007, crude oil prices tripled—from $50 a barrel to $150 a barrel—and as the financial crisis unfolded in 2008, several sovereign wealth funds in Middle Eastern countries took positions in U.S. financial institutions that were in need of additional equity capital.
2 The proportion of all mortgage originations that were subprime increased from near zero in the early 1980s to 20.1 percent in 2006, although not monotonically. Chomsisengphet and Pennington-Cross (2006) described the history of subprime mortgage lending in the United States beginning in 1980 as follows: “Many factors have contributed to the growth of subprime lending. Most fundamentally, it became legal. The ability to charge high rates and fees to borrowers was not possible until the Depository Institutions Deregulation and Monetary Control Act . . . was adopted in 1980. It preempted state interest rate caps. The Alternative Mortgage Transaction Parity Act . . . in 1982 permitted the use of variable interest rates and balloon payments. These laws opened the door for the development of a subprime market, but subprime lending would not become a viable large-scale lending alternative until the Tax Reform Act of 1986 (TRA). The TRA increased the demand for mortgage debt because it prohibited the deduction of interest on consumer loans, yet allowed interest deductions on mortgages for a primary residence as well as one additional home." (p. 38)
3 As an example of foreign banks in the supply chain, consider that the Industrial and Commercial Bank of China bought $1.23 billion in securities backed by mortgages.
4 The interview with Fama titled “Fama on Market Efficiency in a Volatile Market" is at www.dimensional.com/famafrench/2009/08/fama-on-market-efficiency-in-a-volatile-market.html#more.
5 The five case studies are not intended to provide a comprehensive analysis of decision making in the financial crisis. Rather, the case studies are intended to provide examples of behavior that can be classified as rational or irrational. To deal with the issues raised by Posner (2009b), the focus is on financial institutions and government agencies, not on the behavior of individual homeowners. Nevertheless, many homeowners used subprime mortgages to purchase homes with the unfounded expectation that housing prices would continue to increase and that the homeowners would be able to refinance adjustable-rate mortgages in which the future interest rates would reset to a much higher rate.