October 12, 2008
Why Were Warnings Ignored?
Why the Warnings were Ignored: Too many False Alarms-Becker
I will first make a couple of comments on the present situation. It is not yet obvious that the recent steps taken by the Fed and the Treasury have been "failures". One cannot make that determination when the $700 billion plan has not yet been tried, and the Treasury seems to be changing its mind about the approach to use. I agree with the argument in the recent article in the Wall Street Journal by Paul Volker, the distinguished former head of the Fed, that the Fed and Treasury have enough tools to end the financial panic, and to get investment by banks started again.
To be sure, a recession is looming, but is this the biggest economic bust since the Great Depression? It is the biggest financial crisis since the 1930s, but the economic bust of the Great Depression meant a 25 percent unemployment rate for much of a decade, and sharp and sustained falls in GDP. While I expect unemployment to increase significantly from its present 6.1 percent level, and GDP to fall for a while, maybe sharply, there is little chance the downturn will approach anywhere near the 1930s levels. Perhaps it will be the most severe recession since then, although even that remains to be seen.
Consider what happened in Japan during the 1990s when it had a widely discussed major financial crisis that lasted for a decade. Unlike the Great Depression, Japan's real GDP did not fall much but was mainly stagnant: the real value of its GDP was 430 billion yen in 1990, 462 billion yen in 1995, and 482 billion yen in 2000. Nothing in this stagnating Japanese experience approached the economic devastation of the 1930s. I believe we have learned how to avoid such a huge economic disaster, although a decade of world stagnation would be quite bad.
To come to this week's blog topic, I also have a somewhat different take than Posner on why warning signals were ignored. The period since the early 1980s until the crisis erupted involved both rapid economic growth for most of the world, and unprecedented stability in this growth. Inflation rates were low and fluctuations in real output, as measured by the size and duration of recessions, were modest compared to the past. Economists and central bankers like Greenspan believed that we had learned how to keep inflation low, and also had the capacity to smooth out fluctuations in output and employment. The main Central Bank technique was inflation targeting, and a more general set of rules, called Taylor rules, that targeted a combination of the inflation rate and deviations of real output growth from its long term trend rate. These policies did work well for about 25 years, which created considerable confidence that they could handle future economic difficulties as well.
The second relevant development has been advances in financial instruments, such as derivatives, securitization, credit-backed swaps, and other even more esoteric instruments. These instruments seemed to work quite well in managing, spreading and even reducing the risk of the assets held by banks and other institutions. However, in the process they encouraged greater risk-taking ventures, as reflected by the large increase in the leverage-that is, in the ratio of assets to capital- of banks and financial institutions like Fannie Mae and Freddie Mac. What has been insufficiently understood is that the growing use of these instruments, and the growing leverage of financial institutions, created considerable aggregate risk for the system as a whole that could not be diversified away.
This combination of growing central bank confidence in its ability to iron out various wrinkles in economic performance, and the belief that the new financial instruments would help manage and reduce risk, blinded the vast majority of economists (I include myself), bankers, and government regulators to the vulnerability of the whole system. This vulnerability was especially important for aggregate shocks akin to a classical run on banks. When institutions are highly leveraged, they have great difficulty coping with a massive loss of confidence in the system.
While Roubini and others who warned about weaknesses in the mortgage market and other parts of the financial system deserve credit for their foresight, experts predicted numerous disasters during the past several decades that never happened. For example, after the huge one-day stock market collapse in October 1987, Business Week and other magazines and newspapers warned that a Great Depression might soon be coming. I argued against that view in a column I published in Business Week the same week as the market crash (reprinted in The Economics of Life by Guity N. Becker and myself). These dire forecasts turned out to be completely wrong. Similar highly negative but wrong economic forecasts were made during the internet bubble, after the Asian financial crisis of 1997-98 (on this see my post of September 21st), the aftermath of the 9/11 attack, and after other periods of economic distress. In an atmosphere where the world economy showed great capacity to withstand difficult shocks, it is not at all surprising that some forecasts of disaster that turned out to be more correct were ignored.
In addition, one should not minimize the great economic achievements of the past 25 years in the form of rapid growth in world GDP, low world inflation, and low unemployment in most countries. Perhaps these achievements will be overshadowed by a deep world recession, but that remains to be seen. If the impact of this financial crisis on the real economy is not both very severe and very prolonged, and time will answer that question, the combination of the past 21/2 decades of remarkable achievement, and the economic turbulence that followed, may still look good when placed in full historical perspective.
The Financial Crisis: Why Were Warnings Ignored?--Posner
When Becker and I blogged on the financial crisis last Sunday, the bailout had just been announced. The reaction of the stock markets and of senior government officials here and abroad suggests that the premise of the bailout--that the financial crisis is a liquidity crisis that can be resolved by the government's buying the assets of troubled banks at prices equal to the value the assets would have if there were a market for them (that is, if there were adequate liquidity to enable transactions)--was mistaken. The crisis appears to be one of solvency rather than (or perhaps along with) one of liquidity; banks, along with insurers of bonds and other securities, are undercapitalized and so, as I suggested last week, require a capital infusion rather than just a purchase of frozen assets.
All of which merely underscores the enormous cloud of uncertainty that has enveloped the crisis and left economists struggling to understand the causes, magnitude, future course, and cures of what is shaping up as the biggest economic bust since the Great Depression of 1929 to 1933. Last week's stock market crash may also reflect doubts about the government's competence to deal effectively with the crisis. There is a sense that its reluctance to take an equity stake in the banks reflects a doctrinaire hostility to public ownership.
But here is the biggest mystery of all: why was the crisis not foreseen? An article on the front page of the business section of yesterday's New York Times attributes that blindness to "insanity," more precisely to a psychological inability to give proper weight to past events, so that if there is prosperity currently it is assumed that it will last forever. This explanation is implausible--often people fail to adjust to change because they expect the future to repeat the past--and unhelpful, especially when one remembers that the academic specialty of Federal Reserve Board chairman Bernanke is the Great Depression.
We can get more help in answering the question of unpreparedness, or neglect of warning signs, from the literature on surprise attacks, notably Roberta Wohlstetter's great book Pearl Harbor: Warning and Decision (1962). As she explains, there were many warnings in 1941 that Japan was going to attack Western possessions in Southeast Asia, such as the Dutch East Indies (now Indonesia); and an attack on the U.S. fleet in Hawaii, known to be within range of Japan's large carrier fleet, would be a logical measure for protecting the eastern flank of a Japanese attack on the Dutch East Indies, Burma, or Malaya. Among the factors that caused the warnings to be disregarded are factors that may also have been decisive in the neglect of the advance warnings of the financial crisis now upon us: priors (preconceptions), the cost and difficulty of taking effective defensive measures against an uncertain danger, and the absence of a mechanism for aggregating and analyzing warning information from many sources. Most informed observers in 1941 thought that Japan would not attack the United States because it was too weak to have a reasonable chance of prevailing; they did not understand Japanese culture, which placed a higher value on honor than on national survival. Securing all possible targets of Japanese aggression against attack would have been immensely costly and a big diversion from our preparations for war against Germany, deemed inevitable. And there was no Central Intelligence Agency or other institution for aggregating and analyzing attack warnings.
Much the same is true of the warning signs of the current financial crisis. Reputable business leaders and economists had been warning for years that our financial institutions were excessively leveraged. In mid-August of this year the New York Times Magazine published an article foolishly entitled "Dr. Doom" about a perfectly reputable academic economist, a professor at New York University named Nouriel Roubini, who for years had been predicting with uncanny accuracy what has happened. In September of 2006--two years ago--he had "announced that a crisis was brewing. In the coming months and years, he warned, the United States was likely to face a once-in-a-lifetime housing bust, an oil shock, sharply declining consumer confidence and, ultimately, a deep recession. He laid out a bleak sequence of events: homeowners defaulting on mortgages, trillions of dollars of mortgage-backed securities unraveling worldwide and the global financial system shuddering to a halt. These developments, he went on, could cripple or destroy hedge funds, investment banks and other major financial institutions like Fannie Mae and Freddie Mac." By August of this year, when the Times article was published, Roubini's predictions had come true, yet he continued to be ignored. Until mid-September, the magnitude of the crisis was greatly underestimated by government, the business community, and the economics profession, including specialists in financial economics. Bernanke had repeatedly stated that it was unlikely that the mortgage defaults that accelerated after the housing bubble burst in mid-2006 would spill over to the financial system or the broader, nonfinancial economy. In May of 2007, for example, he said: "Importantly, we see no serious broader spillover to banks or thrift institutions from the problems in the subprime market." It has been more than two years since the housing bubble burst. One might have thought that that was enough time to enable the experts to discover that our financial system was in serious trouble.
Why were the warnings ignored rather than investigated? First, preconceptions played a role. Many economists and political leaders are heavily invested in a free market ideology which teaches that markets are robust and self-regulating. The experience with deregulation, privatization, and the many economic success stories that followed the collapse of communism supported belief in the free market. The belief was reinforced, in the case of the financial system, by advances in financial economics, and relatedly by the development of new financial instruments that were believed to have increased the resilience of the financial system to shocks. Borrowing and then lending the borrowed funds is inherently risky, because you have fixed liabilities but (unless you invest in risk-free assets such as short-term Treasury Bills) risky assets. But it was believed that the risks of borrowing had been reduced and therefore that leverage (the ratio of borrowing to capital) could be increased without increasing risk. Bayesian decision theory teaches that when evidence bearing on a decision is weak, prior beliefs will influence the decision maker's ultimate decision.
Second, doing something to reduce the risks warned against would have been costly. Had banks been required to increase their reserves, this would have reduced the amount they could lend, and interest rates would have risen, which would have accelerated the bursting of the housing bubble--and then Congress or the Administration would have been blamed for the fall in home values and the increase in defaults and foreclosures. In addition, it is very difficult to receive praise, and indeed to avoid criticism, for preventing a bad thing from happening unless the probability of the bad thing is known. For if something unlikely to happen doesn't happen (as by definition will usually be the outcome), no one is impressed; but people are impressed by the costs of preventing that thing that probably wouldn't have happened anyway. This is why Cassandras--prophets of doom--are so disliked. It usually is infeasible as a practical matter to respond to their warnings--but if the prophesied disaster hits, those who could have taken but did not take preventive action in response to the warnings are blamed for the disaster even if their forbearance was the right decision on the basis of what they knew.
The deeper problem is that it is difficult and indeed often impossible to do responsible cost-benefit analysis of measures to prevent a contingency from materializing if the probability of that happening is unknown. The cost of a disaster has to be discounted (multiplied) by the probability that it will occur in order to decide how much money should be devoted to reducing that probability. No one knew the probability of a financial crisis such as we are experiencing. Even Roubini did not (as far as I know) attempt to quantify that probability.
Which brings me to the last and most important reason for the neglect of the warning signs, because it suggests the possibility of responding in timely fashion to future risks of financial disaster. That is the absence of a machinery (other than the market itself) for aggregating and analyzing information bearing on large-scale economic risk. Little bits of knowledge about the shakiness of the U.S. and global financial systems were widely dispersed among the staffs of banks and other financial institutions and of regulatory bodies, and among academic economists, financial consultants, accountants, actuaries, rating agencies, and business journalists. But there was no financial counterpart to the CIA to aggregate and analyze the information--to assemble a meaningful mosaic from the scattered pieces. Much of the relevant information was proprietary, and even regulatory agencies lacked access to it. Companies do not like to broadcast bad news, and speculators planning to sell a company's stock short do not announce their intentions, as that would drive the stock price down, prematurely from their standpoint.
In any event, no effort to determine the probability of financial disaster was made and no contingency plans for dealing with such an event were drawn up. The failure to foresee and prevent the 9/11 terrorist attacks led to efforts to improve national-security intelligence; the failure to foresee and prevent the current financial crisis should lead to efforts to improve financial intelligence.
Of all the puzzles about the failure to foresee the financial crisis, the biggest is the failure of foresight of professors of finance and of macroeconomics, with a few exceptions such as Roubini. Some of the media commentary has attributed this to economics professors' being overly reliant on abstract mathematical models of the economy. In fact professors of finance, who are found mainly in business schools rather than in economics departments, tend to be deeply involved in the real world of financial markets. They are not armchair theoreticians. They are involved in the financial markets as consultants, investors, and sometimes money managers. Their students typically have worked in business for several years before starting business school, and they therefore bring with them to the business school up-to-date knowledge of business practices. So why weren't there more Roubinis? I do not know. And why, if not more Roubinis, not more financial economists who took the warning signs sufficiently seriously to investigate the soundness of the financial system? I do not know that either.