In the summer of 2008, wheat and corn prices shot up across the globe. Pundits provided seemingly convincing explanations: grain was becoming scarce and thus more expensive because mainland Chinese were changing their eating habits and needed lots of it to feed their cattle—or perhaps because fear of oil shortages, combined with ecological fads, was leading consumers to adopt corn-based ethanol. Yet one year later, the Chinese are eating basically the same food as last year (feeding habits change very slowly), ethanol production is more or less at the same level, but the price of grain and corn on the Chicago market is back down again. How to explain the volatility of prices when production levels remain essentially the same?
The reason: grain or corn prices may at any point in time be driven more by speculation than by actual harvests. The rule applies to all transactions on financial markets, including oil, stocks, and derivatives. This is one of many examples that Rama Cont offers to describe how the real economy and the financial markets follow different rationales. In the short term—which can mean several years, in practice—the connection can be tenuous at best and difficult to model. If the connection were closer, Cont would know: he is at the forefront of the new science of financial modeling.
Finance itself is a relatively young field of research in which data have been available in large quantities only over the last 20 years. Thanks to electronic trading, it is now possible to quantify and analyze the fluctuations of financial markets on a large scale, but much interdisciplinary expertise is necessary to make headway in understanding it all.
In fact, Cont, an associate professor at Columbia University, conducts his research not in the economics department but in the School of Engineering and Applied Science. “My background is in theoretical physics,” he tells me. He discovered economics by accident while studying in Paris, where he emigrated from Tehran in 1987. “When I first became interested in economics, I was surprised by the deductive, rather than inductive, approach of many economists,” he says. Whereas in physics, researchers tend to observe empirical data and then build a theory to explain their observations, “many economic studies typically start with a theory and eventually attempt to fit the data to their model.” Such an approach might have been justifiable when financial and economic data were scarce, he believes, but with today’s wealth of information it is no longer acceptable.
Traditional economics, Cont argues, has failed to grasp the complexity and dynamic nature of financial markets. This outlook leads him to a distinctive interpretation of the current financial crisis. While the mainstream view explains the crisis by a lack of regulation, Cont believes that misguided regulations, often applied by not-too-smart regulators, were also a major factor. “Bear Stearns was perfectly compliant with regulations on the eve of its bankruptcy,” he observes. It had strictly followed the international banking rules imposed by the so-called Basel II Convention and, in fact, held $2 billion in excess of the amount of capital that the convention required for the bank to weather a major shock. But that capital, the regulations say, can consist of assets, such as bonds or shares, that aren’t liquid—that is, they have a market value but aren’t cash. In a panic situation, a bank needs cash to pay its creditors, and when confronted about its debts, Bear Stearns had only illiquid assets and no time to sell them. The bank went under the next day. The regulation seemed cleverly designed, Cont says, but proved useless in a real-life situation.
Why did the demise of Lehman Brothers generate worldwide financial turmoil? Again, Cont believes, existing regulations were at fault. As Lehman liquidated its portfolio, he explains, it sold off large quantities of stock, pushing overall stock prices down. This resulted in losses for other banks and increased the risk of their portfolios. To comply with Basel II regulations, these banks then had to reduce that risk by cutting back lending and by selling assets. They did both, bringing the stock exchange to its knees and drying up global credit.
Does Cont’s indictment of poorly designed regulations mean that he holds Wall Street bankers blameless? No: the bankers were greedy, he charges—though “greed in itself is not a new phenomenon.” But the Wall Street bonus system, as everyone knows by now, institutionalized that greed by compensating high short-term returns, so that traders and managers had an incentive to take more—and more dangerous—risks. That risk-taking, Cont says, should have been balanced by countervailing forces, such as boards of directors that, in theory, would represent the long-term interests of the banks and their shareholders. But Wall Street bank boards tended to have little say in the day-to-day management of the companies.
Nor did other countervailing forces intercede. Every bank had a risk-management unit, Cont points out, but the risk manager usually wasn’t a major figure, and the quantitative analysts in the risk units typically had little influence on major bank decisions. “Everybody knows the names of the CEOs of major investment banks,” he says. “But who has ever heard of the Chief Risk Officer of these institutions?”
Less sophisticated investors—fund managers and sovereign funds—relied on rating agencies for guidance, but didn’t get good advice, to say the least. The agencies—Moody’s, Standard and Poor’s, Fitch—utterly failed to anticipate the huge risks in the subprime market. “Either they truly ignored the risk of a fall in the housing market or they pretended everything was fine, in order to sustain the bubble and profit from it while it lasted,” says Cont. Ignorance probably played the larger role, he thinks. Rating agencies, like investors and regulators, rely on relatively simple models to forecast the risk associated with future market movements. Those models often assume a “mild randomness” of market fluctuations. In reality, Cont argues, what visionary mathematician Benoît Mandelbrot calls “wild randomness” prevails: risk is concentrated in a few rare, hard-to-predict, but extreme, market events (see sidebar).
Simple forecasts can also be mistaken if they fail to account for the actions of market participants themselves: investor strategies can influence prices, which in turn influence future strategies in a feedback loop that can cause considerable instability. Cont recalls the severe stock-market crash of October 1987, which seemed to strike out of the blue, since nothing significant was happening in the real economy. Subsequent research, though, blamed the crash in part on a new investment strategy, “portfolio insurance,” which a large number of fund managers had simultaneously adopted. Based on the famous Black-Scholes options-pricing model, this strategy recommended that fund managers reduce their risks by automatically selling shares whenever their values fell. But the approach didn’t take into account what would happen if many investors followed it simultaneously: a massive sell-off that could send the market plummeting. The 1987 crash was thus not provoked by events in the real economy but by a supposedly smart risk-management strategy—and the current downturn, of course, also derives at least partly from a global craze for a seemingly foolproof financial innovation.
Yet if the financial markets can become disconnected from the real economy and then generate storms that threaten and damage prosperity, Cont continues, they are nevertheless essential to a flourishing economy. They have dragged us under now; but over the last several decades, they have helped drive unprecedented global growth and innovation.
Investors in financial markets rationally pursuing individual profit, then, can produce outcomes that are globally negative. Doesn’t that contradict classical economic theory? “Both theory and empirical facts do tend to show that, on the financial markets, the Invisible Hand does not always lead to welfare-improving general outcomes,” Cont replies.
He offers another example: the diversification strategies that any wise investor should follow to protect himself from market risk. In the 1990s, in the name of diversification, investors poured money into various emerging markets and did well. When a currency crisis hit some Asian countries in 1997, however, emerging-market funds had to sell off Brazilian assets to dampen their losses on Asian investments, causing a sharp fall in Brazilian stocks. Rational individual choices in the financial markets had amplified a local shock in Asia into a systemic shock felt across the globe.
Is there a way to protect against these disruptions? “To regulate a financial market efficiently, we need first to understand its mechanisms,” Cont argues. Knowledge is thus the priority. Currently, regulatory bodies like the Federal Reserve and the SEC can determine the risk exposure of financial institutions, but only at a national level. The market, however, is global, and at present, we have no monitor to assess risk factors and their interdependence at the global level.
A first step toward rectifying this problem, Cont believes, would be to create a global risk observatory. In previous global crashes, abnormal concentrations of market participants began to engage in similar investment strategies—portfolio insurance back in the eighties and leveraged housing loans more recently. A global risk observatory would be in a position to observe such concentrations and raise a red flag. It would then be up to each national regulator to take these alerts into consideration or dismiss them. Cont doesn’t go so far as to promote the idea of an international regulator, which would potentially conflict with national sovereignty. In any case, the United States has already said that it would not accept such a body. An observatory is feasible, however: the data exist and need only be consolidated.
As for national regulators, Cont maintains that they should focus on ensuring financial stability and protecting against systemic risk, rather than worry about the health of individual financial institutions. Rules that apparently help reduce risk in individual firms can sometimes amplify systemwide risk, as with the Basel II regulations that required banks to reduce lending and sell assets after Lehman’s fall. Further, national regulation should encompass not only banks but all institutions, such as insurance companies and mortgage brokers, that have an impact on the financial system.
If realized, Cont’s proposals would not stabilize the financial markets overnight. But they could help us avoid the kind of regulations that tend to aggravate crises. His proposals may also seem modest, given the scope of the current crisis. “The American public is fond of gurus,” he says—and he isn’t one. He is only a man of science.
The Mandelbrot Line
At 86, creative and witty as ever, Benoît Mandelbrot has grown accustomed to the ups and downs of his scientific reputation. When the Dow Jones touches the sky, economists tend to forget his dark prophecies; when crisis strikes, he is suddenly rediscovered. These days, at his Cambridge apartment overlooking the Charles River, he gets more calls than ever. The last time publishers and conference organizers besieged him with so many requests was in 2000, after the Internet bubble burst—and before that, in 1987, when the stock market unexpectedly crashed. At the time, Mandelbrot seemed to be the only thinker able to explain why crashes could happen without any apparent economic reason. Financial markets, he argued, follow their own internal logic, not necessarily related to actual economic factors.
Mandelbrot has many disciples. Rama Cont, in the field of financial modeling, is one of the most noteworthy. The notion of “black swans”—unpredictable, rare, and massive-impact events—has made the trader and author Nassim Taleb famous, and it is pure Mandelbrot. George Soros’s apocalyptic economic scenarios derive from Mandelbrot, too, though haphazardly quoted.
Mandelbrot does not define himself primarily as an economist: he is, above all, a mathematician. Born in Poland, educated in France, a professor at Harvard in the 1960s and at Yale in the 1990s, with 30 years in between at IBM’s research center in Yorktown Heights, New York, the man is as unconventional as his career. In 1974, he became an instant celebrity with his theory of fractals—a fractal being “a rough or fragmented geometric shape that can be split into parts, each of which is (at least approximately) a reduced-size copy of the whole,” as he once defined it. On college campuses, T-shirts soon appeared adorned with fractal figures, which often appear in nature: think of snowflakes.
While most scientists try to understand and describe what is regular, repetitive, and hence predictable in nature, Mandelbrot mostly interests himself in the accidental, what he calls “monsters.” The fractal mathematics that he invented lets us see the hidden rules of apparent disorder, the order behind monstrous chaos. But can we deduce from the idea of fractals that any seemingly chaotic occurrence—like prices on the stock exchange—can be anticipated? Many financial economists think so and have tried to use mathematical formulas to master market volatility.
Mandelbrot dismisses these economists as hubristic and notes that all of their predictive theories have proved false. They commit two scientific errors, he says. First, they try to transport the theory of fractals into a field where it does not apply. (This is a common temptation—recall Marx, who tried to apply the laws of thermodynamics and Darwinian evolution to history and the social sciences.) Second, they do not start from the empirical data but instead build a curve first, assuming a logic behind volatility. When stock prices are shooting upward, these economists win media praise and even Nobel Prizes. When the market crashes, the same economists suddenly become less visible. It happens that one of them, Robert Merton, who shared a Nobel with Myron Scholes for a theory on predicting financial markets, lives in Mandelbrot’s apartment building. “We do not see him very often these days,” Mandelbrot says, tongue in cheek.
When one looks closely at financial-market data, as Mandelbrot has done throughout his life, unexplained accidents appear to be common—even the rule, so to speak. But if prices prove so erratic, there is no way for investors to become rich by incremental investments, Mandelbrot believes. Any portfolio can only follow the market, not beat it.
But some investors do make fortunes, right? “Yes, this is called ‘luck,’ ” answers Mandelbrot. As the financial market is prone to major accidents, one can strike it rich by being positioned luckily on the right side of the road. All major fortunes on the financial market have basically been made in a day, never on an incremental basis, he maintains. Soros comes to mind: in 1992, he earned $2 billion betting against the British pound. He got lucky and never did it again. Since that day, he has managed his fortune, and his faithful clients’ fortunes, by following the ups and downs of market volatility.
Mandelbrot suggests no alternative approach to the theories that pretend to predict volatility. “My role as a scientist,” he says, “is to demonstrate that available theories are plainly wrong. This does not mean that in my own turn I will invent a substitute snake oil that will make you rich.” The financial market is inherently a dangerous place to be, he emphasizes. “By drawing your attention to the dangers, I will not make you rich, but I could help you avoid bankruptcy. I am a doomsday prophet—I promise more blood and tears than windfall profits.”
Both scientists and the public like to believe in what Mandelbrot calls “mild randomness,” in which ordinary laws of probability apply, as is the case with many phenomena in nature. This happens not to be the case with financial markets. History—as well as Mandelbrot’s own empirical research, beginning with a famous 1960s study of the unpredictability of cotton prices—shows that the law of the financial markets is instead “wild randomness,” and that no mathematical model will ever be able to tame it.
Top Photo: Donte Tatum/iStock