George Soros' new book has breathtaking intellectual ambition, with a scattering of good insights about international economics and politics. But Soros' proven success as an international investor is more likely to attract readers than his stab at economic theory, as he himself notes. I confess to being most interested in the interrelationship of Soros' three roles -- investor, philanthropist, and public intellectual. When Soros the speculator helps force a currency into crisis, what does Soros the philanthropist think about the social or moral implications for the country under attack? The reader might anticipate one of five possible answers.

Stop me before I speculate again. This is what one most expects. As an intellectual, Soros clearly thinks that market values have overshadowed social values, that financial markets have become excessively volatile, and that some government intervention is appropriate. But his book contains no actual proposals, such as James Tobin's turnover taxes on foreign exchange, for calming excitable financial markets.

As an individual speculator, I am too small to have an effect. There is no logical contradiction between supporting tightened financial market regulations and simultaneously exploiting existing ones. This is particularly true for individual speculators whose small investments cannot affect market prices. Soros used to be in this category, but in a footnote he acknowledges that he no longer is. Now he moves markets.

My investments generally push markets in desirable directions; it is the activities of others that are potentially harmful. Soros could plausibly defend this claim. Hedge funds in general and Soros in particular make bets based on economic fundamentals, for example, against overvalued markets; in contrast, most bank foreign exchange traders simply follow the herd. The Quantum Fund's 1992 short sales of sterling arguably helped move Britain's exchange rate toward its proper equilibrium, and its 1997 baht selling should have alerted Thai authorities to a similar problem. This is how investors should operate -- making markets more efficient -- yet too often fail to. But Soros never explicitly says that financial markets would be fine if other investors behaved as intelligently as he.1

I devote a large share of my profits to charity, thus morally compensating for my speculation's negative effects. Soros could sustain this claim as well. His "open society" philanthropy in eastern Europe and elsewhere is extraordinarily generous, innovative, and efficient. But he does not attempt this sort of self-justification, perhaps to his credit.

My speculative self is completely divorced from my philanthropic self. This seems to be Soros' preferred answer. His two distinct personalities need not be consistent: "I try to be a winner as a market participant and to serve the common interest as a citizen and a human being. Sometimes it is difficult to keep the two roles separate." Elsewhere he claims to have synthesized them, which is what one would expect, considering that his analytical self writes about the interconnection of markets and social values: "I have made a conscious effort to integrate the various aspects of my existence, and I am happy to report that I have been successful." But how these personalities view each other remains a puzzle.


The reader can find a fascinating illustration of this puzzle in the chapter "A Real-Time Experiment." On August 9, 1998, Soros interrupts his manuscript, turns to the reader, and in effect confides, "I see things are heating up in Russia. Let me show you how it is done." He then chronicles in real time two weeks of the Russian crisis, including his conversations with Anatoli Chubais, Yegor Gaidar, David Lipton, and Robert Rubin.

Frustrated that his plan to save Russia is not eliciting sufficient reaction from top policymakers, Soros publishes in the Financial Times a letter that includes the recommendation "the best solution would be to introduce a currency board after a modest devaluation of 15 to 25 percent." Speculators read over their morning coffee that the financial wizard considers the ruble overvalued, and some infer that Soros must have a large short position in rubles. In other words, they may have assumed that Soros the speculator was speaking when it was really Soros the public intellectual. In any event, they flee from the Russian currency. On August 26, with the ruble devalued, Russian debts in default, and Prime Minister Sergei Kiriyenko fired, Soros admits that his real-time experiment is a failure. To his horror, he has precipitated the very crisis he set out to prevent.

Incidentally, the market manipulation some suspected Soros of is perfectly legal in international currency markets, although illegal in domestic securities. As it happens, Soros was long in ruble assets, not short, as he takes pains to stress. His funds reputedly lost $2 billion on Russia in August. Clearly, Soros' split personality is partly explained by his now limited involvement in the Quantum Fund's day-to-day decisions. One only wishes he had recorded in the book the conversation with his fund manager following the Russian default.


Economists will be infuriated by Soros' characterization of their financial market models. His dismissal of rational-expectations and efficient-markets theories is disarmingly straightforward: "I have to confess that I am not familiar with the prevailing theories about efficient and rational expectations. I consider them irrelevant and I never bothered to study them." One sympathizes. But since he purports to instruct his readers in what economic theory says, this attitude sits poorly.

The centerpiece of Soros' grand theory is his idea of reflexivity: "On the one hand, the participant seeks to understand the situation in which he participates. . . . On the other hand, he seeks to make an impact. . . . When both functions are at work at the same time I call the situation reflexive." The core application of his theory is to financial markets. Investors form expectations about a stock's or a foreign currency's future value based on the current market price. At the same time, the market price depends on investors' expectations as transmitted through their buy-and-sell orders. Both relationships must be taken into account simultaneously.

It is distressing for an economist to hear it claimed that this idea has been ignored by economists. The simultaneous determination of market prices and investor expectations features in virtually every recent study of financial markets, including orthodox theories based on economic fundamentals or rational expectations, as well as those that challenge the orthodoxy.

But what Soros means by reflexivity is something beyond this rather obvious simultaneity definition. He believes that markets from time to time carry prices far away from economic fundamentals. Any increase in price -- even if it originates in a random blip -- generates expectations of future price increases. Participants respond by placing buy orders, which drive the price up further -- a form of destabilizing speculation. To express this idea, theorists have developed models of bandwagons, rational speculative bubbles, fads, second-generation speculative attacks, multiple equilibriums, and bank runs. In contrast, the most orthodox theorists argue that normal investors respond to rising prices by expecting a future reversal, placing sell orders to take profits, and push the price back down. This is stabilizing speculation.

If one gives Soros the benefit of the doubt, one suspects that by reflexivity he means something more profound than that the typical speculator forms expectations extrapolatively. Contrary to most economists' theories, there is no such thing as the typical speculator. Various kinds of speculators operate with different models and thus produce wide-ranging expectations of future prices. After all, if there were not diverse expectations, there would not be such tremendous trading volume ($1.5 trillion per day in the world's foreign exchange markets alone). It is differences in opinion that make a horse race. The interaction of these varied speculators gives the market interesting dynamics.

Stabilizing speculators estimate currency or stock worth from economic fundamentals such as growth, interest, and inflation rates. In normal times, especially in the long run, their views roughly determine the market price. But many other participants eschew economic fundamentals, believing them to be irrelevant in the short term where their interests lie. Instead, many use technical analysis, which looks for trends within the currency or stock markets.

Soros argues that there has been a noticeable increase in the weight given to short-term technical analysis at the expense of fundamentals. Such a shift has important implications. When traditional models indicate that the stock market is overvalued, those investors who respond by selling stocks depress the market and thereby dampen the overvaluation. But if fundamentals are accorded less weight over time, the pressure to sell diminishes. As a result, market prices continue to rise, generating a self-confirming market bubble.

To be sure, the shift away from fundamentals is not wholly irrational on the part of the individual speculator. Investors are reasonably responding to a market rise that has repeatedly proven fundamentals models wrong and technical analysis right. Every month that the fundamentalists' predictions of doom go unrealized, the more their importance diminishes. It is not that investors necessarily decide that the market is correctly valued; it is that they lose money if they do not follow the herd. There is little use "being right" if everyone else persists in being wrong -- and makes a profit. By the time the bubble reaches its peak, fundamentals models have been proven wrong so many times that they have lost credibility. The market is operating far from equilibrium, outside the range that their models specified. Of course, this may be precisely the point that the momentum to technical analysis is spent and market sentiment begins to turn.


Financial markets do not always work as perfectly as orthodox economic theory assumes -- consider such recent disruptions as the 1982 international debt crisis, the 1992-93 crisis in the European exchange-rate mechanism, the 1994-95 Mexican peso crisis, and the 1997-98 Asian financial crisis. First, large inflows can suddenly give way to large outflows, with little news to explain why. Second, contagion sometimes spreads even to countries with relatively strong fundamentals and sound government policies. Third, the recessions that have hit emerging markets in such crises have been so large that it is difficult to argue that the system is working well.

With this in mind, most readers will come to Soros' book expecting a proposal for regulating international financial markets. But he offers little indication of which measures he favors. He argues that markets should not be left to themselves, but devotes only brief passages to specific controls on capital inflows or hedge funds. Soros gives no clue to how controls would be enforced or how they could weed out destabilizing flows while still "weeding in" desirable capital.

It is not as if there were no suggestions already: taxes on all foreign exchange transactions, taxes on all domestic securities transactions, Chile's tax penalties on short-term capital inflows, Malaysia's quantitative restrictions on capital outflows, Argentina's high reserve requirements on short-term bank borrowing, or regulation of hedge funds. But the debate has been sterile, consisting largely of arguments for and against the virtues of free and unfettered capital markets. Proponents of putting "sand in the wheels" point to inefficiencies, anomalies, bubbles, speculative attacks, and crashes. Opponents -- until recently almost all academic economists -- argue that any such measures would inevitably inhibit desirable capital flows. Neither group bothers to distinguish among the various proposals' details, which are quite different.

Soros offers a short but sympathetic discussion of the penalty on short-term capital inflows that Chile imposed in the early 1990s in its financial boom phase. Statistical evidence suggests that the lower the short-term or bank-intermediated foreign currency borrowing, the lower the probability of crisis. Bank flows are particularly susceptible to "moral hazard" problems, more than are foreign direct investment and longer-term securities. A mismatch of short-term bank liabilities with longer-term bank assets, such as real estate, leaves a country vulnerable to crisis. Chile's experience has suggested that the controls have succeeded in encouraging longer-term maturities while having little effect on the total amount of capital inflows. Although some analysts described Chile's 1998 removal of some inflow controls as evidence against their utility, it could be argued that Chile saw controls as a measure relevant only during a phase of excessive inflows. A promising alternative places higher reserve requirements on banks' short-term borrowing in foreign currency. Unfortunately, Soros' total discussion of capital controls is only two pages long.

Soros also offers a one-page discussion of hedge funds and argues for comprehensive regulation. But his proposed regulation is unspecified. Hedge funds, especially Soros', are becoming the popular image of speculation at its worst. Currency collapses from London to Kuala Lumpur are blamed on them. But in fact they are smaller collectively than other speculators (such as international banks) involved in foreign exchange markets or local residents taking their money out of crisis zones. More important, hedge funds tend to bet on economic fundamentals, not on trends, and are a source of stabilizing speculation. Assume that regulators succeeded in clipping the hedge funds in some way (although since some funds are registered offshore it is unclear how). The weight of stabilizing speculation might shrink and the deviation from fundamentals thereby grow.

Soros argues that, in two respects, some hedge funds may merit their reputation for volatility and points a finger at Long-Term Capital Management, contrasting it with Soros Fund Management. First, he argues that LTCM relied too slavishly on efficient-markets theory, especially rational-expectations statistical methods. These methods in effect assume that a recent period's pattern of price movements will hold in the future. But this approach can go badly astray if an unusual event has not occurred during that particular "baseline" historical period. An extreme flight to quality -- where U.S. Treasury interest rates fall while yields on riskier bonds rise -- is sufficiently rare that many investors were caught by surprise when it occurred in August 1998. Similarly, the rise in risk spreads and liquidity spreads simultaneously around the world was a relatively new event historically; it thus devastated hedge funds who had thought they were well diversified.

Second, Soros notes that some hedge funds became too leveraged, borrowing from banks to raise the stakes massively. Bets that go wrong, even if only because the market has not yet caught up with fundamentals, can threaten the entire financial system. But this calls for regulating the banks -- where the public interest is most clearly involved -- more than for regulating the hedge funds.


A disappointment is that The Crisis of Global Capitalism contains only a single sentence referring to Soros' famous January 1998 Financial Times article proposing an International Credit Insurance Corporation (ICIC). He offers no explanation for the proposal here, only a statement that now would be an ideal time to enact it since it would encourage much-needed flows to emerging markets. On the surface, this seems to contradict the rest of the book's belief that capital flows tend to be excessive. The Financial Times proposal would create an authority that would guarantee international loans for a modest fee. It could be part of the International Monetary Fund (IMF) or a new agency. The guarantee would provide a borrowing country access to international capital markets at prime rates, up to an amount set by the ICIC. Beyond that, creditors would have to beware.

Soros is not very explicit about timing. Probably it is only during the "bust" part of an economic cycle (e.g., from 1997 to 1999) that the ICIC would increase emerging markets funding. During the "boom" part of a cycle (e.g., from 1991 to 1996), the institution would presumably somehow limit lending. If Soros intends international support to be limited to the ICIC and support for loans to disappear as soon as national aggregate borrowing goes over the approved limit, this might work. But it might not, especially if the ICIC were to guarantee only specific loans, with additional loans left to the mercy of the marketplace. The famous "moral hazard" problem -- that you lend carelessly if you know that governments will bail you out, which many believe to be the crux of the current malady -- could remain. First, public statements that no loans will be bailed out beyond those insured would be no more credible than current statements that no loans at all will be bailed out. When a crisis hits a meritorious and systemically critical debtor, the international community will still feel pressure to rescue it. Second, countries would have an incentive to have the ICIC guarantee politically favored borrowings while reserving their most creditworthy projects for the market. In that case, the ICIC guarantees would only exacerbate the current moral hazard dilemma.

Admittedly, the moral hazard point has been overdone by those wishing to keep governments out of the rescue business. Airbags tempt people to drive a little faster, but that is no reason to forswear them. At the start of the East Asian crisis, the moral hazard chorus in Washington was deafening. As the crisis spread to Russia and Latin America, the arguments subsided a bit. Perhaps Congress feared being blamed for a global recession. In any case, it eventually restored to the White House the tools that previous administrations had possessed: a usable Exchange Stabilization Fund and U.S. contributions to the IMF.

Overall, Soros' preferred combination -- slowing down capital flows during booms and simultaneously preserving the international community's capability to cushion crises -- makes good sense. Still, one wishes he had offered a clearer and more detailed proposal.


The book's only overarching theme may be that it is tough work being brilliant when everyone else is not. It does little good if an individual speculator correctly diagnoses that a currency is overvalued and takes a short position accordingly if trend-following behavior by other speculators leaves the currency even more overvalued when the time comes to close out his position. Alas, corrections are not likely to come from politicians, intellectuals, or the media, who embark on self-reinforcing wrong turns as often as the financial community does.

Soros recognizes this and laments that enlightened thinking is no more likely to dominate the political process than it does the market process. Policymaking is riddled with unintended consequences, self-interest, voter disaffection, money-dominated politics, and candidates promising what voters want to hear. As a result, politics regularly carries policymaking far from the sensible path, and it takes policy there on bandwagons analogous to those afflicting the financial community. Ultimately, Soros concludes, "the political process is less effective than the market process in correcting its own excesses." Having critiqued the failings of both markets and politics, Soros leaves little reason to believe his implicit claim that the path of reform is straightforward and clear.

1 This reluctance is unlikely to be due to modesty. My two favorite sentences in the book are "But there is a big difference between Einstein's theory and mine" and "Marx proposed a deterministic theory that is diametrically opposed to my own."

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Jeffrey A. Frankel is the New Century Chair at the Brookings Institution. Until March 1999 he was a Member of the President's Council of Economic Advisers, and he will become a Professor at Harvard University in July.
  • More By Jeffrey A. Frankel