by John Williamson, Peterson Institute for International Economics
Globalization and its Discontents by Joseph E. Stiglitz
New York: Norton
© Peterson Institute for International Economics
The title and the cover and the rhetoric and the repeated attacks on the IMF and the author's avowed sympathy for the anti-globalization protesters may give the impression that this book is just a polemic intended to capitalize on the author's Nobel Prize. If one can disregard his taste for playing to the gallery, however, one will find abundant good sense on how to use solid economic analysis to improve the way the world works for the benefit of the poor. Much of the substance is in fact rather reassuring to those of us who have regarded ourselves as mainstream and basically pro-globalization, but his critique of the IMF demands serious consideration.
Stiglitz notes that the original mission of the IMF was that of ensuring global economic stability. He argues repeatedly that it has essentially abandoned this mission, and is now driven instead "as though" its mission were to advance the interests of financial capital. He does not charge it with taking orders from Wall Street, but he does point to incidents like Stanley Fischer's move from the Number 2 position in the Fund to Citicorp to suggest that there can be potent incentives for the Fund management to pay heed to the interests of Big Finance, as well as pressures from the U.S. Treasury—an institution long regarded as subject to strong influence by the US financial sector. Such an objective function would be consistent with sundry actions of the IMF of which he disapproves, from its enthusiastic support of capital account convertibility to its efforts to restructure the financial sector in East Asia to its penchant for increasing interest rates to its willingness to furnish big loans that are to be used for bank bailouts while pressing countries to cyclically-destabilizing cuts in public spending on food subsidies to its aversion to inflation. (However, bankers would nowadays find the charge that it leans over backwards to help them collect their debts rather quixotic.) He charges it with a dogmatic dependence upon simplistic macroeconomic models in which all markets clear, and then makes mirth of the reluctance of the Fund's bureaucrats to trust the markets to set exchange rates. He criticizes its imperialistic ambitions, as displayed most recently in its insistence that it run the Poverty Reduction and Growth Facility (a lending facility for poor countries, whose management requires expertise on development that has not traditionally been the IMF's comparative advantage). He is highly critical of its support for shock therapy in the economies in transition. How many of his criticisms make sense?
As one of the rather few mainstream economists to have opposed the pressure for vastly premature capital account liberalization in the emerging markets during the first half of the 1990s, I can only welcome his belated support on that issue. The IMF surely deserves censure for its role in having helped create the East Asian crisis.
His critique of the Fund's policies on financial restructuring in East Asia during the crisis is also on the mark. If some banks had to be closed because their net worth was negative, one should not have left the impression that this was a first batch and there might be some more later, as happened in Indonesia: one should have given a firm guarantee that those that remain open will be supported through thick and thin, to avoid creating an incentive for a bank run. While it may have been necessary to close some banks, one should have recognized that good banks are institutions that have built up knowledge of their borrowers that could not easily be replicated, so that closing them was going to impose a long-term cost on the economy. (Stiglitz's discussion of this point draws heavily on his own research, and raises an issue that was largely overlooked during the crisis.) Selling them off to foreign banks may have had some advantages in terms of giving them deeper pockets to withstand crises, but it may also result in this institutional capital being allowed to deplete-and to force countries to sell off their banks at firesale prices in the midst of a crisis is almost guaranteed to create resentment. And while it is important for banks to hold adequate capital to deter opportunistic behavior, it was foolish to force them to restore a capital adequacy standard quickly in the midst of a crisis, since that could be done only by calling in loans and further intensifying the recession.
Stiglitz makes a powerful case for the costs that a policy of high interest rates can impose on the domestic economy, especially in countries like those in East Asia where the corporate sector was heavily dependent on bank loans, so that high interest rates can easily lead to bankruptcy (thereby creating more assets available for foreign buyers at firesale prices). But he is less impressive when contending that this policy did not even have a corresponding benefit in terms of strengthening the domestic currency. On p.111 he tells us that "higher interest rates [in Indonesia]…actually drove capital out of the country" and weakened the currency, but then on p.157 we learn that "IMF-driven high interest rates led to an overvaluation of the exchange rate" in Russia. Now maybe the relationship between the interest rate and currency values obeys some sort of a Laffer Curve, but the analytical point that one needs some such postulate to reconcile the two passages is not even recognized, let alone argued.
His criticisms of the procyclicality of the IMF's fiscal recommendations, and the contrast between its willingness to see funds used for debt service in contrast to its pressure to cut other spending during a crisis, raises several issues. Yes of course we would all like to see an anticyclical fiscal policy, but suppose that a country has not given itself leeway to be expansionary when times turn bad because it splurged when times are good and the only way it can finance a budget deficit internally is by printing money (as in Argentina today). Then one might want the Fund to lend so as to finance some modest budget deficit, and one might accept some moderate monetary expansion, but that may still leave a need for cutting public expenditure or raising taxes. In many cases the Fund's lending has indeed enabled countries to limit their cuts in public expenditure during crises, although one can of course discuss whether the scale should have been bigger. As regards its enthusiasm for maintaining debt service while other forms of spending were being cut, the hope was that this would quickly reestablish confidence and so limit the need for austerity. For quite a while that seemed to work reasonably well, but Stiglitz is right in arguing that in recent years it has ceased to work and therefore needs to be supplanted by some sort of international bankruptcy procedure. Perhaps the IMF was a bit slow in drawing that conclusion, but in my view Stiglitz gives it far too little credit for the initiative it finally did take, in Anne Krueger's speech, last Fall. But for all that, there have unquestionably been times and places-again East Asia leaps to mind-when the IMF sought unnecessary and counterproductive fiscal deflation.
How about the charge that the IMF suffers from an obsession with controlling inflation? No economist nowadays believes that there are long-run output gains to be had by running a higher inflation rate; Stiglitz's concern is that insisting on reducing inflation quickly when it is already reasonably low will have an output cost that is incommensurate with any benefit. In principle I share this position, but I have to admit that his and my concept of what is a "reasonably low" rate of inflation are pretty different: I would not want to put it higher than 4 percent, whereas I worry that he might be happy with 40 percent. But even if one regards Stiglitz as altogether too tolerant of inflation, it may be true that the IMF (not to mention the ECB) has sometimes gone to the other extreme.
Does the Fund rely on simplistic macroeconomic models that assume away all the interesting problems? I could not agree more with Stiglitz that the replacement of Keynesian economics by new classical models in many university curricula is a sad example of scientific retrogression: it is as though phlogiston had swept aside Lavoisier's theory of combustion in the chemistry textbooks. But he offers only assertions, not evidence, that IMF economists are more prone to this malady than others, or that they fail to out-grow their silly training once they find themselves on the job.
My own complaint about the Fund's policy toward exchange rates is less that it is too ready to support "interference" in the market (although I agree that it has at times erred in acquiescing in countries trying to defend the indefensible), but that it has become altogether too gung-ho about floating. One of the most stimulating aspects of Stiglitz's book is his take on the state-versus-markets debate. Rather than claiming that the issue is how to get the state out of economic life, or that every market failure demands government intervention irrespective of the dangers of government failure, he regards the typical problem as being how to get state and market to work together in a constructive partnership. Yet strangely he abandons this sensible posture when it comes to the foreign exchange market, which is prone to massive failure when the markets are left to themselves with no one with a responsibility for taking a long-term view, in favor of simple floating irrespective of how misaligned that may at times lead rates to be. It is a shame that he did not use this book to preach the importance of applying his general philosophy toward markets in the foreign exchange market, instead of reinforcing the Fund's current enthusiasm for floating.
Stiglitz also has a legitimate point in charging the IMF with an excessive willingness to take on new missions and add correspondingly to its conditionality, including its insistence on retaining control of what is now called the Poverty Reduction and Growth Facility. But it has partially heeded the critics on this issue, and made an effort in the last couple of years to cut back on the extent of structural conditionality in other facilities while bringing the World Bank into a more central role in the management of the PRGF.
Nowhere was the willingness to take on new roles for which it lacked comparative advantage more evident than in the central responsibility it acquired for managing the transition in the former Communist economies. And on no subject does Stiglitz direct such strong criticism at the Fund. By now it is pretty difficult to deny that rapid mass privatization was a disaster. More interestingly, he argues that gradual liberalization on the Chinese model would have been feasible in the former Soviet Union and Eastern Europe. The essence of this alternative model of transition (which he claims credit for having suggested to the Chinese, along with Kenneth Arrow) was that state enterprises retained responsibility for delivering their state orders at controlled prices to other state enterprises, while being allowed to sell their marginal output on free markets. This got incentives right at the margin without destroying the old economy, so that a new market economy could grow up and expand without the old system imploding. The results of this alternative strategy were massively better. What a shame that this was not the way the transition was managed in Eurasia as well as the Far East. But perhaps we should not blame just the Fund for this. The fact is that no one in 1989-91 was putting this set of ideas on the table. When critics spoke of gradualism, they created the impression that they wanted to decontrol prices one at a time, like things came off the ration in postwar Britain, which made absolutely no sense in a context where nothing except a couple of spices could be readily bought in the supermarkets. Perhaps the Fund and other Western advisors would have brushed Stiglitz's ideas aside if they had known of them, but it is hard to feel that they were the only guilty party when Stiglitz was not, so far as I am aware, making any effort to inject his proposals into the policy debate at the time.
So where does this leave his critique of the IMF? As I have suggested in several contexts, I do not see the Fund as being nearly as unresponsive to criticism as is portrayed by Stiglitz. (Well, as unresponsive as he portrays it most of the time, though there are the odd places where he acknowledges that Fund economists are also capable of intelligent professional debate.) On the other hand, the collective critique is more coherent and wide-ranging than has been developed before, and all the more so because he does not pretend that abolishing the Fund would be a sensible response. On the contrary, he says quite explicitly that if the Fund did not exist we would have to invent it, because the world needs an organization to help countries respond to macroeconomic crises. His critique is a challenge to the Fund's new Independent Evaluation Office rather than a set of denunciations that will get cheers from the anti-globalization crowd if they look behind the rhetoric.
But Stiglitz clearly believes that returning the IMF to something closer to its original mission is going to require reforms in its governance rather than simply internal debate within the organization as it exists at the moment. He sees this as needing a rebalancing of the voting rights between the main industrial countries and the developing countries, though he expresses skepticism about achieving any major changes in the short run. That is probably right, but it is a pity that he did not discuss other changes in global governance that might help promote the sort of world that he would like to see. In fact, the Zedillo Report (at www.un.org/reports/financing/) got a lot further in sketching the sort of changes that would be needed than Stiglitz does. (None of the Zedillo Report suggestions got as far as the Monterrey Consensus, if only because their squelching was the price that the United States demanded as the condition for President Bush to go to Monterrey.)
While the critique of the IMF is the most persistent and coherent theme running through the book, the attack on globalization "done the wrong way" is broader. The right way is, for example, to liberalize trade slowly (and without side-agreements like TRIPS); to privatize carefully; and to recognize that there are myriad different forms of market economy, including the Asian way in which governments rely on markets but take an active role in creating, shaping, and guiding them (p.10). The wrong way is to follow the dictates of the Washington Consensus, which is also variously referred to as market fundamentalism or neoliberalism. Since I invented the term Washington Consensus I used to imagine that I had some sort of intellectual property rights in defining what the term meant, but I was told long ago that I was very naïve to imagine that. So let me appeal instead to my credentials as a former student of Fritz Machlup to assert that there should at least be a measure of consistency in the way that terms are used. For example, a consensus should be a consensus. Stiglitz tells us on p.16 that the Washington Consensus is a consensus between the IMF, World Bank, and the US Treasury about the "right" policies for developing countries. Since Stiglitz was the World Bank's Chief Economist for 3 of the years about which he is writing, and therefore determined the Bank's economic policy insofar as it is coherent enough to have one, one would think that he must have been a party to any Washington Consensus. But of course the truth is that there never was a consensus that the set of policies he defines as characterizing the Washington Consensus-fiscal austerity, privatization, and market liberalization (p.53)-would provide any sort of panacea for the problems of developing countries.
What I originally intended the term to recognize was that in the 1980s there had been a major change in attitudes toward economic policy in developing countries (or actually in Latin America, the part of the world about which I as writing), and that an important part of the Reagan-Thatcher agenda had survived to win general acceptance, even though a good part of the original agenda was pretty nutty (monetarism, supply-side economics, minimalist government and therefore minimal taxes, capital account convertibility, and so on). As it happens, that is pretty much the intellectual position staked out by Stiglitz in this book, except that he doggedly refuses to recognize that many of the ideas that were widespread before 1980 were about as misguided as the market fundamentalist agenda that he attacks. Import-substituting industrialization, not gradual trade liberalization, was the conventional wisdom in developing countries.1 They should unilaterally be granted trade preferences, not bargain for trade concessions (as a direct result of which the trade regime became strongly biased against the interests of developing countries). Inflation provided a mechanism for increasing savings by exploiting the inflation tax. Markets do not work so state planning and public enterprises were needed to compensate. The danger is that Stiglitz's denigration of the Washington Consensus will serve to undermine the long-overdue consignment of this load of nonsense to the dustbin of history by those who do not realize what a narrow concept of the Washington Consensus he is using.
Perhaps Stiglitz hopes that the empathy he expresses for their position will win the anti-globalists to his side, so that they will endorse the sensible positions he espouses in favor of gradual trade liberalization, careful privatization, some version of the market economy, and so on. It would be wonderful if he were right in this, but I have more confidence in the robust position he stakes out on p.x: "…the need for politicians to behave more like scholars and to engage in scientific debate, based on hard facts and evidence. Regrettably, the opposite happens too often, when academics involved in making policy recommendations become politicized and start to bend the evidence to fit the ideas of those in charge." I have no complaints about his handling of evidence, but his frequent resort to polemical language is only one degree less unscientific.
1. Incidentally, is it really the speed of trade liberalization that is the defining issue, as opposed to the need for import liberalization to be accompanied by a competitive exchange rate and adequate market access?