Created: 1999-05-24
Last Modified: 1999-05-24
Go to
Brad DeLong's Home Page

Teaching | Writing | Career | Politics | Book Reviews | Information Economy | Economists | Multimedia | Students | Fine Print | Other | My Jobs

is always very welcome...

Review of Stiglitz book proposal: The Economic Role of the State

J. Bradford DeLong
U.C. Berkeley

May 20, 1999

Answering the standard questions-from-the-publisher is the easiest part of writing this review. The author has successfully accomplished his major goal--to present a book that both collects his essays of the past decade on the role of the state, and to provide a serious intellectual underpinning for the "third way": a view of the market and the state that neither puts excessive trust in markets nor in governments, but that sees government and market as strategic complements to be used to achieve social-democratic goals. This work is a very significant contribution: no one else has attempted an intellectual justification of the "third way" of such scope. There are, I think, no comparable books as of yet--certainly none of such scope.

The audience to which this book is addressed is that of professional economists. Although the pieces in the book are written at a variety of levels, much of the meat is contained in works of professional economic theory. This limits the audience for the book. I would give it to my graduate students in political economy. I would not recommend it to most journalists.

What else to say about this book manuscript is harder to determine. There is no point in going through an idea-by-idea critique of the chapters. These have all been published before, and one of the principal objectives of a book like this is to collect a bunch of already existing stuff in one place. Thus it is somewhat destructive to the project as a whole to demand revisions of individual chapters.


Theory and Application

So let me give a few thoughts that may guide Uncle Joe as he writes the two remaining pieces of this book--the introductory Essay 2, and the concluding Essay 20. And let me phrase them as aggressively as possible: in the battle between the third-wayers and the neo-liberals, I am 80% on Stiglitz's side. But there is that 20%. And I think that this book does have to do a much better job of dealing with the inevitable neo-liberal critique that will be mounted against it.

To begin with, Joseph Stiglitz is right when he says that the conventional wisdom is certainly that economists who work in government leave with a very strong sense of the magnitude and modalities of "government failure"--that those who suspected that markets were almost always preferred modes of social organization leave with their suspicions confirmed, and that those who believed that markets could always be improved by a judicious government intervention leave with their beliefs shaken.

Joseph Stiglitz appears to be an exception. He tries to set forth the intellectual core of the "third way"--a pragmatic approach that sees governments and markets as complements, with the government having a particularly important role in ensuring eqauality of opportunity (while resigning itself to enormous inequality of outcomes).

Yet I think that Stiglitz's targets are sometimes the wrong targets. He talks of the "Lucas" and "Coase" critiques--that whatever can be done by the public sector can also be done by the private, and that whatever is done by the public sector can be undone by the private sector. He claims to "dispense" with them by showing that they are not "general propositions." Yet I do not think that either Coase or Lucas would ever have claimed that they were.

The Lucas policy-irrelevance point and the Coase "theorem" hold true in very simplified and stylized models (the information-imperfections theory of the level of aggregate economic activity, the world with no "transactions costs"): break the simplicity in any of a wide number of ways, and the results disappear. Similarly, Stiglitz wants to argue that "the general argument for privatization is in fact based on highly restrictive assumptions" and that "the real cautionary tale... is that privatization will only achieve the claimed benefits if the preconditions for an effective market economy (such as competition) are put into place..." But once again the unanswered question is "compared to what?" Andrei Shleifer has an aggressive defense of Russia's privatization in his ms. Without a Map. Stiglitz's book would be, I think, much stronger if it took on not the claim that markets are perfect, but the claim that in the world we live in market faiilure is likely to be less damaging than government failure.

After all, theoretical critiques of market optimality are of dubious use in helping us set public policy. Stiglitz is correct in his claim that "competitive economies failed to produce efficient outcomes essentially wherever information was imperfect of markets were incomplete... that is, always." But once again he misses the target, for Stiglitz's real argument is not the theoretical one of Lucas and Coase, but the practical one of Friedman and Stigler: that in most situations government failures are likely to be more damaging than market failures, and that the situation in which government failure is likely to be most destructive is that of deciding when government intervention is practical and useful. Stiglitz says that this is not true. But throughout the ms. he begs the question of how government can be made more effective and how to reduce the consequences of its limitations. Essays 2 and 20 have to carry this load somehow--and if they do not it will be a much weaker book.



Writing as an economic historian, I want to condemn Stiglitz for his slighting of... history (what else?). For example, Stiglitz finds himself somewhat surprised at the emergence of robber barons: "rent-seeking... in the private sector. Given information imperfection and broader imperfections in market discipline, managers have enormous discretion in diverting resources.... In the absence of effective legal structures, such corporate theft can reach the status of grand larceny--as it did in some of hte former socialist economies.... A whole new vocabulary was invented--assets were 'tunneled' out of corporations by the controlling manager, leaving the shareholders with nothing but a shell."

This made me sit up, surprised. A new vocabulary? Asset-stripping by managers has a long and distinguished history. Where does Stiglitz--a Professor of Economics at Stanford University, on leave at the World Bank--think that the endowment of Stanford University came from? It came from the assets of the Central Pacific railroad, tunneled out of the corporation after the Civil War by the controlling manager, Leland Stanford. Some more attention to economic history might pay very large dividends.

I also found myself unsatisfied with the drawing of analogies between the S&L crisis and the East Asian crisis. In the S&L crisis depositors escaped whole--their commitment of funds to institutions going for variance seems to have been rational both ex ante and ex post. In the East Asian crisis a very large share of foreign and domestic investors took an enormous hit--their commitment of funds is hard to rationalize ex post, even though ex ante optimism about Asia's future was very bright.

In the S&L crisis funds were injected into the system on a large scale. In the East Asian crisis the IMF insists on getting paid back--so that there is no net injection of funds. It is hard to see how there can be moral hazard without a net injection of funds when you wind up in the bad state. And the only possible source of moral hazard I can see--that the taxpayers in East Asia ultimately fund the bailout of banks in East Asia--seems to me to be largely orthogonal to the debates over IMF policy.

Yes, it would be nice if East Asian countries avoided domestic sources of moral hazard. But I do not see what straightforward bearing this has on IMF action.

Perhaps attention to the last wave of crises back before World War I in the last era of globalization would pay dividends. (I am, in fact, trying to write such a paper for Brookings in the fall).


Institutions and Politics

I also believe that Stiglitz needs to think more deeply about the relationship between economics and politics. For example, he claims that the "S&L debacle in the United States" is largely attributable to the "failure of government to perform key regulatory rules: too little government, not too much government regulation." Yet this government "failure" to properly fulfill its regulatory mission had political causes--was itself the result of not government but political action. It seems to me that you cannot call for government to do more when the more that you want the government to do is blocked by the political process that controls the government.

Now Stiglitz recognizes this: he recognizes that there are "limitations of government arising from the force of special interest groups." He then claims that there are three directions one can take.

  • One is "wholesale contraction of the scope of government activity to... core functions"--the Friedman-Stigler approach.
  • The second is "to reform the political process." But this is not an area in which success is guaranteed, or even likely....
  • The third is "to look for areas in which government can intervene in ways that both increase economic efficiency and are politically sustainable..." And this is the direction that Stiglitz prefers.

Now phrased in this way few would object to what Stiglitz calls for (although Milton Friedman and the editorial page of the Wall Street Journal certainly would). But many would object to the answers that Stiglitz comes up with, and say that he gives only lip-service to the point that government failure is likely in a large range of situations and is extremely damaging. I think that if Stiglitz's book is going to have an enduring impact, he has to take on those who agree with his goals but think he is on the wrong road. He has to take on the "neoliberals"--his old friends Larry Summers and Stanley Fischer.



As with any intellectual and policy movement named by its adversaries, it is sort of hard to figure out what neoliberalism is. But I have a few ideas about just what it is that Stiglitz might make his target in Essays 2 and 20. Let me, once again, phrase them sharply and aggressively--not because I agree (completely) with this neoliberal point of view, but because it is a powerful argument that Stiglitz needs to engage...

Neoliberalism is many things. It is:

  • A counsel of despair with respect to the possibility of social democracy today (outside of the global economy's industrial core).
  • A counsel of hope with respect to the prospects for rapid market-generated economic development outside the global economy's industrial core--if governments adopt market-conforming policies.
  • A bet that improvements in transportation and communication--the shrinking world--"globalization"--gives us today an extraordinary opportunity to rapidly reduce global inequality by incorporating more and more people and more and more more regions into the global economy.
  • The only live utopian program in the world today.


A counsel of despair...

After World War II in Latin America, and at the achievement of independence elsewhere outside the global economy's core, there were high hopes that social democracy (or something further to the left) could be successfully instituted. And there were high hopes that such social democratic or socialist regimes would enable peoples living outside the core to cut a generation or more off of what had been a lengthy, bloody, and cruel three- or four-generation process of industrialization and democratization in northwest Europe and its settler colonies.

Social democratic or socialist governments would from the beginning establish strong redistributive social insurance states to severely reduce the income and wealth inequalities that had been characteristic of Bismarckian Germany or the Gilded Age United States. They would put into place the physical infrastructure to reduce infant mortality and disease that the aristocracies and bourgeoisies of northwest Europe had not thought profitable. They would spend money like water on education.

Moreover, they would use Keynesian policies to make sure that growth was free of the recessions and depressions that characterized industrialization in the industrial core. They would carefully manage their connections with the global economy--choosing the level of the real exchange rate, controlling imports so that imported goods were those of high social utility, preventing artificial drives for export success from raising the prices of necessities to the people, and establishing national independence from imperial capital.

They would nationalize at least the monopolistic commanding heights of the economy (if social democratic) or nationalize far more of the economy (if socialist) in order to take full advantage of the massive economies of scale in industry, and to make sure that investments in capacity and productivity that made sense from the social point of view were made--as they would not be if large-scale industry remained private, and if it proved difficult for the private monopolists to make a profit off of such investments. And all these economic decisions would be made by democratically-elected governments responsible to an electorate that had learned by exercising power what the trade-offs were and how to choose the best path forward that led by the quickest way to utopia.

By the end of the 1970s, however, it was clear to all except blindered ideologues that something had gone very wrong with social democracy at the periphery. (And that even more had gone wrong with really-existing socialism at the periphery.)

Stable political democracy proved far more to be the exception than the rule. Authoritarian rule by traditional elites, dictatorship by impatient army officers, and charismatic populist politicians ruling by virtue of carefully-prepared and carefully-staged plebiscites were much more common than were stable parliamentary or separation-of-powers democracies. Those aspects of social insurance that were installed seemed to do more to redistribute income from (poor) rural peasants to (richer) urban workers and (rich) urban civil servants than to moderate income and wealth inequalities. With some exceptions (many of them among really-existing-socialist countries) high government spending on education and on physical infrastructure seemed to produced less in the way of actual education or infrastructure--and more in the way of sweetheart contracts to the Minister of Regional Development's nephew's cement factory--than one would have hoped.

The nationalized commanding heights of the economy turned out more often than not to become employment bureaus for the politically well-connected: under Juan Peron in Argentina the number of employees of the (newly nationalized) Argentinian railroad system close to tripled, while the number of trains and the volume of goods carried fell. It seemed that while the state was superior as an instrument of social evolution, it was not very good as a bank, or as a stock exchange, or as a nursery for inefficient enterprises.

Too-great a reliance on Keynesian policies of demand stimulation turned out to generate high- and hyper-inflation, with the recessions that came with the crash of the monetary system proving arguably larger than the booms-and-busts Keynesian policies were supposed to avoid. Import restrictions turned out to limit imports not to those of social utility, but to those profitable to the import companies owned by the son-in-law of the Vice Minister of Finance--and to the Vice Minister himself. High real exchange rates turned out to do less to amplify the purchasing power of the country abroad than to artificially shrink exports, and to divert employment and investment away from sectors of comparative advantage.

There were exceptions: places outside the core where the social-democratic program was a stunning success. India managed to hang on to political democracy (albeit with disappointing economic growth). East Asia managed to limit corruption and maximize investment in infrastructure and export capacity, achieving the fastest economic growth rates ever seen in world history (albeit with disappointingly slow progress toward political democracy, and civilian blood on the hands of the military in massacres ranging from the thousands (in South Korea) to the tens of thousands (in Taiwan) to the hundreds of thousands (in Indonesia). In Brazil rapid growth in measured GDP was associated with the most hideous income distribution ever seen. Southern Europe alone managed to "converge" to the industrial core of northwest Europe, its ex-settler colonies, and Japan.

It seemed that the key was political democracy. With stable political democracy--in France, in Italy, even in Spain after the fall of Franco--social democracy could work and achieve great successes. Without political democracy it seemed that the chances of success were low (unless somehow the poorly-understood foundations of East Asia's low corruption could be duplicated). And it also seemed that the prospects for achieving stable political democracy on the periphery were rather low. After all, France experienced its first democratic revolution until 1789, yet depending on who you talk to it was not until 1871 or 1958 or 1981 that France truly achieved stable democracy.

Hence neoliberalism as a counsel of despair. As Marx wrote, the executive branchy of the modern state is nothing but a committee for managing the affairs of the ruling class--meaning, among other things, that a democratically-elected legislative branch turns the state into something better. But the prospects for stable political democracy in the periphery are slim. And thus the government becomes the tool of the ruling class--a ruling class that may be made up of army officers, or landlords, or urban elites, or those who profit as middlemen from the traditional channels of trade and exchange--who are not terribly interested in the success of social democracy or in rapid broad-based economic growth.

Hence the policy advice of neoliberalism as a counsel of despair: get the state's nose out of the economy as much as possible. When the state is neither an instrument of positive redistribution nor an instrument of growth-boosting investment, its interventions in the economy are likely to go strongly awry. And to the extent that a reduction in the economic role of an elite-controlled state can be required as a price for rapid incorporation of an area into the global economy, such a reduction should be required.


A counsel of hope...

Yet neoliberalism is not just a counsel of despair, it is a counsel of hope. The hope is that the prospects for rapid market-generated economic development outside the global economy's industrial core are very bright.

The prospects for rapid market-generated economic development are very bright for three reasons. First, the productivity gap between the periphery and the industrial core has never been larger. Second, governments now have a large number of positive examples to copy (as well as negative examples to avoid) in planning market-conforming development strategies. Third, investors in the industrial core now have the confidence and the resources to materially assist in peripheral development.

First, because the productivity gap between the periphery and the industrial core has never been larger, what Alexander Gerschenkron called the "advantages of backwardness" are now uniquely great. In 1870 an Indian or a Chinese textile-making entrepreneur could perhaps quadruple labor productivity by importing the modern capital goods of the British industrial revolution. Today any entrepreneur on the periphery has the prospect of being able to amplify labor productivity tenfold or more by investing in latest-generation or latest-but-on-generation capital equipment and factory organization. The stunning multiplication of productivity in Mexican automobile manufacture gives a clue to how quickly productivity can be amplified--if the capital is there to do so.

Second, all governments everywhere are now aware--from the examples of northern Europe, southern Europe, and East Asia--of those government interventions and policies that appear to be powerful boosters of growth. They are aware of the centrality of education (especially female secondary education) in accelerating the demographic transition. They are aware of the importance of making it easy for domestic producers to acquire industrial core technology (embodied in capital goods or not). They are aware of the importance of administrative simplicity and transparency. They are aware of the value of the transportation and communications infrastructure that only the government can provide. In those areas in which the government's nose should and must stay deeply embedded in the economy, even those states controlled by elites that have only a limited interest in growth and development now have many positive models to imitate.

Third, improvements in communications and transportation have made investors in the industrial core more willing than ever before to consider placing their capital in the periphery. The pre-World War I wave of international investment was largely limited to regions in which there were lots of white guys--guys who could play polo (never mind that polo in its original form was a sport played by central Asian nomads using a goat carcass as a ball)--plus the French geostrategic commitment to Russia as an ally against the Second Reich. The U.S. benefited enormously from Britain's willingness to lend capital to industrializing America in the years before 1900. The inflow of capital cut a decade or two off of the time it took the U.S. to industrialize (and crony capitalists like Jay Gould, Colis Huntington, and Leland Stanford took British investors to the cleaners as well). Now that investors in the industrial core are willing to commit their money to regions in which there are not lots of white guys, an opportunity to speed industrialization that used to be limited to a relatively narrow part of the non-European world is now open to many more--if their governments undertake the steps needed to reassure industrial-core investors, and if those who make economic policy in the G-7 can limit the destructive effects of the financial crises generated by the manic-depressive swings of opinion in Manhattan, London, Frankfurt, and Tokyo.


A bet on globalization...

And this is where the neoliberal view of the world is weakest: in its bet on globalization--its bet that a tightly-integrated global economy, with large flows of capital and goods (and, to the extent industrial core governments permit, of labor) is a richer and faster-growing global economy. John Maynard Keynes and Harry Dexter White would disagree with the proposition that large flows of capital are good: they would call them too dangerous to be risked.

Nevertheless, neoliberals today are more impressed with the gains from capital flows than the risks. The quadrupling of real wages in Indonesia from 1965 to 1997 would have been significantly lower without capital inflows which carried technology and enabled higher domestic investment (even though real wages in Indonesia have fallen by at least a quarter since 1997). Lowered transportation and communications costs have amplified the gains from expanded international trade by an order of magnitude over the past generation. And it is next to impossible to have large international flows of goods while excluding the possibility of large international flows of capital as well. Small changes in the timing of payments and in the extension of trade credit add up to large swings in the capital account.

Thus neoliberalism is not only a bet that increasing economic integration is a good thing--that an integrated global economy will see much more levelling-up than levelling-down--but that successful stabilization policy can be pursued by the G-7 on a global level. It is thus a claim about the economic environment (that the gains from globalization are large) and about the state capacity of the G-7 (that they can successfully carry out global-level stabilization policies).


The only live utopian program...

Perhaps not all of the principles of neoliberalism are correct.

Successful development in East Asia suggests that the counsels of despair are perhaps somewhat overstated: East Asia is an example if not of successful social democracy at least of a successful developmental state. On the other hand, as Lant Pritchett has observed, there is nothing worse than state-led development led by an anti-developmental state. And pending a better understanding of what has gone right in East Asia or much greater success in institutionalizing political democracy, the risks of a government turning away from the neoliberal path and attempting to duplicate East Asian developmental states appear very high. The belief that the opportunities for market-conforming development are now uniquely great appears to be almost certainly correct. But the jury is still out on whether the free-capital-flow part of "globalization" is a good thing: the odds are 80% that the G-7 do have the state capacity to successfully manage a world economy with large-scale capital flows, but there is a 20% chance that they do not.

Nevertheless, the neoliberal program is the only live utopian program in the world today.

Opposition to neoliberalism on the left seems to call in some cases for a return to effective autarchy. But if there is one lesson from economic history over the past hundred years, it is that there has been one decade--the 1930s--when economic autarchy was the road to relative prosperity, while there have been nine decades in which the more open to trade a country's economic policy, the faster has been economic growth. Opposition to neoliberalism on the left in other cases seems to call for a return to state control of the economy--to the pattern of Peronism or of the PRI--in the hope that this time the state will be not the tool of elites with little concern for development and growth but instead the faithful servant of the interests of the masses, and that this time social democracy of the kind that works in western Europe can be successfully installed in the periphery.

But that is not very likely. The state can be the servant of the people only if political democracy is well-established, and not always then. To place one's chips on the maximization of the power of a not-very-democratic and not-very-developmental state does not seem a promising path for either democratization or successful industrialization. It seems to embody a remarkable unwillingness to learn from world history since the end of World War I, and an ideological-blinded refusal to ever mark one's beliefs to market.

Opposition to neoliberalism on the right seems based on a fear that neoliberalism will bring with it a breakdown of social order: peasants will no longer fear landlords, workers will no longer be the clients of bosses or of the leaders of government-sponsored puppet unions, and voters will no longer respect the views of notables.

All the rest of us certainly hope that the right-wing opponents of neoliberalism are correct, and that neoliberalism this generation will begin structural transformations that will make social democracy on the periphery possible next generation. But then, not now, will be the time for the positive economic role of the state--as provider of social insurance, guarantor of equality of opportunity, moderator of market-generated income inequalities, and complementer of private economic activity--that Stiglitz advocates.



As I said, I am not sure that I agree with this neoliberal perspective. (I would rather be a social democrat; I'm just not sure that social democracy is yet possible outside the industrial core.) But I am sure that Stiglitz needs to take on neoliberal beliefs about the limited capacity and benevolence of developing-economy governments--and that he needs to face them head on, grabbing the bull by the horns...

Proposed Table of Contents for Joseph Stiglitz, The Economic Role of the State

Essay 1: The Economic Role of the State (1989)
Essay 2: [A Critique]
Essay 3: Privatization, Information, and Incentives (1987)
Essay 4: The Economic Role of the State (1991)
Essay 5: Discouraging Rivals: Managerial Rent-Seeking and Economic Inefficiencies (1995)
Essay 6: The Invisible Hand and Modern Welfare Economics (1991)
Essay 7: Information and Economic Efficiency (1994)
Essay 8: The Role of the State in Financial Markets (1993)
Essay 9: Perspectives on the Role of Government Risk-Bearing (1993)
Essay 10: Markets, Market Failures, and Development (1989)
Essay 11: Development Strategies (1991)
Essay 12: The Role of the Government in the Economy of Less Developed Countries (1996)
Essay 13: A Framework for a Development Strategy in a Market Economy (1996)
Essay 14: The Role of Government in Economic Development (1996)
Essay 15: Rethinking the Economic Role of the State: Publicly Provided Private Goods (1992)
Essay 16: [Bellagio]
Essay 17: Looking Out for the National Interest: The Principles of the Council of Economic Advisers (1997)
Essay 18: The Private Uses of Public Interest (1998)
Essay 19: Liberalization, Moral Hazard in Banking, and Prudential Regulaton (1998)
Essay 20: [The New Pragmatism]

Unpublished Book Reviews

is always very welcome...

Professor of Economics J. Bradford DeLong, 601 Evans Hall, #3880
University of California at Berkeley
Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax

This document:

Search This Website