|The Atlantic Monthly | October 2001
Notes & Dispatches
Thanks for Nothing
A former chief economist at the World Bank offers a case study in how
heavy-handed interference can break what doesn't need fixing
by Joseph Stiglitz
During the recent demonstrations in Seattle, Quebec City, and elsewhere
denouncing the International Monetary Fund and the World Bank, the press
tended to dismiss the protesters as fringe reactionaries ignorant of the
benefits of globalization. But although no one condones the violence in
Genoa, for example, it would be wrong simply to reject many of the
protesters' concerns. As the chief economist at the World Bank from 1997 to
2000, I have seen firsthand the dark side of globalization-how the
liberalization of capital markets, by allowing speculative money to pour in
and out of a country at a moment's whim, devastated East Asia; how so-called
structural-adjustment loans to some of the poorest countries in the world
"restructured" those countries' economies so as to eliminate jobs but did
not provide the means of creating new ones, leading to widespread
unemployment and cuts in basic services. The media and the public have since
become concerned about this dark side as well-globalization without a human
face, it is sometimes called.
However, the issue that is commonly debated-namely, whether we should be
"for" or "against" globalization-is not the salient one. As a practical
matter there is no retreating from globalization. The real issue is the
conduct of the international economic organizations that steer it. If we
continue with globalization as it has been managed in the past, its agenda
driven by the North for the North, reflecting the North's ideologies and
values, the future will not be bright. There will be a backlash in the
developing world and increasing conflict with the developed world. There
will be greater global instability and rising doubts about the value of a
market economy. Those doubts are already reflected in a pervasive hostility
toward the IMF in the Third World: in Thailand and Korea, for example,
ordinary citizens refer to their countries' debilitating recessions as "the
IMF." Yet well-managed globalization has enormous potential for improving
the lives of people in poor countries.
Events in Ethiopia offer a case study of the ways in which globalization can
go awry, and they highlight the need for reform. In March of 1997, barely a
month into my job at the World Bank, I went to Ethiopia to meet with Prime
Minister Meles Zenawi. Meles came to power in 1991, after a seventeen-year
guerrilla war against a bloody Marxist regime. His victory left him facing
seemingly intractable problems. Ethiopia, at the time a nation of 58 million
people, had a per capita income of around $100 a year. Droughts had killed
millions. Though he trained in medicine, Meles had studied economics at the
Open University, in England, and knew that only major changes in economic
policy could bring his country out of poverty. During our discussions he
showed a deeper and more subtle understanding of economic principles (not to
mention a greater knowledge of the circumstances in his country) than many
if not most of the international economic bureaucrats I would deal with in
the succeeding three years.
These intellectual attributes were matched by integrity: Meles was quick to
investigate any accusations of corruption in his government. He was
committed to decentralization-to ensuring that the center did not lose touch
with the various regions.
At the time of my arrival Meles was engaged in a bitter dispute with the
International Monetary Fund, which had suspended its program in his country.
At stake was not just some $125 million of IMF money but potentially
hundreds of millions of dollars in World Bank loans as well. Traditionally
the World Bank is reluctant to lend money unless the IMF certifies that the
country in question has a solid macro-economic framework. The provision is
well intentioned: history has shown that governments that cannot manage
their overall economy do not do a good job managing foreign aid.
The IMF is supposed to judge performance by results. Ethiopia's results
could not have been better. It had no inflation; in fact, prices were
falling. Output was growing steadily. Meles was demonstrating that with the
right policies even a poor country recovering from civil war and famine can
experience sustained economic growth. After years of struggle and
rebuilding, Ethiopia was beginning once again to receive assistance from
Judging by results, then, the IMF should have given Ethiopia an A+. And
there were other positive indicators, such as direct evidence of the
competence and commitment of the government. For instance, it had cut back
dramatically on military spending-a remarkable feat for a government that
had come to power by military means-in favor of spending to fight poverty.
This was precisely the kind of government to which the international
community should have been directing assistance. Yet the IMF had suspended
its aid. Why?
The Fund was worried, first, about the role of foreign aid in the
government's budget. A poor country like Ethiopia has two sources of
revenue-taxes and foreign assistance. The government's budget is balanced as
long as those revenues equal expenditures. This may seem like elementary
economics-but it is not IMF economics. Although Ethiopia's budget was
balanced, the Fund argued that the country's budgetary position was
untenable: what would happen if foreign assistance suddenly dried up?
Ethiopia should act immediately, the Fund argued, to prevent the possibility
of disaster. That meant cutting spending or raising taxes-a difficult action
in any country, but especially in a desperately poor one.
An argument against long-term reliance on foreign aid may be superficially
appealing, but in reality it dictated that Ethiopia's expenditures could be
paid for only by tax revenues. That is a fundamentally unsound policy: it
means that foreign aid does not lead to more schools or health clinics.
Instead the money is, in effect, simply added to reserves. Surely this was
not the intention of the international donor community. Surely donors wanted
to see those new schools and health clinics built in Ethiopia. Meles put the
matter to me passionately: he said that he had not fought so hard for
seventeen years to be told by some bureaucrat that he could not actually
provide improved services for his people once he had persuaded donors to pay
I cannot adequately describe the emotional force of his words or the impact
they had on me. I had taken the World Bank job with one mission in mind-to
work to reduce poverty in the poorest countries of the world. I had known
that the economics would be difficult, but I had not fathomed the depth of
the bureaucratic and political problems imposed by the IMF.
Meles provided an economically sound response to the IMF's concerns about
the stability of foreign aid: flexible spending. Building schools and
clinics does not require long-term commitments. If donors provided money to
build schools, Ethiopia would build schools; if they stopped providing funds
(as they had for a while), Ethiopia would stop building schools. But the IMF
would not be swayed.
he IMF had other bones to pick with Meles. In 1996 Ethiopia repaid a U.S.
bank loan early, using some of its reserves. The transaction made perfect
sense. In spite of the solid nature of its collateral (an airplane),
Ethiopia was paying a far higher interest rate on its loan than it was
receiving on reserves. But the United States and the IMF objected. They were
bothered not by the logic of the strategy but by the fact that Ethiopia had
undertaken this course without consulting the IMF. The IMF used this failure
to consult as one of the grounds for suspending its program. But why should
a sovereign country-one whose policies had convincingly demonstrated its
capability-have to ask permission of the IMF for every action it undertakes?
Another point of contention related to financial markets. Even after the
United States experienced the ruinous consequences of financial
deregulation, in the form of the savings-and-loan debacle, the IMF preached
the gospel of rapid deregulation around the world, to countries far less
able to withstand its negative consequences. Earlier deregulation in Kenya
had led to soaring interest rates there. Meles sensibly resisted such a move
in Ethiopia. But the IMF continued to insist on deregulation, and not even a
panel of scholars I assembled, most of whom supported Meles's position,
could budge the organization.
These episodes highlight two troubling aspects of the IMF's characteristic
behavior. The first concerns secrecy. Because so many of its decisions are
reached behind closed doors, the IMF leaves itself open to suspicions that
power politics, special interests, or other agendas unrelated to its stated
purposes are at play. For example, some critics questioned whether it was
just a coincidence that Ethiopia's early repayment deprived a U.S. bank of a
high-interest, secure-collateral loan on which it was making large profits
and that the United States was the country most vociferously protesting. A
second, closely linked aspect concerns the subordination of matters of
substance to matters of process. The processes themselves, with the numerous
conditions that are often attached, not only infringe on national
sovereignty but also tend to undermine democracy.
I returned to Washington from Ethiopia gravely upset by what I had seen.
During the following weeks I convinced the World Bank that the IMF's
position made no sense, and the Bank tripled its lending to Ethiopia. In the
ensuing years the country has been beset by political problems and war. It
is impossible to know whether some of its travails could have been avoided
or mitigated if aid had been more forthcoming.
The debate over globalization has already had an impact: the IMF's rhetoric,
and in some instances its actions, have changed. The IMF talks more about
poverty and participation than it used to. Last year it finally offered a
number of poor countries meaningful debt relief. Still, these are only
The biggest problems afflicting the IMF and other instruments of
globalization concern governance. At the United Nations five countries can
exercise veto power. In the IMF only one-the United States-can do so. At
both the IMF and the World Bank voting rights are allocated not according to
population but according to economic power, and the various countries'
representatives are typically finance ministers or members of central banks,
not officials with broader outlooks and concerns. Most of the debate about
reforming the international economic architecture has occurred within these
same small, elite circles. The voices of those most affected by
globalization are barely audible in discussions about how the table should
be reshaped and who should have a seat at it.