. 30
( 53 .)


deadline passed without action. The deadline was reset to 1999, and
was missed; to 2003, and was missed; to 2004, and was missed yet
again. In June 2004 the lead negotiator on DSM reform wisely decided
against setting a new deadline.82

Laying siege to the Crystal Palace
The politics of disclosure differed substantially in the case of the
World Bank. In the European Union, powerful states hostile to trans-
parency had been compelled to bend to smaller states with the
momentary ability to block the project of European integration. In the
WTO, powerful states supportive of transparency had been stymied
by the consensus rule. In the case of the World Bank, by contrast,
motive and power coincided: American legislators faced domestic
pressure to improve transparency and had the means to force com-
pliance by the Bank. Even under these favorable circumstances, how-
ever, movement toward transparency again encountered ¬rm limits.
The World Bank actually has several components. Two of the
most important are the International Bank for Reconstruction and
Development (IBRD) and the International Development Association
(IDA). The IBRD was established in 1944 to support postwar recon-
struction in Europe, but now provides loans to developing coun-
tries for major development projects. The IDA was established in
1960 to provide interest-free loans and grants to the poorest coun-
tries. Together, the IBRD and the IDA make the World Bank the
most important public development ¬nance agency for developing
By the early 1980s, the World Bank was beginning to suffer its
own crisis of legitimacy “ certainly among environmental and social
activists, and particularly in the United States. At issue was the Bank™s
support of projects undertaken by developing countries, the effects
of which on the environment and disenfranchised peoples could
be catastrophic. A coalition of U.S.-based activists “ led by groups
such as the Environmental Policy Institute, the Natural Resources
Defense Council, and the National Wildlife Federation84 “ decided to
draw attention to the worst cases of World Bank-funded mismanage-
ment. One of these was the Polonoroeste project, a massive Brazil-
ian program for settlement of Amazon frontier that resulted in reck-
less deforestation and the deaths of thousands of indigenous people.

Remote Control

Another was the Akosombo Dam in Ghana “ one of dozens of “big
dams” built with World Bank support whose impact on the environ-
ment and nearby communities had spawned a transnational antidam
movement.85 The aim of the U.S. activists, one of its leaders later said,
was to “lay siege to the Crystal Palace” “ by which he meant the obses-
sion with technocratic planning epitomized by the World Bank.86
The activists proved to be skilled in seizing opportunities pre-
sented by the United States™ fractured legislative process. Between
1983 and 1987, activists persuaded American legislators to organize
over twenty hearings in which witnesses testi¬ed to the damage done
by World Bank-funded projects.87 They also built an unusual coali-
tion of liberal Democrats concerned with environmental and social
causes and conservative Republicans concerned with the account-
ability of multilateral institutions. This coalition echoed the activists™
concerns (“Congress,” Larry Summers later said, “is the megaphone
of the NGOs”88 ), but also “ through its in¬‚uence over legislation that
regulated the executive branch™s involvement with the World Bank “
had the capacity to compel a Bank response.
Throughout the late 1980s, Congress experimented with legisla-
tive reforms intended to put pressure on the World Bank. One tech-
nique consisted of statutory directions to the director who repre-
sented the U.S. Treasury Secretary on the Bank™s executive board. A
1986 congressional directive advised the American director to encour-
age borrower countries to “fully inform” affected communities about
new projects.89 Congressional advice quickly became more pointed.
In 1989, Representative Nancy Pelosi “ a liberal Democrat from San
Francisco “ obtained an amendment to U.S. law that instructed the
American director to abstain from any vote on a project that would
have a signi¬cant environmental impact unless an Environmental
Impact Assessment had been made public at least four months before
the vote.90 The Pelosi Amendment (as it became known) was impor-
tant because decision rules within the World Bank gave substantial
power to the American director. The voting power of each national
director is weighted according to the amount of ¬nancial support
given to the Bank: The United States, contributing roughly 15 per-
cent of IBRD and IDA resources, also controlled one-seventh of the
Executive Board vote. (Pelosi, as a member of Congress, could also
call on the General Accounting Of¬ce to act as a monitor of Bank
compliance with the Amendment, as it did in a 1998 study that found

Blacked Out

shortfalls in World Bank practice regarding disclosure of environ-
mental impact statements.91 )
Members of Congress were also prepared to use a more powerful
tool for in¬‚uencing disclosure policy: control over the United States™
¬nancial contributions to the Bank. In 1985, Republican Senator
Robert Kasten bluntly warned World Bank President A. W. Clausen
that its inattention to NGO complaints put congressional support for
the Bank™s ¬nancing at risk.92 The threat was repeated in a Senate
report in 1989.93 In 1990, and again in 1992, Congress temporarily
withheld funding for Bank activities because of dissatisfaction with
its response to congressional requirements, including directions on
access to information about World Bank-funded projects.94
The IDA, whose ¬nances were “replenished” through appropria-
tions from member countries every three years, was particularly sus-
ceptible to congressional in¬‚uence. Historically, about one-¬fth of
IDA funding has been provided by the United States. During review
of a replenishment request in 1993, Democratic Senator Patrick
Leahy warned World Bank President Lewis Preston of Congress™s
“waning tolerance for a public institution supported with public
funds that denies the public access to relevant information.”95 Leahy,
discouraged by a recent report that found “fundamental failures”
in the Bank™s handling of the Sardar Sarovar Dam in India, also
pressed the Bank to establish an independent body to investigate
complaints about Bank mismanagement. When the Bank balked at
taking stronger action, Democratic Senator Barney Frank threatened
to delay approval of the IDA replenishment.96
The threat impelled the World Bank to establish an independent
inspection panel and adopt its ¬rst Information Disclosure Policy,
released in March 1994. It also began distributing some documents
through its website and through newly established “public informa-
tion centers” in its of¬ces in major Quad capitals.97 Congressional
pressure did not ease. The Bank made more commitments on
disclosure in anticipation of Congress™s consideration of another
replenishment round in 1999,98 and a revised policy was released
in September 2001. The next replenishment prompted promises of
further action,99 but no revision of the policy has been undertaken
since 2001.
The 2001 Disclosure Policy establishes a “presumption in favor
of disclosure” of World Bank documents.100 In fact, the reality is

Remote Control

quite the opposite. Bank policy, like the WTO policy, is a publication
scheme that proceeds on the assumption that documents are not
accessible unless they are explicitly listed in the policy.101 It excludes,
for example, most internal documents relating to the management
of the World Bank and internal papers relating to the formulation of
Bank policy “ subjects that would often be the subject of information
requests under national disclosure laws. There are some particularly
critical omissions. Draft versions of key documents that outline the
terms of the Bank™s ¬nancial assistance “ that is, the versions that have
not yet been approved by the Executive Board “ are not accessible.
While the approved versions of these documents may be released, it
may be too late for nongovernmental organizations to in¬‚uence their
terms.102 Nor were minutes of the board™s meetings made available
under the 2001 policy, a practice that allowed the Executive Board to
make decisions in “near total secrecy.”103 (In March 2005, the Bank
revised its position to allow the release of some basic information
about board meetings.104 ) While preparing the policy, the Executive
Board had apparently debated the wisdom of releasing more infor-
mation about its work, according to the Financial Times, which saw
leaked minutes of the board™s debate over the policy; the discussion
revealed a split between industrialized countries who supported NGO
demands for openness and emerging market countries who feared
that transparency would “invite external actors to become involved
in issues discussed by the board.”105
The policy is essentially a compact: the product of two decades of
dif¬cult negotiations between the World Bank, borrower countries,
the U.S. Congress, and nongovernmental organizations. It gives evi-
dence of the realities that have shaped the negotiations. For example,
the disclosure requirements that relate to Bank lending to its poorest
clients through the IDA under the 2001 policy were more demanding
than the requirements attached to IBRD lending106 “ a product of
the fact that IDA ¬nancing is more open to congressional in¬‚uence
through the replenishment procedure. The ability of middle-income
countries “ those borrowing from the IBRD “ to choose against dis-
closure of key lending documents was a key limitation of the 2001
policy. The dif¬culty was illustrated in 2002, when the Uruguayan
government exercised its prerogative to block the release of a letter
to the Bank that outlined the steps the country was prepared to take
as conditions for receiving a $252 million loan. The letter “ which

Blacked Out

summarized Uruguay™s commitments to cut public sector salaries,
pensions, and unemployment bene¬ts “ was later leaked, stirring con-
troversy in Uruguay. The Bank Executive Board had been told that the
Uruguayan government had undertaken extensive consultations with
nongovernmental organizations about its plans; critics complained
that the government™s assurances were “completely untrue.”107
In early 2005, the World Bank amended its policy in an effort to
remedy the inconsistency in disclosure rules for poor and middle-
income countries. Under the revised policy, there is a presumption
that key lending documents eventually will be released for all borrow-
ers. However, consistency has been purchased at the price of rigor: All
countries have now regained the ultimate discretion to block release
of key documents. Furthermore, they may choose to put sensitive
material in separate memoranda not affected by disclosure rules.108
In principle the policy of separating highly sensitive material is unob-
jectionable, if the test of sensitivity is fairly applied. But there is evi-
dence that World Bank of¬cials stretch the interpretation of disclo-
sure requirements “ a not surprising result “ and there is no effective
recourse (such as an ombudsman) for individuals who suspect that
information is being improperly withheld.109
Paradoxically, the World Bank now promotes more rigorous dis-
closure standards for the countries to which it lends. The Bank™s
recent emphasis on the need to foster “good governance” as a pre-
requisite to social and economic development has included a call for
improved transparency on the part of borrowing governments; it calls
“information access” one of the key elements of its “empowerment
framework” for the public in borrowing countries.110 Its research arm
promotes the adoption of disclosure laws that conform to “interna-
tional standards,” recognizing a right to government information and
establishing a procedure for independent review of decisions to refuse
access to information.111 In 2004, the adoption of a national disclo-
sure law was made a condition for loans granted by the Bank to two
countries “ Honduras and Nicaragua.112

A transparency revolution?
The story of openness at the World Bank™s sister institution, the Inter-
national Monetary Fund, seems at ¬rst glance to follow a similar path.
Also established in 1944, the IMF originally aimed to coordinate the

Remote Control

currency exchange rate policies of its member states and provide
¬nancial aid to countries dealing with balance-of-payments prob-
lems. Throughout its ¬rst four decades, says a former IMF of¬cial,

Institutional transparency was not high on the agenda of the
IMF. The IMF generally followed the practices of member coun-
tries, particularly their central banks and ministries of ¬nance,
which valued the con¬dentiality of their relationship with the
IMF. The IMF saw itself as a technical institution, accountable
to its member governments and with little need to explain itself
to the broader public.113

In the 1980s, however, the role of the IMF began to change, and so
did the public perception of its legitimacy. In 1982, the IMF responded
to Mexico™s debt crisis by providing assistance that was conditioned
on Mexico™s pursuit of signi¬cant ¬scal and economic reforms. It fol-
lowed by negotiating similar “structural adjustment” agreements with
other Latin American countries, and later with Russia and the for-
mer Soviet Bloc states. By the end of the 1990s, the IMF was involved
in structural adjustment programs in seventy countries.114 In many
countries, critics complained about the harshness of the reforms
required by the IMF; moreover, they complained about the secrecy
with which their own governments had negotiated agreements with
the Fund.115 Dissatisfaction mounted as the IMF responded to a
series of ¬nancial crises throughout the 1990s “ in Mexico again
in 1994, in East Asia in 1997, and in Russia in 1998. Ten thou-
sand protestors blocked the streets of Washington during an IMF
and World Bank-hosted meeting of ¬nance ministers in April 2000;
another 5,000 protested during the institutions™ annual meeting in
Prague in September 2000.
Popular protests, particularly in developing countries, were one
thing; dissent among policy makers within lending countries was
another. The public relations of¬ce of the IMF tabulated the num-
ber of times in which it was described as “secretive” in major press
outlets: The count had been insigni¬cant before 1997, but rocketed
upward afterward.116 The Fund (as the Wall Street Journal editorial-
ized in the ¬rst months of the East Asian crisis) was “one of the most
secretive institutions this side of the average missile base. . . . What
they do, or learn, or exactly what guides their decisions, is largely
kept secret. For the most part, the IMF has moved for decades

Blacked Out

in an off-the-record mist of internal deliberations and closeted
discussions.”117 Joseph Stiglitz, the former chief economist of the
World Bank, said that secrecy was the root cause of the bad advice
that the IMF had given to governments in crisis. If the Fund™s decision
makers had opened themselves to outside scrutiny, Stiglitz argued,


. 30
( 53 .)