. 31
( 53 .)


“their folly might have become much clearer, much earlier.”118
Many members of the U.S. Congress, already uneasy in their sup-
port of organizations such as the IMF, echoed these complaints. “The
issue of transparency really goes to the heart of the legitimacy of the
IMF . . . and arguably, in the long term, its survival,” an advisor to
the Senate Banking Committee said. “If it is not seen that the institu-
tion as a whole has some measure of accountability and transparency,
then it is very hard to justify the extraordinary in¬‚uence that it exer-
cises and to hold it accountable in some way.”119 Congress had twice
given a statutory direction to the American representative on the IMF
Board to pursue the question of transparency,120 with little effect.
But in January 1998, Congress™s capacity to in¬‚uence IMF policy
improved substantially when the Clinton administration approached
it with a request for an $18 billion contribution to support the IMF™s
efforts at crisis management. Congress was hostile to the proposal;
three years earlier, it had rebuffed the administration™s request for
assistance in dealing with the Mexican crisis.
In April 1998, Representative Jim Saxton, the Republican chair
of Congress™s Joint Economic Committee, responded to the adminis-
tration™s request by proposing the IMF Transparency and Ef¬ciency
Act, which would deny support to the IMF until it agreed to institu-
tional reforms that included the publication of board minutes and
key lending documents.121 (Saxton also commissioned a General
Accounting Of¬ce study that concluded that it was impossible, given
publicly available information, to make a timely assessment of the
Fund™s ¬nancial position.122 ) Some of the disclosure requirements of
Saxton™s bill were included in the appropriations bill that eventually
provided the requested $18 billion in October 1998.123
In Fall 1998, the IMF™s managing director, Michel Camdessus,
gave a speech that warned about the institutional dif¬culties in
achieving greater openness. “The pace of change is largely in the
hands of the IMF™s members,” Camdessus cautioned. “Calls for
more IMF transparency are, in many respects, calls on the member

Remote Control

countries. . . . Once consensus is established, we will be enthusiastic to
proceed with the necessary adaptation of procedures and policies.”124
But the IMF wasted little time when the appropriations bill put $18
billion at risk: ten days after the bill™s passage, Deputy Treasury Sec-
retary Larry Summers reported to Congress that representatives of
the IMF™s major donor countries had agreed to adopt a more expan-
sive disclosure policy.125 (The IMF™s commitment, Saxton claimed,
“would not exist if Congress had not made an IMF reform effort over
the last twelve months.”126 )
The IMF announced substantial revisions to its disclosure rules
in April 1999,127 and made incremental changes to its policy in sub-
sequent years.128 But as the funding crisis passed, the IMF™s old dif-
¬dence about radical reforms resurfaced. “We have to strike a bal-
ance between openness and the members™ desire for candid and
con¬dential advice,” Camdessus™ successor, Horst Kohler, told the
National Press Club in 2000. “Civil society has serious questions,
and I take these seriously. But we should also be ¬rm. We have a
membership “ governments “ that is accountable. We cannot have
responsibility ˜transferred™ from these institutions to nongovernmen-
tal organizations.”129
The IMF had been put under unprecedented pressure to be less
secretive, a senior IMF of¬cial recalled in 2003. “To respond to those
demands, the IMF started to publish documents that had been kept
outside the public eye. We used to publish virtually nothing, now we
publish everything.”130 This was tantamount to saying, as the World
Bank had said, that there was a presumption of disclosure for doc-
uments. However, this was far from true. While the IMF had taken
important steps to improve openness, it had not recognized a general
right to information; instead, it had negotiated a publication scheme
for a limited number of listed documents. There was no process for
requesting access to administrative or other internal papers of the
IMF. The list of records covered by the IMF™s publication scheme
also omits critical “of¬cial” documents, such as the Board™s minutes
(still accessible only after a ten-year waiting period) and drafts of
lending documents sent to the Board for approval. The scheme pre-
sumes that countries will agree to the publication of the ¬nal versions
of these documents, but it does not require publication; and in many
instances governments, particularly those of developing countries,

Blacked Out

have not agreed.131 Governments may also negotiate about the word-
ing of publicly disclosed agreements and ask for politically sensitive
material to be contained in inaccessible “side letters” to the Fund.
(In 2003, Argentinian activists ¬led a legal complaint in national
courts alleging that a side letter to a recent IMF agreement contained
a concession on the raising of utility prices.132 The IMF refused to
con¬rm or deny the existence of the letter.133 ) IMF staff are told to
avoid language “that would exacerbate domestic political challenges
to implementing reforms.”134
The Fund™s new disclosure rules represented a signi¬cant and
important change in practice. Yet IMF spokesmen represented its
new attitude on transparency in more dramatic terms. The Fund had
undergone a “transparency revolution,” its chief economist, Stanley
Fischer, said in 2001.135 Another IMF of¬cial agreed: The commit-
ment to transparency constituted an “understated revolution, . . . a sea
change” in the way the Fund does business.136
This was not simply hyperbole: Rather, it was a testament to the
plasticity of the concept of transparency itself. The revolutionary
aspect of changes in transparency at the Fund referred mainly to
the extension of its own effort to monitor the behavior of its member
states. This was motivated by a widespread perception that the ¬nan-
cial crises of the 1990s had been caused by ignorance about the state
of ¬nancial sectors in the crisis countries, and that governments in
those countries had been (in the words of a senior IMF of¬cial) “eco-
nomical with the truth” in reporting their ¬nancial positions.137
The result was an effort to improve the IMF™s capacity to col-
lect information about ¬nancial and regulatory conditions in mem-
ber states. In 1977, the IMF began a routine of completing regular
reviews of the domestic policies of each member state that might
affect exchange rates. Authorized by Article IV of the IMF™s Articles
of Agreement, the scope of these “surveillance” exercises (the IMF™s
own phrase) broadened substantially in the 1990s, to include more
detailed scrutiny of each country™s ¬nancial sector and institutional
arrangements that might make a country vulnerable to crisis.138 In
1999, the IMF added another routine, producing “¬nancial system
stability assessments” for member states that include reports on the
extent to which countries conform to internationally recognized stan-
dards in the management of ¬scal and monetary policy, banking and
securities regulation, and corporate governance.139 In 1999 an IMF

Remote Control

review group called this “a potentially major expansion” of its surveil-
lance functions.140
These measures were all aimed at improving transparency, but
not of the IMF itself. Rather, they were intended to address the threat
of economic instability posed by liberalized capital ¬‚ows. The IMF
had no interest in new restrictions on capital ¬‚ows; on the contrary,
it regarded capital ¬‚ow liberalization as one of its main objectives, a
point it reaf¬rmed at the height of the Asian ¬nancial crisis.141 But
investors who were ignorant of the true conditions of the economies
in which they were investing were prone to dangerous herd behav-
ior. More extensive surveillance by the IMF would alert policy mak-
ers to conditions that might trigger investor panic, and at the same
time give investors the information to make more rational and less
¬ckle decisions.142 IMF of¬cials themselves characterized their new
initiatives as a form of informational regulation “ “a new kind of
r´ glementation . . . [that will] reduce the risk of abrupt changes in mar-
ket sentiment through greater transparency.”143
The effectiveness of these surveillance efforts is heightened if the
products of surveillance are accessible, so that lenders and investors
can make decisions that re¬‚ect risk more accurately. This has pro-
duced added pressure from advanced economies, as well as the
IMF itself, for publication of key surveillance documents. In prac-
tice, however, developing countries have often refused to consent to
disclosure.144 A 2003 report by the IMF™s evaluation of¬ce observed
that three countries caught in recent ¬nancial crises “ Indonesia,
South Korea, and Brazil “ had refused to publish the IMF™s most
important surveillance document; the of¬ce observed that the IMF™s
in¬‚uence would have been strengthened if the document had been
published, to “promote better risk assessment by private investors
and lenders.”145 Efforts to persuade less advanced economies about
the merits of disclosure have been accompanied by warnings that the
market may punish states that do not cooperate.146

Common language, separate purposes
The contest over transparency in these four institutions “ the EU,
WTO, World Bank, and IMF “ gives a sense of the dif¬culty that will
confront advocates of openness as the power of the traditionally struc-
tured state is diffused, either domestically to private or quasi-private

Blacked Out

actors, or to supranational or international organizations. The mul-
tiplication of centers of authority means that one large battle over
openness is replaced by many smaller battles.
In the international arena, the situation is even more complex than
suggested here. There are many other international organizations
that wield in¬‚uence but are not bound by disclosure policies compa-
rable to those imposed on national governments “ ranging from the
obvious cases (such as the United Nations and its agencies) to a host
of little known organizations (such as the Codex Alimentarius or the
Bank of International Settlements) that play critical roles in encour-
aging convergence in national regulatory practices.147 Progress in
advancing disclosure rules in this sphere will be uneven, contingent
upon the structure of decision rules in each body, the predispositions
of in¬‚uential states, and the extent to which an organization ¬nds
itself embroiled in controversy. Experience shows us that the norm
of diplomatic con¬dentiality will prove durable “ even when organi-
zations become the object of extraordinary public protests.
The contest over openness in the international ¬nancial institu-
tions “ that is, organizations like the WTO, World Bank, and IMF “
also reminds us about the dangers of accepting what might be called
the na¨ve view of transparency, in which openness is regarded as
a single commodity, and an unalloyed good. Transparency can be
employed as a tool by different players for dramatically different pur-
poses. In fact, the ¬ght over transparency in the WTO and IMF might
be said to present a clash between two doctrines of transparency.
One “ the neoliberal doctrine “ deploys transparency as a tool for
advancing the project of global economic liberalization. The neolib-
eral doctrine is ¬rmly embedded in WTO trade agreements and IMF™s
surveillance policies. Most immediately, it serves the states and cor-
porations in a position to exploit the opportunities presented by such
liberalization.148 Opposed to this is a rights-based doctrine of trans-
parency, which pursues disclosure as a tool for protecting the political
and economic rights of citizens affected by the process of liberaliza-
tion, and which is manifested in policies intended to lay open the
decision-making processes of the organizations themselves so that
their decisions can be more easily in¬‚uenced.
There is a second sense in which the na¨ve view of transparency is
rebutted. As developing countries have pointed out, stakeholders in
the ¬rst world “ nongovernmental organizations as well as business

Remote Control

stakeholders “ are in a much better position to exploit the oppor-
tunities for in¬‚uence that are created by improving transparency in
international organizations. Transparency is not a neutral concept,
leaders of some weaker states argue. Rather, it shapes the balance of
forces that in¬‚uence policy, and it may do this in ways that are unfa-
vorable to the interests of their citizens. Indeed, one can imagine the
frustration of a policy maker in a weaker state “ subjected on the one
hand to a system of surveillance promoted by dominant states to
encourage economic liberalization, and subjected on the other hand
to a rights-based rhetoric of transparency (most forcefully articulated
by interest groups of the same dominant states) that complicates their
ability to negotiate effectively and to implement policies at home.
Assertion is not proof, however, and it is equally possible to imag-
ine reasons why this account may be mistaken or misguided. It may
underestimate the extent to which the community of nongovern-
mental organizations has internationalized and the manner in which
groups rooted in the advanced economies act as proxies for citizens
in other countries. Or it may simply be an excuse for policy mak-
ers in developing countries who have never accepted the principle
of democratic accountability. Furthermore, it neglects the possibil-
ity that the best way to remedy imbalances in in¬‚uence may be to
¬nd ways of enfranchising the citizens of weaker states, rather than
denying information to the citizens of stronger states.


During a recent visit to Britain™s National Archives, I spent time
reading ¬les produced by the Foreign Of¬ce in the early 1950s. The
subject “ negotiation with Americans about the balance to be struck
between security needs and civil liberties while making decisions on
security clearances “ was fascinating. Equally fascinating, from the
point of view of a researcher who learned his craft in the computer
age, was the form of the documents themselves. There were relatively
few, and generally concise. They took a limited number of forms “
a letter, a memorandum, a short report, the minutes of a meeting.
Obviously all were on paper. Some were typescript, but many were
written in ink, in clear longhand script. Related documents were held
together in a folder that provided, on its cover, a longhand summary
of the material within. Each folder was bound with a red ribbon “ the
proverbial bureaucratic red tape.
Government documents had been produced and stored in much
the same fashion for perhaps the preceding two centuries. The tech-


. 31
( 53 .)