International Political Science Review Revue

International Political Science Review

Revue internationale de science politique

Volume 24 Number 3 July 2003

Issue Title

The Political Economy of International FinanceL’Economie politique de la finance internationale

Issue Editor

JAVIER SANTISO

Contents/Sommaire

In This Issue 283

From Here to EternityERIC BARTHALON 285

Why Do Governments and the IMF Enter Into Agreements? Statistically Selected Cases

JAMES RAYMOND VREELAND 321

Private Actors and Public Policy: A Requiem for the New Basel CapitalAccord

MICHAEL R. KING AND TIMOTHY J. SINCLAIR 345

Financial Markets and Politics: The Confidence Game in LatinAmerican Emerging Economies

JUAN MARTÍNEZ AND JAVIER SANTISO 363

Abstracts/Résumés (in the alternative language) 397

In This Issue

The articles presented here investigate the political economy of internationalfinance, exploring the boundaries between economics and politics.

The first article, contributed by Eric Barthalon, offers a comprehensive reviewof theories of financial crisis, focusing on the dynamics, resolution and preventionof these turbulent episodes in international financial interaction. As a formerchief economist of Banque Paribas and now chief economist and strategist ofAllianz Asset Management in Zurich, the author has considerable practicalexperience with financial markets. His article focuses, in particular, on the role ofthe lender of last resort, and the three dilemmas such institutions confront: moralhazard, the absorption of losses by the community, and the handling of the assetsobtained through intervention.

The second contribution deals with the political economy of InternationalMonetary Fund loan agreements. James Vreeland asks, “Why do governments andthe International Monetary Fund (IMF) enter into agreements?” Distancinghimself from the conventional argument that governments accept IMF termssimply because they need loans, Vreeland emphasizes that governments oftenwant the IMF to impose conditions in order to help them push though domesticallyunpopular economic reforms. The analysis draws upon a comprehensive data setinvolving 7011 country-year observations in 199 countries from 1952 (or year ofindependence) to 2000. But the discussion particularly focuses on two key casestudies: Uruguay and Tanzania.

The third article also invites us to focus on actors and institutions, this time inorder to understand reforms to the international financial architecture thatemerged from the turbulent economic environment of the 1990s. Michael Kingand Timothy Sinclair consider the “privatization” of mechanisms for internationalfinancial regulation, focusing their attention on two major debt rating agencies:Moody’s Investors Service (Moody’s) and Standard & Poor’s (S&P). They revealthat such “quasi-regulatory institutions” are part of a privatization of internationalfinancial management which is narrowing the sphere of influence andintervention of governments.

The final article also emphasizes actors, this time by prizing open the “blackbox” of financial markets and exploring the interactions between politics andfinance in emerging economies. Juan Martínez and Javier Santiso consider how Wall Street financial analysts react to major political events, focusing on the case of the 2002 Brazilian presidential elections. They examine the behavior ofleading Wall Street investment firms, and track interactions between political

International Political Science Review (2003), Vol 24, No. 3, 283–284

0192-5121 (2003/07) 24:3, 283–284; 033542 © 2003 International Political Science AssociationSAGE Publications (London, Thousand Oaks, CA and New Delhi)

www.sagepublications.com

developments in Brazil and movements on the financial markets. The authorsshow that interaction between politics and economics is central to understandingfinancial crises. Indeed, they propose that an “emerging market” can be defined asan economy whose political outcomes and political uncertainties (such as apresidential election or a cabinet reshuffle) tend to have disproportionate impactson financial variables and stock markets.

The purpose of this thematic issue of International Political Science Review istherefore less to offer a comprehensive overview of the international politicaleconomy of finance, and more to shed light on particular areas of a vast andrelatively unexplored terrain that links economics to politics, finance to sociology.It follows on from previous efforts by International Political Science Review (forexample, the 1999 thematic issue on “States and Markets: Essays in Trespassing”)to explore the parameters of the growing field of research in internationalpolitical economy.

Javier Santiso

284 International Political Science Review 24(3)

From Here to Eternity

ERIC BARTHALON

ABSTRACT. This article offers a critical reflection on the character andsignificance of financial crises, drawing on the work of various thinkerswho have examined this issue. After defining a financial “crisis,” thearticle asks three questions: Are financial crises foreseeable? Can andshould they be contained? How can they be prevented? The first sectionreviews the various mechanisms that lead to a crisis situation. The secondanalyzes the role of the lender of last resort and the three dilemmas itfaces, relating respectively to moral hazard, the bearing of losses by thecommunity, and the handling of the assets obtained from such a lender’sintervention. The last part presents the various proposals that have beenmade for the prevention of crises—some of them based on the existingmonetary institutions, others more radical.

Keywords: • Financial crises • International political economy

For almost three centuries, economic history has been punctuated by financialconvulsions, whose appearances have generally coincided with the peaks ofeconomic cycles. It would take too much space here to draw up the long list ofsuch crises (see Kindleberger, 1989). For the economist prepared to regard historyas a legitimate source of economic knowledge, financial crises therefore constitutean abundant mine of raw material. Indeed, Adam Smith, John Stuart Mill, KarlMarx, Vilfredo Pareto and Irving Fisher, to mention but a few, have all taken aninterest. And yet the space devoted by the economic textbooks to financial crises issmall or non-existent. In most cases, no trace of the word “bubble” is to be foundin the tables of contents of books on macroeconomics. At best, they contain a fewlines on the subject. Even The New Palgrave, the economics dictionary par excellence,devotes barely more space to the subject. This may seem extraordinary. Wouldmedicine have made the progress it has if sound and healthy organs had been theonly topic of research? At the risk of disappointing the reader, it has to be said thateconomists, at least those who take pleasure in analyzing equilibrium situations,are not a priori best placed to discuss financial crises—probably less well placed




than historians, psychologists or sociologists. Indeed, from an orthodoxstandpoint, the question of financial bubbles does not even arise. It is onlyrecently that economic analysis has recognized their existence (Davis, 1995;Kindleberger, 1989).

However transparent and “unidentified” bubbles may be while they areinflating, they generally end by bursting and so providing irrefutable, if belated,proof of their existence. By adding further episodes to an already long list, recentfinancial crises therefore revive three old questions. Are financial crisesforeseeable? Can they be (should they be) contained? Lastly, how can they beprevented? These will be the three main thrusts of the discussion that follows. Thefirst section will review the various mechanisms, some rational, others not, thatlead to a crisis situation. The second will analyze the role of the lender of lastresort and the three dilemmas that this lender faces, relating respectively to moralhazard, the bearing of losses by the community and the handling of the assetsobtained from the intervention. The third and last part will present the variousproposals that have been made for the prevention of crises—some of them basedon the existing monetary institutions, others more radical. The discussion drawsfreely from the work of numerous authors, especially Charles Kindleberger, whoseMania, Panics and Crashes (1989) is the standard work on the subject. Beforeproceeding, however, we have to define what a financial crisis is, by brieflydescribing the function and the main organizational principles of banking systemsand financial markets.

There Are Crises and CrisesA financial crisis can be defined as a large-amplitude oscillation affecting all orpart of a set of financial variables: issue volumes and prices of bonds and equities,volumes of outstanding loans and bank deposits, and exchange rates (Pareto,1964). It is only when the value of these variables collapses that one can speak of afinancial crisis. But the crisis in fact begins during the upswing, also very marked,which generally precedes its inception (Aglietta, 1997; Pareto, 1964). Thisasymmetry in the identification and perception of upswing and downswingphenomena could itself probably be the subject of considerable discussion. The“crisis” phase in the normally accepted sense of the term (in other words, amarked and rapid fall in the quantities and prices that measure banking andfinancial activities) is characterized by a desertion of those assets previously held.Suddenly and simultaneously, all operators want to exchange their holdings forinstruments which seem to them to offer both greater negotiability (the ability tobe traded at short notice) and liquidity (the ability to be traded without risk ofcapital loss). Generally speaking, a financial crisis takes the form of a sudden andsteep increase in the demand for money for precautionary purposes(Kindleberger, 1989). There are, however, a number of possible variations on thiscentral theme, fitting into each other like so many Russian dolls: if there is doubtas to the value of equities or bonds, investors will try to take refuge in bankdeposits; if it is the soundness of the banks issuing the deposits that is in doubt,they will turn to money issued by the central bank; if the national currency has lostits credibility, to foreign currencies; and lastly, if it is nonconvertible papercurrencies that are the subject of generalized mistrust, investors will turn to goldor any other form of commodity-based currency that does not involve a claim onanother party. Hence the distinction made by the monetarists between authentic


crises, that is, those which degenerate into bank panics, and “pseudo-crises,” whichare limited to sharp swings in the prices and trading volumes of financial assets,but are, nevertheless, liable to have systemic repercussions (Davis, 1995; Friedmanand Schwartz, 1963). Shocks are regarded as systemic if they are sufficientlyimportant to lead to the failure of one or more financial intermediaries, upset thepayment system or inhibit the process of allocation of capital via the financialsystem (Davis, 1995). There will probably be agreement that this is only adifference of degree.

The Necessary But Illusory Promise of LiquidityThe investment vehicles listed earlier can be broken down into two main families:those which are recorded as liabilities in bank balance sheets, on the one hand,and on the other, those which are traded on the financial markets and establish adirect relationship between agents with liquidities to invest and agents withfinancing requirements. The banks are linked in quasi-irreversible fashion witheach of their debtor clients, just as they are with each of their depositors, forexplicitly stated periods, these being generally short—even very short in the caseof deposits. They guarantee the nominal value of the deposits they take in. On theother hand, the assets they hold (non-negotiable loans or securities) are generallyrepayable at a much more distant date than deposits and are, moreover, lessliquid. Bank assets are in fact far from consisting uniquely of notes and coinsissued by the central bank. In the case of securities (equities or bonds), theirnegotiability on the secondary market is also aimed at favoring the mobilization ofresources, but over the longer term (Keynes, 1936). This negotiability of securitieson the secondary market, which is in fact a secondhand market, enables theholders to convert them into money if needed, but at an unfixed price. All theyhave to do is sell their securities on to other agents. For its part, the company orother entity issuing the securities does not have to sell the productive assets(factories, machinery and so on) that the issue of the securities enabled it tofinance. In both cases (bank deposits and financial assets), considerable instabilityis therefore technically possible if all the depositors or all the holders of securitiestry to liquidate their holdings simultaneously. Given the fundamental uncertaintythat exists regarding the distant future, it would be very difficult, even impossible,to finance certain types of investment if the providers of the funds were not giventhe possibility of converting their securities into money by selling them to possiblerepurchasers on the secondary market (Keynes, 1936). The uncertainty in thiscase is not only that relating to the return on the investment being financed(factory, highway and so on), it is also, and perhaps mainly, the uncertaintyrelating to the future situation of the provider of the funds. What purchaser ofsecurities can be absolutely sure that he will not be in need of liquidity at somestage during the coming 15, 20 or 30 years? It is here therefore that the difficultylies: if one wants to be able to raise capital in substantial volume and at minimumcost, the holders of securities have to be given the possibility of liquidating them atany moment (Keynes, 1936).

An Ideal Terrain for SpeculationAs it happens, financial assets are speculative instruments par excellence (Kaldor,1939). The term “speculative” can be applied to any sale or purchase whose sole

BARTHALON: From Here to Eternity 287

motive is the expectation of an imminent change in the ruling price (Kaldor,1939). Admittedly, securities can be purchased for their usage value to the holder,as in the case of the saver who builds up a portfolio in order to benefit from thefinancial income. For the speculator, all that counts is the exchange value. If hemakes a purchase, generally with borrowed money, it is in the hope of reselling ata higher price; if he sells, most often after borrowing the securities, it is torepurchase later at a lower price. Financial assets are speculative instruments parexcellence in that they are traded on markets that are technically perfect or semi-perfect (“efficient” to use the contemporary vocabulary) and their holding costsare low. They are standardized; they are of interest to a numerous and wide public;unlike certain commodities, they are unaffected by the ravages of time; they have ahigh market value in relation to their physical volume, which is increasingly beingreduced to a computerized entry; lastly, they bring in a current yield (Kaldor,1939).

The Normal and the Pathological StateMost of the time, banking systems and financial markets function smoothly.Admittedly, the equilibrium point on the capital markets is constantly shifting, butwithout major discontinuities. The banking systems and the financial marketsserve the purpose of allocating that rare resource, capital, which makes it possibleto finance investment that is itself the source of productivity gains. In so doing,they contribute to the growth of the economies they finance. Of this there is notthe slightest doubt: over the long term, the least developed countries are alsothose where the capital markets are the most embryonic or even nonexistent.Financial crises can therefore be regarded as the pathological state of a systemwhich generally functions in a healthy fashion. They are, in a way, a caricatureexaggerating the essential features of economic life (Rueff, 1989). Analyzing themtherefore is of both practical and theoretical interest. The practical interest isobvious: financial crises tend to discredit the free-market economy and provoke atemptation to “throw the baby out with the bath water.” From the theoreticalstandpoint, financial crises cast a particular and sometimes disturbing light oncertain essential assumptions and questions in the field of economic analysis(Kindleberger, 1989) as regards agents’ behavior and expectations, money andcredit, and policies to be implemented. The view defended in what follows is thatfinancial crises are not foreseeable, if this means knowing in advance the date, thehour of their occurrence and, above all, their amplitude. But if foreseeing a crisismeans identifying a risky terrain and deducing the subjective probability of anaccident, then crises are probably foreseeable. The intervention of a lender of lastresort makes containment possible, on condition that such intervention does notbecome systematic. Lastly, if these interventions are not to become, in time, thesource of constantly increasing costs for the community, it has to be rememberedthat our monetary institutions are man-made and therefore open to reform.

(1) Are Crises Foreseeable?The economic literature proposes two main interpretations of financial crises. Thefirst is based on a set of assumptions, the main ones being the rationality andhomogeneity of economic agents; the second explicitly rejects these two“founding” assumptions (Davis, 1995). For the perfect-market and rational-


expectations school, crises are due to random exogenous shocks: wars, naturalcatastrophes and so on. Depending on our knowledge of non-financialphenomena, crises are therefore regarded as foreseeable. In our present state ofknowledge, it is impossible, for example, to predict earthquakes. And yet a majorcatastrophe in Japan or California could easily upset certain financial equilibria.Crises are therefore, on this view, unforeseeable by the economist and differfundamentally from each other (Davis, 1995). The other viewpoint is not strictlythat of a school, although it too is in line with a long tradition. It is thereforedifficult to give it a generic title. Based more on historical observation, this viewstresses the financial instability that it regards as inherent in the functioning of theeconomy (Kindleberger, 1989). It could probably therefore be called the“financial-instability school.” For those taking this view, financial crises areendogenous. Behind their apparent diversity lies a deep-seated unity, with criseshaving as their first cause shifts in the collective psychology of those involved(Pareto, 1964). The practitioner having to deal, often painfully, with the financialmarkets cannot fail to have some sympathy for the financial-instability school, butnot to the point of ignoring the rational explanations for crises, since these give agood account of what can happen once the tide has started to ebb.

All Rational, All Identical?In economics, the term “rationality” is used in two different but related senses:that of the classical economists, on the one hand, and that of the neoclassicists andthe rational-expectations school, on the other (Arrow, 1987; Phelps, 1987).Neither of these meanings matches what the man in the street means byrationality, namely, the consistent utilization of appropriate means for attainingprecise objectives. For the classical economists, the rationality postulate meansthat agents make choices in the light of a complete and constant order ofpreference, with perfect information and zero cost (Blaug, 1980). According tothis hypothesis, economic agents’ reasoning is in conformity with the principles ofeconomic calculus; they have a set of logical and well-established preferences; forequivalent objective risk, they prefer the more to the less; they look for the highestrate of return; they minimize costs; they act in their own interest. In other words,they are pure economic beings, experiencing no emotion and subject to noimpulses.

Does Everyone Act Like an Economist?The rational-expectations school is merely an extension of the classical hypothesisof perfect prediction to a world in which the future is admittedly unknown, but inwhich it can nevertheless be described in terms of a set of objective probabilitiesdeduced from observation of the past (Arrow, 1987). Behavior will be described as“rational” if it conforms to what a given model supposes economic agents will do.Why so? Because, in a burst of humility and respect for homo economicus, theeconomist rejects the construction of a model in which agents might have a lessgood understanding of economic phenomena than he himself does (Phelps,1987). It is rare to find an expert claiming to have no particular expertise! Ittherefore emerges that the rationality hypothesis takes on its full dimensions whenassociated with others, starting with that of the homogeneity of economic agents(Arrow, 1987).


At the Frontiers of RationalityRecent theoretical developments have tried to explain banking and financialcrises by adopting the rationality hypothesis, but at the same time also stressing theasymmetry of information (all agents are not necessarily aware of the same things)and uncertainty (cases where it is simply impossible to be all-knowing) (Davis,1995). Uncertainty situations are those in which it is not even possible to deducefrom observation of the past the distribution of objective probabilities (Keynes,1936). This means that one leaves the realm of risk to enter that of pureuncertainty, where only subjective probabilities obtain. Whether the situation isone of risk or uncertainty, the problem posed to the financial markets is how toassess correctly the “insurance premiums” making it possible to face up to thearrival of possible shocks.1 Generically and using a simplicity of language whichcan lead to confusion, these premiums are described as “risk premiums.” Theexamination of “rational” theories for financial crises can suggest that the frontierbetween rationality and irrationality is permeable. Even so, these theories providea convincing explanation of market behavior once a reputedly exogenous shockarrives to upset an established equilibrium. They also make it possible tounderstand part of the sequence of events observed since the outbreak of theAsian crisis.

First Come, First ServedFor the monetarists, for example, under a system of fractional coverage of depositsby central reserves, there is a risk of illiquidity for the banks (Davis, 1995;Friedman and Schwartz, 1963). The fact that depositors can withdraw theirdeposits on a “first-come, first-served” principle provides a rational basis for bankcrises in the event of a shock liable to affect the quality of bank assets (especially asdepositors are less well informed than the banks themselves on the quality of theseassets) or where there is uncertainty regarding this quality (Davis, 1995). Accordingto this approach, even solvent institutions can be confronted with a run on deposits.Adjustment then takes place via quantity rather than price: either there is confidenceand depositors entrust their money to the banks or there is not and depositorswithdraw their funds, regardless of the remuneration offered (Davis, 1995).

The Law of “All or Nothing” Applied to the Distribution of CreditThe same idea can be used to explain certain lending behavior on the part ofbanks and financial intermediaries, behavior describable as credit rationing byquantity rather than by price (Davis, 1995). In some cases exacerbated byderegulation, competition between lenders can, in situations of uncertainty, leadto an incorrect estimation of risk premiums, especially for the “big risks” thatmaterialize only very rarely. It therefore eliminates the most prudent lenders(Davis, 1995). These phenomena are accentuated by psychological factors. Giventheir exceptional character, the big risks are gradually swept aside. Alternatively,without being effaced entirely from memory, their subjective probability may betoo small to be taken into account. There also come into play certain mechanismsfor the suppression of fresh information that might call into question thejustification of existing lending strategies (Davis, 1995). These psychologicalfactors are then compounded by institutional considerations. Incorrect estimation


of risks will be all the more of a danger if any or all of the following conditionsobtain: systems for the valuation of profits lay stress on the short term; staffturnover is high; remuneration packages (especially in the form of bonuses and“golden parachutes”) introduce asymmetry into the sharing of gains and losses;and, lastly, measures of return take inadequate account of the risks incurred(Davis, 1995). On all these points, the development of institutional capitalisminstead of family capitalism heightens the risks. Under this approach, theperception of vulnerability will not increase until such time as there is anexogenous shock. If the borrowers are undercapitalized, if risk premiums aresmall and if loan portfolios are relatively undiversified, as will often be the caseafter a prolonged period of tranquillity, the optimal response of lenders faced witha shock that dents their confidence will be to apply rationing by quantity ratherthan by price. With borrowers no longer having sufficient funds of their own forrationing by price to be effective even the best risks among them will find itimpossible to borrow at any interest rate. The emergence of such a situation is allthe more likely if borrowers, as a matter of policy or habit, deal with a largenumber of banks (Davis, 1995).

When One Knows that One Does Not Know All One Ought to KnowThe approach based on asymmetry of information and agency costs arrives at anidentical conclusion. This approach takes as its starting point the observation thatborrowers always have a better knowledge of their financial situation than isavailable to the lenders, especially if the latter are numerous (Davis, 1995).Because of their inability to distinguish between good and bad borrowers, thelenders demand risk premiums that reflect the average quality of risks. The fear ofselecting bad risks can lead them to ration the quantity of credit. This propensityto impose rationing will be exacerbated if interest rates rise; if uncertaintyincreases; if the value of collateral assets declines or, lastly, if borrowers have littleequity at their disposal. Indeed, the smaller the funds at the disposal of theborrowers, the greater their possible inclination to take excessive risks: they will beable to claim the virtual totality of potential profits, but will bear only a very smallpart of the possible losses. The marginalization of the banks in the financingcircuits, either because of a tendency to disintermediation or following failures,also aggravates this risk of rationing, given the banks’ definite advantages indealing with the problems of asymmetry of information (Davis, 1995).

The Financial Markets’ Evanescent LiquidityWith only minute adjustments the preceding analysis can also be applied to thefinancial markets. Their liquidity depends on the number and diversity ofoperators and “market-makers.” Such liquidity is real if all the holders of securitiesdo not try to realize them simultaneously. If a doubt arises concerning the futureliquidity of a market, it is therefore rational to want to sell before others do (Davis,1995). Such reasoning can be held by the market-makers themselves, obliged asthey are to hold a float of stock to respond to orders from their clients. Undersevere selling pressure, the market-makers can either increase the spread betweenthe buying and selling prices or refuse to quote prices at all, which in either casewithdraws liquidity from the market. This process can last as long as uncertaintyremains regarding the existence of an equilibrium price (Davis, 1995). The same


will be true if the market-makers come to believe that certain operators are betterinformed than they are (Davis, 1995).

Yes, Speculation Can Have Its UsesFor all the theories taking as their starting point the hypothesis of rationality,speculation cannot be a destabilizing factor. Such speculation, because it wouldmove prices away from the equilibrium level, would by its nature be a losingstrategy and hence condemned to rapid disappearance (Kaldor, 1939). Beingbetter informed of the particular temporal and spatial circumstances than theaverage operator, speculators are regarded as merely correcting temporaryimbalances between supply and demand. They would simply be punishingerroneous economic policies or ill-informed private strategies, whether thesepolicies and strategies were in fact applied or merely expected to be applied(Eichengreen, 1996). By their presence, speculators are regarded as bringingliquidity to the financial markets.

“Rational” Bubbles and SunspotsThose who subscribe to the school of rational expectations can neverthelessenvisage the existence of “rational” bubbles (Davis, 1995). Such bubbles appearwhen a majority of operators are convinced that a certain price-formationmechanism is at work, even though this may have nothing to do with what areconventionally called the “fundamentals.” One is then in the presence of self-fulfilling phenomena where it is expectations of future price levels that determinecurrent prices. One can, in this case, imagine a so-called “sunspot equilibrium.”Even though solar cycles have no material influence on the economic situation, ifa sufficiently large number of operators believe they have and act in consequence,then economic fluctuations will indeed reflect the sun cycle, for a time at least(Phelps, 1987). Even so, it is not easy to make money in such a situation because,while the mechanism for the formation of expectations is then known (byreference to sunspots), the behavior of the sunspots is not known in advance andso prices are not predetermined. The operator who thinks he knows that sunspotshave no influence on the economy will not be able to take advantage of thisknowledge, given that the majority believes the contrary. This is one explanationfor the methodological “zapping” of the financial markets. At any moment, theirbehavior always depends on what is regarded as the crucial variable, but thischanges constantly over time, with, for example, the trade balance replacingmoney supply as “flavor of the month.”

To Each His Own Rationality?The assumptions of rationality and homogeneity have been the subject of muchcriticism, with both their logic and their likelihood coming under question(Arrow, 1987). If all the agents are homogeneous and have the same information,how is any trading conceivable? And how can one imagine the existence of pricevariations in the absence of trade? And indeed, can one imagine that agents’behavior conforms to the hypotheses of a given model, when in fact there aremany models—why one rather than another? And how can one justify the fact thatmodels change over time (Arrow, 1987)? There is also agreement that the


hypothesis of rational expectations implies on the part of economic agents acapacity for calculation and information processing that is completely beyondbelief. Along the same lines, the rationality hypothesis in fact neglects the pointthat information and its processing are not cost-free. The most telling criticismprobably lies in the difficulty of forming rational expectations in situations ofuncertainty (Arrow, 1987). Empirically speaking, the volatility of the prices offinancial assets often seems excessive in the light of the fundamentals (and thesethemselves change from one operator to another), which the models take asexplanatory variables. The evolution of the dollar between 1980 and 1985 is amuch-quoted example of an irrational movement (Phelps, 1987). The rationalityzealots’ reply to these criticisms is that the markets could at a given momentlegitimately expect one event or another justifying a rise in the dollar, acircumstance whose probability would later turn out to be nil. In the economicliterature, this is known as the “peso problem.” However, it is difficult to see thistype of argument as more than prevarication (Phelps, 1987). There are alsonumerous cases where prices vary even in the absence of fresh information or,conversely, where such information fails to have an impact on prices.

Irrationality and Endogenous Financial InstabilityThe financial-instability school sets out to go back far beyond the shock, hithertoregarded as endogenous, that triggers the financial crisis (Kindleberger, 1989;Pareto, 1964). The tendency is to see this as merely a catalyst operating in anexplosive cocktail whose fermentation is itself governed by an endogenousprocess. For them, the rationality of economic agents is more a convenientworking hypothesis than a faithful description of reality. In their view, rationality islimited at best, with most markets being rational most of the time (Kindleberger,1989). The theory of financial instability allocates a key role to uncertainty andirrationality. In an uncertain universe, we are condemned to trial and error, andsometimes to retracing our footsteps. This school admits that psychological factorsare to be found underlying the whole economic field and that, generally speaking,people like to hold a common belief in the same things. They recognize the dead-weight of habit and the strength of the herd instinct, which can be moreinfluential than the maximization of income or utility (Kindleberger, 1989). Theyagree that all agents are not similar, and that operators on the financial marketsare not all intelligent, well informed and financially sound. They take the view thatoperators cannot all have access to the same information, as this carries a cost.They stress that operators may not all have the same processing capacity, the sameexperience or the same degree of expertise, and not all display the same aversionto risk, the latter varying, among other things, inversely with their wealth. In anuncertain universe, it may seem therefore perfectly justified to imitate thebehavior of operators whose success may, for a time at least, suggest that they haveparticular know-how (Kindleberger, 1989).

Standard economic theory is by no means devoid of contradictions. It explicitlyaccepts that the behavior of consumers may be irrational, but it seems to assumethat individuals who consume are not the same people who operate on financialmarkets. It is generally admitted that, for most consumer goods, demand isdetermined by sociological factors and especially by imitation of fashion leaders. Ifthis were not true, why should publicity and brand imaging exist in their presentforms?


Other manifestations of irrationality can be envisaged, the most frequent beingthe product of a fallacy of composition: the sum of individually rational behaviorscan produce a collective behavior that is itself irrational (Kindleberger, 1989). Anexample is the illusion of liquidity: the collective belief that it will always bepossible, in case of need, to unload securities before other operators do. Thisfallacy of composition maintains an illusion of liquidity for securities (Keynes,1936). The same fallacy can be applied to government economic policies and bereflected in the markets’ assessments of these policies. Judging whether a givenpolicy is appropriate or not largely depends on the context in which it is tooperate and, in particular, on the policies implemented by other countries. Also,the markets, which are impatient by nature, tend to credit a given country with thesupposed benefits of the latest economic orthodoxy, once that country’sgovernment declares its adherence to it. There may also be interaction betweenprice movements and agents’ expectations, with a price rise becoming anincentive to buy. This is known as “feedback trading,” a situation in whichpurchases are made not because prices fall but because they rise, the predominantsentiment being fear of missing out on an upward movement (Kindleberger,1989). Lastly, agents can show a cognitive bias, unconsciously processinginformation to bring it into conformity with their preconceptions andrepresentations of reality. This leads to the flowering of the notion of a “new era,”that is, the inclination to regard as normal and lasting evolutions that are in factneither or to the propensity to remain invested notwithstanding the exceptionalnature of past performance (Kindleberger, 1989). For its adherents, the belief thata new era has begun is sufficient to justify a total reshaping of the usual valuationmodels. The perception that things have changed, that the world is constantly influx, is certainly justified by reference to the way things really work. But who cansay whether the cycle of enthusiasm and despair really changes throughout thecenturies?

Keynes’s Casino EconomyWhile the reversibility of financial engagements is of major importance, as shownearlier, for the mobilization of long-term resources, at the same time it introducesan element of permanent potential instability into the system. This is because thedominant perception of the future, whether regarding the return on realinvestments or even more regarding the situation of the providers of funds, clearlyhas an influence on the resale price of financial assets on the secondary market.From there it is only a short step to an approach based on second-guessing themarket (the only level of reasoning that counts in certain circumstances) withspeculation no longer involving the intrinsic values of things, but rather the ideathat the average player is going to have of those values (Keynes, 1936). In suchcircumstances (see above), the “rational” player is bound to take account of thereasoning and behavior of the other operators, even when these seem to himtotally irrational. Keynes (1936) has stressed the point: it is highly dangerous to berational in a market that is not. This means that a superior form of rationalitywould be to take account of the irrationality of the other agents. For thoseadopting such an approach, it has the major advantage of not leaving themstranded and isolated in the case of an error of judgment. In matters of finance, itis important to be wrong along with everyone else, this being less blameworthy(Keynes, 1936). How many times has one heard it said since the outbreak of the


Asian crisis that, in any case, no one had seen it coming? It therefore takes notonly a highly critical mind, but also great strength of character not to yield to thecomfortable temptations of imitation and the herd instinct. Over short periods,the return on a stock-market investment can then become totally uncoupled fromthat of the real investment it represents (Keynes, 1936).

Yes, Speculation Can Be DestabilizingIn this analytical framework, the assessment made of speculation will be a matterof circumstance, modalities and proportions (Kaldor, 1939). If speculators are notparticularly clairvoyant, if they use mainly borrowed funds and if their transactionscome to dominate the market, then they will play a destabilizing role because, inorder to survive, they will have no need to examine with precision the evolution ofunderlying factors. All they have to do is to make a correct prediction of theaverage opinion of the other speculators. The success (generally short-lived) of afew will make possible the permanent renewal of the speculator population(Kaldor, 1939).

Are the Markets Macro-Efficient?While the financial markets may be technically efficient, there are grounds fordoubting whether they are also efficient in informational terms. Theinformational efficiency of financial markets is measured by their capacity forinstantaneous incorporation into prices of all available or anticipated information.Since this flow of information is itself unpredictable, it would be impossible to beatthe market except by constituting a more risky (that is, more volatile) portfoliothan that represented by the market as a whole. Hence the recommendation to construct index-tracking portfolios, in which the investor is content with an exact replication of the structure of the market concerned. As Samuelson has stressed, it is more plausible for the financial markets to be micro-efficientthan macro-efficient. It is less exacting to be able to deal instantaneously and correctly with micro-information than with macro-information (Samuelson,1994). Macroeconomic mechanisms are necessarily more complex. Unlike themicroeconomic world, it is never possible to reason on a ceteris paribus basis.Interdependencies and interrelations are numerous and all-pervading. Moreover,even supposing that it were possible to have a correct idea of one or othermacroeconomic phenomena, the possibilities of eliminating a macro-inefficiencythrough arbitrage are more limited than in the case of micro-inefficiencies, sincethe funds that would need to be applied are much larger (Samuelson, 1994).However, for others, informational efficiency is quite simply a logical impossibility.This is because if all traders accept the hypothesis of informational efficiency, theincentive to look for relevant information disappears and the market, by the sametoken, ceases to be efficient. This is the significance of the Grossman-Stiglitzparadox.

How to Inflate an Irrational BubbleFollowing in the footsteps of Mill (1987), Fisher (1996a, 1996c) and Minsky(1975), Kindleberger (1989) proposes a model of the inflation of financialbubbles. In this model, everything starts with an initial shock that increases


earnings prospects (Mill, 1987). The nature of this shock is immaterial:deregulation, the end of a conflict, the opening up of new markets, theintroduction of new technologies, financial innovations, the resolution of anindebtedness problem, and so on. The early 1990s were indeed marked by a largenumber of shocks of this kind. The important thing is that a bubble starts toinflate for perfectly sound reasons. Thereafter, it is just a good idea taken too far.Indeed, the difficulty lies precisely in the fact that at the origin of these cumulativeprocesses there are nearly always real shocks justifying their initiation—shockswhich create fresh opportunities for investment and profit. There is therefore noquestion of casting doubt on the reality of these evolutions. However, from theeconomic standpoint, the question is whether the financial behavior generated bythe new prospects and the prices of financial assets do or do not go too far,especially in extrapolating recent tendencies. Before going further with thedescription of the formation of a bubble, it is useful to dwell on two processes thatplay a role in its emergence: the credit mechanism and the role of the dollar inthe international monetary system.

The Prodigious Credit MiracleIt is profit expectations that motivate the demand for credit (Mill, 1987),regardless of whether this is by nature bank or nonbank, whether it is limited tocredit from the central bank or goes beyond that source of credit. This recourse tocredit may even bypass the banking system entirely (Mill, 1987). For example,futures and options markets can also be used for speculation (on both a rise and afall in prices) without having to find the means of payment corresponding to theprices of the underlying assets. In a way, the sellers and buyers of these derivativeproducts provide credit to each other for that part of the nominal value of thecontracts that is not covered by deposits. The essential points are that the creditpossibilities available to the agents are used to the full and that these promises topay lend themselves to monetization by the private agents (Mill, 1987). In theirefforts to control the quantity of credit and money in circulation, central bankshave a Sisyphean task on their hands. In accordance with Goodhart’s Law, it issufficient for them to try to control growth in an aggregate Mi for the market toadjust and begin to create an aggregate Mj (Kindleberger, 1989). There are two,and only two, ways of buying securities if one does not want to give up makingother expenditures. The first involves handing over money that one hadpreviously saved; the second consists of first obtaining a bank loan and thenhanding over the money borrowed from the banking system, that is, moneycreated by the system for the occasion and of which no other agent is deprived(Mill, 1987). Nor is there any obligation for the link between the purchase ofsecurities and the recourse to lending to be a direct one. If, in order not to give upan expected capital gain on his security portfolio, the buyer of a car borrows inpreference to selling part of his securities, the result in purchasing power terms isjust the same as if he had bought the securities on credit and his car for cash. Inboth cases, borrowing increases in relation to his total resources, as does thepurchase of shares in his total resource use (Mill, 1987). In any case, as long asprices rise, the fact of financing the purchase of securities with borrowed fundsmagnifies the financial result: an annual increase of 20 percent in the price of thesecurities held produces a capital gain of 40 percent if the portfolio is only 50percent financed out of the purchaser’s own funds. Against this a financing cost of


5 percent per year can look ridiculously small. During this period, thanks to therise in prices, the more agents borrow in nominal terms, the less indebted they arein real terms. Credit in fact enables all agents to rid themselves of budgetaryconstraints and purchase securities without reducing other items of expenditure(Mill, 1987). A simplified model of a closed economy makes it possible, simply bysumming the individual accounts of firms, to show in, rigorous fashion, the linkbetween the granting of bank credit and agents’ spending (Allais, 1993). Mindsengrossed by credit find it increasingly difficult to distinguish between money andsavings (Allais, 1988). In fact, on a financial market, the loanable funds (that is,those looking for a place to invest) may represent the counterpart either of savingsor of pure monetary creation. This means that there is nothing easier (andnothing more dangerous) than to confuse the two, especially when internationalcapital movements become substantial.

The More Widespread the International Use of the Dollar, the More Dollars There AreIt is more than likely that the complexity of international financing circuits helpsto maintain the confusion between savings and money. Without great difficulty theapproach consisting of the summation of individual accounts can be extended tothe case of an open economy. This analytical framework shows that the crisis in theemerging economies is in fact a crisis of international monetary regulation.Through the way it functions at present, the international monetary system makesit possible for the USA to finance its external payments deficit in its own currency(Rueff, 1989). The system does not regulate the international economy because,as Jacques Rueff would have said, it permits “demand without supply.” Whether itbe for commercial or financial transactions, the USA pays its international creditorsby handing over dollars, dollars it itself issues or which the international bankingsystem creates for it. In this system, instead of settlement being made by thetransfer of “supranational” means of payment, which would result in a reductionin the quantity of means of payment in circulation in the USA, the central banks ofcountries having overall surplus balances lend to the American financial systemthe dollars received by their nationals. These dollars are in fact instantaneouslyrecycled to the USA by the central banks which accept them as internationalreserves. At the same time, they act as the basis for domestic monetary creation inthe countries receiving them (Rueff, 1989). In these circumstances, how is itpossible to distinguish a dollar that has been genuinely saved from one created“out of thin air” by a simple movement of capital? Means of payment can thereforebe remitted to creditors of the USA without even leaving the American economy,the result being duplication of means of payment (Rueff, 1989). This systemtherefore allows the USA to export capital even though it runs a saving deficit. As ithappens, a good number of emerging countries have chosen the path of export-led growth. To finance the rapid accumulation of productive capital, they havehad recourse to external funds. With a view to establishing financial credibility,they needed to tie their currencies to one that was internationally recognized andfor this purpose they chose the dollar, being the currency in which their exportswere invoiced and with the USA absorbing a substantial share. In an internationalcontext of this kind, any flow of dollars in the direction of the emergingeconomies was bound to increase the means of payment available in thosecountries without any corresponding reduction in the USA (Rueff, 1989). It can infact be seen that since the beginning of the 1990s, private capital flows to the


emerging countries have risen substantially, helped by the increasing liberalizationof capital movements.

In our system of nonconvertible paper currency, where money returns to theUSA and leaves it at the same time, where the dollar is used as a counterpart formoney creation at the periphery of the system without any equivalent reduction inthe stock of means of payment in circulation at the center, the world market forloanable funds can find its equilibrium only at an artificially low level.Furthermore, when the cost of capital is artificially low, the inevitable result is atendency toward overinvestment and poor resource allocation (Allais, 1988;Wicksell, 1936). The self-stabilizing function of the international monetarymechanism has therefore been weakened, with the following consequences in theemerging countries, as well as in the USA: an explosion of money-supply andlending aggregates; a tendency toward overinvestment and the squandering ofcapital; the formation of bubbles in the industrial sector, in real estate and on thestock market; a decline in the quality of bank assets; a recurrent widening incurrent-account deficits at one point or another in the circuit; inflation in the realeconomy; and a rise in the real exchange rate (Rueff, 1989).

Fair-Weather FinanceAt the beginning of the indebtedness process just described, the actors interveneon a modest scale—“just to see.” When the first operations, arranged withprudence, give better-than-expected results, what next comes into play is whatMinsky (1975) has called the “tranquillity paradox.” The early successes establishconfidence and give heart to new entrants. The tranquillity seen thus far sows theseeds of future instability (Kindleberger, 1989). The progressive disregard of themore traumatic past shocks therefore plays a role in the gestation of crises (Pareto,1964). This forgetfulness manifests itself not only in the volume of lending incirculation, but also in the structure of credit operations. As the bubble builds up,equity is increasingly supplanted by borrowing, with the increase in the quantity oflending accompanied by a deterioration in its quality, while the modalities offinancing become increasingly speculative and unstable (Kindleberger, 1989). Inthe development of this “capitalism without capital,” an important consideration isthe fact that interest is generally treated more favorably for tax purposes thandividends. At the beginning of the upswing, the cost of servicing the debt (interestand principal) is covered by cash flow. When this is no longer adequate to meetmore than the interest payments, repayment of the capital is based either on thesale of assets or on the rollover of the initial loan. The final stage is that of“pyramid mechanisms,” in which creditors can only be repaid with the inflowsfrom new subscribers—a pattern that has gone down in posterity under the nameof “Ponzi finance” (Kindleberger, 1995).

The notion of the quality of credit is a complex one (Kindleberger, 1995),being neither deducible from the simple quantity of credit nor reduced to anumeric indicator. To see this, one only has to remember the very large number ofparameters that go into a loan contract: amortization modalities, maturity, thepossibility of early repayment, collateral, guarantees from third parties, theinterest rate (fixed or floating), the currency of denomination, and debt ratios inrelation to equity or cash flows. In the case of financial intermediaries, this list islengthened by the addition of the ratios of deposits to equity and of deposits toreserves, the degree of diversification of the loan or securities portfolios and the


nature of relations between the corporate officers and the principal debtor clients(Kindleberger, 1995). Lenders can be tempted to pass certain risks on to theborrowers, this being generally the case for interest-rate or exchange-rate risks.While understandable, this approach can turn out, in the end, to be illusory inthat, in the final analysis, its effect is to increase the risk of default by the borrowerin the event of major swings in exchange rates or interest rates (Kindleberger,1995). Last and most important is what one might call the “repayment culture.”Some private agents and governments make it a point of honor to repay theircreditors at the agreed time, others do not (Kindleberger, 1995).

Across-the-Board EuphoriaIt is not hard to imagine the euphoric impact of the ability to borrow when thisleads to the creation of fresh means of payment and becomes generalized, theresult being that economic agents find themselves collectively rid of all budgetaryconstraint (Mill, 1987). The less-well-informed observer, largely unaware of theadmittedly tortuous mechanisms of monetary creation, will then be tempted toput the rise in security prices, especially of equities, down to the simultaneousexpansion of real activity, even though the creation of means of payment has longceased to have as its exact counterpart the creation of real wealth. It is importantfrom a psychological standpoint that the build-up of financial transactions beaccompanied by increasing prosperity in the real economy. At the very least, itmust take place without jeopardizing the latter, which is not always the case towardthe end of the rise. Before that occurs, two phenomena of exceptional size andmutually self-supporting arise simultaneously: rapid economic expansion and anexplosion in the nominal value of financial transactions (Kindleberger, 1989).Both the volume of transactions and the prices at which they are struck increaserapidly.

Ever More Profit?During this period, profits also rise rapidly (Aglietta, 1997). Being more markedthan for other types of income, this rise in profits leads to a substantial shift in theallocation of national income. This shift in the allocation of value added poses amajor problem: how far is it compatible with the solvency of final demand(Aglietta, 1997)? It can in fact be assumed that the natural tendency for holders ofcapital will be to reinvest their profits in productive activities, with a growing shareof final demand therefore involving capital goods, the risk being that these goods,which, in part at least, produce consumption goods, have increasing difficulty infinding outlets. Toward the end of the upswing phase, this shift in the distributionof value added is viable only through increased recourse to lending, either byconsumers or by the producers of the investment goods. The system then starts torun away with itself (Aglietta, 1997; Mill, 1987). According to this Marxist-inspiredinterpretation, the financial crisis and debtor insolvency reveal that the tendencyin the allocation of value added has attained its limits. The risk is that expectationsof long-term corporate earnings growth (the only real factor justifying growth inequity values) may then be formed by reference to the period when the allocationof value added was substantially lopsided.


Think of a Sisyphus Showing Sophistry, Credulity and HappinessThanks to the availability of credit and to the general euphoria, prices of financialassets find themselves pushed to extreme levels, at least by traditional valuationcriteria. And when the prices of all financial assets become increasingly difficult tojustify in the light of traditional valuation criteria, the response strategy is all toopredictable: ignoring any monetary explanation for the phenomenon, thepartisans of a new golden age place the emphasis entirely on real factors. Thesenew-age Sisyphuses do not hesitate for an instant in inventing a fresh valuationmodel every time an additional rise in prices refutes the one they were previouslyusing. Moreover, they have no option but to give increasing importance to themost cyclical variables, although these are, in theory, no more than negligible“noise.” In these periods, the creation of wealth tends, furthermore, to beconfused with its circulation (Claudel, 1993). The diplomatic correspondence ofPaul Claudel, French Ambassador to the USA between 1928 and 1932, shows astriking parallel between our amazement at the Internet and that of ourgrandparents toward telegraphy. The peak of a financial cycle is often preceded,and later followed, by the emergence of financial scandals, of indelicate or evenfraudulent financial practices (Kindleberger, 1989). Greed for rapid self-enrichment is characteristic of periods of frenetic speculation. As Kindlebergerpoints out, nothing is more disturbing than to see one’s own friends getting richon the stock market. This state of mind predisposes part of the public to rushheadlong into hazardous transactions, thus exposing themselves to becoming thevictims of indelicate behavior. It is possible therefore that in criticizing Asianfinancial morals (corruption, nepotism, cronyism and cover-ups) one is merelyconfusing cause with effect. Kindleberger (1989) reports having discovered anissue prospectus dated 1720 stating that the profit outlook for the plannedoperation is so exceptional that it cannot be revealed for fear of exploitation byothers. The “hedge funds” industry, with its known jealousy of operating secrets(some of them concocted by Nobel prizewinners for economics, no less) and littlesubject to any obligation of financial transparency, is merely a particularlyoutstanding variation, although within the law, on a well-worn theme. While, inthe upswing phase, it is demand that leads to sharp practice and financialdelinquency, at the beginning of the downturn it is, on the contrary, spontaneoussupply that provides the source. Certain operators finding themselves in difficultycan then be tempted to try to sort it out by failing to live up to certain standards ofhonesty—hence the flowering of Ponzi-style schemes.

Beauty Contests in the Global VillageIn the investor’s defense, it should be noted that the global village is the scene ofconstant confrontation between financial nationalisms—competition that canimpair the objectivity of information. International competition to attract capitaloften takes on the air of a beauty contest between “models,” which in fact changewith the evolution of economic and financial cycles. In these circumstances, theofficial monetary and financial authorities stand shoulder to shoulder with thenational financial intermediaries. “Self-criticism” of the national model is thenalways more muted than the criticisms of competitors’ models. More or lessdeliberately, information will be screened, or even suppressed, and the investorwill have to show an exceptional critical faculty to be able to unravel myth from


fact, however, the logical contradictions involved are too flagrant to be ignored.The run-up to a financial crisis is also something of a crisis of lucidity. Thisproblem will be intensified if, as is often the case, research services are notthemselves global in their working methods. If they are constituted simply as agrouping of local specialists, fragmentation of conclusions and lack of any realoverview become the rule, with, as a result, a consequent risk of ignoring a certainnumber of common factors.

The Greater the Velocity of Circulation of Money, the More There Seems to Be of ItAnother sign that a crisis is gestating is a substantial increase in the velocity ofcirculation of money, especially within the financial sphere—a phenomenonwhich the classicists termed “overtrading” (Kindleberger, 1989; Mill, 1987; Pareto,1964). An acceleration in the velocity of circulation on the financial markets neednot, at least in the initial stages, have as a counterpart a slowdown in the realsphere. It is not easy to measure these respective velocities. The overall velocity ofcirculation, taking all markets together, can be directly calculated as the ratio ofthe sum of debits and credits entered in the accounts of the banks’ clientele to theaverage size of deposits (Pareto, 1964). An investor who has sold his shares in acompany will be tempted to reuse the newly acquired liquidity by investing inanother company whose prospects he sees as better. It makes no difference to himwhether the origin of this liquidity was a counterpart of real saving or the result ofmoney creation. In the USA between 1925 and 1929, the velocity of circulation ofdeposits in the New York banks was multiplied by a factor of 2.5 (before beingdivided by 4.0 between 1929 and 1932), whereas it was 1.3 in the rest of thecountry (Fisher, 1996a).

This increase in the velocity of circulation of money effectively increases thequantity of money in circulation, that is, what is offered in exchange for securitiesor goods. It can therefore give the illusion of an absolute abundance of liquidity.Some will certainly say that by reason of the absolute need to invest this liquidity arise in share prices is irresistible, inevitable and (why not?) unlimited. They willoverlook the fact that stock-market capitalization is at its height in relation to thestock of liquidity, which amounts to saying that, measured by the yardstick ofsecurity prices, the purchasing power of the stock of money is actually in decline.Instead of absolute abundance, it is relative scarcity that has become the rule. Thesame sophists also fail to see that, in a system where it is loans that create thedeposits, this mass of liquidity can equally well contract: means of payment cansimply disappear from the system if borrowers decide to repay their debts or iflenders choose not to renew their loans. At bottom therefore the liquidityargument is nothing more than the expression of a high appetite for risk. Sinceeverything becomes possible at the same time (prosperity in the real sphere just asmuch as in the financial sphere), irrational behavior now starts to take over.Emotion replaces rationality; fear of missing an opportunity of making moneyreplaces that of losing money. It becomes possible to make one’s fortune in amatter of days by purchasing stocks at random (Pareto, 1964).

In the Ending was the Word ... and the TumultIt is by no means rare to hear authorities of all kinds (central bankers, politicalleaders and well-known businessmen) uttering warnings and cautions (Kindleberger,


1989). The “irrational exuberance” referred to by Alan Greenspan is reminiscentof statements by Paul Warburg or by Governor Young in 1928. Experience suggeststhat the “croaking” of these Cassandras is generally in vain (Keynes, 1936). Therethen comes an event which acts as a catalyst for a revision of expectations, usuallytaking the form of a rise in the cost of short-term money (Kindleberger, 1989),that is, the kind used for speculation. This rise can be the result either of adecision by monetary authorities wanting to calm things down and curb therunaway progress of the economic cycle or a market mechanism by which thedemand for funds for speculative purposes throws the market for loanable fundsout of balance. The mechanism involving the financing of the rise in pricesthrough borrowing can then be sufficient to produce a rise in interest rates. Inmany cases, too, what happens is that international creditors, more lucid thantheir debtors, signal “game over” (Kindleberger, 1989).

The Fall Is More Rapid and More Intense than the RiseJust as the rise fed on the rise, the fall then feeds on the fall. However, in a crisissituation, the ebb tends to be more rapid and more intense than the flood. Thereare several technical reasons why this should be so. In a system based on credit, themechanism by which money is created is based on a dual confidence on the partof both lenders and borrowers. On the other hand, all that is needed to block themechanism is a withdrawal on the part of either lenders or borrowers (Keynes,1936). The volume of outstanding lending therefore generally decreases morerapidly than it increases, especially if a substantial number of operators findthemselves obliged to unwind their positions at any price in order to meet theircommitments (Fisher, 1996c). The modalities of the formation of futures pricesintroduces a second element of asymmetry between price rises and price falls(Kaldor, 1939). Just as it is easy for arbitrageurs to take advantage of an excessivelyhigh forward price, so it is difficult for them to do the same thing in the event thatthe price is too depressed. Professional speculators use intervention techniquesthat accentuate still further this asymmetry between price rises and price falls.They know that they are liable to be caught wrong-footed by the evolution inprices. While they can see their profits swell without any particular concern, theyhave to be quick off the mark in containing losses. This is illustrated by a simpleobservation. To wipe out a negative price movement of 10 percent, it takes apositive movement of 11.1 percent, a gain of 25 percent to compensate for a lossof 20 percent, and so on. The influence of this arithmetical asymmetry is obviouslygreater when the positions taken are financed by borrowing, with the leverageeffect amplifying both losses and gains. If a large number of speculators operate inthis way, the selling pressure can suddenly become very strong and produce steepfalls in prices and, above all, major discontinuities in price quotations.

What Use Are Economists?At the risk of causing disappointment, it has to be said that there can be littlecredence, not to say none at all, in economists’ capacity to predict with precisionthe occurrence and amplitude of crises of an economic, monetary or financialnature. What economists can perhaps hope to do is to identify risk factors anddisequilibria (of which there are always some) that make major turning pointsinevitable (the only kind that are really important for the conduct of business)


(Pareto, 1964). This identification does not generally make it possible to put aprecise date on the outbreak of a crisis, as what seems to happen involves suddendiscontinuities in the perception of phenomena. In other words, evaluating a riskis in no way equivalent to predicting the occurrence of an event. It is highlyunlikely that a proportional linear relationship exists between the evolution of themain economic variables and their perception by economic agents. There arephases of accumulation of observations that lead to situations where doubt is nolonger possible and then suddenly the logjam of previous beliefs bursts. Nor is itcertain that most of the economists’ audience really expect them to predict thesemajor turning points, given that this type of prognosis can upset both establishedpreconceptions and strategies already being implemented. Any economist voicingcriticism of the Asian miracle in the 2–3 years preceding the crisis or whoexpressed doubts over the reality of the “new American paradigm” wascondemning himself to ostracism, especially if he was not privy to the New Yorkand London “in-groups.” But this did not stop those same people from criticizingeconomists after the event for not having been more clairvoyant, for not havingsounded the alarm more vigorously; because it is the suddenness with which criseshappen that is perplexing (Eichengreen, 1996).

In the belief that markets are efficient and that speculation has a stabilizingeffect, so-called speculative movements are there to punish inappropriateeconomic policies either in the past or yet to come. If this model was valid, oneshould see progressiveness in the way disequilibria are taken into account(Eichengreen, 1996). The structure of futures prices should gradually becomedistorted, thus heralding the possibility of a slippage: the more a current accountwidens, the greater the likelihood to be attached to devaluation, and the moreinterest-rate spreads should reflect this risk. It therefore has to be supposed thatsome crises may become self-fulfilling. The mere fact of a speculative attack on acurrency may render nonviable a parity which would have remained viable if theattack had not been launched (Eichengreen, 1996). Speculation in fact raises thestakes, obliges the authorities to defend the exchange rate with measures (interest-rate hikes, in practice) which, because they are self-destructive, are not viable inthe long term. If financial crises are predictable, this is so only in terms of a moreor less probable event at a date which is itself difficult to determine. But precisedating is not possible and the potential amplitude difficult to evaluate.

Once the crisis has broken out, its progress in fact depends on the adroitness ofthe reactions of a large number of players in the public and private sectors.Depending on whether a financial crisis is well or badly handled (the challengebeing all the more difficult to face when it is international and when it affects awide range of financial instruments held by a wide range of players), the cyclicalfluctuation which generally ensues will be either only a modest recession or, onthe contrary, a major depression. The script of a financial crisis is never written inadvance (Kindleberger, 1989). This is why an observer who has identified therising dangers and is not taken entirely off guard by the outbreak of the crisis willhesitate to make somber predictions, at least in public. Even when he findsbrought together all the ingredients of a major economic and financial upheaval(high and largely poor-quality debt, excessive investment, excess capacity,deflationary tendencies, deformation of the allocation of value added, absence ofconsensus on the causes of the crisis and lack of leadership), he will contenthimself with stressing what he refers to as the risks. He will keep his baselinescenario to himself, at the risk of suffering near schizophrenia. But perhaps, in the


end, he is to be excused: knowing as he does that the inflation of the bubble hasbeen largely an irrational process, he will assume that its bursting may also not berational. He will not take on the heavy responsibility of strongly supporting acollective psychology that has suddenly turned pessimistic.

(2) Can Financial Crises Be Contained?When a crisis breaks out, long-term considerations take a back seat to immediateurgency. In all logic, the question of what purpose financial crises serve has to beasked. Are they a necessary stage in the development of a capitalist economy? Dothey constitute the phase of “creative destruction” referred to by Schumpeter? Dothey serve to eliminate the more imprudent decision-makers and wipe out ill-conceived investments?

The Cost of Financial CrisesThe problem is that the immediate cost of a crisis is sufficient to scare those livingthrough it into ignoring the longer-term prospects. Once a process of debtliquidation has started, the soundness of the banks and the integrity of currencyare threatened (Fisher, 1996c; Keynes, 1931). Forced liquidations, a decline in thevelocity of circulation and reductions in lending and bank deposits combine toproduce a general decline in prices. Profits fall, while the real value of debtsincreases, pushing some firms into bankruptcy. These evolutions are particularlyharmful for the banks, since they find themselves in the middle, between the legalowners of the goods and the economic “owners,” the latter lending the former thefunds they need (Keynes, 1931). To reduce the risk they incur in this way, thebanks apply a discount on the market price to the assets they take as guarantee forthe loans. Banking systems can therefore only bear a fall in price that does notexceed this margin of security. Any further fall will mean that their debt, in otherwords, the mass of deposits (that is, the major element in total money supply) willexceed the market value of the assets. The banks are then insolvent and themoney they have issued becomes worthless. Sooner or later, the public becomesaware of this imbalance and the resulting run on the banks by depositors merelyaccentuates the problem. Bank failures can then occur in knock-on fashion,forcing the authorities to bail out the banking system at the expense of thetaxpayer. The quantity of money can decline and its quality deteriorate at the sametime. Distribution circuits for credit and capital allocation can cease to function(Fisher, 1996c). In a money economy, such a development is bound to paralyzethe normal course of trade. The macroeconomic consequences can therefore besubstantial, possibly involving a contraction in national income and employment(Fisher, 1996c). Even though it has to be assumed that not all agents entered atthe top of the cycle, the fall in prices from the peak will erode their confidence. Ifthe fall in prices fuels expectations of further falls, the process becomescumulative. Interest rates may decline in nominal terms for high-qualityborrowers, but in real terms they rise for everyone, starting with the weakerdebtors. Irving Fisher (1996c) has clearly shown the extent to which borrowingand a fall in the general level of prices form an explosive mixture: the moreborrowers repay their debts, the more indebted they become in real terms.


Multiple Transmission ChannelsIn a situation in which national economies are mutually open, the channels forinternational transmission of a shock occurring at any one point in the system arethemselves multiple (Kindleberger, 1989). In this field, contrary to a certaintheoretical sectarianism, historical experience leads to eclecticism (Kindleberger,1995). In certain cases, it is the strictly monetary phenomena and the short-termcapital flows that predominate, as monetarists tend to think. In a gold exchangestandard or currency-board system, the evolution of the money supply isdependent on capital movements. The defense of a currency by means of highinterest rates may not be seen as credible by the market if this policy seems out ofstep with the evolution of the economic situation. In other situations, the shock istransmitted through movements in long-term capital. Traders who have alreadysuffered losses on a given market will be inclined to take their profits in otherswhere they still can. It should be pointed out, incidentally, that in a crisis situation,when mechanisms of forced liquidation are in play, the market value of a givensecurity depends less on the financial soundness of its issuer than that of itsholders. There then arises the risk of a total absence of discrimination betweendifferent borrowers, which is bound to be highly frustrating for the moredeserving of them. Somewhat paradoxically, the globalization of portfolios, if thisis done using borrowed funds, can therefore eliminate its prime justification,namely, diversification. In other circumstances still, it is the direct investmentchannel that will come into play. Keynesians prefer to stress fluctuations inexchange rates—fluctuations which disorganize trade flows and have an impact onthe prices of goods traded on international markets, notably for commodities. Allthis can help to bring down the receipts of exporting countries or can lead toincreased competition for third countries. However, not all crisis countries canhope to obtain, simultaneously and successfully, a rise in their savings through afall in their demand. At any given moment, the contraction in imports threatensto produce a contraction in their exports. Lastly, there is the intangible, but veryreal, channel based on collective psychology, that of entrepreneurs’ “animalspirits” (Keynes, 1936). Every age has its promised land, of a more or lessimaginary character. Disillusion is bound to match the illusion itself. A backlash isby no means rare: a crisis triggered at the periphery of the system by a change inmonetary policy, in the exchange rate or the export of capital on the part of thecentral players, can very well rebound on them. The high potential cost and themultiple transmission channels explain why the political, monetary and financialauthorities generally choose to intervene when a financial crisis blows up. Thissaid, they do not always do so or, in any case, not immediately. The sequence isthen usually as follows: denial that the crisis exists, friendly pressure on thestrongest elements involved to induce them to support the weaker, controls onprice movements, closure of markets, postponement of repayment dates and thedeclaration of a moratorium (Kindleberger, 1989).

Everyone has their ScapegoatUnder the heading of the expedients often used to “manage” a crisis, there mustalso be included the search for a scapegoat, preferably foreign. This seems to be anecessary and early phase in the development of any international financial crisis(Kindleberger, 1989). The fact that the global village can be the scene of a dialog


of the deaf, even in the era of CNN, is nothing new. A crisis is also and always abreak in communication between players whose fates are nevertheless linked.Whether this initial non-interference is based on a philosophical view (“the law ofthe market is hard, but it is the law”) or on the hope of a spontaneous and quasi-miraculous recovery in the situation, it fails to work. If one wants to stem afinancial crisis, more energetic action is required, in particular the intervention ofa lender of last resort.

The Lender of Last Resort Is Now a BuyerOne way of looking at a financial crisis is to see it as a sudden and steep increase inthe demand for cash balances for hoarding purposes, at least until such time asconfidence returns (Kindleberger, 1989). Should one (and if so, on whatconditions) accommodate this demand by lending money created for the occasionby the lender of last resort, in most cases the central bank? Will the operators,knowing that they are liable to be bailed out through the intervention of thelender of last resort, not be inclined to take more risks than if they were able tocount only on themselves? This is an old debate to which practical experienceprovided a positive response long before economists took it over (Kindleberger,1989). The monetarists and their Currency School ancestors favor this type ofintervention, while insisting on the long-term risks it implies. Keynesians and theirpredecessors in the Banking School are also in favor of the intervention of thelender of last resort, being above all concerned with short-term stabilization. Forthe ultra-liberals, on the other hand, there should be no intervention and marketforces should be left to do the work. Once the principle of intervention is decided,the questions for the lender of last resort are as follows: lend, yes, but how much?To whom? On what conditions? When? It was during the 19th century in Englandthat the theory of lender of last resort took shape. The answers to the abovequestions being, yes, the sudden demand for money should be accommodated; itmust be done fast, so as not to allow deflation to take root, through unlimitedlending to solvent agents offering solid guarantees, but at a punitive interest rate,in order to scare off potential free riders. But this is easier said than done, as therapidity and effectiveness of the intervention can themselves redefine the frontierbetween a trader’s illiquidity and his insolvency. In fact, the intervention by thecentral bank becomes that of a “buyer of last resort” when it involves firmpurchases on the open market, as is the case today, rather than rediscounting, asin the past (Kindleberger, 1995). The same observation applies to the centralbanks’ interventions on currency markets, when, for example, they sell foreigncurrencies to protect their own currency or buy dollars in order to curb the fall. Inthat case, the possibility of applying a discriminatory rate disappears. Thearchetype of a successful intervention is that of the Federal Reserve Bank at thetime of the 1987 crash. By declaring itself ready to provide the markets with all theliquidity they might need, the central bank successfully quelled the panic.

The Problem of Moral HazardAbove all, however, the intervention of the lender of last resort must not be takenfor granted (Kindleberger, 1989), because, if so, the problem of “moral hazard”arises, meaning that when financial operators feel that they can always count onthe central banks coming to the rescue, they may take excessive risks. Even Charles


Kindleberger (1995), an advocate of the lender of last resort if ever there was one,has come to wonder whether the idea has not been taken too far, whether thetrade-off between the short and the long term is not too often in favor of theformer. The same problem of moral hazard arises through the mechanisms for the protection of depositors (Kindleberger, 1995). The propensity of the centralbanks, the IMF and the World Bank to fly to the rescue of financial markets whenthe latter find themselves in difficulty has a lot to do with their appetite for risk. Itis possible that the repeated interventions of the various lenders of last resortsimply result in a displacement of the problems in space and time, amplifyingthem as they go. This would explain why crises are becoming more and moredifficult to stem. In addition, it could mean that a lasting restoration of order tothe situation could require the destruction of a certain amount of financial andmonetary capital. Moreover, the cumulative process just described could wellgradually modify the statistical parameters used in the models for the evaluation ofrisk on financial markets. All the current models in fact assume these parametersto be stationary. It may be that the confusion referred to earlier between savingsand money is fueling another equally dangerous one, namely, the confusionbetween risk and uncertainty.

Who Pays for the Losses?The intervention by a lender of last resort is a highly political act inasmuch as itresults in an imposed transfer of wealth which can mean that society at large bearsthe losses incurred by a relatively small number of private operators(Kindleberger, 1995). As a result, such an intervention is equivalent in fact tosubsidizing the prices of financial assets. It therefore presupposes politicallegitimacy and leadership as well as a minimum of consensus within thecommunity. Whatever the cause, greater inequality in the spread of income andwealth is not favorable to the emergence of such a consensus. The problem ofspreading the losses is highly political, since it involves deciding who should bearthe cost of the errors of just a few agents when these errors pose a threat to thewelfare of an entire community. Any solution that totally relieves one category ofprotagonist or another creates an unfortunate precedent and is an open invitationto irresponsibility. The identification of agents liable to pay their dues for thecommon cause is a simple matter: they may be investors, their creditors, thoseguilty of financial delinquency, bank shareholders, bank managers, depositors, thecentral bank or even rentiers of all types if the writing-off of the debt is obtainedthrough inflation (Kindleberger, 1995). It is much more difficult to determine theproper apportionment among them, with each category only too ready to forgetits own share of responsibility for the inflation of the bubble.

The political dimension stemming from the redistributive character of theintervention is particularly evident when this takes place at international level. It iseven more so when the crisis results from a flight of domestic capital rather than awithdrawal of foreign capital—a distinction which is in any case not always easy toestablish. This is because, in the absence of international political institutions andof a central world bank, the responsibility of lender of last resort can only fall onthe financial center which predominates in the world today (that is, the USA) viathe IMF and its own central bank, possibly backed up by the central banks offriendly or allied countries (Kindleberger, 1989). As the function of lender of lastresort at the international level is a heavy responsibility and one which can possibly


harm its own currency, no country ever takes it on through altruism, but alwaysout of political or economic interest, if only so as not to leave a clear field foranother major power. Furthermore, even this requires that the dominant countryhas the will and resources to intervene and that this intervention is compatiblewith its own domestic constraints. Another requirement is that its solicitudeshould not generate in those to whom it is addressed any feeling of resentment orhumiliation. International rescue packages therefore always have a high politicalcontent: the risk is that they may become too politicized (Kindleberger, 1995).

In contrast to the interventions by the US Treasury and the Federal Reservesystem in favor of Mexico in 1982 and 1995, it has to be pointed out that the IMFlacks resources, that it intervenes only slowly and, above all, that its lending is onconditions that are not always transparent. It is therefore merely a shadow of agenuine international lender of last resort (Kindleberger, 1989). If the IMF actsrelatively slowly, this is because its intervention depends on the debtor countries’respect for the conditions set out in the adjustment programs negotiated withthem: monetary tightening, fiscal austerity and, more recently, structural reforms,especially in the financial sector. This conditionality means that it is permanentlyrunning the risk of being accused of favoritism, of unfair discrimination. It is alsoin this light that one should interpret the reactions of disgust heard in Asia at thetime of the bailout of LTCM: why advocate pushing insolvent debtors intobankruptcy and sacking their managers in Asia, if these same principles are notapplied at the very heart of the system? And yet, the stakes are high for thedebtors: to default on bond debt, to fail to pay, not to succeed in re-establishingconfidence means cutting themselves off from the international capital marketsfor at least a generation.

What Should Be Done with Recovered Assets?At national level, a further problem arises for the lender of last resort, namely,what to do with the assets that come onto its books following its intervention(Kindleberger, 1995). Should they be sold on the market and, if so, at what rate, orshould they simply be written off? Here too, depending on the choice made, thecrisis can take a more or less serious turn.

Lender of Last Resort and the “Liquidity Trap”Intervention by the lender of last resort cannot be expected to solve everything(Keynes, 1936). It is indeed reasonable to think that the process of money creationitself (the way in which money “enters the market”) has at least as muchimportance as the quantity of money effectively created. This becomes apparentwhen one considers the mechanisms by which expectations are formed. It ispossible that price expectations (for goods and services or financial assets) are notindependent of the process by which the expansion of the monetary base takesplace. In the framework in which monetary institutions operate today, the centralbank increases the supply of money by purchasing from the commercial bankscredit instruments issued by the public authorities. All things remaining equal,this type of monetary expansion has the result of lowering interest rates. If thedominant mechanism is one of rational expectations, such a fall in the cost ofborrowing is bound to stimulate overall demand. But if expectations are adaptivein nature (and why should they not be at this stage of the cycle as well?), then the


stimulating effect may well be non-existent (Keynes, 1936). One can even imaginethe possibility of an effect contrary to the one expected. This will be the case, forexample, if the majority of players remain convinced that the prices of goods andservices are still headed downward, that a further dose of monetary stimulus willtherefore be required, involving further purchases by the central bank of treasurynotes and bonds. In this case, any injection of liquidity (not, as one often hears itsaid, into the economy, but in reality into the financial markets) will remain thecaptive of the financial markets. In an environment in which agents fear a fall intheir income, it is necessary to eliminate the expectation that buying and holdingfixed-interest securities will remain profitable regardless of the level of interestrates.

Activism in Fiscal MattersMore direct injections of liquidity into the real economy can therefore also be ofuse. These injections must bypass the financial markets—technically possible,although running counter to contemporary monetary orthodoxy (Allais, 1988;Fisher, 1996b; Friedman, 1948). All it takes for this is that the government pay partof its expenditure using an overdraft with the central bank, commonly referred toas the “printing press.” By such a mechanism, the government allocates to itselfthe income corresponding to the quantity of money created, and this can evenpossibly enable it to cut taxes. Direct financing of public expenditure via moneycreation has the advantage of increasing the flows of income that the financialassets are meant to represent. The result is a reduction in the risk (mentionedearlier) of raising the value of these assets to levels that have no relation to reality.At the same time, the disturbing influence exerted by the central bank or thegovernment on the market for loanable funds would be reduced, which should, apriori, improve the process by which capital is allocated. There is also anotheradvantage, particularly appreciable for governments whose public debt, made upof treasury notes and bonds, is already high in relation to national income. Moneyissued by the central bank is in fact a form of public debt that costs nothing ininterest and does not have to be repaid. The experience of the USA in the 1930s orof Japan in the present decade nevertheless shows that the real economy takes along time to recover from the bursting of a major financial bubble. This is becausethe counter-cyclical policies in support of demand are probably fairly impotentwhen the deficiencies of monetary self-stabilizing mechanisms have enabled thebuild-up over a long time (too long) of excessive stocks of debt and productivecapital. Hence a further interest in trying to prevent financial crises.

(3) Can Financial Crises Be Prevented?Through its strictly economic and financial dimensions, but also its political andsocial aspects, the crisis that has broken out in Asia comes as a reminder of theextent to which monetary and financial stability is a public good both useful andnecessary. The nature of the stakes involved are as old as the monetary economyitself, but recent events have distinctly increased their importance. Thegeneralized freedom of capital movements, the predominance of private flows in the capital going to the emerging economies since 1990, disintermediation and its corollary (that is, the increase in the number of players), increasinginstitutionalization of fund management, the development of derivatives markets


and the spread of techniques for real-time telecommunication and dataprocessing all combine to make the problems more complex. Taken together,these six factors require modern financial markets to observe the three unities ofclassic tragedy (unity of time, unity of place and unity of action), while putting onstage a number of characters worthy of the commedia dell’arte. For the sametechnical reasons, the financial markets and the real economy give the impressionof not operating within the same timescale.

The Bold Wager on Market EfficiencyIf one is prepared to accept that financial markets are always efficient (oncondition, naturally, that they are well informed), one can only approve thevarious proposals aimed at promoting transparency, that is, the rapiddissemination of data that are both precise and exhaustive. This applies, inparticular, in the following fields: volumes of outstanding foreign-currencyborrowings by banks, but also by nonbank intermediaries (investment banks,institutional investors, fund managers and “hedge funds”) as well as industrial orcommercial firms; interbank activities; the maturity structure of commitments;capital movements; off-balance-sheet commitments and, more generally, positionsin derivative products; and forward sales of currencies by central banks and netexchange reserves. Imposing on lenders an obligation to make a more detaileddeclaration of their operations to the Bank for International Settlements (BIS)seems in this respect to be the indispensable minimum, in parallel with thepressure exerted by the IMF to obtain greater transparency on the part of theemerging economies. And yet, whether one takes the standpoint of the borrowersor the lenders, it has to be recognized that such a degree of transparency wouldnot bring immediate advantages only. For example, the so-called market-riskcontrol systems do not necessarily bring out the nominal value of thecorresponding assets and commitments. They therefore do not lend themselves tothe monitoring of credit risk. In addition, producing the required informationcarries a high cost, which has to be borne by either the lenders or the borrowers.

Imagining the UnimaginableIn any case, no international body is prepared to guarantee the quality of theinformation made available to the public. The IMF carries out an a posteriori checkon the quality of the information it receives from its member countries.Frequently, the incomplete or defective nature of the information put out onlyemerges once the crisis has erupted. Unfortunately, everything suggests that themarkets develop according to a logic in which they will always be ready to conceiveof new operations, instruments and techniques that, for a time at least, slipthrough the statistical nets. The statistical offices will therefore always be “fightingthe last war” and no statistical series taken in isolation will ever succeed in sendingout the appropriate warning signals. One reason for this is that, in a system of freecirculation of capital, all short-term or negotiable securities issued by private orpublic agents in a given country, whether held by residents or non-residents,whether denominated in local or in foreign currency, are a potential threat to theexchange reserves. For bank economists, probably the best way forward involvesfar-reaching inquiries and cross-checking of information, with the process basedon the largest amount of data possible, which themselves are derived from


numerous sources, including the private sector. It is by comparing a large numberof pieces of information that one can hope to identify possible inconsistencies andpick out certain disequilibria.

Putting a Price on the Absence of InformationIn these circumstances, the absence of information is information in itself, andprobably the most precious item of all, or at least the one that markets shouldincorporate in their risk premiums. The publication by the borrowing countries ofa provisional timetable for the supply of certain information could help players tohave a better idea of what they do not know, possibly with the aid of the ratingagencies. But if the markets do not impose a punitive price on the absence ofhigh-quality information, it is useless to hope that the borrowers will be preparedto incur the costs relating to the production of this information. Moreover, theinformation not only has to be available, it also has to be processed. It is clear, forexample, that the data published in good time by BIS regarding tendencies in thebank borrowing of Asian countries were simply ignored by the markets. With thesame aim of transparency, one can only approve the proposals aimed at improvingthe supervision and regulation of financial systems, especially for offshore centersand nonbank intermediaries. However, here too one has to expect that theregulators will generally be several steps behind reality. Yet another good idea isthat a code of good behavior be imposed on issuers and borrowers as regards thecirculation of the information concerning them. Yet another is to encouragemultilateral surveillance and coordination of macroeconomic policies in capital-receiving countries. However well founded these various proposals, their politicalimplications should not be underestimated. Their effective implementationpresupposes active cooperation from the borrowing countries, where they willundoubtedly be perceived as interference in their internal affairs, not to mentionin the affairs of the handful of private or family groups close to the seat of politicalpower.

The Importance of Not Confusing Risk and UncertaintyAs regards so-called market risks (that is, risks related to the volatility of financialasset prices), particular attention has to be paid to the problem of the stability ofthe statistical parameters on which are based all calculations for the assessment ofthe risks incurred. More or less explicitly, all these models assume that there willalways exist a market price at which it will be possible to deal, that there are nomajor discontinuities in quotation and, lastly, that the various statistical parameters(volatility and covariance) are stationary, in other words, constant over time. Thisis not impossible, but is by no means certain, all the less so in that substantialpositions are taken by numerous operators on the basis of these assumptions.

Strengthening the Means of InterventionEven if one accepts the hypothesis of informational efficiency on financialmarkets, the possibility has to be considered that unexpected exogenous shocksmay trigger a financial crisis (Davis, 1995). In order to avert the engagement ofcumulative mechanisms involving forced liquidations, the sudden increase in thedemand for precautionary cash balances following such shocks must be satisfied


by an institution or a group of institutions acting as the lender of last resort at theinternational level. In no country in the world is water rationed to the fire brigade.In no country in the world are firemen asked to a draw up planning regulationsand ensure that they are respected before they go into action. It goes withoutsaying that the IMF was not designed to perform such a mission. This poses thequestion, to which there is no reply for the moment, of knowing which are thepolitical institutions capable of providing the markets with the institutionalframework without which they are unable to function properly. Although thiswould not be easy, it is also certainly necessary to reduce to a minimum the moralhazard associated with rescue operations. This is true in particular of the implicitguarantees provided by public authorities on the commitments of their nationals.It is also true of the interventions of lenders of last resort: the private sector mustbecome more involved in the resolution of the crisis itself than it has been so far.Not all losses can be passed on to the community. The private sector must bear atleast part of the cost of its errors of judgment. The sheer scale of the problems tobe resolved implies that the whole of the private sector (banks and nonbanks)must participate, as was the case in 1982, on a voluntary basis and in proportion toindividual exposures, in the rescue packages put together by the IMF. It can do sovia rescheduling operations, new money, debt–equity swaps, a freeze on securityportfolios and so on. The aim would be to mobilize sufficient resources to financecurrent-account deficits and normal debt servicing, the impact of the adjustmentprograms having been estimated and making it possible to hope for theachievement of a minimum growth rate by debtors. All this is only conceivable ifthe participants find it in their interest, with the IMF opening its books so that theycommit themselves in full knowledge of the facts, and with the beneficiarycountries guaranteeing them the preferential transfer of hard currency.

Not Forgetting More Radical MeasuresIf, on the contrary, one takes the view that there are times when markets are notefficient and can be victims of preconceptions, cognitive biases and, moregenerally, waves of excessive optimism or pessimism (especially as money creationis the work of private agents monetizing promises to pay), then no doubt it will benecessary to envisage much more radical reforms. For a long time now, analysishas suggested that excessive development of credit and of the monetization ofdebts has contributed to the amplification of cyclical fluctuations (Mill, 1987). Fora long time also, economists have wondered which institutions might bestguarantee economic stability. Three paths seem worth exploring. The first consistsof examining the pertinence of the price indices normally monitored by thecentral banks. The second relates to the elements of a reform of our monetaryinstitutions. The third concerns the generalization of indexation clauses in allcommitments on the future.

Inflation and Its Many FacesInflation is not the word generally used when the price of assets (real or financial)surge, but there is no hesitation in referring to the threat of deflation when theyfall. This asymmetry says much about the way in which we are used to approachingthe problems and suggests a question: Should our price indices not also takeaccount of the prices of financial assets instead of limiting themselves, as at


present, to the prices of goods and services? This is an old question which JacquesRueff had already posed in 1928 in a memorandum to the League of Nations, aquestion which is generally raised only too late (Rueff, 1989). If an imbalanceemerges on the money market between the suppliers of money (purchasers ofgoods, services, securities and so on) and those demanding it (sellers of goods,services, securities and so on), the securities market is indeed one place where thisexcess liquidity can find a home. This spillover will induce an increase in tradingvolumes (thanks to new issues) or a price rise, or both. The rise in prices willpermit the establishment of a new equilibrium point between actual and desiredcash balances. There is therefore a strong reason here for the central banks tokeep a close eye on fluctuations in the prices of financial assets, especially whenactivity on the financial markets (both primary and secondary transactions) issubstantial in relation to GDP.

Contrary to the suggestions made by certain critics of this proposal, it does notinvolve attributing to the central banks a capacity for clairvoyance denied toordinary mortals—remember the numerous injunctions addressed to AlanGreenspan after he had referred to irrational exuberance! It simply involvesincorporating changes in the prices of financial assets into the measurement ofchanges in the general price level. But it goes without saying that such anapproach would be nothing less than revolutionary for certain financial circles.This is because an inflation index calculated in this way would be undeniably morevolatile than the usual indices. It is precisely here that the central bankers’reservations are greatest. What credibility as guarantors of the value of currencywould they have in the light of such price indices? After all, an “abnormal” rise inthe price of financial assets can probably be regarded as an erosion of thepurchasing power of currency. In addition, how should they react to a fall in thelevel of prices that such an instrument would not fail to register more often thanthe usual indices? The type of reaction required in the face of deflation does notcorrespond to the more measured style beloved of central bankers. Theseapprehensions will be all the greater in that the problem of the correctmeasurement of the general price level comes to be posed at a late stage in thecycle. The best way of dealing with it would therefore be not to wait until it is toolate; in other words, when stock-market speculation has become a nationalpastime and is fueled by expectations of capital gains of the order of 15, 20 or 25percent, making a rise of one-quarter of a point in leading rates seems derisory.The redefinition exercise is a difficult one, both technically and politically, butdeserves to be attempted.

Dissociating Money Creation from the Distribution of CreditThe second area for reflection involves re-examining the proposals for reform ofour monetary institutions made by Henry Simons, Irving Fisher (1996b); MiltonFriedman (1948); and James Tobin, and Maurice Allais (1988). These proposalsregularly resurface at the time of major financial upheavals. They take asinspiration the teaching of Knut Wicksell, for whom “with regard to money,everything is determined by human beings themselves” (1934). In a monetaryeconomy, the means of payment do not just appear providentially. Someone has tohave introduced the appropriate quantity. But who? How? How much? These arethe questions to which our monetary institutions have to reply, questions of primeimportance remembering that the creation of money is a source of income for


those who carry it out (Allais, 1988). By giving the central bank the monopoly forthe issuing of money, proposals for monetary reform try to dissociate moneycreation from the distribution of credit, so that this should not depend any longeron the simultaneous and unstable confidence of private lenders and borrowers.Another aim is to prevent the creation of money permitting the creation ofpurchasing power out of thin air by private individuals. Dissociating moneycreation from the distribution of credit also means preventing loan repayment (orborrower default) from destroying money (Allais, 1988; Fisher, 1996b; Friedman,1948). The proposals for monetary reform cover: specialization of the banks asbetween deposit banks and lending banks; the deposit banks would ensure,against billing their clients, cashier services (that is, holding and transferringcurrency) to the exclusion of all lending operations; deposits would also be 100percent covered by central bank money, in other words, by banknotes or reservesheld with the central bank; the central bank would create money by financingpublic expenditure; the income derived from money creation would be passedback to the state; the lending banks’ resources would consist of remuneratedfunds borrowed on the markets; and they would be forbidden to operatetransformation, in other words, to finance uses in the future using resources ofshorter maturity. This centralization of overall money creation would mark a breakwith the present situation of “spontaneous” creation on the basis ofmicroeconomic decisions that are entirely independent of each other. The wholecircuit for the creation of money would be modified.

What Kind of Globalization for Money?In order for this monetary reform to function, it has to be international.Otherwise, the banks could continue to create money on the euro markets.Monetary reform therefore presupposes a reshaping of the system of internationalpayments around a clearing mechanism that would be responsible for ensuringinternational settlements. In this case, the questions that arise concern whatmeans of payment should be used between countries and who should beresponsible for their issuance. There is no need to insist on the highly politicalnature of these questions. History teaches us that international monetary systemshave until now been born by chance (Eichengreen, 1996; Rueff, 1989). The gold-standard system emerged in the 19th century in line with the conversion of theprincipal economic powers at the time to monometallism and their adoption ofthe principle that their national currencies had to be covered by their goldreserves. In this manner, by a kind of spontaneous generation, a unique system ofworld money, and hence of fixed exchange rates, was created without anysupranational central bank. The particular social and political environment rulingat the time meant that it was capable of being long lasting (Eichengreen, 1996).

The financing requirements of the First World War shattered the system. Sincethe bulk of the world stock of gold had now become the property of the USA, asolution had to be found for the reorganization of international settlements. Thiswas carried out in 1922 at the Genoa Conference, which adopted an arrangement,anodyne at first sight, aimed at economizing the use of gold in internationalsettlements (Rueff, 1989). This provision enabled central banks to issue theirnational currency against holdings in foreign currencies, provided that the latterwere themselves covered by gold holdings. Thus was born, “by the back door,” thesystem known as the “gold exchange standard.” When substantial private capital


flows headed from the USA and the UK to continental Europe from 1925 on, thissystem was at the origin of an inflation of money supply which, according toJacques Rueff (1989), “organized the Great Depression.”

In 1944, at the time of the Bretton Woods conference, it was clear that thelesson had been only half learnt. The British delegation, led by Keynes, proposedthe creation of a Clearing Union that would have been a genuine supranationalcentral bank. According to this proposal, international settlements would becarried out using an international currency, the “bancor,” issued by the ClearingUnion. It would have implied the USA giving up the income procured through theissue of international currency. Not surprisingly therefore the British proposal wasrejected by the American delegation and the Bretton Woods agreements made thedollar the linchpin of the international monetary system. However, persistentAmerican external deficits finally undermined the value of the dollar, while at thesame time being a recurrent source of financial and monetary instability (Rueff,1989). All the zones in the world which, at a given moment in history, haveaccumulated substantial dollar balances have come, sooner or later, to regret it.The rapid build-up of exchange reserves denominated in dollars is probably themost reliable sign of trouble to come.

The Need for a Self-Stabilizing Currency MechanismIn an ideal world, the international currency system should operate as a self-stabilizing mechanism tending to eliminate external payments deficits, as anarrangement providing order and stability for the currency markets (Eichengreen,1996). This is what is referred to by the “price-specie flow mechanism” describedby Hume: when a country is in deficit, it loses means of payment, thuscorrespondingly reducing the purchasing power in circulation. This diminution inthe means of payment depresses domestic demand and prices, bringing theexternal accounts back toward balance, especially as the same mechanism, butreversed, comes into play in the surplus country. Our international system doesnot function in this way. As shown earlier, the more internationally widespread theuse of the dollar, the more dollars there are. In the current debate over thebenefits and risks of the globalization of capital markets, there is permanentconfusion between genuine savings and money creation from nothing. Butinternational capital flows are by themselves generators of money creation and aremuch larger than the current-account deficits they are supposed to finance. This isa very far cry from David Hume’s “price-specie flow mechanism.”

Are We Seeing a Trend Back to Controls on Capital Movements?The emerging-country crisis is a fresh illustration of an old dilemma in the form ofthe incompatibility between freedom of capital movement, a system of fixedexchange rates and autonomy for a monetary policy having both internal andexternal objectives. In a somewhat special social and political context, the goldstandard solved the problem by setting a unique objective for monetary policy,that is, maintaining the exchange rate (Eichengreen, 1996). This is also what thecurrency-board systems are doing today. Another way of breaking out of theincompatibility triangle just described is to adopt a system of floating exchangerates. For economies that are both large and relatively closed (as is the case for theUSA, Japan and now Euroland), this is a viable solution. But for economies that are


both small and wide open (such as the European countries before the euro or theAsian economies), floating is not without its risks or its cost. The experience of the1930s clearly shows that free trade and unstable exchange rates are not goodbedfellows. The geographic zones where intra-regional trade is highly developedtherefore have no other solution but to adopt a single currency (Eichengreen,1996).

This leaves the third apex of the triangle, namely, capital movements. Thearguments in favor of freedom of capital movement are well known: it is desirablethat the saving surpluses in the aging countries should be capable of beinginvested in countries with rapid population growth and a scarcity of capital; it isalso desirable that the financial systems and governments of the emergingcountries should find themselves subjected to discipline via the financial markets;all this is bound to improve overall economic efficiency and accelerate worldgrowth. While these arguments deserve to be taken into consideration, it isimportant not to overlook either the very special way in which financial marketsfunction or the very inadequate self-stabilization operated by what passes for aninternational monetary system. Total mobility of capital is therefore notunanimously approved of by specialists in international economics. Moreover, thetheory of comparative advantage subscribed to by most of the advocates of freetrade presupposes (a fact frequently forgotten) that the factors of production areimmobile. Given the international monetary disorder, selective control of capitalmovements is not necessarily a bad thing. This implies preferring slow-but-steadygrowth to the rapid-but-irregular variety. Such arrangements are probably easier toapply than the proposal for an international financial tax, whose paternity Tobinhimself attributes to Keynes (Keynes, 1936). In the immediate postwar period anduntil quite recently, western Europe and Japan had a regime of controlled mobilityof capital, but this did not prevent them from posting remarkable growth in realincome per head. Advocates of total freedom of capital movement object thatcontrols are impossible to apply satisfactorily. Admittedly, any control system willalways be permeable at the margins. However, if the political will is there, if thesystem is administered by competent and honest officials (an assumption which isnot necessarily stronger than that of rationality on financial markets), a selectivecontrol of capital movements is feasible, the aim possibly being merely todiscourage inflows of capital that are short term in nature or otherwise inherentlyunstable. One can only hope that this will not be a case of “out of the frying panand into the fire.” All in all, the analysis of the functioning of the internationalmonetary system clearly shows the dangers of a mechanism based on themonetization of debts and deficits.

But Since There Is Monetary Instability...The reshaping of our price indices is a long-haul task with an uncertain outcome.As for monetary reform, this is a “revolutionary” proposal for which the conditionsfor its international adoption are not ripe and will not be as long as the perceptionprevails that the upheavals resulting from such reform would outweigh theinconveniences of the state of crisis. This makes it interesting to evoke asupplementary proposal, less radical in nature, namely, to generalize the inclusionof indexation clauses in commitments on the future (leases, annuities, loans andso on) (Allais, 1988). The most urgent aspect of this would be to index the interestand capital of loan contracts to the general price level, a technique that has


already developed considerably in a certain number of countries in the past 20years. Historical experience suggests that, within the framework of the currentmonetary institutions, the stability of the purchasing power of a currency is by farthe least likely hypothesis, especially if the general price level is measuredcorrectly. Independence for the central banks makes little difference. Moreover,for want of having tried it, one cannot be certain that monetary reform willproduce all the hoped-for effects. If inflation and deflation are risks which are, ifnot inevitable, at least chronic, the indexation of debts makes it possible to protectboth lenders and borrowers from the results. It is also an assurance provided bythe public authorities that they will abstain, for lack of objective interest, from theexpedient consisting of inflation (Allais, 1988). Indexation amounts todissociating the functions of a unit of account and a means of payment attributedto a currency—a dissociation already practiced in the past. It would instill aconsiderable element of flexibility into the functioning of our monetaryeconomies. For lack of such flexibility, we are left exposed to the possibility thatthe natural volatility of the prices of financial assets may become unbearable andlead inevitably to mechanisms involving subsidies on the part of the lender of lastresort. The UK showed the way in this matter by issuing indexed bonds as long agoas 1981. Others have since followed in its footsteps: the USA issued its first indexedbonds in 1997 and France has also recently adopted this technique. It can only behoped that it will become more general, especially in countries where the publicdebt ratios are high. An extension of this technique to long-term commitments bythe private sector is also desirable.

By Way of Conclusion“From here to eternity” there will be financial crises. Pareto (1964) drew theconclusion that it was possible to conceive that crises would occur even in aplanned economy, since planners are fundamentally no different from other men,capable of oscillating between enthusiasm and despair. The somber history of thiscentury has shown that he was not entirely wrong. It is not the market economythat is in question, but rather the framework in which it operates (Allais, 1988).This observation, according to which economic and financial crises are, in thefinal analysis, a product of our all-too-human frailty, does not signify resignation,far from it. It is more a call for lucidity in the face of the type of dynamic that canmove financial markets in certain circumstances and with certain monetaryinstitutions. It is also a call to develop, promote and exploit the teachings ofeconomic and financial history, the sole laboratory open to economists. This isperhaps the means (the most reliable means) of countering the tendency to forgetthe major crises and the trap (into which we so often fall) set by the tranquillityparadox. The free market, it might be said, is the worst system except for all theothers. Granted. But, if Lord Acton is to be believed, “power drives men mad,absolute power drives men absolutely mad.” A market economy cannot function ifmoney is unable to play a self-stabilizing role.

Note1. These “insurance premiums” take the form of additional remuneration over and above

the risk-free interest rate: that of short-term treasury notes.


ReferencesAglietta, M. (1997 [1976]). Régulation et crise du capitalisme, 2nd edn. Paris: Editions Odile

Jacob.Allais, M. (1988 [1977]). L’impôt sur le capital et la réforme monétaire, 2nd edn. Paris: Hermann.Allais, M. (1993 [1954]). Les fondements comptables de la macro-économique, 2nd edn. Paris: PUF.Arrow, K.J. (1987). “Economic Theory and the Hypothesis of Rationality.” In The New

Palgrave (J. Eatwell, M. Milgate and P. Newman, eds). London: Macmillan.Blaug, M. (1980). The Methodology of Economics. Cambridge: Cambridge University Press.Claudel, P. (1993). La crise—Correspondance diplomatique Amérique 1927–1932. Paris: Métailé

Transition.Davis, E.P. (1995). Debt, Financial Fragility and Systemic Risk. Oxford: Oxford University Press.Eichengreen, B. (1996). Globalizing Capital: A History of the International Monetary System.

Princeton, NJ: Princeton University Press.Fisher, I. (1996a [1932]). “Booms and Depressions.” In The Works of Irving Fisher (W.J.

Barber, ed.), Vol. 10. London: Pickering & Chatto.Fisher, I. (1996b [1935]). “100% Money.” In The Works of Irving Fisher (W.J. Barber, ed.), Vol.

11. London: Pickering & Chatto.Fisher, I. (1996c [1933]). “The Debt-Deflation Theory of Great Depressions.” Econometrica.

Reprinted in The Works of Irving Fisher (W.J. Barber, ed.). London: Pickering & Chatto.Friedman, M. (1948). “A Monetary and Fiscal Framework for Economic Stability.” American

Economic Review, June, 38(3).Friedman, M. and A. Schwartz (1963). A Monetary History of the United States 1857–1960.

Princeton, NJ: Princeton University Press.Kaldor, N. (1939). “Speculation and Economic Stability.” Review of Economic Studies.Keynes, J.M. (1931). “The Consequences to the Banks of the Collapse of Money Values.” In

Essays in Persuasion. London: Macmillan.Keynes, J.M. (1936). “The State of Long-Term Expectations.” In The General Theory of

Employment, Interest and Money, Ch. 12. London: Macmillan.Kindleberger, C. (1989 [1978]). Manias, Panics and Crashes, 2nd edn. New York: Basic

Books.Kindleberger, C. (1995). The World Economy and National Finance in Historical Perspective. Ann

Arbor: University of Michigan Press.Mill, J.S. (1987 [1848]). “Influence of Credit on Prices.” In Principles of Political Economy,

Book III, Ch. 12. Fairfield, NJ: Augustus M. Kelley.Minsky, H.P. (1975). John Maynard Keynes. New York: Columbia University Press.Pareto, V. (1964 [1896–97]). “Les crises économiques.” In Cours d’économie politique, Vol. II,

Book II, Ch. IV. Geneva: Librairie Droz.Phelps, E. (1987). “Marchés spéculatifs et anticipations rationnelles.” RFE, summer, II(3).Rueff, J. (1989). “Various Texts Gathered by François Bourricaud and Pascal Salin.” In

Présence de Jacques Rueff. Paris: Plon.Samuelson, P.A. (1994). “The Long-Term Case for Equities and How It Can Be Oversold.”

Journal of Portfolio Management, fall.Wicksell, K. (1934 [1906]). Lectures on Political Economy, Vol. II. New York: Augustus M.

Kelley.Wicksell, K. (1936 [1898]). Geldzins und Güterpreise bestimmenden Ursachsen (Interest and Prices.

A study of the Causes Regulating the Value of Money). London: Macmillan.

Biographical NoteERIC BARTHALON graduated from the Ecole Supérieure de Commerce de Paris in1979 and holds a law degree from Paris University (1979). He started hisprofessional life with Paribas’ Private Banking Department in 1980 as a bond


portfolio manager, becoming Head of Asset Allocation Research and then ChiefEconomist at Group Paribas. In 2000, he joined Allianz Asset Management to headits economics and strategy unit. He has published several articles in Conjoncture,Paribas’ monthly economic bulletin, including: “The Post-1990 Surge in WorldCurrency Reserves” (October 1996), “When Does the Dollar Not Fall?” (April1997), and “What if Inflation were to be Around 10% in 2002?” (February 1999).Contributions to edited volumes include: “Nouvelle économie et capacité d’oubli”(“New Economy and Memory Decay”) in Crises Financières, published byEconomica in Paris during 2001; and (with Jacques de Larosière) “TheRestructuration of the European Banking Industry,” in Adapting to FinancialGlobalisation, edited by M. Balling, E.H. Hochreiter and E. Hennessy, andpublished by Routledge in London and New York in 2001. ADDRESS: AllianzDresdner Asset Management, Nymphenburger 112–116, Munich 80366, Germany[email: [email protected]].

Acknowledgments. An earlier version of this article appeared in Conjoncture, Paribas’ monthlyeconomic bulletin. The author wishes to thank Pascal Blanqué and Olivier de Boysson fortheir very critical and yet friendly reviews of the first drafts, Francis Wells (whose translationfrom French to English is a testimony to the depth and breadth of his economic culture)and, last but not least, Patricia Chanoinat and Elisa Laik for their demanding and yetsmiling dedication to their editing responsibilities.


Why Do Governments and the IMF Enter intoAgreements? Statistically Selected Cases

JAMES RAYMOND VREELAND

ABSTRACT. Why do governments and the International Monetary Fund(IMF) enter into agreements? Conventional wisdom holds thatgovernments turn to the Fund for a straightforward reason: they need anIMF loan. An alternative argument is that governments want IMF

conditions to be imposed to help push through unpopular economicreforms. To illustrate how the desire for IMF conditions can play a role inthe decision to enter into an agreement, this article considers twoanalytically significant cases. Drawing upon a data set of 7011 country-year observations of 199 countries from 1952 (or year of independence)to 2000, two outliers are selected: the country observed with the leastneed for an IMF loan participating in an IMF program (Uruguay in 1990)and the country with the strongest need for an IMF loan not participatingin an IMF program (Tanzania in 1983).

Keywords: • Africa • Financial markets • International Monetary Fund• International political economy • Latin America

(1) IntroductionWhy do governments enter into programs sponsored by the InternationalMonetary Fund (IMF)? IMF programs are supposed to be imposed by the Fund as acondition for providing a government with access to foreign exchange. Accordingto the IMF Articles of Agreement, a country may enter into such a program when it“represents that it has a need to make the purchase because of its balance ofpayments or its reserve position or developments in its reserves” (IMF, 1997: ArticleV, Section 3). Suppose one could measure the “need” a country has for foreignexchange. One could imagine the different types of cases presented in Figure 1.Cases found in Cells 1 and 4 of Figure 1 would support the understanding thatgovernments turn to the Fund when they need a loan, while cases in Cells 2 or 3




would not: countries that do not need foreign exchange should not participate inIMF agreements, and countries that need foreign exchange should participate.

Drawing upon a data set of 7011 country-year observations of 199 countriesfrom 1952 (or year of independence) to 2000, I find cases that fit into each of thecells of Figure 1. In particular, there are substantial numbers of observations thatfit into Cells 2 and 3. This is consistent with what other studies have found. AsGraham Bird notes in a review of recent empirical research, while foreign reservesinfluence the demand for IMF arrangements, “neither a current account nor anoverall balance of payments deficit provides sufficient motivation for seeking Fundsupport” (1996b: 1755).

An alternative argument can explain such cases. According to this view,governments bring in the Fund to help push through unpopular economicreforms (Spaventa, 1983; Remmer, 1986; Vaubel, 1986; Putnam, 1988; Stein, 1992;Edwards and Santaella, 1993; Bjork, 1995; Dixit, 1996; Przeworski and Vreeland,2000). Governments may enter into IMF agreements not simply because they needan IMF loan, but because they want specific conditions to be imposed. As Putnam(1988: 457) explains, “International negotiations sometimes enable governmentleaders to do what they privately wish to do, but are powerless to do domestically ...this pattern characterizes many stabilization programs that are (misleadingly) saidto be ‘imposed’ by the IMF.” An executive may enter into IMF agreements to pushthrough unpopular domestic policies—a phenomenon Gourevitch (1986) calls“the second image reversed.” Thus, some governments may sign an agreementeven if they do not need a loan (Cell 2). Other governments may not want IMFconditions imposed upon them. They may choose not to sign an IMF agreementeven if they need a loan (Cell 3). To explore how the utility of IMF conditions caninfluence the decision to enter into an IMF agreement, I study two analyticallysignificant cases: (1) the case observed with the highest level of foreign reservesparticipating in an IMF program, that is, Uruguay in 1990, and (2) the case observedwith the lowest level of foreign reserves not participating in an IMF program, that is,Tanzania in 1983.

Can one generalize from such cases? While only statistical tests can establishtypical patterns about the population (see Vreeland, 2001, 2003), the analyticallysignificant cases presented in this article indicate that governments entering intoIMF agreements derive some utility from IMF conditionality. The case of Tanzania in1983 sets a minimum bound on the utility derived from IMF conditions, and thecase of Uruguay in 1990 indicates a range. Governments that enter into IMFarrangements derive more utility from conditionality imposed by their IMFagreements than the government with the strongest need for an IMF loan but withno IMF agreement (Tanzania), and they may even desire conditionality to be


Participating Not Participating

Need

foreign exchangeCell 1 Cell 3

No need for

foreign exchangeCell 2 Cell 4

FIGURE 1. Participation in IMF Agreements According to Need for Foreign Exchange

imposed, as the case of an agreement with no need for a loan illustrates(Uruguay). The purpose of exploring these cases in detail is to establish that, infact, the preferences of the governments over IMF conditionality were decisive. Thearticle is organized as follows. The second section describes how I chose thesecases. In the third and fourth sections, I focus on the two cases of interest:Tanzania and Uruguay. The fifth section concludes.

(2) Statistically Selected CasesHow should cases be selected? Consider the different types of cases that a questionidentifies. The substantive question of this article is why governments enter intoagreements, and the standard answer is that they need foreign exchange. Thus,the “types” have to do with a country’s need for foreign exchange.

Suppose one starts with “level of foreign reserves” as a first-cut proxy for thisneed. There are, of course, other measures of a country’s need for foreignexchange that the IMF uses, such as the country’s balance of payments position, butsuch measures indicate a flow of exchange. Ultimately, what matters togovernments is whether they have on reserve enough foreign exchange tocontinue to purchase necessary imports. Thus, the foreign-reserve position,measured in terms of monthly import requirements, is a good starting point. Iaddress other economic indicators when exploring the specific cases below.

Table 1 divides 3262 country-year observations of lagged foreign reservesavailable from the World Bank for 168 countries from 1952 to 1998 into fourtypes: country-years with high and low reserves observed participating or notparticipating in IMF agreements.

Note that 56 percent of the observations in Table 1 are consistent with the viewthat governments sign when facing a foreign-exchange crisis (those observationsalong the downward-sloping diagonal). In 689 country-year observations, reservesare low and the government participates in an IMF agreement. In 1144observations, reserves were high and the government does not participate in anIMF agreement.

But what about the remaining 1429 observations? Why are there 487 country-years in which governments have strong foreign reserves, but choose to participatein an IMF program the following year? Why are there 942 country-years with lowreserves and no IMF agreement? The need for an IMF loan may not be sufficient to

VREELAND: Governments and the IMF 323

TABLE 1. Participation in IMF Agreements According to Lagged Foreign Reserves

Country-Year Observations

Reserves Lagged One Year Participating Not Participating Total

Low 689 942 1631High 487 1144 1631Total 1176 2086 3262

“Reserves” are foreign reserves measured in terms of monthly imports. “Low” reserves areset at less than 2.6 times monthly imports. This arbitrary cut-off point is the approximatemedian level of the reserves of the 3262 observations available on the 168 countriesconsidered in this study, and is also the average level of reserves for countries participatingin agreements. Source: World Bank, World Development Indicators 2000.

explain these observations. To gain analytical leverage over how the desire for IMFconditions may enter into the decision-making process, I choose two starkexamples from the sets of observations noted above.

My full data set includes 7011 observations of 199 countries from 1952 (thedate of the first IMF agreement) or date of independence to 2000—nearly allpotential country-years in which an IMF agreement could have taken place. Iobserve 142 countries that participated in 334 separate spells of IMF agreements.(There were actually 928 agreements signed, however, most of them were signedconsecutively.) Uruguay in 1990 had the strongest reserves of any country ever toenter into an agreement, with both the balance of payments and the currentaccount in surplus.

Regarding the other type of interesting case, in which countries do notparticipate in IMF programs despite low reserves, Table 1 indicates there are 942country-year observations. The most extreme case for which data is available isTanzania in 1982, which averaged foreign reserves of less than 0.05 times monthlyimports did not enter into an IMF agreement for the next 3 years.1

In the next two sections, I focus on the cases of Uruguay and Tanzania. Uruguayis the starkest example of a country with an IMF program and no strong need for anIMF loan, while Tanzania is the starkest case of a country without an IMF programand a strong need for an IMF loan. I argue that one can best understand theiractions by looking at their stances toward IMF conditionality rather than their needfor an IMF loan.

(3) UruguayUruguay did not need foreign currency when it signed the 1990 agreement.Indeed, of the US$136.7 million provided by the 15-month standby arrangement,Uruguay drew down less than 10 percent of the credit provided. This amounted toabout one-tenth of 1 percent of gross domestic product (GDP). The overall balanceof payments of Uruguay was in surplus in 1989 and 1990 as was the currentaccount balance (see Figure 2).2 Uruguay also held a strong foreign-reservesposition. The 1988 reserves were 10.2 times the average monthly importrequirements. This level dropped to 7.7 by 1990, but remained strong bydeveloping-country standards. The rest of Latin America held reserves of only 3.7times average monthly import requirements; reserves in Uruguay were more thandouble this (see Figure 3).

But Uruguay faced other economic problems. From 1951 to 1990, inflationaveraged 47.9 percent per annum. In 1990, it was more than double this, reaching112.5 percent, the highest level since 1968. Uruguay also faced severe foreigndebt, which had shot up in the early 1980s, reaching 89.7 percent of gross nationalproduct (GNP) in 1985. Investment also suffered. After reaching a high point of23.3 percent of GDP in 1979, investment fell almost every year thereafter. By 1990,it had dropped to 9.9 percent of GDP—the lowest level since 1969.

Uruguay faced low investment, high debt and the worst inflation it had seen inmore than 2 decades. These economic indicators prompted the newly elected3

administration of President Luis Alberto Lacalle Herrera to push for economicreform, particularly calling on the congress to cut the budget deficit. Under theIMF programs of the late 1970s, the budget deficit had been eradicated (see Figure4). However, no sooner had the budget gone into surplus than governmentconsumption began to rise. By 1981, government expenditure once again outpaced


government revenue. Still under IMF programs, the government brought thedeficit down again to less than 1 percent of GDP in 1986, but it left the IMFprograms in 1987. After the country left, the deficit again began to grow. By 1989,it had reached 3 percent, the highest deficit since 1984. The new governmentsought to reduce the size of the public sector, but it faced opposition.


-8

-6

-4

-2

0

2

4

6

75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90

Year

Pe

rce

nt

of

GD

P

Current Account Balance of Payments

FIGURE 2. Uruguay 1975–1990 Current Account and Overall Balance of Payments

0

2

4

6

8

10

12

14

16

75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90

Year

Fo

reig

n

Re

se

rve

s

Uruguay Rest of Latin America

FIGURE 3. Uruguay and the Rest of Latin America 1975–1990 Foreign Reserves (In Terms of MonthlyImports)

Why Did Lacalle Need the IMF?To understand the tenuous political position of the reform-oriented Lacalle, it isuseful to review the historical development of the welfare state in Uruguay as wellas the political institutions and parties that supported the large role of the state inthe economy. Early in the 20th century, Uruguay became one of the first welfarestates in the world and the first in Latin America (Mesa-Lago, 1978, 1985; cited inFilgueira and Papadópulos, 1997: 363). The flourishing economy generatedgovernment surpluses, which “helped finance the development of protectedimport substitution industries and the expansion of the state apparatus” (Filgueiraand Papadópulos, 1997: 369). A patron–client system developed in which thepolitical parties used the state to redistribute resources to their constituencies.Filgueira and Papadópulos (1997: 381) define this system as one “of redistributionof political goods sustained by the material and regulatory resources of the state.”Thus, both main parties, the Blancos and the Colorados, advocated a strong rolefor the state in the economy, promoting state-owned enterprises, protectionistpolicies and generous social benefits.

Due to “high levels of incorporation,” large segments of the population came tohave a vested interest in maintaining the welfare state (see Filgueira andPapadópulos, 1997: 368, 379). Thus, when the economy stagnated in the late1950s and crisis exploded in the 1960s, large blocs of voters opposed the austeritymeasures suggested to cope with the difficulties. Both the Blanco and theColorado parties found it more convenient to continue doling out rewards to theirpolitical clientele than to reduce state spending. Nevertheless, with the resourcesof the state diminished, the government turned to the IMF, entering austerityprograms from 1961 to 1963 and 1966 to 1973.

The austerity measures suggested under the IMF programs mobilized


-2

0

2

4

6

8

10

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93

Year

De

fic

it

Not Participating ParticipatingEntering Program

FIGURE 4. Uruguay 1972–1993 Government Budget Deficit (Percentage of GDP)

opposition. Small left-wing parties began to work together in the early 1960s(Gonzalez, 1995: 152). In 1971, the “Frente Amplio” (a coalition of the parties ofthe left) was born and took 18 percent of the national vote (Gonzalez, 1995: 152).In addition to this success at the polls, outright insurrection of the left wing in thelate 1960s and early 1970s pressured the government to resist austerity. In the faceof economic crisis and rising demands from the left, the military took over in1973, restricting legal political activities. The military government slashed thebudget deficit when it signed an IMF agreement in 1975, and it continued to signIMF agreements until stepping down in 1985. When the military held elections fora new civilian regime in November 1984, the fiscal deficit had risen again. By thistime, both the Blanco and the Colorado parties had changed their views on thestate’s role in the economy—they wanted it reduced. The Colorado Party won thepresidency and pressed for rapid liberalization, signing with the IMF in September1985.

With the end of dictatorship, however, opponents of IMF-style austerity wereallowed to organize legally again. Labor leaders called general strikes opposingthe influence of international organizations in domestic politics. The FrenteAmplio re-emerged. In addition, 13 of the most important associations of retireesformed the Plenary of Associations of Retirees and Pensioners of Uruguay(Filgueira and Papadópulos, 1997: 365). Overall, organizations representingretired persons and pensioners constituted about 20 percent of the totalpopulation, and an additional 10 percent were public servants (Rial, 1986: 136).

In November 1989, in the first fully free election since 1972, the opponents ofeconomic reform did well at the polls. Frente Amplio gained a presence incongress and its candidate was elected mayor of the capital city, which includes 42percent of Uruguay’s 3.2 million people. Nevertheless, the reform-orientedLacalle of the Blanco Party took the presidency. Upon election, Lacalleannounced plans for economic reform “if necessary” (New York Times, 1989). Hetalked of raising taxes and cutting public spending. He wanted to privatizenational industries, reform the state pension system and reduce the size of thestate.

But Lacalle had hardly received a mandate. His party received only 37 percentof the presidential vote (Schooley, 1997), and Lacalle himself received only 22percent (Financial Times, 1994b).4 The Colorados received 30 percent of thepresidential vote, and Frente Amplio received 21 percent. The main oppositionparty received nearly as many votes as Lacalle himself, and overall, 78 percent ofthe electorate had not voted for Lacalle.

In the congressional elections, the Blanco Party only won 40 percent of theseats (Schooley, 1997). The main opposition party, Frente Amplio, had enoughseats in congress to force Lacalle to form a coalition government between hisBlanco Party and the Colorado Party. This coalition was tentative at best: Lacalle“obtained ... a vague agreement that did not create politically binding obligations,with non-political Colorado cabinet ministers.... [T]he loose agreement wastermed the coincidencia nacional, or ‘national coincidence’” (Gonzalez, 1995: 161).

There were three further reasons why the newly elected president needed aninternational ally to get past domestic opposition: (1) institutional factors leadingto a lack of party discipline, (2) the patron–client promotion system that failed toreward technocrats, and (3) the institution of the national referendum.

1. Lack of party discipline. Lacalle could not depend upon the loyalty of his


governing coalition. Gonzalez (1995: 147) explains that the electoral votingsystem, with its “double simultaneous vote,” weakened party discipline becausethe system allowed for intra-party preferences at the voting booth. Primariesand elections were held simultaneously, such that the party with the most voteswon the election and the candidate with the most votes from that party tookoffice. Thus, Gonzalez argues, “Candidates cannot rely solely on their party;they must distinguish themselves from competitors within their own party. Theymust develop at least a minimal organizational base of their own” (1995: 147).5

2. Patron–client promotion. Furthermore, Lacalle could not look to nationaltechnocrats for support because there were so few of them in positions ofpower. Promotion within political parties was based not on merit or training,but on patron–client relationships (Filgueira and Papadópulos, 1997: 380).Thus, unlike other developing countries, western-trained economists (that is,IMF-style economists) had not been promoted to the upper echelons of theparties.

3. National referendum. Lacalle was also up against an additional institution thatcould work against his reform efforts: the national referendum (Filgueira andPapadópulos, 1997; Lupia and McCubbins, 1998). Opponents could petition tohave legislation put to a national plebiscite that could override any reformspushed through congress. As Huber and Stephens (2000: 20) point out, “theinstitution of the referendum constituted a veto point where opponents ofprivatization could successfully mobilize.”

Not surprisingly, Lacalle faced difficulties right from the start of hisadministration. Labor leaders called four general strikes and numerous smallerstrikes to oppose Lacalle’s policies, promising him “not one day of truce” (New YorkTimes, 1990a).6 Some 6 months into their terms, while Frente Amplio’s mayor ofMontevideo enjoyed a 55 percent approval rating in the capital, Lacalle’snationwide approval was a mere 18 percent (New York Times, 1990b). FrenteAmplio, labor leaders and even members of Lacalle’s coalition governmentconfronted his program. Even members of Lacalle’s own party publiclydenounced parts of his economic policy.

Indeed, “The party identification of the leaders did not prevent them fromopposing the positions of their parties ... during the administration of PresidentLacalle” (Filgueira and Papadópulos, 1997: 365). By mid-1991, one of thepresident’s cabinet ministers resigned, protesting the economic program. InJanuary 1992, Lacalle decided to reorganize his cabinet to increase support fromwithin his government. The plan backfired. No one was willing to participate inthe government, and the president could fill only three cabinet positions. Neitherthe Blanco nor the Colorado Party wanted to be associated with Lacalle’s program.The administration correctly feared that so many compromises would be necessarythat economic restructuring would be diluted (New York Times, 1990a). Lacking thedomestic political support to put forth his program of economic reform, Lacalleturned to an international ally: the IMF.

The 1990 IMF ArrangementThe Lacalle government proved extremely eager to have IMF conditions imposed.Less than 2 weeks into his term of office, Lacalle announced the preparation of anew letter of intent for an IMF agreement. He said he expected to sign the


agreement within 2 weeks (Busqueda, 1990; Financial Times, 1990a). This was inmid-March 1990. In July, the Lacalle administration announced that thegovernment had clinched an 18-month standby arrangement for US$150 million(Financial Times, 1990b). In fact, however, the arrangement had still not beensigned. It was not signed until December, and it was for less money and fewermonths than the government had announced when it declared the deal“clinched.” Thus, the statements of the government were premature andmisleading. By announcing the agreement with the IMF early, however, Lacalle wasable to use IMF conditionality to push through reforms right away.

The original announcement of the IMF letter of intent revealed specificconditions attached to the loan. The Lacalle administration reported that the IMF-sponsored program required increasing value-added tax by 1 percent to 22percent, raising the basic income tax from 17 percent to 20.5 percent, andincreasing the basic import duty from 10 to 15 percent (Financial Times, 1990a).The goal of the program was to cut the fiscal deficit to 2.5 percent of grossdomestic product. Under the guise of conditionality, the government exceededthis goal, pulling off a budget surplus of 0.37 percent for 1990 (recall Figure 4). Byeffectively putting the country under IMF conditions ahead of schedule, Lacallesuccessfully changed the balance of the budget by 3.4 percent of GDP. Meanwhile,the IMF itself made no press release regarding the standby arrangement until thearrangement actually began on 12 December 1990.

What did the IMF say about Uruguay’s need for a loan? The IMF made nomention of the reserve position of Uruguay—perhaps because Uruguay’s reserveswere conspicuously strong. Regarding the balance of payments situation, the IMFhighlighted that the agreement would help to continue the efforts of thegovernment toward “strengthening the balance of payments” (IMF, 1991: 12–13).The Fund did not point out that Uruguay had actually maintained a balance ofpayments surplus for the past 2 years and a current account surplus for the past 3years.7

What did the IMF announce regarding Uruguay’s need for conditions? TheFund “stated that its intention was to assist the government’s efforts to strengthenpublic finances and tighten credit policy” (Economist Intelligence Unit, 1991a: 11).Officials attributed the economic problems in Uruguay to a rising governmentbudget deficit (IMF, 1991: 12–13). Hence, a key element of the program wasreducing the public sector. The agreement also called for improving publicfinances, linking public-sector wages to the projected decline in inflation,deregulating and privatizing public enterprises, and reorganizing the socialsecurity system (IMF, 1991: 12–13). The last condition was not met during Lacalle’sadministration, but Uruguay complied with the other conditions of theagreement. Indeed, when the 1990 agreement ended, the IMF declared the1990/91 program a success (IMF, 1992: 238–39). By raising taxes and cuttingspending, the government built on its 1990 budget surplus, increasing it to 0.91percent of GDP for 1991, the highest surplus recorded in Uruguay’s history (recallFigure 4).

Despite ultimately giving Uruguay its seal of approval, the IMF stepped up thepressure to reform the economy over the course of the arrangement. For example,during the first half of 1991, the Fund withheld its approval of the disbursement ofUS$65 million in World Bank loans and US$800 million in Inter-AmericanDevelopment Bank loans that were to be used to buy back foreign debt(Economist Intelligence Unit, 1991b: 13). This increased the pressure on congress


to approve Lacalle’s proposed new taxes and adjustments to public and privatewages (Economist Intelligence Unit, 1991c: 14). Even though the Fund approvedof the 1990/91 program, it noted that public wages had gone over target by 5percent, and insisted that they be cut before it would enter into furtheragreements with Uruguay (Economist Intelligence Unit, 1992a, 1992b).

Some of Lacalle’s measures were not successful. He won the approval ofcongress for the controversial privatization of the last state-owned telephonecompany. Yet, this measure was so unpopular that after it was passed, opponentsgathered the signatures necessary to hold a national referendum on thelegislation. When put to the national electorate, the motion to repeal theprivatization legislation was carried with a 70 percent majority.

This referendum highlights just how unpopular Lacalle’s policies were andshows why he wanted IMF conditionality to be imposed. Lacalle needed allies. Thefact that he won the approval of congress for a measure opposed by more than 70percent of the voters indicates that the IMF alliance may have helped. Theincreased leverage from the IMF may not have been enough to push throughpolicies when put “directly to citizens on the basis of specific policy preferences”(Filgueira and Papadópulos, 1997: 380), but it was enough to get it past substantialopposition in congress.

Note that 30 percent of the electorate voted in favor of privatization in thereferendum. Lacalle was not entirely without supporters. Some groups actuallybenefited immediately under the IMF program. Unfortunately, due to lack of data,it is not easy to ascertain the distributional consequences of the 1990 IMFagreement. The United Nations Development Program (UNDP) World IncomeInequality Database (WIID) does not report income-inequality data for Uruguayafter 1989. There is evidence, however, that IMF programs in general have negativedistributional impacts. In his study of IMF programs in Latin America, Pastor(1987) finds that labor’s share of income declines under IMF programs. Similarly,Garuda (2000) shows that income inequality (as measured by Gini coefficients)increases for typical countries participating in IMF programs. Looking specificallyat the manufacturing sector (about 25 percent of GDP for Uruguay), Vreeland(2002) shows that even if IMF programs have contractionary effects overall, thefavorable shift in income toward some groups is large enough to mitigate lowergrowth. The income of the owners of capital can actually increase in the short run.

This, in fact, was the case in Uruguay in 1990. Figure 5 illustrates the pattern.Throughout the 1970s and 1980s, every year that Uruguay signed an IMF program,labor’s share of income from manufacturing dropped (with the exception of 1985,when democracy was reinstated). In 1988, the first year with no IMF program inmore than a decade, labor’s share reached its highest level in 5 years (the highestlevel since the previous democratic regime). After peaking in 1988 at 27 percent,labor’s share of income from manufacturing dropped to 26 percent the yearLacalle was elected and then dropped to 23 percent the year the IMF agreementwas signed. Labor’s share remained at about 23 percent during the course of thisand the following IMF agreement. In 1990, the economy experienced a contractionof 1.03 percent and earnings from manufacturing dropped from US$3722 millionto US$3667 million, but because labor’s share of income from manufacturing alsodropped, the income going to capital increased from US$2762 million to US$2820million. Despite negative growth for the economy as a whole, the income of theowners of capital increased by 2 percent. This group was made better off by thechanges in policy.


The defeat in the referendum was, nevertheless, a major defeat for Lacalle’sreform program. It indicated extremely low public support for the continuation ofreforms. Yet, when the 1990 agreement with the IMF expired, Lacalle signedanother agreement in July 1992. This arrangement provided Uruguay with US$72million, but nearly 70 percent of this loan went untouched. For this agreement,while there was no immediate need for the IMF loan, the case can be made thatthere was a prospective need.8 The IMF explained that certain policies of theprogram would hurt the current account surplus because they called for openingthe economy to imports (IMF, 1992: 238–39). By 1992, export duties were cut bymore than half of what they were in 1990 and import duties were also reduced.Recall, however, that this prospective need was not the reason for the first IMF loan.Indeed, under the 1990 agreement, import duties were increased 5 percent.

Beyond the opening of the economy, the Fund reported frankly that the goal ofthe 1992 program was to “consolidate gains made in the fiscal area” (IMF, 1992:238–39). And despite the lack of domestic support, Lacalle implemented much ofthe reform program. In addition to what had already been forced through, thepresident succeeded in ending the state monopoly on insurance and pushedthrough port reform, as well as re-privatizing all four banks absorbed by theprevious administration and selling the last state-owned meat-packing plant(Financial Times, 1994a). Lacalle lost the big battles over social security and the saleof the telephone company, but he won many other unlikely victories.

By the end of his term, Lacalle gained the approval of investors and creditors.In May 1994, the Economist Intelligence Unit improved the political risk rating ofUruguay. The change was issued because of “continued efforts at structural reformby the Lacalle administration—and its partial success—despite a hostile congress”(Economist Intelligence Unit cited in Financial Times, 1994a). In particular,foreign direct investment experienced a dramatic increase, going from zero in1991 to less than 0.01 percent of GDP in 1992 to 0.74 percent of GDP in 1993 andnearly 1 percent of GDP in 1994. Overall, investment grew every year after 1990,going from 12 percent of GDP in 1991 to 13 percent in 1992 to more than 14percent in 1993 and 1994. Economic growth (as a percentage of GDP) went from


Signed IMF agreements

Signed IMF agreements

No IMF program

20

25

30

35

40

45

50

76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93

Year

Lab

or

Sh

are

of

Man

ufa

ctu

rin

g

(Perc

en

tag

e o

f T

ota

l E

arn

ing

s)

Entered new IMF agreement IMF agreement in place No IMF program

FIGURE 5. Uruguay 1976–1993 Labor Share of Manufacturing Earnings

less than 1 percent in 1990 to 3 percent in 1991 and nearly 8 percent in 1992.Growth slowed to about 2.5 percent in 1993, but then shot back up to more than6.5 percent in 1994.

Note that sending a signal of commitment to reform may indeed have beenpart of the motivation of Lacalle.9 Investors who fear that economic reforms willbe undermined by domestic forces or that promises of reform are mere “cheaptalk” on the part of government officials may be reassured by the presence of anIMF arrangement. Because IMF agreements are unpopular, entering into them iscostly, and because the IMF can punish non-compliance, failure to meet the termsof the agreement can also be costly. Both of these costs can enhance the credibilityof a government promising reform. This represents an additional reason forgovernments to desire conditions to be imposed. Rather than bringing in the IMFto push reform past opposition, the government may need the agreement toreassure investors that the government desires reform and believes it can achieveit.

As for the causal link between the IMF agreement and the economicimprovements, the relevant counterfactual to consider for my argument is notwhether these changes would have occurred had no agreement been signed.Rather, consider the following counterfactual: given that an agreement was signedby the Lacalle administration, had the IMF program been rejected in its entirety,these improvements would not have come to fruition. The IMF would not haveissued the 1992 press release declaring the 1990 program a success, and a negativesignal would have been sent to creditors and investors. Investment and growthwould have suffered from a rejected IMF program.

(4) TanzaniaThe case of Uruguay is clear: the government did not need an IMF loan, and itsigned because it wanted IMF conditions to be imposed. Other cases in whichgovernments enter into IMF agreements, however, are more ambiguous. They maysimply want a loan, or they may want a loan and want conditions imposed. In thissection, I seek to establish that governments derive some minimum utility from IMFconditionality. Governments that sign with the IMF at least derive more utility fromconditions than did the government of Tanzania in 1983. This is not to say that allgovernments entering into IMF programs gain positive utility from the conditionsattached to the loan. But they at least do not disdain the particular conditionsimposed as much as did the government of Tanzania. The case of Tanzania is thestarkest example of a government that has a strong need for an IMF loan, but doesnot sign an IMF agreement. Not once did the government argue that it needed alarger loan from the IMF. Rather, the government argued for weaker conditions.Such a case is a clear illustration of a government that decides not to sign an IMFagreement because it does not want conditions to be imposed.

Consider Figure 6. According to the level of foreign reserves in 1983, Tanzaniawould appear to be a prime candidate for an IMF loan. Indeed, the first part ofFigure 6 illustrates that the early IMF experience of Tanzania closely follows theconventional story of participation: international reserves dropped in 1974, andthe government signed an IMF agreement the next year. Reserves dropped again in1978 and 1979, and the government signed in 1980. What is strange about the caseof Tanzania is that reserves continued to drop, reaching an all-time low in 1982,and yet the government allowed its 3-year agreement with the IMF to expire


without signing another agreement in 1983. Reserves remained extremely low in1983, 1984 and 1985, and the country ran a balance of payments deficit rightthrough this period. It is even more remarkable that reserves could hit such a lowwhen one considers that Tanzania received much more foreign aid than otherdeveloping countries (see Lancaster, 1999). From 1983 to 1985, the country wastruly in dire straits. The government did not return to the Fund, however, until1986. Why did the government not sign when it needed the IMF loan? Was itbecause of conditionality? This question is particularly puzzling since Tanzaniahad entered previous agreements, submitting to IMF conditions when it neededforeign exchange.

The 1975 AgreementAccording to Campbell and Stein (1992: 1): “The IMF and the World Bank’s statedgoal of economic liberalization was publicly resisted in Tanzania until 1986.”President Julius Nyerere based his rule on a particular form of Tanzaniansocialism that stressed independence from world powers. His famous ArushaDeclaration, issued in 1967, called for egalitarianism, socialism and self-reliance.The government was accordingly reluctant to enter into IMF agreements.

When the government needed foreign exchange in the early 1970s, it avoidedIMF conditions for as long as possible by taking out loans from the IMF that did notentail conditionality. First, the government drew down the maximum amount offoreign exchange allowed without entering into a special arrangement. (All IMFmember governments keep a specified amount of their national currency, thecountry’s quota, on deposit at the Fund. Some 25 percent of a member’s quotacan be drawn down in terms of foreign exchange without conditions attached.)

When the government faced further need of foreign exchange, it negotiated an


0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88

Year

Fo

reig

n

Re

se

rve

s


FIGURE 6. Tanzania 1970–1988 Foreign Reserves (in Terms of Monthly Imports)

agreement from the Oil Fund Facility of the IMF for SDR6.3 million.10 The Oil FundFacility was a special facility designed to provide foreign exchange to countriesfacing shortages due to the impact of increased petroleum prices (IMF, 1974: 86).Because rising oil prices were viewed as a random shock and not the result of badpolicy, these agreements entailed no conditions. Thus, the Tanzanian governmentwas able to obtain foreign exchange through this agreement without having theFund impose conditions. Economic problems persisted, however, and Tanzaniafinally entered into a 1-year standby arrangement for SDR10.5 million in August1975. The IMF press release stated that the program was “in support of thegovernment’s economic and financial policies of expanding output andtightening fiscal and monetary measures” (IMF, 1975: 254). Stein (1992: 63)reports, however, that the actual conditions associated with the arrangement wereweak. The IMF required only that domestic credit usage by the public sector beconstrained. The government did not want any conditions imposed, but itdesperately needed foreign exchange. The IMF granted an agreement with softconditions, and the government accepted.

The 1980 AgreementForeign reserves plummeted 4 years later.11 The government returned to the Fundfor assistance in 1979. This time, however, the Fund offered a severe package thatthe government refused despite its shortage of foreign reserves. The IMF insistedon restrictions over imports, foreign exchange and price controls, devaluation ofthe national currency and an end to the growth of the sizable public sector.Tanzania refused the agreement and walked away from the negotiation table.President Nyerere announced, “People who think Tanzania will change hercherished policies of socialism because of the current economic difficulties arewasting their time” (New York Times, 1979). He criticized the IMF for attempting totake advantage of the economic crisis to cut public expenditure, freeze wages,promote the private sector and reduce the size of the state (Kiondo, 1992: 24). In1980, he co-hosted the Arusha Initiative Conference (with President MichaelManley of Jamaica) at which governments criticized the Fund for forcing austeritymeasures upon them.

The need for foreign exchange continued, however. President Nyerere used hisprominence as a world figure to negotiate an IMF agreement with weakerconditions. He also took advantage of a move by the new Managing Director ofthe Fund, Jacques de La Rosière, to reach out to Africa. Following a series ofpersonal communications between Nyerere and de La Rosière, negotiationsresumed (New York Times, 1980). Subsequently, the government succeeded ingetting extremely soft conditions.

According to the 3-year arrangement, Tanzania would receive access toSDR179.6 million, equivalent to 327 percent of Tanzania’s quota (IMF, 1980: 328).In return, the government had to do very little. It got around devaluing thenational currency by agreeing to “a joint Tanzanian-IMF study of the exchangerate” (New York Times, 1980). The other previously demanded conditions regardingforeign exchange and price controls were abandoned. It turned out that the onlyactual demand that the Fund made was that a ceiling be placed on governmentborrowing. Even this single condition was too much for the government. Theprogram fell apart by November 1980, after the government exceeded the limit onpublic borrowing and the IMF suspended disbursement of loans (Stein, 1992: 64).


At this point, Tanzania had drawn a small fraction of the credit set aside by the1980 agreement. To regain access to the foreign exchange remaining under thearrangement, the Fund demanded that Tanzania meet a new list of conditions: acurrency devaluation of 50–60 percent, reduction of the government’s budgetdeficit, a freeze on wage increases, “commercially sound” parastatal policies(removing subsidies on gas and petroleum products), increased interest rates,increased producer prices and the removal of import controls (Stein, 1992: 65).Tanzania rejected these conditions and allowed the agreement to expire in 1982without entering a new IMF program. Thus, Tanzania did not participate in an IMFprogram in 1983, despite shortages in reserves, because the IMF demanded moreconditions than the government was willing to accept.

Note that rejecting the IMF was costly. Figure 7 shows that once it became clearTanzania would not renegotiate a new program, investment dropped from 11.8percent of GDP to 8.9 percent, the lowest level since 1964 (Heston and Summers,1995). Consistent with statistical evidence from Edwards (2000) and with Stone’s(2002) argument, the failed IMF program was associated with a drop in investment.Interestingly, a successful program does not increase investment, however, in thiscase, investment seems to have rebounded without a new agreement in 1984 and1985. Nevertheless, the program failure in 1983 was associated with a reduction ininvestment. The government preferred this alternative to complying with IMFconditions.

The 1975 and 1980 agreements called for conditions that were soft enough forthe government to sign. Why did the negotiation posture of the IMF become moresevere after the 1980 agreement fell apart? In 1975, the Fund faced a loose budget


5

10

15

80 81 82 83 84 85 86

Year

Inv

es

tme

nt


FIGURE 7. Tanzania 1980–1986 Investment (Percentage of GDP) Penn World Tables 5.6

constraint. The number of countries participating in IMF programs had droppedfrom 32 in 1969 to 21 in 1974. Indeed, originally intended to provide exchange-rate stability, the IMF faced a crisis of purpose when the USA went off the goldstandard in the 1970s. It was during this period that the Fund changed its focusfrom monitoring exchange rates to “providing [members] with [the] opportunityto correct maladjustments in their balance of payments” (De Vries, 1986: 14). Thenegotiation posture of the IMF was weak, and it was willing to grant soft conditionsto the government of Tanzania. Essentially, the Fund needed business. By the timeof the 1980 agreement, the number of countries with IMF agreements had climbedback up, and the negotiation posture of the Fund was tougher. In 1981 and 1982,the number of countries participating in IMF programs went up even further. By1983, the number of countries participating in IMF agreements had reached an all-time high. Furthermore, after the 1982 world debt crisis, the Fund was undermore scrutiny than ever before. Critics of the Fund began to argue that thestabilization programs sponsored by the IMF did not go far enough—deeper,structural adjustment was needed. Thus, not only was the budget tighter, thetechnocratic position of the IMF also became more severe. When Tanzania failed tolive up to the lenient conditions granted in the original 1980 agreement, the Fundinsisted on stricter conditions than Tanzania had ever faced.

Changing Views of ReformWhen the 1980 IMF program fell apart, Nyerere undertook his own economicprogram that relied on mere directives to state entities and on the moralexhortation of peasants and workers (Kiondo, 1992: 23–24). The program failedand Tanzania’s economic problems deepened. During the years 1982–85, thegovernment of Tanzania changed its views on economic reform. In 1982, thegovernment introduced a much more aggressive program and implemented manyof the measures called for by the IMF. The currency was devalued, first by 12percent in March, and then by 20 percent in July (Kiondo, 1992: 24). By 1984, thegovernment removed subsidies, raised producer prices, introduced new taxes tofinance the budget deficit, froze certain civil service hiring, and devalued anadditional 26 percent (Stein, 1992: 68–69).

Another important area of reform involved trade liberalization. Thegovernment allowed people to import goods with their own foreign exchange andsell these imports domestically. This represented an important area forcompromise between the IMF and the government, since it effectively opened asecond window allowing the Tanzanian currency to operate on a dual exchange rate(Kiondo, 1992: 26). Hence, this reform really represented a further devaluation.

Other reforms were envisaged: adjusting market and producer prices, cuttingback government expenditure, reducing monetary expansion and improving theefficiency of the state sector. Yet, unable to garner enough support from themembers of the ruling (and only) political party (the Chama Cha MapinduziRevolutionary Party or CCM), supporters could not push these reforms through(Kiondo, 1992: 24). As Kiondo (1992: 28) explains:

One of the contradictions of the liberalization process in Tanzania was theconstant tug of war between those leaders in the government and party whowanted to remain true to the spirit of [Tanzanian socialism] and those whopushed for reforms.


This tug of war manifested itself in the struggle for leadership of the country. Withthe 1985 elections, Nyerere stepped down as president after more than 2 decadesof rule. Ali Hassan Mwinyi became the new president. Although the two camefrom the same political party (the CCM), Mwinyi, who favored deeper economicreforms, represented a shift in the agenda of the government.

Yet, while the reform-minded Mwinyi was elected president of the country, themembers of CCM re-elected Nyerere (who still opposed economic reform) aspresident of the party to curb Mwinyi’s agenda (New York Times, 1987).Furthermore, they voted out two members of the CCM central committee who werepro-economic reform (Matthews, 1998: 1036). Facing such resistance, Mwinyi wasunable to bring about his reforms. Indeed, as Figure 8 shows, while thegovernment deficit had been steadily reduced since the reforms of 1982–85, thedeficit increased dramatically in 1986, after Mwinyi took office.

Lacking the support of his party to bring about further reform, Mwinyi broughtin the IMF. In September 1986, the president signed an 18-month agreement withthe Fund for SDR64.2 million (60 percent of Tanzania’s quota at the time).

Kiondo (1992) argues that the Fund was brought in as an ally for thosesupporting economic reform to use against the elements in the CCM that opposedMwinyi’s economic program. Over time, “The owners of [foreign exchange] ...gained supporters in the party and the government while forging an alliance withthe IMF and the World Bank” (Kiondo, 1992: 26). Therefore:

by the time of the IMF agreement [in 1986] there were elements in Tanzaniansociety with an objective interest in the reforms. However, because they did notyet dominate the state there was subterfuge in the approach taken by Tanzania


0

5

10

15

20

25

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88

Year

De

fic

it


FIGURE 8. Tanzania 1972–1988 Government Budget Deficit (Percentage of GDP)

in implementing the reforms. Indeed, Tanzania’s reforms are implemented insecretive, unplanned, and nondemocratic ways. (Kiondo, 1992: 35)

Kiondo (1992: 35) goes on to argue that “the IMF found domestic allies who inturn found supporters within the state.”

How could an alliance with the IMF help push through the reforms? On the onehand, the government could still argue that the country needed foreign reserves.The need, in fact, was desperate (recall Figure 6). Furthermore, the IMFagreement increased the pressure for reform, since a failure to do so would nolonger be a mere rejection of Mwinyi, but a rejection of the IMF. A rejection of theIMF would send a negative signal to creditors and investors. Recall from Figure 7how investment dropped when the 1980 IMF agreement fell apart. A drop ininvestment would hurt all of Tanzania, including those opposed to the IMF.

So even though Tanzania turned to the Fund because it needed foreignexchange, the government also wanted IMF conditions to be imposed. As Stein(1992: 59) puts it:

Critics typically argue that the [IMF’s] policies tend to come at the expense of“socialist” and nationalist governments’ autonomy to pursue their ownagendas. [However,] the IMF agreement with Tanzania is not inconsistent withthe agenda of the state.

Thus, even though Tanzania needed an IMF loan, the same logic that applied toLacalle’s move to bring in the IMF applies here as well. Indeed, when the countryneeded an IMF loan, but did not want IMF conditions (1982–85), the governmentrejected Fund agreements. When the government of Tanzania finally returned tothe IMF, it did so not only because it wanted foreign exchange, but also because itwanted conditions. The government sought political support through IMFconditionality to push through its preferred policies.

Why did Tanzania not turn to the Fund in 1983? Tanzania did not sign in 1983because it did not get the conditions it wanted. Regarding the 1975 and 1980agreements, the government signed reluctantly and only after negotiating forconditions close to its ideal point (which was effectively zero conditions). Whenthe government finally signed another IMF agreement in 1986, it did so because itspreferences over conditions had changed. The new government gained greaterutility from the increased conditions demanded by the Fund.

(5) ImplicationsConventional wisdom holds that the IMF uses conditionality to force governmentsto accept painful austerity measures. Otherwise, governments would reform ontheir own. As Fischer puts it, “Policy conditionality can be interpreted as a ...penalty, as seen from the viewpoint of the borrower country’s policy makers”(1999; see also Bird, 1995: 94–96). As the 1990 Uruguay agreement demonstrates,however, governments may seek IMF agreements because they want particularconditions to be imposed on them. In addition, the case of Tanzania in 1983illustrates that when governments disdain IMF conditions, they may forgo an IMFagreement even if they need an IMF loan. Surprisingly, not only the continuedneed for an IMF loan, but also the desire to have conditions imposed eventuallydrove Tanzania to enter into the 1986 agreement. Both the Lacalle and the Mwinyigovernments sought to impose painful, unpopular economic austerity and facedtough opposition.


By entering into IMF agreements, these governments tied their preferredpolicies of economic reform to the conditions imposed by the IMF. For opponentsof economic reform, this move raised the cost of rejecting the government’sproposals because a rejection was no longer a mere rejection of the president, butalso of the IMF. A total rejection of the IMF would have sent out costly negativesignals to creditors and investors. Thus, the governments were able to pushthrough more of their reform programs than they would have without the IMF.This argument is similar to Schelling’s (1960: 22) contention that “the power toconstrain an adversary may depend on the power to bind oneself.”

Moreover, conditions may be desired by a government as a credibility-enhancing mechanism. Entering into an IMF agreement sends a costly signal toinvestors and creditors that the government desires reform and believes it canachieve it. IMF arrangements can be costly because they represent an unpopularsacrifice of national sovereignty and because failure to comply with them mayresult in punishment from the IMF. Investors who fear that domestic forces willundermine economic reform or that promises of reform are mere “cheap talk” onthe part of government officials may be reassured by the costly signal of bringingin the IMF.

What broader lessons can one draw from these cases? What do these cases tellus about the observations of participation in IMF programs where countries needan IMF loan and participate in IMF programs? Ostensibly, these governments mayhave entered into IMF agreements simply for the IMF loan. But they might also havewanted conditions to be imposed. Indeed, after studying the cases presented here,one can conclude the following about countries participating in IMF programs:governments may not always want conditions imposed, but at least they do notdisdain the specific conditions imposed as much as did the 1983 Nyereregovernment in Tanzania, which desperately needed an IMF loan, but refused IMFconditions. This may be because the IMF is demanding only weak conditions beimposed (as the participation of Tanzania in the 1975 agreement illustrated) orbecause the government wants strong conditions imposed to help push throughunpopular policies (as the Uruguayan case and the 1986 IMF agreement withTanzania illustrated). While only statistical tests can establish typical patterns, onegets a broad picture of the level of utility derived from IMF conditions from thecases presented here because they were selected according to specific analyticalcriteria: the case of Tanzania in 1983 sets a minimum bound on the utility derivedfrom IMF conditions and the case of Uruguay in 1990 indicates a possible range.

This conclusion has broad implications about how one thinks about IMFconditionality. Conditionality is not always a “penalty,” but may be, rather, amechanism useful to governments for changing policy. In light of this, rethinkingthe role of conditionality is in order. Some argue, for example, that seeking outand assisting only reform-oriented governments should become the explicit policyof the IMF (IMF, 2001: 65; citing Dollar and Svensson, 2000). This would essentiallymake my argument (that governments enter into IMF programs to force their ownreform agendas) the explicit policy of the Fund. Instead of “making” reformers,the IMF should look for reformers and extend assistance to them. Others mightargue, however, that this approach will exacerbate problems already present in theimposition of IMF programs, by increasing the alienation of those groups within acountry who are “left out.”

Indeed, while there is no conclusive evidence that IMF programs improveeconomic growth or economic stability (see Bird, 1996a; Przeworski and Vreeland,


2000), Pastor (1987: 89) finds that “the single most consistent effect the IMF seemsto have is the redistribution of income away from workers.” Thus, the income ofthe owners of capital tends to increase under IMF programs—at the expense oflabor. If IMF programs consistently affect distribution, it should not be surprising tofind that politics play a role in the decisions of governments to bring in the IMF.Some groups stand to gain by pushing through the policies supported by the IMF,while others stand to lose.

This speaks to the view that the IMF should reach out to a broad spectrum ofconstituents within a country when designing an economic program. A recent IMFreport (2001: 42) suggests that, “subject to the guidance of the authorities, theFund staff can ... play a role ... by holding substantive discussions with othergroups, including other ministries, trade unions, industry representatives, andlocal non-governmental organizations, especially at a stage at which the design ofthe program is still under consideration.” The goal of such meetings is to “helpgroups within the country to participate meaningfully in the process” (IMF, 2001:42). The new leadership at the Fund has made this a cornerstone of recentnegotiations, a trend that may change the way IMF programs are designed andimplemented.

Regardless of the direction that the Fund takes (whether it seeks out reformistgovernments and helps push their agenda through, whether it seeks broad-basedsupport from various constituencies or whether it pursues a combination of thetwo), the IMF should continue to make explicit the role it plays in domestic politics.

Notes1. The only cases of countries having lower reserves without turning to the IMF are

Azerbaijan in 1992–93 (which turned to the IMF in 1994), Equatorial Guinea in 1996–98,Liberia in 1987 and Sudan in 1991. I chose to study Tanzania over these cases because ofthe greater availability of data on this country.

2. In all of the country-specific graphical figures throughout this article, the source of thedata is the World Bank’s World Development Indicators unless otherwise noted.

3. The government was elected in 1989.4. Until recently, Uruguay had a rare electoral system in which primaries and elections

were combined. The party with the most votes won the election, and the candidate withthe most votes from that party took office.

5. Note this does not imply that there was no discipline or loyalty within the party system.The intraparty preferences exacerbated by the double simultaneous vote system led tomany distinct factions within the party. Since a candidate not only depended on hisparty, but on his faction within the party, ultimately legislators voted along “faction”lines (See Buquet et al. 1999). I am grateful to Juan Andrés Moraes for pointing thisout.

6. By the end of his term, Lacalle had faced eight general strikes.7. The press release was misleading in other ways. Reviewing Uruguay’s economic history,

it stated that subsequent to strong growth and balance of payments surpluses in the1970s, economic performance in Uruguay had declined in the early 1980s. The pressrelease leads one to believe that the IMF addressed Uruguay’s economic problems forthe first time with the 1983/84 program. It fails to mention that Uruguay was under IMF

programs right through the period from 1975 to 1987.8. I am grateful to Beth Simmons for raising this possibility.9. I am grateful to the anonymous reviewer for this suggestion.

10. SDR (or Special Drawing Right) is a pseudo-currency used as a common denominationfor the foreign exchange held on deposit at the Fund. It is currently valued at US$1.25.

11. This was because of the Tanzanian intervention in Uganda.


ReferencesBird, Graham (1995). IMF Lending to Developing Countries, Issues and Evidence. London:

Routledge.Bird, Graham (1996a). “The International Monetary Fund and Developing Countries: A

Review of the Evidence and Policy Options.” International Organization, 50: 477–511.Bird, Graham (1996b). “Borrowing from the IMF: The Policy Implications of Recent

Empirical Research.” World Development, 24: 1753–1760.Bjork, James (1995). “The Uses of Conditionality.” East European Quarterly, 29: 89–124.Buquet, Daniel, Daniel Chasquetti and Juan Andrés Moraes (1999). Fragmentación Política y

Gobierno en Uruguay. ¿Un enfermo imaginario? Montevideo: Facultad de Ciencias Sociales.Busqueda (1990). “Se suscribirà el mes próximo una nueva Carta Intención con el FMI.” 28

February.Campbell, Horace and Howard Stein (1992). “Introduction: The Dynamics of

Liberalization in Tanzania.” In Tanzania and the IMF: The Dynamics of Liberalization(Horace Campbell and Howard Stein, eds), 1–20. Boulder, CO: Westview Press.

De Vries, M. (1986). The IMF in a Changing World: 1945–1985. Washington, DC: IMF.Dixit, Avinash K. (1996). The Making of Economic Policy: A Transaction-Cost Politics Perspective.

Cambridge, MA: MIT Press.Dollar, David and Jakob Svensson (2000). “What Explains the Success or Failure of

Structural Adjustment Programs?” Economic Journal, 110: 894–917.Economist Intelligence Unit (1991a). Uruguay, Paraguay Country Report 1. London: Economist

Intelligence Unit.Economist Intelligence Unit (1991b). Uruguay, Paraguay Country Report 3. London: Economist

Intelligence Unit.Economist Intelligence Unit (1991c). Uruguay, Paraguay Country Report 4. London: Economist

Intelligence Unit.Economist Intelligence Unit (1992a). Uruguay, Paraguay Country Report 1. London: Economist

Intelligence Unit.Economist Intelligence Unit (1992b). Uruguay, Paraguay Country Report 2. London: Economist

Intelligence Unit.Edwards, Martin S. (2000). “Reevaluating the ‘Catalytic’ Effect of IMF Programs.” Paper

prepared for the Annual Meeting of the American Political Science Association,Washington, DC, 31 August–3 September.

Edwards, Sebastian and Julio A. Santaella (1993). “Devaluation Controversies in theDeveloping Countries: Lessons from the Bretton Woods Era.” In A Retrospective on theBretton Woods System (Michael D. Bordo and Barry Eichengreen, eds), 405–455. Chicago:University of Chicago Press.

Filgueira, Fernando and Jorge Papadópulos (1997). “Putting Conservatism to Good Use?”In The New Politics of Inequality in Latin America: Rethinking Participation and Representation(Douglas A. Chalmers, Carlos M. Vilas, Katherine Hite, Scott B. Martin, Kerianne Piesterand Monique Segarra, eds). New York: Oxford University Press.

Financial Times (1990a). “Uruguay Prepares IMF Letter of Intent.” 13 March.Financial Times (1990b). “Uruguay Clinches $150m Standby Credit from IMF.” 2 July.Financial Times (1994a). “Defiant Uruguay Leader Holds Out for Free Market.” 25 May.Financial Times (1994b). “Uruguay Elections Marked by Factionalism.” 8 November.Fischer, Stanley (1999). “On the Need for an International Lender of Last Resort.” Paper

prepared for the joint luncheon of the American Economic Association and theAmerican Finance Association, New York, 3 January. Available at www.imf.org.

Garuda, Gopal (2000). “The Distributional Effects of IMF Programs: A Cross-CountryAnalysis.” World Development, 28: 1031–1051.

Gonzalez, Luis E. (1995). “Continuity and Change in the Uruguayan Party System.” InBuilding Democratic Institutions: Party Systems in Latin America (Scott Mainwaring andTimothy R. Scully, eds), 138–163. Stanford, CA: Stanford University Press.

Gourevitch, Peter Alexis (1986). Politics in Hard Times. Ithaca: Cornell University Press.


Heston, Alan and Robert Summers (1995). Penn World Tables 5.6. Cambridge, MA: NationalBureau of Economic Research.

Huber, Evelyne and John D. Stephens (2000). “The Political Economy of Pension Reform:Latin America in Comparative Perspective.” Paper prepared for the 2000 meeting of theLatin American Studies Association, Miami, 16–18 March.

IMF (1974). IMF Survey 1974. Washington, DC: IMF.IMF (1975). IMF Survey 1975. Washington, DC: IMF.IMF (1980). IMF Survey 1980. Washington, DC: IMF.IMF (1991). IMF Survey 1991. Washington, DC: IMF.IMF (1992). IMF Survey 1992. Washington, DC: IMF.IMF (1997 [1944]). Articles of Agreement of the International Monetary Fund. 6 November.

Available at http://www.imf.org/external/pubs/ft/aa/index.htm and http://www.imf.org/external/pubs/ft/aa/aa.pdf.

IMF (2001). “Conditionality in Fund-Supported Programs—Policy Issues.” Paper preparedby the Policy Development and Review Department (in consultation with otherdepartments), approved by Jack Boorman. 16 February. Available at http://www.imf.org/external/np/pdr/cond/2001/eng/policy/021601.pdf.

Kiondo, Andrew (1992). “The Nature of Economic Reforms in Tanzania.” In Tanzania andthe IMF: The Dynamics of Liberalization (Horace Campbell and Howard Stein, eds), 21–42.Boulder, CO: Westview Press.

Lancaster, Carol (1999). Aid to Africa: So Much To Do, So Little Done. Chicago: University ofChicago Press.

Lupia, Arthur and Mathew D. McCubbins (1998). “Conditions for the Stability of PoliticalAgreements.” Paper written for the World Bank.

Matthews, Graham (1998). “Recent History of Tanzania.” In Africa South of the Sahara 1998,1035–1038. London: Europa Publications Limited.

Mesa-Lago, Carmelo (1978). Social Security in Latin America: Pressure Groups, Stratification, andInequality. Pittsburgh: University of Pittsburgh Press.

Mesa-Lago, Carmelo (ed.) (1985). The Crisis of Social Security and Health Care: Latin AmericanExperiences and Lessons. Pittsburgh: Center for Latin American Studies, University Centerfor International Studies, University of Pittsburgh.

New York Times (1979). “Uganda War Broke Tanzania’s Treasury but not its Spirit.” 20December.

New York Times (1980). “Tanzania Reportedly About to get Major IMF Loan.” 4 September.New York Times (1987). “Nyerere, in Shift, Remains Tanzania Party Leader.” 1 November.New York Times (1989). “Uruguay’s Mixed Election Result May Lead to Conflict.” 28

November.New York Times (1990a). “Freer Economy Resisted in Uruguay.” 24 August.New York Times (1990b). “In Uruguay, Two Leaders and Two Ideologies.” 29 August.Pastor, Manuel (1987). The International Monetary Fund and Latin America: Economic

Stabilization and Class Conflict. Boulder, CO: Westview Press.Przeworski, Adam and James Vreeland (2000). “The Effect of IMF Programs on Economic

Growth.” Journal of Development Economics, 62(2): 385–421.Putnam, Robert D. (1988). “Diplomacy and Domestic Politics: The Logic of Two-Level

Games.” International Organization, 42: 427–460.Remmer, Karen L. (1986). “The Politics of Economic Stabilization, IMF Standby Programs in

Latin America, 1954–1984.” Comparative Politics, 19: 1–24.Rial, Juan (1986). “Uruguay: From Restoration to the Crisis of Governability.” In Transitions

from Authoritarian Rule (Guillermo O’Donnell, Phillippe C. Schmitter and LaurenceWhitehead, eds), 133–146. Baltimore: Johns Hopkins University Press.

Schelling, Thomas C. (1960). The Strategy of Conflict. Cambridge, MA: Harvard UniversityPress.

Schooley, Helen (1997). “Uruguay: History, Economy.” In South America, Central America, andthe Caribbean, 6th edn., 642–660. London: Europa Publications Limited.

Spaventa, Luigi (1983). “Two Letters of Intent: External Crises and Stabilization Policy,


Italy, 1973–77.” In IMF Conditionality (John Williamson, ed.). Washington, DC: Institute forInternational Economics.

Stein, Howard (1992). “Economic Policy and the IMF in Tanzania: Conditionality, Conflict,and Convergence.” In Tanzania and the IMF: The Dynamics of Liberalization (HoraceCampbell and Howard Stein, eds), 59–83. Boulder, CO: Westview Press.

Stone, Randall W. (2002). Lending Credibility: The International Monetary Fund and the Post-Communist Transition. Princeton: Princeton University Press.

Vaubel, Roland (1986). “A Public Choice Approach to International Organization.” PublicChoice, 51: 39–57.

Vreeland, James (2001). “Institutional Determinants of IMF Agreements.” Paper preparedfor the Annual Meeting of the American Political Science Association, San Francisco, CA,30 August–2 September.

Vreeland, James (2002). “The Effect of IMF Programs on Labor.” World Development, 30:121–139.

Vreeland, James (2003). The IMF and Economic Development. New York: Cambridge UniversityPress.

World Bank (various). World Development Indicators on CD-ROM. Washington, DC: World Bank.

Biographical NoteJAMES RAYMOND VREELAND, Ph.D., New York University, 1999, is Assistant Professorof Political Science at Yale University. His research concerns the effects ofinternational institutions on domestic politics. His publications have appeared inthe Journal of Development Economics, Political Analysis and World Development, and hehas recently completed a book entitled The IMF and Economic Development(published by Cambridge University Press in 2003). ADDRESS: Department ofPolitical Science, Yale University, New Haven, CT 06520-8301, USA [email:[email protected]].

Acknowledgments. I thank Rossana Castiglioni, José Cheibub, Martin Edwards, JenniferGandhi, Geoffrey Garrett, Barbara Geddes, James Meadowcroft, Covadonga Meseguer, JuanAndrés Moraes, Adam Przeworski, Javier Santiso, Beth Simmons, Allan Stam, RandolphStevenson, James J. Vreeland, Thomas Willett and the two anonymous referees for helpfulcomments. This study was made possible in part by a grant from the Carnegie Corporationof New York. The statements made and views expressed are solely the responsibility of theauthor. This article was drawn in part from Chapter Two of The IMF and Economic Development,reprinted with the permission of the Cambridge University Press.


Private Actors and Public Policy: A Requiem forthe New Basel Capital Accord

MICHAEL R. KING AND TIMOTHY J. SINCLAIR

ABSTRACT. After the Asian crisis of 1997–98, policy-makers invested muchenergy in designing a new international financial architecture. However,many of the policy proposals that have emerged from think tanks and the multilateral agencies have proven unworkable or politicallyunpalatable. The debate focuses on state-led initiatives. But theassumption that public policy is by definition an output of publicinstitutions is difficult to sustain in an era of global change. This articleconsiders specialized forms of intelligence gathering and judgmentdetermination which seem increasingly important as sources ofgovernance in this era of financial market volatility: Moody’s InvestorsService and Standard & Poor’s, the major bond rating agencies. Morespecifically, we examine a proposal of the Basel Committee on BankingSupervision to reform the existing capital adequacy framework byincorporating banks’ own internal ratings and external bond ratingsfrom the rating agencies, in order to calculate bank risk-weighted capitalrequirements. The article identifies a series of negative implications fromthe use of private rating agencies as a substitute for state-basedregulation, premised on the organizational incentives that shape theratings industry. Cementing these organizational incentives into theemerging financial architecture will, we argue, lead to negative social andeconomic consequences.

Keywords: • Banking • Basel capital accord • Bond rating agencies• International financial regulation

(I) IntroductionMost political scientists spend their time studying obviously political phenomenasuch as elections, political parties and parliamentary debates.1 We offer a politicalreinterpretation of what are more usually thought of as mundane entities and




processes in the world of global finance. In this article, we argue that the major debt rating agencies, Moody’s Investors Service (Moody’s) and Standard & Poor’s (S&P), serve to privatize policy-making, narrowing the legitimate sphere of government intervention. More specifically, we suggest recent efforts to incorporate the outputs of the rating agencies into plans to make global finance less volatile are flawed, and likely to generate unexpected, unwantedoutcomes.

We have organized this article in five sections. In the next section, we examinethe key dynamics of the new global finance and some of the main characteristicsof the rating agencies. In the section that follows, we investigate public–privatelinkages, as exemplified by the rating agencies. Subsequently, we focus on oneaspect of the emerging international financial architecture, namely, the new Baselproposals on capital adequacy (Basel II), which mandate incorporating bondratings into the international regulation of global finance. Criticisms of thecurrent proposal and an alternative agenda are identified. The final sectionexamines the political economy of Basel II, identifying who wins and who losesfrom the proposal. This analysis identifies the private interests blocking theprocess and examines their motivations.

(II) Global Finance and the Rating AgenciesThe rating agencies operate in what Sinclair (2000) has called the New GlobalFinance (NGF). The NGF amounts to a new form of social organization of moneyand finance (Cohen, 1996). Most of us are familiar with bank lending (Sinclair,1994a, 1994b). Banks traditionally acted as financial intermediaries, bringingtogether borrowers and lenders. They borrowed money, in the form of deposits,and lent money at their own risk to borrowers. However, in recent yearsdisintermediation has occurred on both sides of the balance sheet. Borrowershave increasingly obtained money from nonbank sources. By the mid-1990s,mutual funds, which sweep depositors’ money directly into the financial markets,contained around US$2 trillion in assets, not much less than the US$2.7 trillionheld in US bank deposits (The Economist, 1994: 11). The reasons for thisdevelopment lie in the heightened competitive pressures generated byglobalization, and the high overhead costs of banks (The Economist, 1992). Fortheir part, banks and other financial institutions increasingly bypass depositorsand borrow funds directly in the capital markets.

Disintermediation is a key feature of the NGF. Disintermediation changes therole of banks and creates an information problem in the capital markets. Inbanking, lenders can depend on the prudential behavior of banks to preservetheir wealth. However, with disintermediation there is an adverse selectionproblem, because lenders must make judgments about whether borrowers willrepay them or not. Given the costs of gathering information with which to make acreditworthiness judgment, institutions have developed to generate economies ofscale and provide centralized judgments on the ability and willingness ofborrowers to repay their obligations.

Bond rating agencies developed first in the USA. From around 1850 until WorldWar I, there was considerable growth in private information provision for theAmerican financial markets. Poor’s American Railroad Journal appeared in mid-century. In 1868, Poor’s produced the Manual of the Railroads of the United States. Bythe early 1880s, this publication had 5000 subscribers (Kirkland, 1961: 233). John


Moody first began publishing his Manual of Industrial Statistics in 1900. It proved tobe a “gold mine” for Moody (Kirkland, 1961: 234).

Issuing ratings (as opposed to selling information) developed between the 1907financial crisis and the Pujo hearings of 1912. The consolidation of thecontemporary rating system occurs between this period and World War II. Duringthis time, capital market information and analysis radically changed. TheSecurities and Exchange Commission (SEC) imposed standardization on informationto help make information comparable across US corporations. Accounting firmsflourished (Smith and Sylla, 1993: 42). New rating firms appeared to compete withMoody’s, and the rating processes themselves were extensively elaborated. Duringthe 1930s, rating became a standard requirement for the sale of any issue in theUSA after many state governments incorporated rating standards into theirprudential rules for investment by pension funds. A series of defaults by majorsovereign borrowers in the 1930s narrowed the bond business largely to US firmsand government agencies until the 1980s. This period was dominated by Americanblue chip industrial firms (Toffler, 1990: 43–57).

Rating agencies share a number of characteristics with the Law Merchant of themiddle ages. The Law Merchant developed as a way of enforcing contracts bymaking judgments on trade disputes and keeping records of these actionsavailable for scrutiny by merchants engaging in intra-European trade. Thismechanism backed up merchants when their names were not well known topotential new trade partners in geographically distant places (Cutler, 1998;Milgrom et al., 1990). Rating agencies are also responsible for keeping an eye onwho is violating the norms of financial and commercial practice. What isinteresting here is that such an ancient form of market governance should, in thecontemporary form of rating agencies, once again assume such prominence.

The two major agencies dominate the contemporary market in ratings, eachmaking judgments on approximately US$30 trillion of securities (Moody’sInvestors Service, 2001a). Both Moody’s and S&P are based in New York. Moody’swas made a stand-alone corporation by parent Dun and Bradstreet in 1998, whileS&P remains a subsidiary of publisher McGraw-Hill, owner of the company since1966. Both agencies have numerous branches in the USA, other Organization forEconomic Cooperation and Development (OECD) states, and in emerging markets.A distant third in the market is Fitch Ratings. Fitch Ratings may emerge as a majorplayer in the future, but currently lacks breadth of coverage and the reputationalassets to be considered on the same level with the “Big Two.” An increasingnumber of domestically focused agencies in developed countries, and especiallyemerging markets, opened during the 1980s and 1990s (BIS, 2000; Greenberg,1993; Everling Advisory Services, 2001).

The outputs of the rating agencies are consumed by key capital market actors,including pension funds, investment banks, other financial institutions andgovernment agencies. Moody’s has 4000 clients for its publications and estimatesaround 30,000 people read its output regularly (Chmaj, 2000). Annualsubscription fees range from US$15,000 to US$65,000 for heavier users, who also have the opportunity to talk to analysts directly. Increasingly, outputs areproduced online. The “relationship-level clients” may also attend conferences and take part in other events related to credit quality. Moody’s actively puts its analysts in front of journalists and, like S&P’s, regularly issues press statementson credit conditions. S&P’s produce a wider range of published products in both traditional and digital format. Their core weekly publication, CreditWeek,

KING/SINCLAIR: Private Actors and Public Policy 347

has some 2423 subscribers. Global Sector Review is bought by 2988 clients (Bates,2000).

During the 1990s and subsequently, perceived rating miscalls have become asignificant issue for the rating agencies, as these potentially erode the reputationalassets the rating agencies have built up since the 1930s. The 1990s saw more ofthese events as financial volatility grew in an increasingly liberalized worldeconomy, including the Tequila crisis of 1994–95 and the Asian financial crises of1997–98 (King, 2001). At the same time, derivatives and other new financialtechnologies and associated accounting strategies stimulated a number ofcorporate scandals and collapses in the USA, including Enron in December 2001and WorldCom in June 2002. Like other financial industry institutions, the ratingagencies ran to catch up with financial innovation, spending money on stafftraining and hiring. They pushed harder for analytical innovation in their ownproducts. S&P created new symbols to indicate when, for example, ratings werebased on public information only and did not reflect confidential data. Second,the rating agencies, especially Moody’s, have sought to change their cloistered,secretive image and become more transparent and willing to justify their ratings.This latter strategy may be more to do with reducing market and publicexpectations about rating than with actually improving the rating product.

(III) Public–Private RelationsPublic utilization of ratings in US regulation goes back more than 70 years (BIS,2000: 54). The Depression, the consequent sharp decline of credit quality and theproblems for domestic financial institutions it brought about led the US Office ofthe Comptroller of the Currency (OCC) to rule that bank holdings of publicly ratedbonds had to be “BBB” or better to be carried on bank balance sheets at their faceor book value. Otherwise, the bonds were to be written down to market value,imposing losses on the banks (Cantor and Packer, 1994: 6). Numerous statebanking departments also adopted this rule. Ratings are incorporated intoregulation in 11 out of the 12 countries that make up the Basel Committee onBanking Supervision, with only Germany an exception (BIS, 2000: 41). Thisregulatory role has increased the importance of ratings by making the judgmentsof rating agencies more significant in the transactions of investors and traders.

Government regulation in the USA has reinforced an oligopolistic ratingsmarket and made it harder for new entrants to launch ratings businesses. A keyregulatory development occurred in 1975 with the adoption of Rule 15c3-1 by theSEC. Under this rule brokers who underwrote bond issues had to keep a certainpercentage of their financial capital in reserves: a haircut. However, the rule gave“preferential treatment if the instruments had been rated investment-grade by atleast two ‘nationally recognised statistical rating organizations’ (NRSROs),” whocould get a “shorter haircut” (Edwards, 1994). The SEC did not specify whatdefines an NRSRO or eligibility criteria for new entrants. Despite this, the NRSROconcept has subsequently been incorporated into many regulatory initiatives.Recently, the United States Senate Intergovernmental Affairs Committee hearingsinto the Enron bankruptcy have revived prospects for clarification of the NRSROconcept.

The initiative to make the NRSRO status more transparent reflects the intensifiedcompetitive conditions of the global economy and concerns over barriers to entry.In these conditions, state intervention is generally becoming more codified,


institutionalized and juridified. Rules tend to be more elaborate and formal, withfewer tacit understandings (Moran, 1991: 13). This tendency tends to devolve stateactivities onto nominally private institutions, such as the rating agencies, whichnow find themselves increasingly part of disclosure rules, and sets the terms inwhich these institutions operate (Moran, 1991: 14). Self-regulation of this type is away for the formal political system to distance itself from the increasinguncertainty of globalized markets. The latest example of this tendency is the BaselII capital adequacy proposals which mandate the use of rating agency outputs forless sophisticated banks.

(IV) A Requiem for Basel IITraditionally, banking supervision has relied on bank examiners to test the qualityof a bank’s assets (Stevens, 2000). The growth of cross-border financial flowscombined with a series of domestic and international shocks during the 1970s and1980s exposed the weakness of this form of supervision. One study documents asmany as 86 episodes of banking insolvency in 69 countries between the late 1970sand the early 1990s, as banking crises hit the developed and developing countriesalike (Caprio and Klingebiel, 1997). While some of these banking crises wereisolated to the country in which they occurred, many of these crises wereinternational in character and threatened the stability of the internationalfinancial system, such as the Latin American debt crisis.2 Policy-makers recognizedthat increased competition among financial institutions combined withgovernment-sponsored deposit insurance and lender of last resort facilities werecreating incentives for banks to take on greater risk than was optimal—the so-called moral hazard problem. An international agreement was required to allowbetter supervision of internationally active banks in order to create a level playingfield among different jurisdictions. In response to this systemic failure, the G-10central banks sat down under the auspices of the Bank for InternationalSettlements (BIS) and formulated the 1988 Basle Capital Accord.3

The 1988 Accord sets out standards to address credit risk (the main riskincurred by banks) by agreeing common minimum capital standards for all banksoperating in G-10 countries.4 The Accord consists of two main sections, with thefirst outlining the definition of capital and the second elaborating a system of riskweights used to calculate the minimum capital applied to each asset class.5 Theidea behind the Accord is simply that a bank’s capital should be commensuratewith the riskiness of its business (Stevens, 2000). Capital is divided into twocategories. Primary or Tier 1 capital consists of the amounts paid in byshareholders, including retained earnings. Secondary or Tier 2 capital includescertain classes of preferred shares and subordinated debt obligations. The Accordset the minimum level of bank capital at 8 percent of the risk-adjusted exposure ofassets, of which Tier 1 capital must make up at least half of this amount.6

While the 1988 Accord addressed a number of key shortcomings plaguing thesupervision of international banking, the politics surrounding negotiation of theAccord and the actions taken by private-sector actors following its release quicklyundermined the Accord’s effectiveness.7 Oatley and Nabors (1998) document thepolitics of the negotiations, and view the Accord as an example of a redistributiveagreement promoted by US politicians that provided domestic rents at the expenseof foreign banks. As their discussion details, G-10 policy-makers outside the USA didnot believe their commercial banks needed higher capital requirements—only the


US commercial banks were overleveraged with risky assets. The Germans inparticular opposed international regulation because their universal bankingsystem would be disadvantaged relative to other players in the international capitalmarkets. Similarly, Japanese banks lobbied to have special treatment of theirequity holdings for capital purposes. For their part, private-sector actorsresponded to the incentives in the accord in a fashion that increased, rather thandiminished, the riskiness of their loan portfolios. Banks pursued lending andsecuritization activities designed to reduce the impact of binding capitalrequirements, by taking risky assets off-balance sheet through securitization and byloading up on the riskier assets within each risk category—a practice termed“cherry picking” (Stevens, 2000).8 Lastly, the nature of BIS meant that the 1988Accord did not apply to nonbank financial institutions such as investment dealers,insurance companies and asset managers. These actors successfully resistedregulators’ efforts to impose minimum capital standards under the auspices of theInternational Organization of Securities Commissions (IOSCO), distorting theplaying field among international financial actors (Porter, 1993).8

The new capital adequacy framework introduced in June 1999 (and modifiedin January 2001) was designed to address these shortcomings (BIS, 1999, 2001a).The first pillar of the new proposal replaces the arbitrary risk weights with a morerefined assessment of credit risk and addresses a wider set of risks involved in banklending. This proposal puts more weight on private-sector assessments of riskiness,and gives credit for the use of risk-mitigation techniques. More significantly, theproposal extends the scope of the Accord to include banking supervision under asecond pillar and the risk-management practices of banks under a third pillar.Taken together, these three pillars of the new capital adequacy proposal areexpected to enhance the soundness of the financial system, and provide a morelevel playing field while putting more reliance on private actors active ininternational financial markets.

The new accord, dubbed “Basel II,” would replace the 1988 Capital AdequacyAccord, which was found to have a number of faults. In particular, the favorablerisk weightings assigned to OECD countries relative to non-OECD membersencouraged investors to lend disproportionately to the weakest credits in theOECD, with little or no capital put aside against the risk of their default.9 Basel IIwas proposed to remove previous distortions like this, to create incentives forprivate actors to develop better risk-management practices, and to address riskssuch as interest-rate risk and operational risk which were ignored in the 1988Capital Accord. The original schedule requested comments by March 2000, withthe committee expecting to publish a comprehensive set of proposals by late 2000(McDonough, 1999).

Basel II quickly got bogged down in the detail. The schedule has slipped, notonce, but twice. After two disastrous rounds of consultations with the privatesector, the regulators were sent back to the drawing board for a third time in late2001.10 Part of the delay is economics. Two quantitative impact studies conductedby the committee to gauge the impact of the proposals for capital requirementsgenerated some unexpected results. Whereas the committee had begun theprocess with the goal of maintaining a similar level of capital prescribed under the1988 Accord, the proposals under Basel II led to an increase in the capitalrequirements for all banks. Perversely, the internal ratings approach that wasdesigned for the most sophisticated banks led to higher capital requirements thanthe external ratings approach, removing any incentive for banks to invest in a


more sophisticated internal ratings system.11 This was not satisfactory and theproposals needed to be recalibrated.

The other part of the delay is political. Countries such as Germany protestedthat the proposals would discourage lending to small- and medium-sizedenterprises, particularly the Mittelstand that make up much of German industry.German Chancellor Schroeder threatened to veto any new European directivebased on the latest proposals unless changes were made (The Economist, 2001).Other vested interests have similar concerns. As a consequence, the BaselCommittee is now working on a third consultative package to be issued in 2002,with a final agreed accord to be finalized in late 2003 and implemented by 2006.Recent commentary suggests, however, that even this deadline is unlikely to beachieved.

Basel II is ambitious and complex, with the new accord and supportingdocuments running to close to 500 pages (BIS, 1999). Basel II consists of threepillars that address different aspects of banking regulation. Pillar 1 outlines a newsystem of minimum capital requirements, with banks given the choice amongthree methodologies for assigning capital against their risky portfolios. Thestandardized approach calculates capital based on the external credit ratingsassigned to a security. Capital charges are determined based on an ascending scaleof “risk buckets,” with riskier loans requiring greater capital to be held againstdefault. Alternatively, banks may calculate capital requirements by using internalratings (IRB approach), with banks choosing among a basic (IRB Foundation) or anadvanced (IRB Advanced) methodology based on their ability. Pillar 2 deals withsupervisory reviewing of a bank’s capital adequacy and internal assessmentprocess, and lays out a set of standards for banking regulators to ensure consistenttreatment across jurisdictions. Pillar 3 proposes a wide range of disclosureinitiatives that would enhance the effective use of market discipline to encouragesound banking practices.

We assume here that non-metropolitan and most emerging-market ordeveloping-country banks will be subject to rating agency review rather than theirown internal ratings. Our view is that further incorporating these de facto or quasi-regulatory institutions into de jure regulation will produce perverse outcomes forthe financial markets and global public policy. We make eight points in support ofthis negative appraisal of these nascent, private makers of global public policy.

1. Rating Agencies are Pro-Cyclical. The primary difficulty with basing the capitalallocation of banks on credit ratings is that these same ratings have been foundto be pro-cyclical (Fight, 2001: 187–97; IMF, 1999: 121; Karacadag and Taylor,2000: 27). While the rating agencies claim to rate firms “through-the-cycle,” inpractice their ratings reflect a point-in-time approach. BIS has confirmed that, infact, the rating agencies give only modest weight to cyclical economicconditions, and their ratings exhibit systemic changes over the course of thebusiness cycle (BIS, 2000: 8, 142). Ratings of lower-rated firms are more volatile,and downgrades are typically associated with further downgrades, whileupgrades are not necessarily associated with further upgrades. This pro-cyclicalbehavior will generate perverse results for the goal of enhancing financialstability. Banks that use external ratings for allocating capital will always havethe least amount of capital put aside at the point when they need it most, whenthe business cycle is about to turn and non-performing assets are set to rise. Adownturn in the business cycle will cause the ratings on firms to deteriorate,


forcing banks to put aside greater capital at the worst possible time, when theeconomy is heading into recession. Should ratings decline unexpectedly, asoccurred in 2001 with Californian utilities or in 1997 with Asian sovereigns,banks will not have the capital in place in advance of the event. Thus, the trend-following behavior of ratings will exacerbate, not reduce, the risk of financialcrisis and contagion as banks will be undercapitalized when market conditionsare most difficult.

2. Basel II will Raise the Cost of Capital in Developing Countries. The new capitaladequacy framework will stunt the growth of firms in developing countries byraising their cost of capital relative to firms in the wealthy countries. The newframework raises the cost of capital in several ways. First, it allocates a higherrisk weighting to non-rated assets than rated assets. More than 80 percent ofrated corporations and 70 percent of rated banks are located in OECD countries,and a higher per capita income is associated with higher ratings (Ferri et al.,2000). By default, most firms in emerging markets are unrated and so loans tothese borrowers will accrue a higher capital charge, which will be passed on toborrowers. A World Bank study found 581 OECD-based industrial firms would seetheir cost of capital decline by 1 percent under the new framework, whereasonly 15 non-OECD firms would pay less (Ferri et al., 2000: 6). Instead, non-OECDborrowers would face an average increase of 1.5 percent. In the banking sector,the disparity is even greater with OECD banks paying 2.4 percent more ascompared to 6.4 percent for non-OECD banks. This higher cost of capital willcreate an incentive for borrowers in developing countries to seek other meansof financing that are potentially risk enhancing, such as off-balance-sheetfinancing (as implicated in Enron’s collapse), use of financial derivatives, orborrowing from unregulated, nonbank financial companies. It will also createdisincentives to adopting more sound risk assessments, increasing the volatilityof emerging-market banks’ capital requirements and worsening the availabilityof credit to cash-strapped firms in a crisis. In the past, this combination offactors created financial crises that ended with large bailouts from the richcountries who were affected by contagion. Second, non-US borrowers aresensitive to the rating on sovereign debt, which usually sets the rating ceiling forall firms in its jurisdiction. This sovereign ceiling penalizes well-managed firmslocated in countries with a low sovereign rating. This sovereign ceiling becomesmore punitive when you consider that studies have found that sovereign ratingsof low-income countries are typically downgraded excessively relative to OECDcountries, and that bank and nonbank ratings in these low-income countriesdid not recover when the sovereign rating was upgraded (Ferri et al., 2000).

3. Rating Agencies Lack Economic Accountability. Implicit in the idea of incorporatingbond ratings into a system of global regulation is the view that these privateagencies will be held accountable for their judgments and their mistakes(Karacadag and Taylor, 2000: 27–34). Up to now, rating agencies have not beenheld accountable under the law for negligent behavior, despite attempts to holdthem accountable following the collapse of Orange County and other publicand private entities (Husisian, 1990). The rating agencies have been able toavoid financial liability by claiming that they are offering only an opinion oncreditworthiness, not a measurement, allowing them to defer prosecutionunder freedom of speech statutes such as the First Amendment to the UnitedStates Constitution. Whereas the accountancy profession has been held liableunder Generally Agreed Accountancy Principles (GAAP), no parallel exists for


credit ratings, and any attempt to standardize ratings in this way has beenresisted by the rating agencies. This lack of a common standard makes itdifficult for a regulatory agency or a court to establish an independentbenchmark for measuring a rating agency’s performance. However, even if therating agencies were found liable for negligence in their duties, they would nothave the financial capital available to settle such disputes without rapidly goingbankrupt. As a result, the cost of rating agency failure would continue to fall oninvestors, or on taxpayers following a government-sponsored bailout. Such wasthe case during the Asian financial crisis (King, 2001).

4. Rating Agencies have Authority, but Lack Political Accountability. Rating agenciesenjoy an “accountability gap” (Kerwer, 2002). Rating agencies are hybrid formsof authority, operating between the state and the market, which have acquiredpublic authority due to their professional expertise, their specialist knowledge,their reputation and acceptance by market actors (Sinclair, 1994a, 1994b,2000). As noted, in the USA, credit ratings have been explicitly incorporated inregulation since 1931, with ratings restrictions written into laws governingbanks, pension funds, insurers, broker dealers and mutual funds (BIS, 2000: 54;Sinclair, 1999: 157). Ratings are also incorporated into regulation in Canada,Britain, France, Italy and Japan. Despite this official role for ratings, nosubstantive public oversight of the rating agencies actually exists. They haveescaped democratic mechanisms of accountability despite playing a part in theoperation of credit markets. This lack of political accountability underminestheir legitimacy and has led to calls from the Investment Company Institute (aninterest group representing institutional investors) for the supervision and legalaccountability of rating agencies (Investment Company Institute, 1998). Theonly form of accountability in place is that of market acceptance.

5. Weak Institutions in Emerging Markets will Undermine Ratings. The reputation ofthe global rating agencies depends on a track record in the USA, where bothmarket and non-market institutions have supported and enhanced theirsuccess. Can this track record be repeated in the context of emerging marketswhere the judicial system, the supervision of banks, the sophistication ofinvestors and the level of financial disclosure are far different from those inindustrialized countries? Under Basel II, domestically certified rating agencieswill provide the risk assessment used to determine capital requirements in theirjurisdiction. In the future, each country can be expected to have a local ratings“champion” which will compete with the global agencies. Domestic ratings willrisk becoming politicized as private interests lobby in order to be awarded thehighest possible rating. Given the evidence of corruption and bribery in manyregimes around the world (termed “crony capitalism” during the Asian crisis),do international regulators wish to use the outputs from potentiallycompromised firms to minimize risks to the international financial system?

6. There are Incentives for Ratings Shopping. More generally, the multiplication ofratings for firms combined with the structure of Basel II will create incentivesfor ratings shopping (Karacadag and Taylor, 2000: 32). Studies of the existingrating agencies have shown that the smaller agencies consistently providehigher ratings than the global leaders. This problem will only worsen as banksshop around to get the highest ratings available on the market in order toreduce the capital charge on their loans. Any rating agency which does not wantto lose market share will be forced to compete in this market.

7. The New System will be Unwieldy and Create Moral Hazard. Basel II may look good


on paper, but in practice its full implementation will be highly bureaucratic andcostly. The external ratings framework is intended as a step toward the internalratings approach in which individual banks determine their own capitalallocation. No one doubts that building these proprietary credit-risk models willbe an expensive and time-consuming endeavor, requiring the accumulation ofcredit-risk expertise and a vast investment of senior management time.Assuming these obstacles are somehow overcome and the models are put inplace, Basel II will still be infeasible for a number of operational reasons. First,adequate historical statistics of the probability of default and recovery givendefault do not exist, even among the most sophisticated banks. Second,mapping the internal credit ratings of thousands of individual banks onto auniform and consistent set of global capital “buckets” will be a monumentaltask. The Basel supervisors will be required to evaluate, test and approve eachbank’s internal credit-risk model before then deciding how to graft the bank’sinternal ratings onto a global template. The functioning of this system willfurther require periodic checks to ensure it is being implemented as planned.Third, the completion of these tasks will require the Basel supervisors to makean equal investment in expertise, management time and resources—oftenrelying on local regulators or central banks to fulfill this role. It is easy toimagine that many banks will be left to their own devices, with loopholes andexotic accounting methods being deployed in order to generate the lowestpossible capital charge. Fourth, this arrangement is fraught with moral hazard,and is not likely to work in practice (particularly in developing countries), giventhe experience of past failures of banking regulators in even the mostdeveloped countries.

8. Negating the Market in Reputation. Institutionalizing rating agencies in Basel II will(given the high barriers to entry) undermine the reputational constraintotherwise enforced by the capital market (Partnoy, 1999; Karacadag and Taylor,2000: 33; White, 2001). Any degrading of reputational enforcement by marketoperatives will loosen inhibitions on inflating ratings to satisfy issuers. Acomplacent, parasitic rating industry will result.

While market participants agree that regulators are constantly playing catch upwith financial market developments, private actors need to see a return frommoving to the new regulatory framework in order to have an incentive to adopt it.Currently, these private actors see only higher costs and increased competitionresulting from the current proposals. This approach to making regulatory policywill not lead to the creation of a safer global financial system. As a result, the BaselII process is in deadlock. The committee are unwilling to rework or reduce theproposals, given the significant work that has gone into producing them. Instead,the committee is trying to address the concerns of individual interest groups, atthe cost of an increased complexity that will render the new accord unworkable.The real cost of this process will be a riskier financial architecture that encouragesprivate actors to find new ways around the rules. In other words, through thispolitical process the main principle behind a new capital adequacy accord is likelyto be lost.

If the process could be restarted from the beginning, how might Basel IIproceed in order to reach a successful conclusion? We argue that the answer is tostart small, focusing on minimum capital requirements and working out a moreappropriate mapping of risk and capital that provides incentives for market actors


to implement the new accord. The detail of the accord is crucial, such as the roleassigned for external ratings, the definition of risk buckets, the treatment ofunrated entities and the determining of implementation incentives facing users.

We argue that the Basel Committee should be less ambitious. The new accordshould start small and focus on problems with the 1988 Accord, such as theperception among G-10 bank supervisors that the financial industry was innovating“around” the Accord via securitization, credit derivatives and similar instruments(Karacadag and Taylor, 2000: 13). The committee should also give more attentionto the calculation of minimum capital requirements addressed in Pillar 1. Ashortcoming of the minimum capital requirements under the 1988 Accord is theunequal treatment of OECD versus non-OECD borrowers. This distortion has beenclearly identified, with the worst fears played out during the Asian financial crisis.Given that membership in the OECD was never meant to be an indication of theability to repay a bank loan, policy-makers should resolve to reform this distortionin the 1988 Accord. A simple framework would be to replace the currenttreatment based on membership in the OECD with the use of external ratings asproposed in Basel II. In other words, loans to sovereign borrowers would incurcapital charges in line with the sovereign’s credit ratings, not its membership ofthe OECD. While Pillars 2 and 3 of Basel II deal with substantive issues, these issuesshould be left off the table until this first fault of the 1988 Accord is addressed.

Second, the more contentious issue relates to the treatment of private-sectorborrowers and the calibration of the risk buckets used to assign capital in thecurrent proposal. The choice of which rating categories to include in any givenrisk bucket under the current proposal does not reflect the documented history ofratings defaults and the ratings migration of borrowers. For example, S&P’sanalysis suggests that the current system will leave banks undercapitalized (S&P’s,2001: 4). Instead of arbitrarily grouping different external ratings intoincompatible risk buckets, regulators should focus on the principle that capitalcharges should be allocated based on the underlying risk of default. Capitalcharges should rise as the risk of default rises. This principle has been violatedunder the current proposal as categories of ratings with widely differing defaultprobabilities are included as the same risk. Reshuffling risk buckets to add more“granularity” and to bring them in line with the default history of credit ratingswould be a second step to a new accord.

Third, a major weakness with Basel II is that it still does not adequately addressthe issue of unrated entities. The current treatment is unsatisfactory for all parties,as corporations with a rating of BB– and lower incur a 150 percent capital charge,while unrated borrowers have a 100 percent capital charge. This classificationcreates a disincentive for riskier borrowers to get a rating if they have reason tobelieve their rating will be below BB–. The new accord should encourageborrowers to get a rating, by forcing riskier entities to bear higher capital charges.At the same time, unrated entities should not be punished for the fact that theyhave not acquired a credit rating. The solution would be to create an incentive forunrated entities to gain a rating, by phasing in higher capital charges over time. Atthe end of some predetermined phase-in period, unrated entities would incur thehighest capital charge. This arrangement would provide an incentive for theseborrowers to get a rating, while not punishing them in the short term for a lack ofrating. The constraint on this process will be the ability of the rating agencies toprovide these new ratings. In all likelihood, this phase-in process may require up to5 years to complete. If the end result is a safer financial system, it is worth the wait.


(V) The Political Economy of Basel IIThis discussion of Basel II has so far ignored the political economy questionregarding who benefits from a new capital adequacy accord. This question goes tothe heart of the problem, as it focuses on the incentives facing the different actorsfor agreeing and implementing a new international regulatory framework. Todate, the saga of “the good tailors of Basel” is one of rent seeking preventing theprovision of public goods (The Economist, 2002). While there is widespreadconsensus that the 1988 Accord is flawed, no one can agree on how to fix it due tocontesting private interests that undermine the public agenda. Vested interestssuch as banks and banking groups, small- and medium-sized enterprises (SMEs),borrowers in developing countries, unrated borrowers and the credit ratingagencies are all focused on the negative implications for their businesses.

Basel II will be costly for banks to implement, and will benefit some institutionsat the expense of others. Banks expect to incur higher capital charges under BaselII, as well as high costs in terms of management time, staffing and IT expenses inorder to implement the new accord (PriceWaterhouseCoopers, 2001). The Swissuniversal bank Credit Suisse has estimated that the cost of Basel II for the world’s30,000 banks will be US$2.5 trillion over 5 years, a cost that they warn will be passedon to consumers (Credit Suisse, 2001). The International Monetary Fund (IMF) isalso critical of the complexity of the proposals, arguing that they have been writtenwith the most sophisticated banks in mind. As a result, they may not be realistic forall banks in all countries and some banks will benefit while others will not (IMF,2001). Lastly, Basel II will encourage an uneven playing field by offering acompetitive advantage to financial services firms not covered by the Accord,namely, investment banks, brokers, asset managers and finance companies, whowill benefit at the expense of the regulated banks.

Borrowers will suffer. SMEs argue that they will pay a higher cost to borrow dueto the higher capital charges related to loans to this sector. Germany is particularlyworried, as Basel II penalizes long-term bank lending that is the backbone ofborrowing by the Mittelstand companies. Developing countries will be hurt due tothe relative absence of external ratings. Unrated borrowers are assigned a higherrisk weighting under the new proposals, which will lead borrowers in manydeveloping countries to be penalized relative to the rich countries (Asiamoney,2001). This treatment has not been overlooked by the IMF, which stressed in itssubmission the need for a level playing field in capital charges that would not besystematically biased against any class of borrowers (IMF, 2001).

Lastly, the credit rating agencies themselves (who stand to see a significantincrease in business as a result of Basel II) express reluctance to have their ratingsincorporated into regulation. Both Moody’s and S&P fear the problem of ratingsshopping, as well as the risk that regulatory competition will erode rating agencyobjectivity (Moody’s Investors Service, 2001b; S&P’s, 2001). They see a risk thatcountries will use national recognition to reward and punish the credit ratingagencies, with the creation of national rating agencies of questionable credibilitydiminishing investor confidence in this product. The end result will be greatercompetition for the main rating agencies and an inevitable erosion of thereputation of credit ratings to the detriment of the leading firms.

This debate over Basel II may be analyzed from the point of view of the theoryof redistributive cooperation, which seeks to explain the forces that lead to thecreation of international institutions (Oatley and Nabors, 1998; Richards, 1999).


The theory of redistributive cooperation explains the creation of internationalinstitutions by focusing on demand and supply. This theory argues that actorsmust receive private rents from a new institution in order to have an incentive todeal with it. The new institution in turn provides a public good that is otherwiseunderprovided by the markets due to problems of collective action and freeriding.

The demand for a new international institution comes from domestic interestgroups that see an opportunity to capture rents from abroad. In the case of BaselII, the US credit rating agencies will capture rents in the form of expandedrequirements for their services by all participants in the bank lending markets.Given the number of unrated entities in this market, the potential furtherrevenues from this business are large, but few actors are in a position to providethis service in a timely fashion other than the major US rating agencies. Bankslocated in higher-rated countries will also benefit as they will receive favorablecapital treatment relative to competitors in lower-rated countries. Neither of theseactors will have an incentive to adopt a new accord if they foresee a decline intheir profitability or an increase in competition.

The supply for a new international institution is provided by politicians in themember states who view international institutions as a means to extract wealthfrom the international sector that can be redistributed domestically to theirconstituents (Oatley and Nabors, 1998: 40). While regulators may draft the newcapital accord, in order for it to be implemented politicians in the member statesmust pass the enacting legislation. Germany has made this point clear in its threatto veto the new proposals at the EU level. To be successful, policy-makers in thewealthy nations must see a benefit from the provision of more rigorous capitalstandards. At the international level, this appears to be the case. The wealthycountries have been forced to come to the aid of sovereign borrowers in Asia,Latin America and eastern Europe over the past decade following currency crisesthat have ultimately been tied to faulty domestic banking systems. Wealthycountries have distributed significant sums through bailouts following these crises,but they have also faced the threat of financial contagion through theinternational banking system leading to the collapse of their domestic banks.These two sources of international rents provide strong incentives for the richestcountries to sew together the obvious holes in the current accord. The mainobstacle to the current accord comes from other domestic actors who see the costsoutweighing the benefits. This dynamic needs to be addressed for a new proposalto be successful.

(VI) ConclusionsAs we have argued, rating agencies offer to solve the asymmetric informationproblem between borrowers and creditors. In providing this function, the ratingagencies adjust the “ground rules” of international capital markets, therebyreshaping the internal organization and behavior of those institutions seekingfunds. Rating agencies’ views on what is considered “acceptable” shape thedecision making and actions of those dependent on ratings, due in large part tothe inclusion of ratings in regulation. This process of interaction between privateactors in the capital markets narrows the expectations of creditors and debtors to aset of norms shared among all parties. At the same time, it limits the scope ofconcrete policy initiatives available to policy-makers who are accountable for the


stability of global finance. Global change makes the public–private distinction atthe heart of traditional studies of public policy increasingly invalid. As this articleshows, public policy can also be made by private institutions or networks when theoutputs of these private institutions shape the basic norms that produce action ingovernments and business organizations.

Rating agencies were examined in this article as an example of the privatemaking of public policy. Their specific structural power (and hence theirinfluence on public policy) is derived, first, from the disintermediation trend inglobal finance and the asymmetric information problem it produces in capitalmarkets, and second, from their internal construction (and outward behavior) aspurveyors of judgments perceived as endogenous and therefore legitimate byother actors. This structural power is made concrete when policy-makersincorporate the decisions of these private actors into regulation at the domesticand the international level.

According to S&P’s President, Leo C. O’Neill (1992), the rating agencies seethemselves as “quasi-regulatory institutions.” They are well placed to defer ormodify government challenges to their authority, as the hesitancy with which anynew effort (including Basel II) to pull them further into regulation demonstrates.Nevertheless, a significant feature of their relationship with public authority is thetendency of government to use quasi-regulatory outputs such as these assubstitutes for government action. This quasi-regulation poses increased risksbecause the incentives mapped out above do not suggest a framework that willbring greater stability to global finance. Just as Basel I created incentives forbehavior in the financial markets which undermined the intent of the initialCapital Adequacy Accord, the political economy of bond rating, as analyzed here,may be the weakest link in the Basel II proposals.

Notes1. In part, this article builds on some of the concepts developed in Sinclair (2000, 2001). It

was first presented at the annual meeting of the American Political Science Association,San Francisco, California, September 2001. The authors would like to thank Louis Paulyand the other participants on this panel for helpful comments.

2. Summers (2000) provides a concise discussion of this literature, with references tonumerous studies. For the dates of crises, see Glick and Hutchison (2001).

3. The Basel Committee on Banking Supervision is a committee of banking supervisoryauthorities established by the G-10 countries in 1975. It consists of senior representativesof bank supervisory authorities and the central banks of Belgium, Canada, France,Germany, Italy, Japan, Luxembourg, the Netherlands, Spain, Sweden, Switzerland, theUK and the USA. The committee usually meets at the Bank for International Settlementsin Basel, where its permanent secretariat is located (see http://www.bis.org).

4. In recent years, five amendments to the Accord have been agreed to incorporatematters such as off-balance-sheet items and market risks.

5. Two shorter sections define the target standard ratio and the transitional andimplementing arrangements. There are four technical annexes covering the definitionof capital, the counterparty risk weights, the credit-conversion factors for off-balance-sheet items and the transitional arrangements. A copy of the 1988 Accord is availablefor download from the BIS website at http://www.bis.org.

6. Under the weighting scheme, cash and government securities of a bank’s own homecountry have a risk weight of zero, obligations of banks incorporated in OECD membercountries have a risk weight of 20 percent, fully secured mortgages on property have arisk weight of 50 percent and all other claims on private-sector entities have a risk


weight of 100 percent. A bank’s risk capital is then calculated by multiplying each assetclass by its associated risk weighting, then taking 8 percent of this risk-weightedportfolio.

7. For international political economy (IPE) studies of the politics of the 1988 Basel CapitalAdequacy Accord, see Coleman (1996), Kapstein (1994), Oatley and Nabors (1998),and Porter (1993). For IPE studies of other international financial regulation, see Cerny(1993), Coleman and Porter (1994), Eichengreen (1996), Germain (1997), Goodmanand Pauly (1993), Simmons (2001), Sobel (1994), and Strange (1998).

8. For more details on IOSCO and other regulatory bodies in the international financialmarkets, see Filopovic (1997).

9. The Accord encouraged loans to countries such as Mexico and South Korea, as loans tothese countries had the same risk weighting as other OECD loans, despite both Mexicoand South Korea having a significantly higher risk of default.

10. The committee received more than 250 comments on its January 2001 proposal (BIS,2001b). Most of these submissions are published on the committee’s website athttp://www.bis.org/bcbs/index.htm.

11. Other problems were revealed by the quantitative impact studies. Many banks were notable to calculate capital requirements under each of the three methods, with only 22out of 138 banks able to calculate the advanced approach for all portfolios. Banks foundit difficult, if not impossible in some cases, to overcome data limitations (BIS, 2001c).

ReferencesAsiamoney (2001). “New Basel Rules Under Scrutiny,” September, 12(7): 72.Bates, Abbe (2000). S&P’s official, New York. Email communication to author. 13

November.BIS (Bank for International Settlements) (1999). A New Capital Adequacy Framework. June.

Basel: BIS.BIS (2000). Credit Ratings and Complementary Sources of Credit Quality Information. Basel

Committee on Banking Supervision Working Paper No. 3. August. Basel: BIS.BIS (2001a). Overview of the New Basel Capital Accord. January. Basel: BIS.BIS (2001b). “Update on the New Basel Capital Accord.” BIS press release, 25 June. Basel:

BIS.BIS (2001c). “Results of the Second Quantitative Impact Study.” 5 November. Basel: BIS.Cantor, Richard and Frank Packer (1994). “The Credit Rating Industry.” Federal Reserve Bank

of New York Quarterly Review, 19(2).Caprio, Gerard and Daniela Klingebiel (1997). “Bank Insolvency: Bad Luck, Bad Policy, or

Bad Banking?” In Annual World Bank Conference on Development Economics 1996 (M. Brunoand B. Pleskovic, eds). Washington, DC: World Bank.

Cerny, Philip G. (1993). “The Deregulation and Re-regulation of Financial Markets in aMore Open World.” In Finance and World Politics: Markets, Regimes, and States in the Post-hegemonic Era (Philip G. Cerny, ed.). Brookfield, VT: Edward Elgar.

Chmaj, Andrew (2000). Moody’s Investors Service official, London. Email communicationto author. 31 July.

Cohen, Benjamin J. (1996). “Phoenix Risen: The Resurrection of Global Finance.” WorldPolitics, 48(2): 268–296.

Coleman, William D. (1996). Financial Services, Globalization and Domestic Policy Change.Basingstoke: Macmillan.

Coleman, William D. and Tony Porter (1994). “Regulating International Banking andSecurities: Emerging Cooperation Among National Authorities.” In Political Economy andthe Changing Global Order (R. Stubbs and G. Underhill, eds). Basingstoke: Macmillan.

Credit Suisse (2001). “The New Basel Capital Accord—Consulative Paper of January 16,2001.” 30 May. Zurich: Credit Suisse.

Cutler, A. Claire (1998). “Locating ‘Authority’ in the Global Political Economy.”International Studies Quarterly, 43(1): 59–81.


Economist, The (1992). “Time to Leave: A Survey of World Banking.” 2 May.Economist (1994). “Recalled to Life: A Survey of International Banking.” 30 April.Economist (2001). “The Basel Perplex.” 10 November: 65.Economist (2002). “The Good Tailors of Basel.” 23 February: 75.Edwards, Ben (1994). “Will the Agencies be SEC Puppets?” Euromoney, November: 26–27.Eichengreen, Barry J. (1996). Globalizing Capital: A History of the International Monetary System.

Princeton, NJ: Princeton University Press.Everling Advisory Services (2001). “List of Rating Agencies.” July. http://www.everling.de.Ferri, Giovanni, Li-Gang Liu and Giovanni Majnoni (2000). “The Role of Rating Agency

Assessments in Less Developed Countries: Impact of the New Basel Guidelines.” April.Washington, DC: World Bank.

Fight, Andrew (2001). The Ratings Game. Chichester: Wiley.Filopovic, Miroslava (1997). Governments, Banks and Global Capital: Securities Markets in Global

Politics. Aldershot: Ashgate Publishing.Germain, Randall D. (1997). The International Organization of Credit: States and Global Finance

in the World-Economy. Cambridge: Cambridge University Press.Glick, Reuven and Michael Hutchison (2001). “Banking and Currency Crises: How

Common Are Twins?” In Financial Crises in Emerging Markets (Reuven Glick, RamonMoreno and Mark Spiegel, eds). New York: Cambridge University Press.

Goodman, John B. and Louise Pauly (1993). “The Obsolescence of Capital Controls?Economic Management in an Age of Global Markets.” World Politics, 46(1): 50–82.

Greenberg, Susan (1993). “New Rating Agency Causes a Stir.” The Guardian, 13 February.Husisian, Gregory (1990). “What Standard of Care Should Govern the World’s Shortest

Editorials? An Analysis of Bond Rating Agency Liability.” Cornell Law Review, 75: 411–461.IMF (International Monetary Fund) (1999). International Capital Markets: Developments,

Prospects, and Key Policy Issues. Washington, DC: IMF.IMF (2001). “IMF Staff Comments on the January 2001 Proposals of the Basel Committee on

Banking Supervision for a New Capital Adequacy Framework for Banks.” 9 May.Washington, DC: IMF.

Investment Company Institute (1998). “Letter to Mr J.G. Katz, Securities and ExchangeCommission.” 2 March. http://www.ici.org/issues/nrsro_comment.html.

Kapstein, Ethan B. (1994). Governing the Global Economy: International Finance and the State.Cambridge, MA: Harvard University Press.

Karacadag, Cem and Michael W. Taylor (2000). The New Capital Adequacy Framework:Institutional Constraints and Incentive Structures. IMF Working Paper WP/00/93.Washington, DC: IMF.

Kerwer, Dieter (2002). “Rating Agencies: Setting a Standard for Global Financial Markets.”Economic Sociology—European Electronic Newsletter, June, 3(3): 1–6.http://www.siswo.uva.nl/ES.

King, Michael R. (2001). “Who Triggered the Asian Financial Crisis?” Review of InternationalPolitical Economy, autumn, 8(3): 438–466.

Kirkland, Edward C. (1961). Industry Comes of Age: Business, Labor and Public Policy1860–1897. New York: Holt, Reinhart and Winston.

McDonough, William J. (1999). “Global Financial Reform: A Regulator’s Perspective.” InFederal Reserve Bank of New York 1999 Annual Report, 2–12. New York: FRBNY.

Milgrom, Paul R., Douglas C. North and Barry R. Weingast (1990). “The Role of Institutionsin the Revival of Trade: The Law Merchant, Private Judges, and the Champagne Fairs.”Economics and Politics, March, 2(1): 1–23.

Moody’s Investors Service (2001a). “About Moody’s.” July. http://www.moodys.com.Moody’s Investors Service (2001b). “Comment on the Second Consultative Package of the

New Basel Capital Accord.” 30 May. New York: Moody’s Investors Service.Moran, Michael (1991). The Politics of the Financial Services Revolution: The US, UK and Japan.

New York: St. Martin’s Press.Oatley, Thomas and Robert Nabors (1998). “Redistributive Cooperation: Market Failure,

Wealth Transfers, and the Basle Accord.” International Organization, 52(1): 35–54.


O’Neill, Leo C. (1992). S&P’s President, New York. Interview with the author. 18 August.Partnoy, Frank (1999). “The Siskel and Ebert of Financial Markets?: Two Thumbs Down for

the Credit Rating Agencies.” Washington University Law Review, 77(3).Porter, Tony (1993). States, Markets and Regimes in Global Finance. New York: St. Martin’s Press.PriceWaterhouseCoopers (2001). “The New Basel Capital Accord.” Letter to the Basel

Committee on Banking Supervision, 28 May.Richards, John E. (1999). “Toward a Positive Theory of International Institutions:

Regulating International Aviation Markets.” International Organization, 53(1): 1–37.Simmons, Beth A. (2001). “The International Politics of Harmonization: The Case of

Capital Market Regulation.” International Organization, 55(3): 589–620.Sinclair, Timothy J. (1994a). “Between State and Market: Hegemony and Institutions of

Collective Action Under Conditions of International Capital Mobility.” Policy Sciences,27(4): 447–466.

Sinclair, Timothy J. (1994b). “Passing Judgement: Credit Rating Processes as RegulatoryMechanisms of Governance in the Emerging World Order.” Review of InternationalPolitical Economy, spring, 1(1): 133–159.

Sinclair, Timothy J. (1999). “Bond-Rating Agencies and Coordination in the Global PoliticalEconomy.” In Private Authority and International Affairs (A. Claire Cutler, Virginia Hauflerand Tony Porter, eds). Albany, NY: State University of New York Press.

Sinclair, Timothy J. (2000). “Reinventing Authority: Embedded Knowledge Networks andthe New Global Finance.” Environment and Planning C: Government and Policy, August,18(4): 487–502.

Sinclair, Timothy J. (2001). “The Infrastructure of Global Governance: Quasi-RegulatoryMechanisms and the New Global Finance.” Global Governance, 7: 441–451.

Smith, George David and Richard Sylla (1993). “The Transformation of FinancialCapitalism: An Essay on the American Capital Markets.” Financial Markets, Institutions andInstruments, May, 2(2).

Sobel, Andrew C. (1994). Domestic Choices, International Markets: Dismantling National Barriersand Liberalizing Securities Markets. Ann Arbor, MI: University of Michigan Press.

Standard & Poor’s (2001). “Standard & Poor’s Response to the New Basel Capital Accord.”New York: Standard & Poor’s.

Stevens, Ed (2000). “Evolution in Banking Supervision.” Federal Reserve Board ofCleveland Economic Commentary, 1 March. http://www.clev.frb.org/research/com2000/#0301.

Strange, Susan (1998). Mad Money: When Markets Outgrow Governments. Ann Arbor, MI:University of Michigan Press.

Summers, Lawrence H. (2000). “International Financial Crises: Causes, Prevention, andCures.” American Economic Review, 90(2): 1–16.

Toffler, Alvin (1990). Powershift. New York: Bantam.White, Lawrence J. (2001). The Credit Rating Industry: An Industrial Organization Analysis.

Working Paper S-01-7. New York: New York University Salomon Center.

Biographical NotesMICHAEL R. KING completed his Ph.D. in International Relations at the LondonSchool of Economics, and is a Principal Researcher at the Bank of Canada. Priorto beginning his doctoral work, he worked in the capital markets for 7 years,primarily with investment bank Credit Suisse First Boston. His research focuses oninternational financial markets and monetary policy frameworks. ADDRESS: Bank ofCanada, Financial Markets Department, 234 Wellington Street, Ottawa K1A 0G9,Canada [email: [email protected]].


TIMOTHY J. SINCLAIR is a political scientist at the University of Warwick in England,where he is Director of the graduate program in International Political Economy.A former New Zealand Treasury official, Dr Sinclair was a visiting scholar atHarvard University during 2001/02, where he completed a book manuscript onthe political economy of bond rating. His scholarly articles have appeared in GlobalGovernance, New Political Economy, Environment and Planning, Policy Sciences andReview of International Political Economy. He has edited a number of books, mostrecently Structure and Agency in International Capital Mobility (with Kenneth P.Thomas), published by Palgrave in 2001. His research is concerned with thesurveillance and regulatory mechanisms associated with global finance, and theconsequences of global change for everyday life. ADDRESS: Department of Politicsand International Studies, University of Warwick, Coventry CV4 7AL, UK [email:[email protected]].


Financial Markets and Politics: The ConfidenceGame in Latin American Emerging Economies

JUAN MARTÍNEZ AND JAVIER SANTISO

ABSTRACT. This article focuses on the interactions between politics andfinancial markets in emerging economies. More precisely, it examineshow Wall Street reacts to major Latin American political events. The casestudy focuses on the 2002 Brazilian presidential elections. The firstsection of the article provides a critical review of the available literature.The second section presents an empirical study of Wall Street analysts’perceptions of the 2002 presidential elections in Brazil, based on reportsproduced by leading Wall Street investment firms. The final section usespolling and financial data from previous Brazilian elections to place theevents of 2002 in comparative historical perspective.

Keywords: • Emerging economies • Financial markets • Internationalpolitical economy • Latin America

The overriding objective of policy must ... be to mollify market sentiment. But,because crises can be self-fulfilling, sound economic policy is not sufficient togain market confidence; one must cater to the perceptions, the prejudices, andthe whims of the market. Or, rather, one must cater to what one hopes will bethe perceptions of the market.

Paul Krugman (2002)

IntroductionAt the heart of financial transactions lies a question of confidence. Economistsfrom Smith to Coase have emphasized the importance of confidence, whether toexplain the wealth of nations or the birth and death of firms. More recently, PaulKrugman has highlighted the contemporary “games of confidence” that liebehind financial turbulence. Given financial needs and a lack of savings, LatinAmerican emerging markets are highly dependent on international capital flows.




The trick for policy-makers is to keep debt premiums low by maintaining investorconfidence in their countries’ respective economies. “The peculiar thing,”explains Krugman (2002: 113), “is that, because speculative attacks can be self-justifying, following an economic policy that makes sense in terms of thefundamentals is not enough to assure market confidence. In fact, the need to winthat confidence can actually prevent a country from following otherwise sensiblepolicies and force it to follow policies that would normally seem perverse.”

Part of the confidence game lies in the interaction between politics andeconomics, or to be more precise, in the perceptions of analysts, fund managers,national and international investors, and rating agencies engaged in emerging-market political games (Santiso, 2003). In emerging markets, financial turbulenceand politics are closely linked. Elections are frequently associated with significantincreases in financial market spreads. Not only do bond spreads tend to increaseduring electoral periods, but agency sovereign risk ratings also tend to deteriorate(Reihnart, 2002; Reisen, 2002). Both agencies and bondholders view elections asperiods of uncertainty, and hence of increased risk. In the same way, suddenswings in capital flows and currency devaluation are also associated with electoralyears. Moreover, output losses from financial crises tend to be larger in emergingmarkets than industrial countries. One explanation lies with politics and the so-called “weak government hypothesis”: weak governments delay theimplementation of necessary, but politically costly economic reform (Calvo, 2001,2002).1

The problem of instability in emerging markets may even have grown worsesince earlier phases of globalization (Bordo and Panini Murshid, 2002; Bordo etal., 2001: 51–82; Eichengreen and Bordo, 2002). With capital flowing more easilyinto and out of emerging markets, following liberalization and the more openfinance of the 1990s, emerging markets have experienced an increased frequencyof crashes. When compared with developed countries, the frequency of crashes indeveloping economies tends to be much higher. According to a recent analysis,the frequency of crashes for a closed emerging country is 25 percent on averageduring a year, against less than 9 percent for a developed country (Martin and Rey,2002). The frequency of crashes for an open emerging country is close to 62percent, compared to less than 10 percent for a developed country. Not only areemerging markets more prone to crashes than developed countries, but also openemerging countries are more exposed to financial crashes than closed emergingmarkets.

The 1990s witnessed a consolidation of democracy in the developing world.With the spread of democracy, politicians became more aware of social pressures,and they tended to delay costly adjustments. Emerging markets are particularlysusceptible to this problem of delayed reform because of more generalinstitutional weakness. As recent research has established, countries with weakinstitutions tend to adopt poor macroeconomic policies, and this leads in turn tobad macroeconomic performance and higher financial volatility (Acemoglu et al.,2002). Indeed, these authors argue that: “distortional macroeconomic policies arenot typically chosen because politicians believe that high inflation or overvaluedexchange rates are good for economic performance. Instead, they reflectunderlying institutional problems in these countries” (Acemoglu et al., 2002).

This article focuses on the interactions between politics and finance inemerging economies. More precisely, we will examine how Wall Street (that is, thefinancial markets) incorporates and reacts to major Latin American political


events. Our case study focuses on the 2002 Brazilian presidential elections. Thefirst section of the article presents a critical review of the existing literature. Thesecond section presents an empirical study of Wall Street analysts’ perceptions ofthe 2002 Brazilian presidential election. This analysis is based on the reports ofWall Street investment firms, focusing on the top fixed-income teams as ranked byInstitutional Investors, Bloomberg and Latin Finance in their league tables.2 Thethird section introduces historical data from previous election cycles in Brazil toset recent events in a comparative historical perspective.

Financial Markets and Politics in Emerging MarketsFinance and politics are closely linked, as can be seen in the history of nationalcurrencies developed by political authorities within specific territorial boundariesor in the systematic effect of electoral policy and political regime on the size andcomposition of government spending (González, 2002: 204–24; Persson, 2002:883–905). Interactions between politics and economics are also central tounderstanding financial crises in emerging markets. Probably the most relevantdefinition of an emerging market is an economy whose political outcomes anduncertainties (such as a presidential election or a cabinet reshuffle) tend to havehigh impacts on financial variables and therefore on stock markets. A leadingemerging-market bond fund manager explained:

At PIMCO, we ask two basic questions whenever confronted with indications ofactual or prospective political noise. First, can this noise undermine theeconomic and financial outlook for the country? Essentially, we were inquiringabout potential impact on debtors’ capacity and willingness to meet theirpayment obligations. The second question concerns the potential reaction ofother market participants. When closely considered, it is not particularlysurprising that, in most cases, the immediate market reactions to increasedpolitical noise has been very pronounced price volatility. After all marketsgenerally dislike uncertainty, especially when it emanates from what is stillviewed as Byzantine emerging markets politics. (El-Erian, 2002a: 3)

The paradox is that, in spite of such evidence, few empirical studies have focusedon the links between political variables and financial markets. One explanationmight be the difficulty that economists have had in formalizing political variables.Another might be the hesitation of sociologists to open the “black box” andexplore the socioeconomics of financial markets. However, the so-called second-generation models of crises,3 in considering optimal government behavior and thepolitical dimensions of economic trade-offs, have begun to introduce politicalvariables into the analysis of financial crises. Eichengreen, Rose and Wyplosz wereamong the first to address the political dimension of financial crises, findingintimate links between political processes and exchange-rate turbulence(Eichengreen et al., 1995: 249–312). Later, Frieden, Ghezzi and Stein argued thatweak governments might be more vulnerable to currency crises. In a detailedstudy of the behavior of real and nominal exchange rates in Latin America, theyconfirmed that changes in exchange-rate regimes coincided with elections(Frieden et al., 2001: 20–63). To be more precise, devaluations were generallypostponed until after elections. In fact, in the months following elections (2, 3 and4 months after), the average rate of nominal depreciation tends to be twice ashigh as in the months preceding elections. In the case of presidential elections,

MARTÍNEZ/SANTISO: Financial Markets and Politics 365

the average rate of nominal depreciation tends to be even higher: 4.5 times higherthan in the preceding months. Overall, the probability of major devaluationsincreases in the run-up to elections, with governments tending to put off theadjustment where possible until after the votes are cast.

Conversely, when governments embark on programs to stabilize the exchangerate, they tend to carry them out during pre-election periods, counting on apolitical dividend from currency stability. For example, between 1970 and 2000there were 34 changes of currency regime aimed at achieving greater stability in18 Latin American countries. In 26 of these cases, the change occurred ahead ofmajor elections; in other words, 76 percent of the cases occurred in the run-up toan electoral process (Schamis and Way, 2001). Overall, strategies for stabilizing theexchange rate are most frequently implemented ahead of presidential elections,with a probability three times higher than at any other point in the political cycle.Another study (focused on 88 speculative attacks on emerging-market currenciesbetween 1985 and 1999) also established that the propensity of politicians todefend a currency regime is greater in the months leading up to an election, butdeclines dramatically thereafter. The defense of the currency is 63 percentstronger in the 4 months preceding an election than in a normal period, but thisdrops to 19 percent in the 3 following months (Leblang, 2001).4

In recent years, there has been increasing interest among economists, politicalscientists and sociologists in this topic and a growing body of literature is nowexploring the connections between financial markets and politics (Bacmann andBolliger, 2001; Barber et al., 2001; Boni and Womack, 2002; Santiso, 1999b:307–30). The emerging literature suggests that political variables are indeedsignificant explanatory factors in emerging markets’ crises. For example,structural political variables are correlated with currency crises: left-winggovernments are more vulnerable to such crises, while strong governments withlegislative majorities and fragmented oppositions tend to be the least vulnerable(Block, 2001b; Drazen, 1999).

If we focus on Latin America, it is clear that political dynamics are a key variablefor understanding financial crises. The three most recent and significant financialcrises in the region (Mexico in 1994, Brazil in 1999 and Argentina in 2001) tookplace during a corresponding presidential or parliamentary electoral year (see, foran analysis of the first crisis, Santiso, 1999a). The same is true of other emergingmarkets: for nine other emerging economies, the financial crises of the 1990shappened during electoral periods or political transitions (Mei, 1999).5 Moreover,among the three types of risk to which financial markets are exposed (financialrisk, political risk and policy risk) political risk appears to be the major driverbehind capital flight from emerging markets.6

Elections in emerging countries have significant effects on market spreads andrating agency decisions. On average, elections in emerging markets tend to beassociated with a decline of 1 rating level on a 17-point scale (0–16). Similarly,sovereign debt due to the sale of US Treasury bills tends to increase by 21 percentagepoints 2 months after a major election in an emerging economy when comparedto the same period without an election. “Together,” underlines Steven Block in hisstimulating study, “these results suggest that at least two key actors in internationalcredit transactions, agencies and bondholders, view elections in developing countriesnegatively and exact a substantial premium on developing sovereigns and sub-sovereign individuals seeking capital” (Block and Vaaler, 2001).7 Overall, these resultssuggest that elections entail a substantial financial cost for developing countries.


In fact, for countries with weak economic fundamentals, political instabilityfurther aggravates financial vulnerability. This is particularly relevant in emergingeconomies, where political-institutional instability tends to be higher. PhilippeAghion et al. (2001) found that the bulk of institutional political changes thatoccurred over a 20-year period in a large sample of 177 countries wereconcentrated in emerging countries. Of 294 significant changes, Africa and LatinAmerica experienced the highest degree of political instability, with 138 and 59institutional changes respectively (compared to 15 changes in all theindustrialized countries). Africa alone accounted for nearly half of all politicalinstitutional changes recorded during the period, and Latin America for some 20percent.

The impact of democratic politics on currency and bond markets is notconfined to emerging markets. Uncertainty about electoral outcomes andgovernment survival also has repercussions for financial markets in Organizationfor Economic Cooperation and Development (OECD) countries, where politicalevents, elections, polls, and cabinet formation and dissolution make it moredifficult for traders to forecast economic conditions (Leblang and Bernhard,2000: 291–324; Lowry, 2001: 49–72). However, the impact of electoral processes isparticularly relevant for emerging markets, for here the associated uncertaintiesare greater and currency or bond traders’ confidence more fragile. For example,during election times, governments tend to increase public spending in order towin political support, potentially hurting investors’ interests.8 Although this is notspecific to emerging countries, fiscal deficits tend to be more critical here becauseof their financial needs and their difficulty in accumulating capital (Alesina et al.,1999: 233–55).

This tendency must be analyzed more carefully, however, since not all electionyears are associated with financial disruptions. Empirical research providesconsiderable evidence of fiscal-policy distortions during election periods inemerging countries. In a study that analyzes 17 Latin American countries over atime period from 1947 to 1982, a panel regression shows an increase of more than6 percent in public expenditure in the pre-election year and a decrease of morethan 7.5 percent in the post-election year (Ames, 1987). Politicians seeking re-election clearly have a strong incentive to manipulate fiscal policy to theiradvantage. The consequence of the electoral cycle is therefore to create the well-known tendency for fiscal performance (carefully monitored by analysts andinvestors) to worsen in election years, leading to macroeconomic instability.According to a study that examined a sample of 123 developing and industrializedcountries, fiscal deficits tended to be on average 1 percent of GDP higher inelection years (Shi and Svensson, 2001). Among emerging economies, the fiscaldeficits in election years were even larger—on average 2 percentage points of GDPhigher.

Another recent study, examining 69 countries, confirms the evidence ofelectorally motivated changes in the composition of public expenditure inemerging countries. In election years, public expenditure tends to shift towardshort-term (and visible) consumption and away from public investment goods.Typically, current expenditure shares show an increase of as much as 2.3percentage points during election years, while long-term expenditure such ascapital investment tends to decline by as much as 1.55 percentage points.However, countries with non-competitive systems exhibit no election-year effect onpublic spending (Block, 2001a). In the same way, political institutions that limit


the discretionary behavior of policy-makers will tend to reduce the volatility ofgovernment expenditures and revenues prior to and after elections (Henisz,2002).

The response of financial markets to electoral and partisan change, examinedin a study of 78 developing countries using monthly data from 1975 to 1998,confirms that speculative attacks are more likely just after an election as comparedto all other periods (Leblang, 2002). The reaction of global financial markets topolitics in new democracies within the same region may also differ, depending onthe degree of democratization and the transparency (or perceived transparency)of the policy-making process, the length of time democratic institutions have beenin existence, and the scale of the government’s legislative majority and its politicalcohesion (Hays et al., 2001; MacIntyre, 2001: 81–122).

Thus politics is far from neutral in financial markets. Moreover, there is clearevidence that financial-market participants take account of political events. Thereturns of certain stocks can be affected directly by expectations of politicaloutcomes (Lin and Roberts, 2001), and the corresponding stock price movementsoccur in the weeks immediately prior to elections (Pantzalis et al., 2000). Themonitoring of political events in emerging markets by financial analysts can beseen in the research products of investment boutiques that constantly ponder overuncertain political outcomes. Some firms, such as Lehman Brothers, have evencreated specific joint ventures to incorporate analysis of political and social factorsmore consistently into their fixed-income research.9 In addition, the Frenchfinancial operator CDC Ixis developed a specific approach for considering theimpact of political risk in emerging markets (CDC Ixis, 2002a).

There are also clear links between lack of transparency, political uncertaintyand financial crises (Drazen, 2000; Haggard, 2000; Chang, 2002). Rationalcontagion in emerging markets can be driven in part by political considerations.Because of their opaque policy processes, less democratic countries can suffermore from contagion in international financial markets. At the same time,because political rumors float freely within globalized political regimes (regardlessof the existence of emerging markets), all countries, democratic or authoritarian,suffer some of the costs.

It is also interesting to consider stock-market behavior when firms announceinvestment in a country with secure or, on the contrary, unsecured property rights.In an empirical study examining shareholder perceptions of the foreign directinvestment (FDI) announcements of large US firms, the authors found that theaverage abnormal return for investments in countries where property-rightsregimes were weak scored negatively (0.0554 percent). In other words,investments in environments with poorly defined property rights were greeted by anegative share-price response (English and Moore, 2002).

Analysts Analyzed: Wall Street and the 2002 Brazilian Presidential ElectionObviously, the monitoring of politics intensifies during crisis periods. This wasunderlined by the coverage of political issues during the Argentine meltdown atthe end of 2001—an economic collapse described by some financial operators asclear evidence of a crisis of political governance, representation and legitimacy(see, for example, J.P. Morgan Chase, 2001). Election years are critical juncturesfrom an investor’s point of view, and at such moments the density of finance-industry reports incorporating political dimensions jumps sharply.


From this perspective, the 2002 presidential elections in Brazil represent aparadigmatic case. The Brazilian episode also underlines that financial turbulenceis contingent—neither inevitable, nor preordained, but rather involving a countrythat has entered a risky zone, a speculative tunnel, where the government lacksthe political and economic capacity to curb market perceptions in order to fendoff financial pressures. As Eichengreen (2002a: 9) has stressed, at such a criticaljuncture:

if investor sentiment turns against the country, for whatever reason, thegovernment of a country with heavy financial burden or a weak economy maybe unable to sustain the harsh policies of austerity needed to deflect mountingpressures. If, on the other hand, market sentiment remains favorable, the sameeconomic and financial fundamentals will be sustainable, and no crisis willresult.

An appreciation of such market sentiment is therefore central to understandingthe dynamics of financial turbulence. In other words, it is necessary to understandhow Wall Street cognitive regimes shift regarding a country. The purpose of thissection is to examine such changing perceptions regarding Brazil during the 2002electoral contest.

It is important to stress at the outset that in 2002 economic fundamentals werealso worrying investors. More precisely, and in contrast to other emerging-marketcrises, investors were concerned less with exchange rates than with debt dynamics.In 1999, Brazil adopted a floating exchange-rate regime, and the recent round ofmarket turbulence has concerned fiscal sustainability rather than the maintenanceof a fixed exchange rate. During the period covered by the present article therewas a major shift in Brazil’s macroeconomic framework, with the pre-1999concerns of investors giving way to very different post-1999 preoccupations. It isalso important to appreciate that “Wall Street” does not form a uniform “epistemiccommunity”: the views of investors and analysts are multiple. For example, theconcerns of investors in equities and investors in debt (government bonds) areunlikely to be perfectly coincident. But in the post-1999 world, and particularlyduring the 2002 elections, it was the possibility of government default on its debt(and not exchange-rate instability) that loomed largest. To understand investorand analyst nervousness, it is also important to keep in mind that Brazil is by farthe largest Latin American economy, contributing 40 percent of regional output.It is also a major equity and bond emerging market for international investors.

At the beginning of 2002, the country’s sound economic track record was beingpraised by almost everybody from New York to London and from Washington toSao Paulo. Brokers’ reports stressed the substantial decoupling between theArgentinian and Brazilian economies, pointing to the weak trade links betweenthe two partners and low Brazilian banking exposure to Argentina. Investmentbank and broker reports lauded Brazilian economic and political achievements,with some analysts emphasizing Brazil’s future potential: “Global decouplingseems to gain ground,” wrote a bullish strategist at the time, “funds are flowing toemerging markets, but have not reached Brazil,” and “Brazil could now be the lastdiversification play” (Morgan Stanley, 2002a). Others stressed the decouplingbetween Brazil and its troubled neighbor Argentina: “the correlation betweendaily returns on Argentina and Brazil in the EMBI+ [Emerging Markets Bond Index Plus], which stood at very high levels between 1995 and 2000, dropped to0.8 in the summer of 2001 when deposits withdrawals accelerated in Argentina,


and slumped to 0.15 in November 2001 when Argentina’s default wasannounced.”10

This celebration of Brazilian success culminated in March 2002 during theAnnual Meeting of the Inter-American Development Bank (IADB) held atFortaleza, Brazil. Arminio Fraga, a major confidence-game player and expert, andappointed as central bank governor in 1999 after a brilliant career in Wall Street atSalomon Brothers and Soros Fund Management, was elected Man of the Year byLatin Finance magazine and dubbed “the man who saved Brazil.” Typical of themood with regard to Latin American emerging markets was the Merrill Lynchreport and summary conclusions of the IADB meetings. Like its competitors,Merrill Lynch organized an investor conference in Fortaleza, attended by portfoliomanagers, officers of the IMF, and US Treasury representatives. The firm’semerging-markets debt strategist concluded that:

The mood in Fortaleza was generally upbeat. In contrast to one year ago inSantiago de Chile, there was no major crisis in the making, such as it was thecase with Argentina back then. Indeed, one of the more remarkable market,economic, and political developments over the past year has been how littleregional contagion has been triggered by the Argentine crisis.... The principalfocus at the meetings was the Brazilian elections, and participants appeared tobe reassured that despite a volatile period ahead, a candidate from thegovernment coalition is still the most likely successor to President Cardoso.(Merrill Lynch, 2002b)

But by late April the currency had slumped, and the risk premium on Brazilianbonds soared, putting them on a par with Nigerian paper. Why such a sudden fallfrom grace? The key was that in October 2002 Brazil faced an importantpresidential election. By March, opinion polls started to appear and to be assessedby major Wall Street brokers. Investors abhor uncertainty, but the 2002presidential election presented Brazilian voters with an important choice.Constitutional provisions prevented President Fernando Henrique Cardoso, whosince 1995 had contributed so much to the modernization of the Braziliangovernment and economy, from competing for another term. There were threemajor contenders for the presidency: two leftist candidates, Luiz Inacio Lula daSilva and Ciro Gomes, and José Serra (Cardoso’s man).11 The first two candidateshad in the past spoken of the possibility of debt restructuring. But in the wake ofArgentina’s default on its public debt just a few months earlier, market sensitivityregarding the debt issue was high. Moreover, some analysts recalled that previouspresidential elections in Brazil had also produced costly devaluations that wreckedthe economy. Between April and July 2002, when investors started to worry thatone of the leftist candidates might actually win, anxiety was transmitted tofinancial variables. By July spread levels jumped above 2000 bps (see Figure 1),and the currency had sunk to more than 3.50 reals per dollar (from a level of 2.30reals at the beginning of the year).

By the beginning of May 2002 a few brokers started to publish cautious reportsdrawing attention to political factors. Later in the month nearly all the Wall Streetboutiques began to crunch numbers to assess the sustainability of Brazilian debtdynamics. Brazilian public debt actually represents a rather unusual case amongemerging markets, which amplifies its sensitivity to financial variables. In mid-2002, Brazil’s debt was 80 percent indexed to either domestic interest rates or tocurrency-exchange levels. Models were used to determine what primary fiscal


surplus would be required to stabilize the debt–GDP ratio, using stress-casescenarios for the major variables affecting debt dynamics (real interest rates, theexchange rate, GDP growth and net contingent liabilities).12

The shift toward a more cautious tone regarding Brazil was mainly triggered byopinion polls revealing strong support for Lula, the candidate of the leftist-leaningWorkers Party.13 As Walter Molano (2002a) of BCP Securities explained in abriefing report issued at the end of May entitled Da Lula Monster: “there seems tobe a sense of panic as economic agents realise that Lula will win the elections. Thebehaviour by the multinationals seems to be the mirror of what is happening inthe bond market. It is too bad that Brazil will not get the benefit of the doubt.”The impact of the more cautious stance on Brazil was exacerbated by the risk-adverse approach to emerging markets that arose because of doubts concerningthe US recovery. But in the case of Brazil, politics reinforced the risk aversion.Lula’s strength in the polls was perceived as a growing risk by investors andanalysts, because it presented a serious threat of discontinuity in economic policy.As a result of the increased political uncertainty, many of the investment banksbegan to shift their portfolio recommendations for Brazil from “overweight” to“neutral.” At this point, Wall Street analysts did not play down the vulnerabilitiesand risks ahead, but neither did they perceive a Brazilian crisis as imminent orinevitable. The rollover of domestic debt was seen as large but manageable, andthe capacity of the Brazilian government to meet its foreign obligations wasregarded as stronger than indicated by foreign obligation spreads. The majorconcern was rather the increasing impact on debt dynamics of prolonged investorrisk aversion regarding Brazil (CDC Ixis, 2002b). Thus at this point, a Braziliancrisis was not seen as the most likely scenario by Wall Street analysts, who were stillenvisioning a happy ending in October.

More extensive political assessments then started to be published by analysts.For example, the BBA economic research team in Brazil published a detailedanalysis of the Workers Party economic program (Banco BBA Creditanstalt, 2002).14


0

500

1000

1500

2000

2500

02/0

1/02

02/0

2/02

02/0

3/02

02/0

4/02

02/0

5/02

02/0

6/02

02/0

7/02

02/0

8/02

Brazil EMBI+

spread

EMBI+ spread

So urce: J.P. Morgan

FIGURE 1. Brazilian and Emerging Bond Spreads

J.P. Morgan Chase (2002a) issued a comprehensive guide to the Brazilian partiesand candidates. Poll results were dissected by all Wall Street firms, with somepointing to television as a crucial determinant of voting intentions. On 8 May2002, UBS Warburg (2002), for example, devoted a complete report to discussingthe impact of free television coverage on the presidential race that would officiallyget under way on 20 August.15 Other firms created specialized analytical devices tocapture the political dynamics. An interesting example was Goldman Sachs(2002a), which developed a “Lulameter”—a model that tried to quantify theprobability of a Lula victory, as priced in by the currency markets.

As markets became preoccupied with Brazil, some boutiques even facedinternal turbulence. On 3 May, for example, Santander Central Hispano’s NewYork research department downgraded Brazilian bonds to a “neutral” from a“heavy” weighting, citing the country’s sluggish economy and the political risksahead of the October presidential election. In Brazil, officials publicly disagreedwith the downgrade and the bank, which has substantial business interests in thecountry, moved toward firing the 12-member sovereign debt research team. In theend, worried about damaging its credibility on Wall Street, the bank decided notto fire its analysts. But the unit was instructed to stop sharing its recommendationswith the press. This episode points to the pressures to downgrade within thefinancial industry, related to the relation between brokers’ sales-side research andother businesses units of investment banks (Boni and Womack, 2002). Severalother investment banks, including Merrill Lynch, ABN-Amro and Morgan Stanley,downgraded Brazil at about the same time as Santander (see Table 1).

By the beginning of July 2002, J.P. Morgan Chase strategists implemented asecond reduction of the “overweight” recommendation for Brazil.16 A month later,the firm’s analysts started to ask how long Brazil could sustain spreads above 2000bps without defaulting. Based on a historical analysis, the quantitative team found11 other examples of such high spreads. The results, however, were inconclusive asseveral countries had sustained spreads at these levels for a period of time


TABLE 1. Wall Street Strategists’ Recommendations on Brazil’s Bond Debt, 2002

Rating change Date

ABN-Amro Neutral from overweight 1 MayGoldman Sachs Neutral from overweight 1 MaySantander Investments Neutral from overweight 3 MayDeutsche Bank Neutral from overweight 9 MayJ.P. Morgan Chase 1st reduction of overweight 4 June

2nd reduction of overweight 1 JulyOverweight to marketweight 22 JulyMoved to underweight 9 December

BCP Securities Sell 8 AugustMorgan Stanley Downgrade to underperform 12 AugustSalomon Smith Barney Changes in macroeconomic forecast 20 AugustUBS Warburg Increased overweight 30 AugustBear Stearns Cuts to underweight 19 SeptemberMerrill Lynch Moved to underweight 25 SeptemberGoldman Sachs Moved to underweight 27 SeptemberMerrill Lynch Moved back to marketweight 4 October

Source: Based on JBIC and Wall Street investment bank reports, 2002.

ultimately avoiding default. Of the 11 countries, 4 defaulted (Argentina, Ecuador,Russia and the Ivory Coast), 2 completed distressed debt exchanges (Pakistan andUkraine) and 5 others (Algeria, Bulgaria, Mexico, Nigeria and Venezuela) tradedhigher than 2000 bps without defaulting. Our calculations show that by the end ofOctober 2002 Brazil had traded above 2000 bps for 35 days (see Table 2).

By the beginning of August, even the more bullish analysts for Brazil started tourge caution. The data were suggesting a “sudden stop capital flow” with portfolioinvestors lowering their Brazilian weighting, sharp cuts in credit lines as financialinstitutions reduced their exposure, and accelerated capital flight. In July, therewere outflows of US$1.1 billion from non-resident CC-5 accounts, twice the total forthe previous month. In mid-July, Barclays Capital strategists were still stressing thedifferences between Argentina and Brazil, seeing Brazil as a moderately positiveplay (Barclays Capital, 2002a). Just a few weeks later, after the Brazilian spreadsovershot 2000 bps and the real slumped to 3.50 to the dollar, the same sourceargued that “the trigger for all crises is now underway,” while public domestic debtrestructuring and private exterior debt default probabilities during the next 3–6months were raised to 45 percent and 35 percent respectively (Barclays Capital,2002b). But by this time the elections were no longer the sole factor behind WallStreet perceptions of Brazilian risks.

By the beginning of August a respected Wall Street analyst, Walter Molano, tooka more radical position, simply advising his clients that “It is time to get out,” andadding:

The outcome in Brazil does not depend on who wins the elections.... Over thepast year, Brazil’s debt to GDP ratio exploded almost 20% as it went to 58% from49%. In 2003, more than 90% of the amortizing domestic debt will be indexedto interest rates or the dollar. Therefore, the incoming administration will need


TABLE 2. Countries that Traded Above 2000 bps

EMBIG countries which Days above traded above 2000 bps spreads 2000 bps and defaulted before default Default date

Argentina 38 December 2001Russia 40 October 1998Ivory Coast 2 March 2000Ecuador 163 September 1999

EMBIG countries which traded above 2000 bps Days above spreads and avoided default 2000 bps Period

Mexico 8 March 1995Venezuela 94 Spring 1995, August 1998Bulgaria 32 Summer 1994, Spring 1995Algeria 10 April 1999Nigeria 308 1994–95, 2000Pakistan 5 September 2000Ukraine 97 June 2002, Spring 2001Brazil 35 June 2002–October 2002

Source: Own estimations for Brazil and J.P. Morgan Chase, Emerging Markets Outlook, 2 August 2002.

to restore investor confidence quickly in order to stabilize the exchange rateand interest rates. However, investors always need time to become comfortablewith the new administration, even if it is Serra. It will be much worse if it is CiroGomes or Lula. Therefore, the odds are stacked against Brazil. (Molano,2002b)

On 7 August 2002, the confidence game was still going on; this time with a newplay from Brazil and the International Monetary Fund (IMF). The countryannounced an agreement with the IMF for a US$30 billion loan to restore investorconfidence in the country’s battered financial markets. In fact, markets werealready pricing in the announcement, rallying strongly in anticipation of the deal.Some leading economists had called for just such an accord (Edwards, 2002). Theexchange rate fell from its high (3.46 reals to the dollar at the end of July) to 3.01reals the day before the announcement, and the country’s risk premium, asmeasured by bond yields over US treasuries, plunged below 2000 bps, tightening bymore than 500 basis points in just a few days. The magnitude of the swing suggeststhe financial impact of the confidence game on volatile sentiments. Most analystsand fund managers welcomed the agreement. “This generous agreement ...combined with O’Neill’s trip to the region and final passage of trade promotionauthority this week, creates a potentially significant positive shock for LatinAmerica” (Bear Stearns, 2002b). Business could go on, the IMF play keeping Brazilalive as a “buy” opportunity. Confidence was further boosted when all the mainpresidential candidates pledged to honor the agreement. In one day, the Brazilianreal rose by more than 5 percent against the US dollar, breaking the US$3.00barrier for the first time since the end of July. Around the world companies withlarge exposures in Brazil and Latin America experienced dramatic increases intheir stock prices, and the Spanish equity market closed 5 percent higher.

However, the confidence game was also seen as a timing game; if the size of theloan (US$30 billion in total) was impressive (and larger than expected by themarket), some 80 percent of the funds were to be released in 2003, after the newadministration assumed office in January. The IMF vote of confidence was clearly acautious one. As a Financial Times columnist observed on the day following theannouncement, the IMF “seems to have bet on binding future governments intothe deal by back loading the disbursement of money heavily into next year—thusrewarding any future administration which agreed to stick to the fiscal targets.” Inthe end, the IMF agreement kept the game going: “The IMF agreement could be theway to give some assurance to the market that sensible policies should be in placein 2003 (maybe beyond), regardless of who would be in charge by that time” (BBAEconomic Research, 2002b: 2).

Walter Molano of BCP Securities was more sanguine, suggesting that while theIMF package represented a win–win solution for the US Treasury and StateDepartment regarding their Latin American policy, and for the Cardosoadministration (allowing the Brazilian president and his economic team “to finishtheir term in office without suffering the embarrassment of a default or aneconomic collapse”), in the end, the IMF’s new money was modest, and left theessential problems unresolved. “It is time,” concluded Molano, “to leave the party.Take the opportunity that there is a bid for Brazilian paper and look for the door”(BCP Securities, 2002a). Eichengreen suggested that geopolitics not economicshad driven the IMF decision, and argued that the “real danger” was no longerBrazil’s elections, “but the global economic climate in 2003.” Thus, “the crisis will


reemerge, and unless the IMF and the US government are prepared to provide yetanother mega-package, which is unlikely, Brazil will be forced to default andrestructure its debt” (Eichengreen, 2002b: 5).

During the month of August, several leading economists criticized the IMFbailout, deploring the absence of a clear strategy (Hausmann, 2002). Othersopenly declared that “options such as an orderly restructuring of Brazil’s nearlyUSD 250 billion debt must be considered with due speed, otherwise a default soonis likely” (Desai, 2002). “The IMF,” wrote a senior fellow at the Institute ofInternational Economics, “must put debt restructuring at the heart of itsconditions for financial assistance” (Goldstein, 2002). In fact, financial marketeuphoria at the IMF rescue package was short lived. After the initial rally, andwithin a few days, bond interest rates once again settled at levels incompatible withlong-term solvency. The risk premium on Brazil’s government bonds rose, whilethe real fell back once again to below 3.00 reals to the dollar. On 12 August,Morgan Stanley published a bearish report on Brazil, downgrading the country to“underperform,” stressing that even with the rescue package it would be hard forBrazil to enter a virtuous circle. This move was synchronized with a shift ofportfolio recommendations in nearly all South American credits to“underperform” (Morgan Stanley, 2002c). By the end of August another leadingWall Street investment bank, Salomon Smith Barney, altered its macroeconomicforecast for Brazil in 2003, anticipating reduced GDP growth, increased inflationand currency depreciation, with its base scenario predicated on an oppositionvictory in the October elections (Salomon Smith Barney, 2002a).

The Brazilian authorities continued their efforts to bolster confidence. At theend of August, Arminio Fraga met a group of market analysts in New York in anattempt to calm jitters (Salomon Smith Barney, 2002c). During the first weeks ofSeptember, while the polls increased concern of a Lula first-round victory, FinanceMinister Malan and Central Bank Governor Fraga hit the road once again toreassure investors in Madrid, London, Amsterdam, Basel and Frankfurt. ThenBrazil announced an upward adjustment in its primary surplus target for 2002,Serra rose in the polls, and trade and current-account data remained positive. Yetthe investment banks remained skeptical,17 and leading international bankscontinued to try to reduce their Brazilian exposure, despite the publiccommitment that had been made in New York in August. Citigroup, in particular,while maintaining a pledge to local officials to keep trade lines open, continued toreduce lending exposure as reported by Bloomberg (Citigroup reported US$9.3billion in outstanding cross-border claims for Brazil on 30 June compared toUS$11.4 billion on 31 March). By September the collapse of confidence becamegeneralized in the cognitive regime of Wall Street operators. Goldman Sachs, whoin mid-September 2002 still believed that analysts had prematurely declared Brazilto be on an unsustainable debt path, had by the end of the month moved to an“underweight” stance. They anticipated that to deal with the confidence shock,“whoever wins will have to appoint a strong economic team, deepen fiscaladjustment, and implement structural reforms to reduce the public debt ratio gapand restore growth” (Goldman Sachs, 2002d).

Polls were confirming financial market fears of a widening gap between Lulaand Serra, the former moving above 40 percent of declared voting preferences,while support for the latter remained unchanged. Arminio Fraga and IlánGoldfajn (the deputy governor of the central bank for economic policy) tried tocalm investor concerns about Brazil’s debt levels with an article in a leading


confidence-game arena, the Financial Times (Fraga and Goldfajn, 2002). ButBrazil’s appeal to financial operators fell on deaf ears. Within a few days Brazilianspreads again crossed the 2000 bps frontier, while the currency depreciated to arecord low. By the end of September, with the real tumbling and risk premiumsrising, another leading Wall Street boutique, Merrill Lynch’s emerging marketsdebt research team, joined Goldman Sachs in shifting to “underweight”recommendations (Merrill Lynch, 2002d).18

With the annual IMF–World Bank meetings approaching, new proposals onmanaging debt crises proliferated, such as those formulated by Guillermo Ortiz(governor of the central bank of Mexico) and Josef Ackerman (chairman ofDeutsche Bank) (Ackerman, 2002; Ortiz, 2002). But after the meetings, the loss ofconfidence was even more pronounced. Walter Molano (BCP Securities, 2002b)was particularly bearish, describing the IMF–World Bank “Monster’s Ball” annualmeetings in these terms: “a gallow’s humor [sic] permeated the Fall Meetings, asinvestors, government officials and future ex-Ministers gathered for what seemedto be the last hurrah of the emerging markets.” Meanwhile, “the fair weatherfriends on the sell-side [that is, Wall Street analysts] abandoned Brazil at itsdarkest moment” (BCP Securities, 2002b). At Bear Stearns, another leading WallStreet firm, the return from Washington was also gloomy. The sovereign debtrestructuring mechanism (SDRM) absorbed most of their attention, and there wasspeculation that Brazil’s next president might be tempted to follow this route: “Butwhat about Lula? Lula might be a policy maker who would use this policyinstrument (SDRM). Under this scenario, investors might be even more nervousthan they already are” (Bear Stearns, 2002c).

By the beginning of October 2002, the political coverage of the Brazilianelections among Wall Street boutiques had intensified. Pimco, manager of thebiggest emerging-market bond fund (US$7 billion of developing-nation bonds invarious funds), assured investors that Brazil would never default on its debt.“Brazil has the willingness and ability to make its debt payments,” said El-Erian(Pimco’s fund manager) to Bloomberg. “Those who are betting on a Brazil defaultare likely to lose. We’re bullish on Brazil because it is a good buy at these levels.”On 4 October 2002, Pimco managers declared that they were adding moreBrazilian bonds to the roughly US$1 billion of the country’s dollar bonds they heldas of 30 June, in what looked to be a desperate move to find some kind of buyers.Pimco’s confidence that Latin America’s biggest country would not default onUS$300 billion in public debt stood in stark contrast to the views of other Brazilianbond investors, as shown by an investor poll conducted by J.P. Morgan at the endof September. In fact, as Bloomberg observed, Pimco bought Brazilian bonds inthe first 3 months of the year, just before the nation’s currency and bonds startedto slide as polls showed Lula widening his lead.

On 27 October, after 13 years and four attempts, Luís Inácio Lula da Silvabecame Brazil’s 36th president. Market sentiment had already started to improveafter 15 October, when bond spreads began compressing 500 bps. By mid-November 2002, the Brazilian real had rallied almost 10 percent from mid-October, and the Brazilian EMBI had tightened 30 percent from the end ofSeptember. A “market friendly” economic team (in the sense of a team committedto fiscal and monetary stability) was by then perceived as helping bridge the gapbetween Lula and the markets. Lula and the financial markets were living under a(brief) “Lula de Mel” (Merrill Lynch, 2002f). But as reports published at the timesuggested, the romance might prove short lived, since most Wall Street boutiques


have maintained their cautious stance, pointing to fiscal weakness, soaringdomestic debt, high interest rates, poor growth and rising inflation (BarclaysCapital, 2002c; Lehman Brothers, 2002; J.P. Morgan Chase, 2002c). On 9December 2002, J.P. Morgan also moved its recommendation on Brazil to“underweight” (J.P. Morgan Chase, 2002d). The confidence game would go on.

Brazilian Elections and Financial Markets: A Case Study in HistoricalPerspective

The analysis of Wall Street views on Brazil during the months preceding thepresidential election illustrates that financial market opinion is dynamic, and evenvolatile. The differences of opinion, while important and reflecting diversitywithin the industry, narrowed as the elections approached. Uncertainty grew withLula’s rise in the polls after April 2002, and the emergence of the “Third Man”(Ciro Gomes) in the polls during July. Brazil provides a perfect example in LatinAmerica of the influence of politics on the markets (Frieden and Stein, 2001). Yetpolitical uncertainty has always had a considerable influence on financial variablesin Brazil.

This section of our argument focuses on the years of the past three Brazilianpresidential elections: 1994, 1998 and 2002. We have chosen these years for tworeasons. First (and this is true of any country), elections by definition always opena period of political uncertainty, as the winner remains to be determined. Thisuncertainty is particularly salient in emerging markets. The second point isspecific to Brazil. Over the past three electoral cycles there have been paralleldevelopments in the voting-intention polls, which have suggested that an anti-establishment candidate (Lula) might break the political monopoly of the rulingcoalition (represented by Cardoso in 1994 and 1998 and by Serra in 2002).

We use the percentage of the intended votes for Lula in the electoral polls as anindicator of political uncertainty. Comparing the three cycles reveals that theseelectoral years are remarkably similar in terms of political uncertainty and timing.In 1994, 1998 and 2002, Lula started his rise in the polls in April, reached amaximum in June, then slipped back until the elections in October. The onlydifference between the years is that in 1994 and 2002 Lula held the lead for quitesome time, while in 1998 he was not able to achieve this (in the June polls he wasvirtually tied with Cardoso). Nevertheless, the fact that in 1998 Lula retainedsignificant support gave him a real possibility of winning, and thus politicaluncertainty was not so different to the other two electoral years.

To confirm whether political uncertainty has an impact on financial variables inBrazil, we analyzed the behavior of the domestic short-term interest rate, for whichwe used the SELIC rate (the real–dollar exchange rate), and the long-term interestrates of the external public debt (represented by the Brazilian bond spread). First,we will look at the domestic short-term interest rate and the real–dollar exchangerate. In Figures 2 and 3, one can discern the high volatility of the cited variables inthe electoral years.

Of course, these two variables are interrelated and are influenced by foreignshocks, as could be the case for the Asian crisis in 1997 or the Russian crisis in1998. Likewise, the domestic interest rate and the exchange rate are seriouslyaffected by economic policy decisions, such as the implementation of the RealPlan in July 1994 (an anchor for the exchange rate implemented to controlinflation) and the devaluation of the real in January 1999. As noted at the


beginning of this article, such decisions are affected by the timing of the politicalcycle, and this makes assessment of the independent impact of electoraluncertainty on these two financial variables difficult. So the profile of theexchange rate during the past three electoral years presents two abruptmovements associated with the beginning and the end of the Real Plan in July1994 and in January 1999, respectively. Logically, during the application of theReal Plan (which included the electoral year of 1998) the exchange rate wasallowed to fluctuate only within a narrow band. What can be recognized is that inthe first half of the other two electoral years (1994 and 2002) there was a gradualdepreciation of the Brazilian real parallel to an increase in electoral uncertaintyassociated with the rise of Lula in the opinion polls.


-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

1993 1995 1998

Electoral

year

Daily change in %, monthly moving average

Real Plan End Real Plan

Source: Bloomberg

Electoral

year

Electoral

year

1994 1996 1997 1999 2000 2001 2002

FIGURE 2. Real–Dollar Volatility

-150

-100

-50

0

50

100

150

jan-9

3

jan-9

4

jan-9

5

jan-9

6

jan-9

7

jan-9

8

jan-9

9

jan-0

0

jan-0

1

jan-0

2

Electoral

year

% m/m

Source: Bloomberg

Electoral

year

Electoral

year

FIGURE 3. SELIC Volatility

With respect to domestic short-term interest rates, reflected here by the overnightrate of the Central Bank of Brazil (SELIC), one can also observe peaks in theelectoral years of 1994 and 1998. Of course, part of the interest-rate volatility wasdue to external shocks (such as the Russian crisis in August 1998), and it isdifficult to ascertain the precise extent to which national electoral uncertaintyaggravated financial volatility. The increased stability of the SELIC rate in 2002 isrelated to the current Brazilian exchange-rate regime. When the exchange-ratesystem was fixed or semi-fixed (as it was in Brazil up until the beginning of 1999),the adjustment to any internal or external shock came on the side of interest ratesand real variables in the economy. In the current situation, the exchange rate isresponsible for absorbing internal and external shocks, reducing the impact oninterest rates and on the level of economic activity. This means that a strongincrease in foreign-exchange volatility is consistent with the stability of the SELICthat was observed in 2002. Furthermore, irrespective of the foreign-exchangeregime, in the first half of each electoral year there was an increase in real interestrates (see Figure 4), which resulted from the higher profitability demanded byinvestors to offset the increased risk premium associated with the possibility of anopposition victory in the presidential elections.

In summary, during the electoral years 1994, 1998 and 2002 the exchange ratewas more volatile and real interest rates increased. Nonetheless, this is aconsequence not only of the uncertainty associated with the evolution of theelectoral polls, but also of the fact that the exchange rate and interest rate areinterdependent and are conditioned by broader economic policy decisions (suchas the Real Plan), which are in turn also influenced by the electoral calendar andthe volatility associated with political uncertainty. In addition, of course, thefinancial volatility of the past three electoral years was in part provoked by externalfactors such as the crises that occurred in other emerging markets.

In an attempt to get a clearer picture of the effects of political uncertainty onfinancial markets, we now turn to long-term external interest rates. Here we used


0

10

20

30

40

50

60

01/0

1/92

01/0

5/92

01/0

9/92

01/0

1/93

01/0

5/93

01/0

9/93

01/0

1/94

01/0

5/94

01/0

9/94

01/0

1/95

01/0

5/95

01/0

9/95

01/0

1/96

01/0

5/96

01/0

9/96

01/0

1/97

05/0

5/97

09/0

9/97

01/0

1/98

05/0

5/98

09/0

9/98

01/0

1/99

05/0

5/99

09/0

9/99

01/0

1/00

05/0

5/00

09/0

9/00

01/0

1/01

05/0

5/01

09/0

9/01

Jan/0

2

05/0

5/02

Source: Banco central do Brazil, IBGE

Pre-stabilization Real Plan Inflation targeting

(floating exchange rate)

Electoral related

uncertainties

FIGURE 4. Real Interest Rates, 1992–2002

the spread of Brazilian debt over US bonds in the EMBI (in 1994 and 1998) andEMBI+ (in 2002) indexes of emerging bonds of J.P. Morgan.

Unlike the domestic short-term interest rates, the spread of Brazilian bonds isnot influenced by the exchange-rate system because the bonds are denominatedin a foreign currency and so provide the best approach to “country risk.”Nevertheless, the Brazilian debt spread over US bonds can be increased by externalshocks, such as a rise in the risk aversion of international investors or a financialcrisis in another emerging country that reduces the general attractiveness ofemerging-country debt. To isolate the impact of such external shocks we used thespread between Brazilian bonds and all emerging bonds integrated in the EMBI+index. An increase of the spread between Brazilian debt and the whole group ofemerging countries cannot be attributed to higher risk aversion by foreigninvestors, nor to the “herd effect” associated with a crisis in another country, giventhat in these two cases the deterioration would be similar for all these emergingbonds. So the cause of any hypothetical increase in the spread between Brazilianand the basket of emerging-country bonds would have to be domestic. In this case,the political variables are the prime suspect, given that they influence not only the“ability to pay” the debt (as an increased premium is demanded at moments ofuncertainty), but also the “willingness to pay” (associated with the electoral victoryof an opposition candidate).

Thus, to find a more effective indicator that shows the influence of Brazilianpresidential elections on financial markets, we studied the correlation betweenpolitical uncertainty, represented by the popularity of Lula shown in electoralpolls, and sovereign risk, represented by the evolution of the spread of Brazilianexternal bonds over the whole emerging-country group within the EMBI (in 1994and 1998) and EMBI+ (in 2002) indexes.19

Comparing the graphs corresponding to each of the past three electoral years,a certain “seasonality” can be observed with regard to the movement of the spreadbetween Brazil and the ensemble of emerging bonds (see Figures 5–7). UntilMarch in each year, the relative spread narrows slightly. This coincides with eitherthe absence of polling data (in 1994) or with a slight decline in Lula’s support (in1998 and 2002). From March onward, a relative deterioration of the Brazilian debtposition occurs in each of the three years. This can be related in 1994 to theappearance of the first polls, which revealed substantial support for Lula at 35percent, or in 1998 and 2002 to the fact that he had acquired majority support.The relative deterioration of the Brazilian debt position peaked in May and Juneof 1994 and 1998, when Lula reached his maximum popularity in the polls. Fromthen on, he lost support: steadily in 1994, and with ups and downs in Septemberand October 1998. Correspondingly, the Brazilian debt spread improved.Although emerging-debt volatility was increased by the Russian crisis in August1998, this did not alter the fact that in the months prior to the presidentialelections in October, the relative spread in Brazil was still being influenced by theopinion polls.

In 2002, the opinion polls registered an evolution parallel to those in 1994 and1998, with a progressive decline of Lula’s popularity from May to June.Nevertheless, from that point on, there is a decoupling between our electoraluncertainty index and the relative spread in Brazil. There are two reasons thatexplain this occurrence, and in each case, the political factor plays a revealingrole. The first element is the emergence of a new challenger in 2002 (CiroGomes) who had a more heterodox program than Lula. This is a situation that did


not obtain in the previous electoral cycles of 1994 and 1998, and helps explain thecontinuing deterioration of Brazilian sovereign risk (see Figure 8).

However, the further deterioration of the Brazilian debt position in theMay–July period is associated with other factors, such as the inherent dynamism ofthe Brazilian debt. Although most of the Brazilian debt is domestic (and thereforedenominated in Brazilian reals), its peculiar composition makes it especiallysensitive to financial variables, and so to the volatility of electoral years. About 50percent of the public domestic debt is interest-rate indexed, while 30 percent isexchange-rate indexed. This composition is a legacy of Brazilian history, ratherthan a result of policy decisions by recent administrations. If Brazil could borrow


20%

25%

30%

35%

40%

45%

03/0

1/94

28/0

1/94

24/0

2/94

22/0

3/94

18/0

4/94

13/0

5/94

09/0

6/94

06/0

7/94

01/0

8/94

25/0

8/94

21/0

9/94

18/1

0/94

14/1

1/94

09/1

2/94

-180

-80

20

120

220

320

420

520

bp.

Voter intention

for Lula in the

opinion polls

FIGURE 5. Spread Brazil-Emerging Countries and Electoral Polls, 1994

20%

22%

24%

26%

28%

30%

32%

02/0

1/98

26/0

1/98

17/0

2/98

10/0

3/98

31/0

3/98

22/0

4/98

13/0

5/98

04/0

6/98

25/0

6/98

17/0

7/98

07/0

8/98

28/0

8/98

21/0

9/98

13/1

0/98

03/1

1/98

25/1

1/98

17/1

2/98

-100

-50

0

50

100

150

200

bp.

Voter intention

for Lula in the

opinion polls


at fixed rates with long-term maturity dates, it would do so. As the graph shows(Figure 9), the Brazilian government was only able to borrow at fixed rates whenthe Brazilian exchange-rate system was fixed or semi-fixed. With a floating real,Brazil has attempted to increase the proportion of fixed-rate debt, but the marketshave not cooperated, limiting the weight of fixed-rate debt to total debt to 10–15percent. Moreover, the small proportion of domestic debt that Brazil has beenable to acquire at fixed rates over the past few years has been with short-termmaturation dates.


20%

25%

30%

35%

40%

45%

02/0

1/02

18/0

1/02

05/0

2/02

21/0

2/02

11/0

3/02

27/0

3/02

15/0

4/02

02/0

5/02

20/0

5/02

10/0

6/02

27/0

6/02

15/0

7/02

01/0

8/02

19/0

8/02

05/0

9/02

0

200

400

600

800

1000

1200

1400

Voter intention

for Lula in the

opinion polls

bp.

Source: Datafolha, J.P. Morgan


20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

02/0

1/02

18/0

1/02

05/0

2/02

21/0

2/02

11/0

3/02

27/0

3/02

15/0

4/02

02/0

5/02

20/0

5/02

10/0

6/02

27/0

6/02

15/0

7/02

01/0

8/02

19/0

8/02

05/0

9/02

0

200

400

600

800

1000

1200

1400

Voter intention

for Lula and Ciro

in the opinion

polls

bp.

Source: Datafolha, J.P. Morgan

FIGURE 8. Spread Brazil-Emerging Countries and Electoral Polls (Lula and Ciro), 2002

So, the negative impact that electoral uncertainty in Brazil has on financialvariables also affects the country’s risk in an indirect way, through an increase inthe domestic debt–GDP ratio, given the increased sensitivity of the domesticnational debt in Brazil to the variations of the interest and exchange rates.

The increase in Brazilian spreads during 2002 can be related not only to thefear of a change in the “willingness” to pay back domestic debt associated with thegrowing possibility of a victory by one of the opposition candidates (Lula or CiroGomes), but also to the deterioration of the “ability to pay,” associated with theinfluence of financial variables on the increment in the domestic debt–GDP ratio.One can appreciate that behind both explanations lie political factors rather thanany underlying change in the fundamentals of the Brazilian economy.

Given this explanation for the apparent disassociation between the popularityof Lula and Brazil’s emerging-debt spread in 2002, we will now proceed to quantifythe influence of political uncertainty on the Brazilian spread. To do this, wecalculated the correlations for the electoral periods of 1994, 1998 and 2002(defined in each case as the time period between the appearance of the first andlast electoral poll) between the Lula popular vote in each of the polls and themean of the relative Brazilian spread over the group of emerging countries. Theresult is illustrated in Table 3. In each of the election years, one can observe a highand positive correlation between electoral uncertainty and the risk premiuminvestors demanded for Brazilian debt. The electoral year has been subdividedinto two periods in order to exclude the months of June, July and August, when adisassociation between Lula’s popularity and the Brazilian spread occurred (whichis explained by political factors mentioned previously). If we were to adjust ourmeasure of political uncertainty by adding together the intended votes for Lulaand Ciro Gomes, the correlation with the relative spread of Brazil is very high bothin the period between January and May (0.97) and also for the whole period (0.88).

Aware of the risks associated with election years, the Brazilian authorities builtup a very accommodating amortization profile for domestic public debt around


0

20

40

60

80

1001994

1995

1996

1997

1998

1999

2000

2001

2002

Fixed rate

Interest rate indexed

Exchange rate indexed

Others

Source: Banco Central do Brazil

FIGURE 9. Brazilian Domestic Public Debt Composition

the holding of the elections in October and in subsequent months (see Figure10). This “valley of transition” aspired to limit the impact of the elections onBrazilian debt.

But the progressive increase in political uncertainty associated with Lula’s stablespot in first place in the opinion polls and the large increase in the popularity ofCiro Gomes completely distorted the amortization profile designed by theBrazilian authorities at the beginning of the year. The beautiful “valley oftransition” created by Brazilian leaders was transformed under market pressureinto a more abrupt “geography” with significant peaks (see Figure 11). The marketdid not accept bonds with maturity dates longer than the term of the Cardosogovernment, because it assumed that there was a high possibility that the newgovernment would have a lower “willingness to pay.” While Brazilian presidentialelections are held in October, the government changeover does not take placeuntil December; thus, in the months before the election, the government wasbasically forced to emit domestic debt with amortization dates prior to the end of2002.


TABLE 3. Cross-Correlation Between the Brazil-Emerging Debt Spread and Popularity Vote for Lula

1994April–October 0.92

1998March–September 0.68

2002 (Lula)January–August 0.54January–May 0.98

2002 (Lula and Ciro)January–August 0.88January–May 0.97

Significance level: 0.50 (1994); 0.63 (1998); 0.60 (2002)

0

5000

10000

15000

20000

25000

30000

35000

01/0

2/0

2

01/0

5/0

2

01/0

8/0

2

01/1

1/0

2

01/0

2/0

3

01/0

5/0

3

01/0

8/0

3

01/1

1/0

3

million reals

Source: Banco Central do Brasil

FIGURE 10. Amortization Profile of Brazilian Domestic Public Debt, January 2002

ConclusionAs the number of crises in emerging markets grows, so do the books and articlespublished on the subject (Dooley and Frankel, 2002; Edwards and Frankel, 2002;Feldstein, 2002; Eichengreen, 2003). However, few studies have been devoted tothe interaction between financial markets and politics. This research has usedempirical evidence to examine these links. The specific case study of Brazil,analyzed through the perceptions of Wall Street analysts and a historical andquantitative economic perspective, has shown that these ties are strong foremerging markets (Chang, 2002). In fact, the essential character of emergingmarkets lies precisely in this intricate link between political uncertainty andfinancial volatility—what could be called the economic fog of democraticuncertainty. In developed countries, political outcomes such as presidentialelections are not perceived, from a financial market point of view, as such criticaljunctures as in developing countries. As this article has argued, electoral outcomesmatter in emerging markets to a much greater degree than they do inindustrialized countries. Further research would be needed in order tounderstand precisely why elections in emerging economies assume this level ofimportance. One possible answer may be provided by recent economics literaturethat emphasizes that the existence of strong institutions dominates other factors inexplaining why countries are economically successful. Indeed, some analystswould argue that democratic forms of government are more likely to deliver stableinstitutions over time than autocratic governments (Olson, 2000; Santiso, 2002).

The question of instability and uncertainty is central to understanding theintricate links between financial markets and politics in emerging economies.More particularly, financial instability and financial market overreactions duringelectoral processes can be explained by uncertainty regarding institutional stabilityand continuity, rather than by uncertainty linked to democracy itself. In otherwords, instability in emerging markets derives not from the democratic process


0

5000

10000

15000

20000

25000

30000

35000

40000

01/0

2/0

2

01/0

5/0

2

01/0

8/0

2

01/1

1/0

2

01/0

2/0

3

01/0

5/0

3

01/0

8/0

3

01/1

1/0

3

million reals

Source: Banco Central do Brasil

FIGURE 11. Amortization Profile of Domestic Public Debt, June 2002

itself, but from the background of institutional weakness in emergingdemocracies. Where institutions are weak, and consequently the government’sability to honor its promises over time is called into question, the identity of thewinning candidate becomes much more important to investors than in countrieswith strong and stable institutions. To put the point another way, in countries withstrong institutions, the government’s ability to honor its commitments is notprimarily dependent on the identity of office-holders. Whether Bush or Gore hadwon in 2000, the US government would have met its obligations to holders oftreasury bills. In Brazil, however, the outcome of the election did (or wasperceived to) make a difference. Some preliminary findings related toinstitutional weakness suggest such a link. Recent research has underlined thatlack of transparency and institutional weakness affect international portfolioinvestments. Not only do international funds invest less in less transparentcountries, but also herd behavior among funds (mimetic volatility) tends to bemore prevalent in less transparent countries (Gelos and Wei, 2002).

At a more fundamental level, examination of market reactions to the 2002presidential elections in Brazil encourages broader reflection on the relationshipbetween democracy and the financial markets. As Georges Soros, a leading “voice”in global finance and Chairman of Soros Fund Management, pointed out:“admittedly, Brazil is going to elect a president who the financial markets do notlike; but if international financial markets take precedence over the democraticprocess, there is something wrong with the system” (Soros, 2002: 13). In aninteresting Wall Street report another analyst made a similar point: “we believethat investors can’t rightfully ask presidential contenders to state everything theyintend to do when and if elected. Trying to figure out what to believe and what notto believe from any presidential candidate is extremely difficult and sometimesunfair to them. Our opinion is that investors should wait for the political dust to settle on what the next elected president intends to do” (Morgan Stanley,2002d: 3). Kenneth Maxwell from the Council on Foreign Relations put it evenmore bluntly: “Wall Street analysts and IMF bureaucrats should leave Brazilianpolitics to its 115m voters and stop confusing fact and fantasy” (Maxwell, 2002).

The problem lies precisely in the fact that financial markets are forwardlooking, and anticipation, and what Albert Hirschman once called “the politicaleconomy of impatience”, are central. Thus financial and political temporalitiescan conflict. With presidential elections coming up, all temporal horizons becomefocused on the very short term. Financial anxiety resonates with political elections,which are by definition dominated by uncertainty. In the Brazilian case, theuncertainty related to the growing probability of a Lula victory—that is to say, onthe increased likelihood of an anti-establishment vote. At a more fundamentallevel, this raises the question of financial market tolerance for electoraluncertainty in emerging democracies. In the end, if we accept Przeworski’sdefinition of democracy (that is, a regime whose essence is the institutionalizationof uncertainty locked in electoral outcomes), we are led to the question of thepolitical preferences of financial markets (Przeworski, 1991).

Further empirical research is needed to explore the intricate links betweenfinancial markets and politics. It would, for example, be interesting to establishwhether there is a “democratic premium” in financial markets. In other words, arefinancial markets “politically correct?” Do markets give a positive premium todemocracy or, on the contrary, are democratic elections in emerging countriessystematically correlated with financial volatility, rising risk premiums and


exchange-rate ups and downs? The analysis of Brazilian presidential electionspresented here invites comparative empirical research on a wider sample ofcountries (for example, the 21 emerging countries composing the J.P. MorganEMBI+ index). Regarding creditworthiness, recent research suggests that there isno “democratic advantage,” because although democracies are supposed to paylower interest rates than authoritarian regimes (given that they are more capableof making credible commitments to repay debts), the evidence shows that thisexpectation is unfounded. Historical research, using a large sample of data onsovereign loans for 132 countries during the period 1970–90, suggests not onlythat dictatorships were less likely to reschedule their debts, but also that the majorsource of better borrowing conditions for emerging democracies was due to thebehavior of multilateral agencies that tend to bail out democracies rather thanenhance the capacity of these emerging democracies to make crediblecommitments (Saiegh, 2001). A closer view of emerging-market fund managers’political preferences and the democratic premium would be interesting in orderto corroborate or, on the contrary, disprove the idea that democracy and marketsalways go hand in hand. In addition, a comparative historical perspective mightshed useful light on the intricate links between emerging democracies (expansionof the franchise) and financial markets during the 19th century.

Further research would also be needed to put names and faces to thisconfidence game. After all, we are not just talking about abstract figures, but alsoabout men and women acting and interacting in this so-called confidence game offinancial markets. The Brazilian case underlines how trust (and mistrust) isembodied by individuals. Pedro Malan, the Brazilian Finance Minister, andArminio Fraga, Brazil’s central bank governor, were both key players—favoritesons of the financial markets and the IMF. Wall Street, the US Treasury andWashington-based multilaterals liked dealing with them, an asset denied to Lula orGomes. Both Malan and Fraga are comfortable in the high-tech world of financialmarkets and international policy-making. They speak the same language oforthodox economics, both are US educated, and both moved in the same circles asIMF officials and Wall Street players. Arminio Fraga was a former fund manager forSoros, while Pedro Malan has been an executive director of the World Bank. Asthe stimulating research done by Barro and Lee (2002) suggests, it would beinteresting to explore the intricate links between finance and politics byexamining the origins of ministry of finance and central bank officials inemerging economies. Most likely this would confirm the intuition that over thepast decade the “right” background (a US education with experience and contactsin the financial circles of New York and Washington) has become a necessary (butnot sufficient) precondition to playing the confidence game successfully.

The presidential candidate from the ruling Brazilian coalition had such apedigree; according to the perceptions of the financial markets, Lula and Gomesdid not. Lula had no Ph.D. or IMF pedigree. Instead, he rose from shoeshine boyto mechanic, and then to leader of the Sao Paulo autoworkers’ union, fightingagainst the military regime and organizing strikes during the 1970s. Lula’s victoryrepresents the (exceptional) victory of democracy and social mobility in a countryknown for its social inequality. It is also a large victory in the world’s fourth biggestdemocracy. Lula won by an ample margin: 53 million to 33 million votes. (Indeed,his triumph is the largest since Reagan’s 1984 total of 54.5 million votes, thelargest-ever vote for a US presidential candidate.) However, in a globalized(financial) world, winning the political vote in your home country is not enough.


For the rest of his mandate, Lula’s challenge will be to restore and maintainfinancial confidence. A few days before the second-round elections, one fundmanager quoted in the Financial Times stated that “to save Brazil” (that is, to winWall Street confidence) Lula would have to implement sound economic policies,make careful appointments to top economic positions, and collaborate closelywith international financial institutions (El-Erian, 2002b).

The consequence of Wall Street’s aversion is that leaders who do not share thisapproved pedigree must find a way to win over the financial markets. Presidentialraces are not only national races, involving citizens, the media, candidates and soon. Presidential candidates also have to win the confidence game, that is, the voteof approval of financial markets. On 5 August 2002, for the first time in 113 years,a candidate for the Brazilian presidency, Lula, visited the Sao Paulo StockExchange, to break some taboos after choosing its former chairman as his vice-presidential running mate. On 19 August, he was back again to seek the approvalof financial operators who were becoming more and more nervous. Lula andGomes met Cardoso in a symbolic move to agree on basic economics and to boostconfidence. Once elected, Lula intensified the “market-friendly” signals directedtoward Wall Street in a continuing effort to win its confidence, naming a formerinternational banker as governor of the central bank and a respected, pragmatictechnician as finance minister.

The political economy of financial markets explains much of the dynamic ofinternational finance. The ups and downs of financial variables can be explainedby economic fundamentals. But part of the story behind emerging-marketfinancial crises is of self-fulfilling prophecies, risk seeking and risk aversion, andchanging perceptions. As we have tried to emphasize, there are complexinteractions between politics and finance, between individual and institutionalinteractions, and macroeconomic and financial variables. In the end, the name ofthe game in emerging markets is confidence, trust and mistrust.

Notes1. On the dynamics of capital flows and crises, see also Eichengreen (2003). However,

there is a fundamental dilemma as the same institutions that can provide a crediblecommitment to stable policies may impede adjustment to external shocks (Rodrik,2002). For a detailed analysis of institutional and political checks and balances andpolicy volatility, see Henisz (2002). On the links between political commitments, checksand balances, and their effects on private investment across countries, see Stasavage(2002: 41–63).

2. According to Bloomberg’s survey on emerging-market bond underwriting, in 2002 theranking was as follows: (1) Salomon Smith Barney (US$10.0 billion of bonds placed foremerging-market borrowers during the year); (2) J.P. Morgan (US$8.6 billion); (3)Morgan Stanley (US$5.7 billion); (4) CSFB (US$5.0 billion); (5) UBS Warburg (US$3.1billion); and (6) Deutsche Bank (US$3.0 billion).

3. After Krugman’s (1979: 311–25) seminal work and the first-generation models, a secondgeneration of financial crisis models have been developed following Maurice Obstfeld’s(1994: 189–213) key work.

4. A careful study of Latin American exchange-rate regimes from 1960 to 1994 shows thatthe probability of the maintenance of a fixed exchange-rate regime increases by about 8percent as an election approaches, while the aftershock of an election conversely increasesthe probability of getting out of the peg by 4 percent. Economic factors such as the sizeof the tradable sectors exposed to international competition also help to explain thepropensity to maintain or leave a fixed exchange-rate regime (Bloomberg et al., 2001).


5. Economic history provides plenty of examples of political events framing economicuncertainty and stock-market volatility. Consider Argentina’s peso exchange rate duringthe 19th century (Irigoin, 2000; Irigoin and Salazar, 2000).

6. See the empirical evidence based on a portfolio-choice model that relates capital flightto rate of return differentials, risk aversion and the three types of risk mentioned for apanel of 47 developing countries over 16 years (Vu Le and Zak, 2001).

7. For similar results, establishing that share prices in a sample of 33 developed anddeveloping countries react negatively to uncertainty in the outcome of elections, seePantzalis et al. (2000: 1575–604).

8. There is a large literature drawing on several traditions that deals with public finance,government expenditure and the size of government, both for developed anddeveloping countries. Among the many works looking at the systematic effects ofelectoral rules and political regimes on the size and composition of governmentspending, consider Persson (2001), Drazen (2000) and Persson and Tabellini (2000).

9. In October 2001, Lehman Brothers’ fixed-income research department and EurasiaGroup’s political science experts launched a joint venture in order to cover politicalrisks in emerging markets, using analytical tools such as the Lehman Eurasia GroupStability Index (LEGSI) which incorporates political factors (65 percent of a country’sweight) and economic factors (35 percent). See http://www.lehmaneurasiagroup.com/.

10. For a balanced assessment of Brazilian economic weaknesses and risks, see CréditAgricole Indozuez (2002).

11. For further information, see their respective websites: http://www.lula.org.br/ andhttp://www.cirogomes.com.br/.

12. See, for example, Deutsche Bank (2002) and Goldman Sachs (2002b). Some brokersdid such analyses before, but the series of simulations realized were mainly to stress theBrazilian capacity to manage turbulence, as underlined, for example, in a reportstressing that “there is ... very ample room for the real to weaken in 2002 beforedynamics could become unstable” (Merrill Lynch, 2002a).

13. See the Workers Party’s website for more information: http://www.pt.org.br/.14. For a more detailed analysis of political data and polls, see also BBA Economic Research

(2002a). In addition, see the reports from Bear Stearns (2002a) and, for an extensiveanalysis of the Brazilian presidential candidates’ economic platforms, Merrill Lynch(2002c). On the specific candidacy of Ciro Gomes, see Salomon Smith Barney (2002b).

15. See also Citigroup and Salomon Smith Barney (2002) and Morgan Stanley (2002b).16. For an extensive analysis of the situation at this time, see J.P. Morgan Chase (2002b).17. See, for example, Morgan Stanley (2002e). Others, however, were more bullish on the

near-term outlook for Brazil, stressing improvements (a growing trade surplus andlower current-account deficits), but noting that “an upside for Brazilian financial assetslargely depends on the new president appointing a strong economic team andimplementing economic policies to reduce the debt to GDP ratio” (Goldman Sachs,2002c).

18. A few days later, just before the Brazilian elections on 6 October 2002, Merrill Lynchissued a new report upgrading Brazilian bonds (Merrill Lynch, 2002e).

19. Note that Brazil is included in the whole EMBI and EMBI+ indexes.

ReferencesAcemoglu, D., S. Johnson, J. Robinson and Y. Thaicharoen (2002). Institutional Causes,

Macroeconomic Symptoms: Volatility, Crises and Growth. Working Paper, 1 August. MIT,Berkeley University and Bank of Thailand.

Ackerman, J. (2002). “There is a Better Way to Manage Debt Crises.” Financial Times, 26September.

Aghion, P., A. Alesina and F. Trebbi (2001). Endogenous Political Institutions. Working Paper,November. Cambridge, MA: Harvard University.


Alesina, A., R. Hausmann, R. Hommes and E. Stein (1999). “Fiscal Institutions and BudgetDeficits in Latin America.” Journal of Development Economics, 59: 233–255.

Ames, B. (1987). Political Survival. Berkeley: University of California Press.BBA Economic Research (2002a). No Free Lunch: Lies, Damn Lies, and Statistics. 19 July. BBA

Economic Research.BBA Economic Research (2002b). No Free Lunch: Getting By With a Little Help from Our Friends.

7 August. BBA Economic Research.Bacmann, J.F. and G. Bolliger (2001). Who are the Best? Local Versus Foreign Analysts on the

Latin American Stock Markets. November. Neuchâtel: RMF Investment Products, QuantitativeAnalysis Group and University of Neuchâtel and FAME Institut de l’Entreprise. Availableat http://www.unine.ch/gge/english/Contenu/Finance/publig.htm.

Banco BBA Creditanstalt (2002). No Free Lunch: Lula in the Sky with Diamonds. 13 May. SaoPaulo: BBA Economic Research on Brazil.

Barber, B., R. Lehavy, M. McNichols and B. Trueman (2001). Prophets and Losses: Reassessingthe Returns to Analysts’ Recommendations. Research Paper 1692, May. Stanford, CA: StanfordUniversity Graduate School of Business.

Barclays Capital (2002a). “Market Drivers. Why Brazil is Different from Argentina.” The Emerging Call, 19 July. New York: Barclays Capital, Emerging Markets StrategyAmericas.

Barclays Capital (2002b). The Emerging Views. 1 August. New York: Barclays Capital,Emerging Markets Strategy Americas.

Barclays Capital (2002c). “Focus: Is the Pressure on Brazil’s Real and Bonds Over?” TheEmerging Call, 28 October. New York: Barclays Capital, Emerging Markets Strategy.

Barro, R. and J.W. Lee (2002). IMF Programs: Who is Chosen and What are the Effects? WorkingPaper, April. Cambridge, MA: Harvard University.

BCP Securities (2002a). Check, Please. 8 August. New York: BCP Securities.BCP Securities (2002b). Overview: Monster’s Ball. 1 October. New York: BCP Securities Latin

American Adviser.Bear Stearns (2002a). Brazilian Election Countdown. Serra’s Chances Increase. Trip Notes. 25

June. New York: Bear Stearns, Emerging Markets Sovereign Debt Strategy.Bear Stearns (2002b). The Brazil IMF Package—What Does it Mean for Emerging Markets? 8

August. New York: Bear Stearns, Emerging Markets Sovereign Debt Research.Bear Stearns (2002c). “Some Impressions from the World Bank/IMF Annual Meetings.” Bear

Stearns Sovereign Latin America Update, 1 October. New York: Bear Stearns, EmergingMarkets Sovereign Debt Research.

Block, S. (2001a). Elections, Electoral Competitiveness and Political Budget Cycles in DevelopingCountries. Working Paper, No. 78, October. Cambridge, MA: Harvard University, Centerfor International Development.

Block, S. (2001b). Political Conditions and Currency Crises: Empirical Regularities in EmergingMarkets. Working Paper, No. 79, October. Cambridge, MA: Harvard University, Center forInternational Development.

Block, S. and P.M. Vaaler (2001). The Price of Democracy: Sovereign Risk Taking, Bond Spreadsand Political Business Cycles in Developing Countries. CID Working Paper, No. 82, December.Cambridge, MA: Harvard University, Center for International Development.

Bloomberg, B., J. Frieden and E. Stein (2001). Sustaining Fixed Rates: The Political Economy ofCurrency Pegs in Latin America. Working Paper, July. Wellesley College, Harvard Universityand Inter-American Development Bank.

Boni, L. and K. Womack (2002). “Wall Street’s Credibility Problem: Misaligned Incentivesand Dubious Fixes?” Brookings-Wharton Papers on Financial Services. Paper presented at the5th Annual Brookings-Wharton Conference on the Future of Securities Markets.

Bordo, M. and A. Panini Murshid (2002). Globalization and Changing Patterns in theInternational Transmission of Shocks in Financial Markets. NBER Working Paper, No. 9019,June. Cambridge, MA: NBER.

Bordo, M., B. Eichengreen, D. Klingebiel and M.S. Martinez-Peria (2001). “Is the CrisisProblem Growing More Severe?” Economic Policy, April, 16(32): 51–82.


Calvo, G. (2001). Globalization Hazard and Weak Government in Emerging Markets. IADB

Working Paper, December. IADB.Calvo, G. (2002). Globalization Hazard and Delayed Reform in Emerging Markets. Working Paper,

February. IADB, UMD and NBER.CDC Ixis (2002a). “Le risque politique: Une composante du risque pays.” Lettre Economique de

la CDC, April, 139: 4–5.CDC Ixis (2002b). “Terminal do Brazil?” Flash Marchés Emergents, 17 June, 2002-10.Chang, R. (2002). Financial Crises and Political Crises. Working Paper, September. New

Brunswick, NJ: Rutgers University, Department of Economics.Citigroup and Salomon Smith Barney (2002). “Overview: The Brazilian Presidential

Campaign.” In Latin American Economic Perspectives (Citigroup), May: 3–12. New York:Citigroup, Economic & Market Strategy Analysis.

Crédit Agricole Indozuez (2002). “Brazil: Volatility or Fragility in 2002?” In Global EconomicMarkets Strategy, 25 February, 15–20. Paris: Crédit Agricole Indosuez.

Desai, P. (2002).”Brazil’s Options for the Here and Now.” Financial Times, 19 August.Deutsche Bank (2002). Brazil: On the Edge. 24 May. New York: Deutsche Bank, Global

Markets Research, Emerging Markets.Dooley, M. and J. Frankel (eds) (2002). Managing Currency Crises in Emerging Markets.

Chicago, IL: University of Chicago Press.Drazen, A. (1999). Political Contagion in Currency Crises. NBER Working Paper, No. 7211.

Cambridge, MA: NBER.Drazen, A. (2000). Political Economy in Macroeconomics. Princeton, NJ: Princeton University Press.Edwards, S. (2002). “Brazil’s Only Hope of Avoiding Collapse.” Financial Times, 4 August.Edwards, S. and J. Frankel (eds) (2002). Preventing Currency Crises in Emerging Markets.

Chicago, IL: University of Chicago Press.Eichengreen, B. (2002a). “Predicting and Preventing Financial Crises: Where Do We Stand?

What Have We Learned?” Paper presented at the Kiel Week Annual Conference,Germany, 24–25 June. Available at http://elsa.berkeley.edu/users/eichengr/website.htm.

Eichengreen, B. (2002b). A Temporary Respite for Brazil. Working Paper, 9 August. Berkeley,CA: University of California. Available at http://elsa.berkeley.edu/users/eichengr/website.htm.

Eichengreen, B. (2003). Capital Flows and Crises. Cambridge, MA: MIT Press.Eichengreen, B. and M. Bordo (2002). Crises Now and Then: What Lessons from the Last Year of

Financial Globalization? NBER Working Paper, No. 8716, January. Cambridge, MA: NBER.Eichengreen, B., A. Rose and C. Wyplosz (1995). “Exchange Market Mayhem: The

Antecedents and Aftermath of Speculative Attacks.” Economic Policy, 21: 249–312.El-Erian, M. (2002a). “Are We There Yet?” PIMCO Emerging Markets Watch, May. Available at

http://www.pimco.com.El-Erian, M. (2002b). “What Lula Must Do to Save Brazil.” Financial Times, 23 October.English, P.C., II and W.T. Moore (2002). “Property Rights Ambiguity and the Effect of

Foreign Investment Decisions on Firm Value.” In 2002 Index of Economic Freedom (G.O’Driscoll, K. Holmes and M.A. O’Grady, eds), 49–57. Washington, DC: The HeritageFoundation.

Feldstein, M. (ed.) (2002). Economic and Financial Crises in Emerging Markets. Chicago, IL:University of Chicago Press.

Fraga, A. and I. Goldfajn (2002). “Trust Brazil.” Financial Times, 18 September.Frieden, J. and E. Stein (eds) (2001). The Currency Game: Exchange Rate Politics in Latin

America. Baltimore: Johns Hopkins University Press.Frieden, J., P. Ghezzi and E. Stein (2001). “Politics and Exchange Rates: A Cross-Country

Approach.” In The Currency Game: Exchange-Rate Politics in Latin America (J. Frieden and E.Stein, eds), 20–63. Baltimore: Johns Hopkins University Press.

Gelos, G. and S. Wei (2002). Transparency and International Investor Behavior. NBER WorkingPaper, No. 9260, October. Cambridge, MA: NBER.

Goldman Sachs (2002a). “The Lulameter.” Emerging Markets Strategy, 6 June. New York:Goldman Sachs.


Goldman Sachs (2002b). “Brazil: Fiscal Shocks as the Cure for a Confidence Crisis.” LatinAmerica Economic Analyst, 28 June. New York: Goldman Sachs.

Goldman Sachs (2002c). “An Update on Brazil and Mexico.” Latin America Economic Analyst,6 September. New York: Goldman Sachs.

Goldman Sachs (2002d). “A Look at Brazil, Oil, and Colombia.” Latin America EconomicAnalyst, 20 September. New York: Goldman Sachs.

Goldstein, M. (2002). “Brazil’s Unwatched Borrowing.” Financial Times, 27 August.González, M.A. (2002). “Do Changes in Democracy Affect the Political Budget Cycle?

Evidence from Mexico.” Review of Development Economics, 6(2): 204–224.Haggard, S. (2000). The Political Economy of the Asian Financial Crisis. Washington, DC:

Institute of International Economics.Hausmann, R. (2002). “Baffled by Brazil.” Financial Times, 14 August.Hays, J., J. Freeman and H. Nesseth (2001). “Democratization and Globalization in

Emerging Market Countries: An Econometric Study.” Paper presented at the AnnualMeeting of the American Political Science Association, San Francisco, 30 August–2 September.

Henisz, W.J. (2002). Political Institutions and Policy Volatility. Working Paper, May. WhartonSchool of Management.

Irigoin, M.A. (2000). “Finance, Politics and Economics in Buenos Aires, 1820s–1860s: ThePolitical Economy of Currency Stabilisation.” Ph.D. Dissertation. London: LondonSchool of Economics.

Irigoin, M.A. and E. Salazar (2000). “Looking at Political Events and Economic Uncertainty:An Examination of the Volatility in the Buenos Aires Paper Peso Rate of Exchange,1826–1866.” Paper presented at Stanford University LACLIO Conference, Stanford, CA,17–18 November.

J.P. Morgan Chase (2001). Argentina: The “Model”—Not Just the Peg—Suffers, 27 and 31December. New York: J.P. Morgan Chase, Economic Research.

J.P. Morgan Chase (2002a). Brazil: A Guide to the October Elections, 17 May. New York: J.P.Morgan Chase, Emerging Markets Research.

J.P. Morgan Chase (2002b). Special Report: Brazil’s Economy and the Upcoming Election, 5 July.New York: J.P. Morgan Chase.

J.P. Morgan Chase (2002c). “Brazil: Three Months that Could Make or Break the LulaGovernment.” Emerging Markets Today, 28 October. New York: J.P. Morgan Chase,Emerging Markets Research.

J.P. Morgan Chase (2002d). Brazil: There is No Quick Fix. Position for Enduring Volatility, 9December. New York: J.P. Morgan Chase, Emerging Markets Research.

Krugman, P. (1979). “A Model of Balance-of-Payments Crises.” Journal of Money, Credit, andBanking, 11(3): 311–325.

Krugman, P. (2002). “The Confidence Game.” In The Return of Depression Economics (P.Krugman, ed.). New York: Penguin Books.

Leblang, D. (2001). To Devalue or Not To Devalue? The Political Economy of Exchange Rate Policy.University of Colorado Working Paper. University of Colorado.

Leblang, D. (2002). “The Political Economy of Speculative Attacks in the DevelopingWorld.” International Studies Quarterly, 46(1): 69–91.

Leblang, D. and W. Bernhard (2000). “The Politics of Speculative Attacks in IndustrialDemocracies.” International Organization, 54: 291–324.

Lehman Brothers (2002). “Brazil: Towards a Definition.” Latin America Weekly, 1 November.New York: Lehman Brothers Global Economics.

Lin, T.M. and B. Roberts (2001). “Parliamentary Politics and Exchange Markets.” Paperpresented at the Annual Meeting of the Midwest Political Science Association, Chicago,IL.

Lowry, R.C. (2001). “A Visible Hand? Bond Markets, Political Parties, Balanced BudgetLaws, and State Government Debt.” Economics & Politics, 13(1): 49–72.

MacIntyre, A. (2001). “Institutions and Investors: The Politics of the Economic Crisis inSoutheast Asia.” International Organization, 55(1): 81–122.


Martin, P. and H. Rey (2002). Financial Globalization and Emerging Markets: With or Without Crash? Working Paper, May. Federal Reserve Bank of New York and PrincetonUniversity.

Maxwell, K. (2002). “Lula Falls Foul of a Foreign Fantasy.” Financial Times, 27 September.Mei, J. (1999). Political Risk, Financial Crisis, and Market Volatility. New York: New York

University, Department of International Business.Merrill Lynch (2002a). The Many Dynamics of Public Debt. 13 February. New York: Merrill

Lynch, Fixed Income Strategy.Merrill Lynch (2002b). What We Learned at Fortaleza and Beyond. 19 March. New York: Merrill

Lynch.Merrill Lynch (2002c). Stress Testing the Candidates’ Economic Programs. 30 August. New York:

Merrill Lynch, Latam Macro Insights.Merrill Lynch (2002d). Emerging Markets Daily. 25 September. New York: Merrill Lynch

Emerging Markets.Merrill Lynch (2002e). Emerging Markets Debt Monthly. 4 October. New York: Merrill Lynch

Emerging Markets.Merrill Lynch (2002f). “Lula de Mel.” Emerging Markets Debt Monthly. 1 November. New York:

Merrill Lynch.Molano, W. (2002a). Da Lula Monster. 24 May. New York: BCP Securities.Molano, W. (2002b). Brazil: Last Chance to Get Out. 5 August. New York: BCP Securities.Morgan Stanley (2002a). “Brazil: Why is this Emerging Market Not Like the Others?” Latin

America Equity Research Strategy, 25 February. New York: Morgan Stanley.Morgan Stanley (2002b). Brazil: ’Til Debt Do Us Part. 19 July. New York: Morgan Stanley,

Latin America Fixed Income Research.Morgan Stanley (2002c). Bearish on Brazil. Model Portfolio Update. 12 August. New York:

Morgan Stanley, Global Fixed Income Research.Morgan Stanley (2002d). Inside Brazil. 12 August. New York: Morgan Stanley, Latin America

Fixed Income Research.Morgan Stanley (2002e). “Still Brazil.” EMD Focus, 6 September. New York: Morgan Stanley,

Fixed Income Research.Obstfeld, M. (1994). “The Logic of Currency Crises.” Cahiers d’Economie Monétaire, Banque de

France, 43: 189–213.Olson, M. (2000). Power and Prosperity: Outgrowing Communist and Capitalist Dictatorships. New

York: Basic Books.Ortiz, G. (2002). “Solving Debt Crises.” Financial Times, 20 September.Pantzalis, C., D.A. Stangeland and H. Turtle (2000). “Political Elections and the Resolution

of Uncertainty: The International Evidence.” Journal of Banking and Finance, 24(10):1575–1604.

Persson, T. (2001). Do Political Institutions Shape Economic Policy? Working Paper, February.Stockholm: Institute for International Economics Studies, Stockholm University.

Persson, T. (2002). “Do Political Institutions Shape Economic Policy?” Econometrica, 70(3):883–905.

Persson, T. and G. Tabellini (2000). Political Economics: Explaining Economic Policy.Cambridge, MA: MIT Press.

Przeworski, A. (1991). Democracy and the Market: Political and Economic Reforms in EasternEurope and Latin America. Cambridge, MA: Cambridge University Press.

Reihnart, C. (2002). Default, Currency Crises, and Sovereign Credit Ratings. NBER Working Paper,No. 8738, January. Cambridge, MA: NBER.

Reisen, H. (2002). Ratings Since the Asian Crisis. Discussion Paper, No. 2002/2, January.WIDER United Nations University.

Rodrik, D. (2002). “Institutions for High-Quality Growth: What They Are and How ToAcquire Them.” Studies in Comparative International Development, 35(3): 3–31.

Saiegh, S. (2001). “Is There a Democratic Advantage? Assessing the Role of PoliticalInstitutions in Sovereign Borrowing.” Paper presented at the Annual Meeting of theAmerican Political Science Association, San Francisco, 30 August–2 September.


Salomon Smith Barney (2002a). Brazil, Economic and Country Market Analysis. 20 August. NewYork: Salomon Smith Barney.

Salomon Smith Barney (2002b). Ciro Gomes: A Candidate in Transition. 23 August. New York:Salomon Smith Barney, Economic and Market Analysis.

Salomon Smith Barney (2002c). Arminio Fraga Meeting Highlights. 26 August. New York:Salomon Smith Barney, Economic and Market Analysis.

Santiso, J. (1999a). “Wall Street and the Mexican Crisis: A Temporal Analysis of EmergingMarkets.” International Political Science Review, 20(1): 49–72.

Santiso, J. (1999b). “Analysts Analyzed: A Socio-Economic Approach to Financial andEmerging Markets.” International Political Science Review, 20(3): 307–330.

Santiso, J. (2002). “Corporate Governance and Political Governance: Multinationals,Portfolio Investors and Institutions in Emerging Markets.” Paper prepared for the OECD

Development Centre.Santiso, J. (2003). The Political Economy of Emerging Markets: Actors, Institutions and Crisis in

Latin America. New York: Palgrave.Schamis, H. and C. Way (2001). The Politics of Exchange Rate-Based Stabilization. Cornell

University Working Paper. Cornell University.Shi, M. and J. Svensson (2001). Conditional Political Business Cycles: Theory and Evidence.

Stockholm: Institute for International Economics Studies, Stockholm University. See alsohttp://www.iies.su.se/~svenssoj/.

Soros, G. (2002). “Don’t Blame Brazil.” Financial Times, 13 August.Stasavage, D. (2002). “Private Investment and Political Institutions.” Economics & Politics,

14(1): 41–63.UBS Warburg (2002). Brazil—TV Time Match. 8 May. Stamford: UBS Warburg.Vu Le, Q. and P.J. Zak (2001). Political Risk and Capital Flight. Working Paper. Claremont

Graduate University, School of Politics and Economics.

Biographical NotesJUAN MARTÍNEZ holds a Master’s Degree in Economics from ComplutenseUniversity in Madrid. Between 1993 and 1998 he worked for the Spanish brokerInterMoney as an Economist and later as Chief Economist for Latin America.Since 1998 he has been a Senior Economist for Latin America in the EconomicResearch Department of Banco Bilbao Vizacaya Argentaria (BBVA), Madrid.ADDRESS: BBVA, Research Department, Paseo de la Castellana, 81, 28046 Madrid,Spain [email: [email protected]].

JAVIER SANTISO holds a BA, MA and Ph.D. from Sciences Po Paris and an MBA degreefrom HEC School of Management. From 1995 to 1997 he completed his doctoralstudies at Oxford University, and was Senior Associate Member of St Antony’sCollege, Latin American Centre. He was a Research Fellow from 1997 to 2002 atthe Paris-based Centre d’Etudes et de Recherches Internationales (CERI-SciencesPo Paris). Since 2002 he has been based in Madrid as Chief Economist for LatinAmerica of Banco Bilbao Vizcaya Argentaria (BBVA). He has served as a Professorat SAIS Johns Hopkins University, Sciences Po Paris and HEC School ofManagement. His recent publications include The Political Economy of EmergingMarkets: Actors, Institutions and Crisis in Latin America. ADDRESS: BBVA, ResearchDepartment, Paseo de la Castellana, 81, 28046 Madrid, Spain [email:[email protected]].

Acknowledgments. First of all, we would like to give very special thanks to Juan AntonioRodríguez Garcia for his tremendously helpful research assistance. Many thanks as well to


those that provided us with pertinent information, data and comments, namely, ManoloBalmaseda, Jorge Blazquez, Juan Carlos Berganza, Santiago Sanz Fuente and Luciana Taft(BBVA Research Department), Pierre Ahlsell (Fortis Investment Management), Olivier deBoysson (Société Générale), Damien Buchet (Barep Asset Management-Société GénéraleGroup), Claire Dissaux (Crédit Agricole Indosuez), Barry Eichengreen (BerkeleyUniversity), José Gijon (SAIS Johns Hopkins University), Marc Flandreau (Sciences PoParis), Martin Grandes (OECD Development Centre and DELTA), Witold Jerzy Henisz(Wharton School of Management), James Meadowcroft (Sheffield University), WalterMolano (BCP Securities), Alvaro Ortiz (Santander Central Hispano), Joseph Oughourlian(Société Générale), Luisa Palacios (Japan Bank for International Cooperation), RiordanRoett (SAIS Johns Hopkins University), Manuel Sanchez (BBVA Bancomer), Jaime Sanz(Merrill Lynch) and Alexandre Schwartsman (BBA Brazil at the time of the research; now atUnibanco). The findings, interpretations and conclusions expressed in this article areentirely those of the authors and do not represent the views of the institutions or personsmentioned before, or reflect the position of any institution with which the authors areaffiliated.


Abstracts/Résumés

Une actualité vouée à l’éternité

ERIC BARTHALON

On trouvera dans cet essai une réflexion critique sur le caractère et la significationdes crises financières, tirée du travail des différents chercheurs qui ont étudié leproblème.

Après avoir défini le concept de «crise financière», trois questions sont posées:Peut-on prévoir les crises financières? Peut-on et doit-on les maîtriser? etComment peut-on les prévenir?

Pourquoi les gouvernements font-ils appel au FMI? Quelques exemples représentatifs

JAMES RAYMOND VREELAND

Pourquoi les gouvernements font-ils appel au Fond Monétaire International? Il estgénéralement admis que les gouvernements font appel au FMI pour une raisontrès simple: ils ont besoin d’un prêt. Une autre théorie est que les gouvernementssouhaitent se prévaloir des conditions imposées par le FMI pour justifier la miseen œuvre de réformes économiques impopulaires.

Pour illustrer à quel point les conditions du FMI jouent un rôle dans laconclusion des accords, nous choisissons deux exemples représentatifs tirés d’uneimportante étude statistique: le cas de l’Uruguay en 1990 et celui de la Tanzanieen 1983.

Politique publique et acteurs privés: La fin des Nouveaux Accords de Bâle

MICHAEL R. KING ET TIMOTHY J. SINCLAIR

Cet article examine une des recommandations du «Comité de Bâle pour laSupervision Bancaire» qui vise à réformer les normes régissant actuellement lesbesoins en fonds propres, en incorporant les évaluations internes des banques etcelles qui sont fournies par des agences spécialisées telles que «Moody’s Investors




Service» et «Standard and Poor’s» pour déterminer le rapport capital-risquerequis.

Cet article met en lumière une série de conséquences négatives, découlant del’utilisation des éléments fournis par les agences privées aux dépens des règlesd’ordre public.

Les marchés financiers et la politique: stratégie de gestion des crisesfinancières dans les économies émergentes des pays d’Amérique latine

JUAN MARTÍNEZ ET JAVIER SANTISO

L’article est consacré à l’analyse des interactions entre la politique et les marchésfinanciers dans les pays d’économie émergente.

En se basant sur des exemples concrets, les auteurs étudient les réactions desanalystes de Wall Street à l’élection présidentielle du Brésil en 2002. Ils se fondeségalement sur les données d’ordre financier ou électoral fournies par les électionsprécédentes pour situer les événements de 2002 dans une perspective historiqueet comparative.


Date post:	11-Feb-2022
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

International Political Science Review Revue

Documents