Preface and Chapter 1 Introduction

THE NEW MASTERS OF CAPITAL

Preface

Global finance is big business. Really big business. The bond rating agencies that are the subject of this book maintain ratings on $30 trillion worth of debt issued in American and international markets. These markets are surely too big for us to ignore if we want to understand how our world works.

Not only is global finance big, but it touches us all. The fortunes of currencies— and of banks and the markets for securities -affect our lives every day. They affect the interest rates we pay for our credit card debt, those for our house mortgage, and the return on our pension fund.

A lot of people are confused by how finance works. It appears very technical. Because finance has this image, many prefer to leave it to "the experts." But we must not allow ourselves to do so. Like war, the institutions and processes of global finance are too important to leave to professionals to figure out. This book is an effort to cross the boundary into expert territory and identify the broader political signifi­cance of these seemingly areane technical institutions.

Sir Robert Muldoon, prime minister and minister of finance of New Zealand, 1975—84, may not be a frequent beneficiary of scholarly thanks, but this book would not exist at all had it not been for him. Although I never met the man, his relations with the bond rating agencies first made me think that understanding these institu­tions might be important in the post-Bretton Woods era. Muldoon, short of stature and wide of girth, was energetic, intelligent, and truculent. Very little cowed him. Noted exceptions were the rating agencies, which Muldoon seemed to think were very important. He left an impression on rating officials that was still evident several years after his tenure ended, as I discovered during interviews on Wall Street. Muldoon's views had an impact on me as well.

I was very fortunate in the intellectual and institutional support that came my way as this book developed. Robert Cox, Stephen Gill, and David Leyton-Brown (DLB) all had a major influence. I could not have asked for more challenging schol­arly training or better mentoring. I also thank DLB for the generous financial sup­port he provided in connection with researeh funded by the Social Sciences and Humanities Researeh Council of Canada, without which my fieldwork in Japan would not have been possible. At York University in Toronto, the faculty and staff of the Department of Political Science, the Faculty of Graduate Studies, Stong Col­lege, and the York Centre for International and Security Studies supported this researeh in many ways, providing office space, fieldwork grants, and conference funding to test my initial ideas.

At the University of Warwick in England, where I have worked since 1995,1 am thankful for two researeh development grants, which allowed me to undertake sup­plemental fieldwork. I also appreciate the financial support for conference partici­pation generously provided by the Department of Politics and International Studies. The Warwick Center for the Study of Globalisation and Regionalisation provided support for attendance at a conference on rating agencies held at New York Univer­sity's Stern School in 2001. At Warwick, I have been fortunate to teach a graduate class on the politics of global finance for some years. Many students have offered valuable insights while reading and discussing my journal articles on rating. Joe Horneck and Belkys Lopez are notable among their colleagues for bringing useful doc­umentary sources to my attention. Paola Robotti, a Warwick doctoral candidate, applied her considerable skill to improve my primitive efforts at drawing figures.

During the closing stages of this project, I was fortunate to spend a sabbatical year at Harvard University, as a visiting scholar at the Weatherhead Center for Inter­national Affairs. Jeffry A. Frieden and James A. Cooney were instrumental in mak­ing this happen and helped make it a most valuable experience. While at Harvard, I was resident at Winthrop House, where Karen Reiber, Martine van Ittersum, David Simms, and Enoch Kyerematen made a big difference to my experience.

Many people in the global academic community helped with this book in one way or another. In Canada, Eric Helleiner, Louis W. Pauly, Chris Robinson, A. Claire Cutler, and Ricardo Grinspun were key. In the United States, I greatly benefited from the interest of James N. Rosenau, Raymond D. "Bud" Duvall, Craig N. Murphy, Rawi E. Abdelal, James H. Mittelman, Timothy J. McKeown, Kenneth P Thomas, Jeffry A. Frieden, Peter Gourevitch, Richard W. Mansbach, Yale H. Ferguson, Michael Schwartz, Mark Amen, Virginia Haufler, Kathryn Lavelle, and "Skip" McGoun. In Europe, I learned much from Ronen Palan, Jan Aart Scholte, Susan Strange, Dieter Kerwer, Torsten Strulik, Helmut Willke, Oliver Kessler, Philip G. Cerny, Donald MacKenzie, Tony Payne, Marieke de Goede, Peter Burnham, Henk Overbeek, and Kees van der Pijl. Frank and Patrick McCann pro­vided expert photographic input.

I have many debts to acknowledge for the help I received during my field researeh. Particularly kind were Leo C. O'Neill and Cathy Daicoff of Standard & Poor's, and David Stimpson of Moody's Investors Service. I met Mr. O'Neill, pres­ident of Standard & Poor's, near the start and the end of the project, and at each meeting he was forthcoming and incisive. Most helpful in the final years of my researeh was David Levey of Moody's. David is a great source of knowledge and good judgment about the rating business and its challenges, as well as a scholar and political economist himself. I learned a great deal from him. Chris Mahoney, also of Moody's, provided important aid. Takehiko Kamo, before his early death a profes­sor in the Faculty of Law at the University of Tokyo, assisted my researeh in Japan, sponsoring my stay at the International House of Japan. Donald J. Daly, Hiroharu Seki, and Seiji Endo helped with contacts in Japan.

In addition to anonymous reviewers, several scholars read the entire manuscript. I am greatly indebted to Benjamin J. Cohen, Tony Porter, Richard Higgott, Randall Germain, and Edward Cohen for their useful comments. They influenced me in many ways.

Friends and associates also supported me during the researeh and writing of this work. Especially important were Steve Patten, Graham Todd, Edward Comor, Robert O'Brien, J. Magnus Ryner, Martin Hewson, Liliana Pop, Randall Germain, Peter Burnham, Shirin Rai, and my IPMS Mercia friends in Warwickshire.

Peter J. Katzenstein and Roger Haydon made a major contribution to my think­ing about how this book should be organized. Their belief in the project and their practical impact cannot be underestimated.

Finally, my deep thanks go to the Wilson sisters. Delphine, Helen, Nancy, and Frances pushed me to defend my early ideas about politics, most memorably around the gas lamps and dinner tables at Mataikona. They inspired me, although it has taken me thirty years to appreciate fully their significance.

Timothy J. Sinclair

Kenihvorth, Warwickshire

Chapter 1 Introduction

We live again in a two-superpower world. There is the U.S. and there is Moody's. The U.S. can destroy a country by levelling it with bombs: Moody's can destroy a country by downgrading its bonds.

Thomas L. Friedman, New York Times, 1995

Contemporary American power is obvious to the casual observer. If you want concrete evidence of U.S. superpower status, take a trip to southern Ari­zona. Outside the city of Tucson is AMARE, the USAF "boneyard," the greatest collection of mothballed warplanes on Earth.¹ If an airplane was a part of the Amer­ican war machine during the past thirty years you will probably find it here, patiently awaiting its fate in the blazing Sonoran desert sun, together with some three thou­sand others. In this place, B-52 Stratofortresses, like those that dropped bombs on Vietnam, Afghanistan, and Iraq, and which were held in readiness for nuclear retal­iation during the Cold War, are broken up, their shattered fuselages and wings dis­played for the benefit of Russian spy satellites documenting the fulfillment of Strategic Arms Reduction Treaty (START) obligations. A-10 Thunderbolt IIs, the venerable "Warthog" tank-busters of Gulf Wars I and II, now expected to be in the USAF inventory until 2028, stand row upon row in the searing desert heat, quietly awaiting redeployment. Other "hogs,”based at nearby Davis-Monthan Air Force Base, fly low overhead, silently circling the University of Arizona campus. In this arsenal, the embodiment of a Tom Clancy or Don DeLillo novel, the basis of Amer­ica's superpower status could not be clearer.

But things are different when it comes to the "second superpowers," the major bond rating agencies—Moody's Investors Service (Moody's), its competitor, Stan­dard & Poor's (S&P), the smaller and less important Fitch Ratings (Fitch), and the multitude of minor domestic rating agencies around the globe. They operate in a very different world. Their arsenal is an occult one, largely invisible to all but a few most of the time.² Financial stress expands the size of the group aware of the agen­cies: in 2002, Europe had its highest-ever level of defaults, up to $15 billion from $4 billion in 2001. To the people directly concerned with matters of financial health— chief financial officers, budget directors, Treasury officials, and increasingly even politicians—rating agencies are well known.³ In this book the world of these second superpowers is explored: the basis of their power, the nature of their authority in financial markets, and implications of their judgments for corporations, municipal governments, and sovereign states.

In examining this world, I argue that rating agency activities reflect not the "cor­rectness" or otherwise of rating analyses but instead the store of expertise and intel­lectual authority the agencies possess. Market and government actors take account of rating agencies not because the agencies are right but because they are thought to be an authoritative source of judgments, thereby making the agencies key organiza­tions controlling access to capital markets. It is the esteem enjoyed by rating agen­cies—a characteristic distributed unevenly in modern capitalism—that this book explores, rather than whether agency ratings are actually valid.

A further claim made here is that this consequential speech has semantic content or meaning. That content, developed within the framework of rating orthodoxy delineated in chapter 3, is not purely technical but is linked to social and political interests. Although it is tempting to suggest that those interests are not related to location, the American origins of the rating agencies are relevant.

Changes on Wall Street and in other global financial centers increased the signif­icance of Moody's and S&P during the 1990s. The destruction of the World Trade Center in 2001 did not reverse this trend.⁴ Since the terrorist attacks, international trade and financial transactions have increased.⁵The broad context for the increased role of rating is the process of financial globalization that began in the 1970s.

Financial globalization encompasses worldwide change in how financial markets are organized, increases in financial transaction volume, and alterations in govern­ment regulation. As discussed here, the concept is more comprehensive than Armijo's specification of financial globalization as "the international integration of previously segmented national credit and capital markets."⁶ In financial globaliza­tion, markets are increasingly organized in an "arms length" way. Institutions that once dominated finance and were politically consequential, as a result, now have other roles.

Cross-border transactions have, of course, massively increased since capital con­trols were liberalized in most rich countries during the late 1970s and 1980s. The regulation of financial markets has also changed form since then. Though increas­ingly detailed, regulation is typically implemented by market actors. Government agencies create and adjust the self-regulatory framework as circumstances merit. In this environment, new financial products and strategies emerge frequently. Market volatility is associated with these developments, as is a sense that governments them­selves are increasingly subject to the judgments of speculators and investors.

The changes in market organization have been significant. Commercial banks used to be the institutions that corporations, municipalities, and national govern­ments sought out in order to borrow money. Today, in a process known as disinter- mediation, bonds and notes sold on capital markets are displacing traditional bank loans as the primary means of borrowing money. In a related process, securitization, mortgages, credit card receivables, and even bank loans are being transformed into tradeable securities that can be bought and sold in capital markets. This does not mean banks are of little importance in global financial markets. It means that judg­ments about who receives credit and who does not are no longer centralized in banks, as was the case in the past.

Over the past decade, the liberalization of financial markets has made rating increasingly important as a form of private regulation.⁷ States have had to take account of private sector judgments much more than in the heavily controlled post­war era.⁸ Liberalization of the financial markets have also increased exposure to risk and therefore the importance of information, investigation, and analysis mecha­nisms. Outside the rich countries, liberalization has been pursued by developing- country governments in Asia and Latin America that have sought to create local capital markets to finance investment in new infrastructure and industrial produc­tion. The importance of these new markets is that their operatives want information about the creditworthiness of the corporations and governments that seek to borrow their money. As things stand, market operatives get some of this information, in the form of bond ratings, from Moody's and S&P.

The two major U.S. rating agencies pass judgment on around $30 trillion worth of securities each year.⁹ Of this $30 trillion, around $107 billion worth of debt issued by 196 bond issuers was in default in 2001—a figure up sharply from 2000, when 117 issuers defaulted on $42 billion.^1,1 Ratings, which vary from the best (AAA or "triple A") to the worst (D, for default), affect the interest rate or cost of borrowing for businesses, municipalities, national governments, and, ultimately, individual cit­izens and consumers. The higher the rating, the less risk of default on repayment to the lender and, therefore, other things being equal, the lower the cost to the bor­rower. Rating scales are described in more detail in chapter 2.

The phenomenon investigated here is usually thought of as a technical matter. But this is largely a nontechnical book. An accurate, meaningful understanding of bond rating requires a broader view than the technical, just as an understanding of war cannot be limited to the analysis of military maneuvers or logistics. Hence, this book considers not just how ratings are done but also the purposes attributable to the rating process, the power and authority of the agencies, the implications of rat­ing judgments, and the problems that may bring change to the world of ratings.

Widespread misunderstandings exist about the way capital markets and their institutions work and shape the world. These markets are complex and seemingly areane. The amount of money involved is titanic and likely awesome to all but the richest inhabitants of the planet. Many think these markets shape economic and political choices in an objective way, much as the laws of physics shape the universe.¹¹ But the unqualified influence of markets and market institutions in recent years has not always been evident. For a time, during the New Deal era of the 1930s and the years of postwar prosperity in the West, a greater degree of public control tempered these global forces. U.S. and other Western governments developed welfare pro­grams and policy measures to insulate their populaces from the vagaries of capital markets. But the constraints, so the story goes, were artificial and, since the 1970s, have been challenged. Financial markets have again opposed the dictates of elected authorities and voters, to assume their "rightful place" in the scheme of things. Now, we are told by the popular and the scholarly press, there is no escaping these imper­sonal forces.

As an explanation of financial globalization, this sort of mechanistic view is not adequate. A technical understanding of the forces that constrain our economic and political choices is necessarily limited. This view assumes markets develop in ways beyond the influence of citizens, that people should simply allow things to take their "natural" course—financial globalization is inevitable. This is a key point. Much that is written about financial markets, even by people who recognize the political consequences of these markets, misses the fundamentally social character of what happens inside the markets and their institutions.¹²

The assumption in established texts is that markets reflect fundamental economic forces, which are not subject to human manipulation. But this view does not take account of the fact that people make decisions in financial markets in anticipa­tion of and in response to the decisions of others.¹³ In this book, the social nature of global finance gets particular emphasis. The social view of finance suggests that in situations of increased uncertainty and risk, the institutions that work to facilitate transactions between buyers and sellers have a central role in organizing markets and, consequently, in governing the world.¹⁴ Financial markets are more social—and less spontaneous, individual, or "natural"—than we tend to believe.

The role of rating agencies is not mechanistically determined, either. Many financial markets survived and flourished in the past without them. Typically, banks assumed the credit risk in the relationship between those with money to invest and those wishing to borrow. Alongside banks, traditional capital markets relied on bor­rowers who were well known and trusted names in their communities. But rating has increasingly become the norm as capital markets have displaced bank lending and as the trust implicit in these older systems has broken down. Rating serves a purpose in less socially embedded capital markets, where fund managers are under pressure to demonstrate they are not basing their understanding of the creditworthiness of investment alternatives on implicit trust in names but use a recognized, accepted mechanism.

At least three other ways of doing the existing work of the rating agencies can be imagined. The first is self-regulation by debtors. Much like the professional bodies for physicians, architects, and lawyers, a debtor-based system of credit information could provide data to the markets. Although this system might not be independent, collective self-interest would mitigate the tendency to self-serving outputs, much as is the case with professional self-regulation. Second, nonprofit industry associations could undertake or coordinate creditworthiness work. Good precedents already exist in countries where nonprofits enforce some national laws, such as in the case of animal welfare. The nonprofit model offers to eliminate some conflict of interest tensions implicit in charging debtors for their ratings. Third, governments could collectively take on the job, perhaps in the form of a new international agency. The International Organization of Securities Commissions (IOSCO) is already involved in discussions about rating standards and codes of conduct.'¹ The World Bank, the International Monetary Fund (IMF), and regional development banks could encourage local rating agencies in emerging markets to issue ratings. Such an arrangement would be independent of particular debtors and less subject to conflict of interest concerns, especially if not funded by rating fees.

John Moody, a muckraking journalist, Catholic convert, and credit analyst, pub­lished The Masters of Capital in 1919. In this volume he chronicled the construction of the railroad and steel trusts in the United States, and the links between these interests and Wall Street during the "robber baron" years, the era between the end of the Civil War and 1914.¹⁶ Moody investigated the capitalism of his day by look­ing at great entrepreneurs. Here, twenty-first-century capitalism is examined through analysis of institutions rather than the actions of "great men," an ontology more appropriate to present conditions.¹⁷

Within contemporary capitalism, rating agencies do not represent the only insti­tutionalization of power, nor are they all-seeing, all-knowing, all-powerful. This vol­ume is not an account of a conspiracy. The issue of power and authority inside capitalism today is its focus, just as Moody sought insight into the business power of his time. Ironically, however, the watchdogs of his day are the subject here.

Characteristics of the Rating Agencics

Rating agencies are some of the most obscure institutions in the world of global finance. Everyone knows what a bank is. Most people can explain what an insurance company does or offer a rough outline of an accountant's activities. But rating agen­cies are specialist organizations whose purpose and operations are little known out­side their immediate environment.

The discussion is not concerned with the merits of the agencies from an economic or policy perspective, to determine whether they are "good" or "had." The purpose, based on the agencies' growing impact, is to evaluate their role in financial globalization. The agencies are influential mechanisms of financial globalization, shaping what governments (at all levels) do and corporate behavior, too. Hence, an understanding of the motivations, objectives, and constraints on these institutions is worthwhile.

Although they are often confused with Moody's and S&P, institutions such as Dun and Bradstreet, which undertake the mercantile rating of retailers for suppli­ers, are excluded from the analysis. Also excluded are corporations that issue credit ratings on individual consumers, such as Experian.¹⁸ Many of the broader processes identified here are evident in these institutions, but these other raters are not cen­tral to the organization of capital markets. Rating agencies are examined in the con­text of their work with institutions in the capital markets, including municipalities, corporations and sovereign states, because that is where rating has the most impact.

What do the raters actually do? The agencies claim to make judgments on the "future ability and willingness of an issuer to make timely payments of principal and interest on a security over the life of the instrument."¹⁹ Ostensibly, this is a narrow remit. The more likely it is that "the borrower will repay both the principal and inter­est, in accordance with the time schedule in the borrowing agreement, the higher will be the rating assigned to the debt security."²⁰ The agencies are adamant about what a debt rating is not. According to Standard & Poor's, a rating is "not a recom­mendation to purchase, sell, or hold a security, inasmuch as it does not comment as to market price or suitability for a particular investor," because investors' willing­ness to take risks varies.²¹ In other words, a credit rating should form just part of the information investors use to make decisions. Rating agencies themselves do not claim to provide more than some of the information investors need.

As noted, financial globalization has widened the scope of the agencies' work. The prevailing objective, for both major agencies, is to achieve globally comparable ratings. If an AA on a steel company in South Korea is equivalent in credit-risk terms to AA on a pulp mill in British Columbia or to a similar rating on a software producer in California, investors can make global choices. In recent years, the agen­cies have also sought to provide ratings that are comparable within specific national contexts. New York, however, very much remains the analytical center, where rating expertise is defined and reinforced internally through the agencies' established training cadres and standard operating procedures.

The agencies produce ratings on corporations, financial institutions, municipal­ities, and sovereign governments in terms of long-term obligations, such as bonds, or short-term ones like commercial paper.

Once issued, rating officials maintain surveillance over issuers and their securi­ties. They warn investors when developments affecting issuers—their tax base, say, or their market—might lead to a rating revision, either upward or downward. As will be seen, this surveillance aspect of rating work is a key one, just as Pauly has shown in the context of International Monetary Fund monitoring.²² Rating agency surveil­lance shapes the thinking and action of debt issuers. It also shapes the expectations of investors, who want the agencies to forensically scrutinize issuers and who com­plain vociferously when this scrutiny seems less than they think it ought to be. Investors seem to expect rating agencies to play the role of the prison guards in Bentham's perfect penitentiary, the panopticon.²³

What product do the agencies sell? They purvey both professional, expert knowl­edge in the form of analytical capacities and local knowledge of a vast number of debt security issuers. The disinter mediation process heightens the role of bond rating agencies. It increases their analytical and local specialization absolutely, because they now rate more issues in more locations, and relatively, because with the growth of capital markets, comparable specialists (bank credit analysts are the obvious exam­ple) have become less important as gatekeepers.²⁴

Both Moody's and S&P are headquartered in New York. Both global agencies have numerous branches in the United States, Europe, and emerging markets. A dis­tant third in the market is Fitch Ratings, a unit of Fimalac SA of Paris. Domesti­cally focused agencies have developed in OECD countries and in emerging markets since the mid-1990s.²⁵

Public panics or crises about rating miscalls are the most significant challenge the agencies face. Crises erode and even threaten to shatter the reputational assets the agencies have built up since the interwar period. The 1990s and the first years of the new millennium saw more of these events, when volatility grew along with financial globalization. Threatening events included Mexico's financial crisis of 1994-95, Asia's financial crises of 1997-98, and Russia's default in 1998. Derivatives and other innovations stimulated corporate and municipal scandals and financial collapses in the United States, including the bankruptcy of Enron Corporation in late 2001. The new millennium was marked by the $141 billion sovereign debt default of Argentina in 2001-02.²⁶

Two main strategies characterize the agencies' responses to these legitimacy crises. Like other financial industry institutions, the agencies try to keep up with financial innovation, spending large sums on staff training and hiring. They push development of their own products. The agencies have created new symbols to indi­cate when, for example, ratings are based on public information only and do not reflect confidential data (in the case of Standard & Poor's). The agencies, especially Moody's, have sought to change their cloistered, secretive corporate cultures and, since 1997, have become more willing to set out a clear rationale for their ratings. That strategy may have much to do with managing public expectations of the agencies.

How do the agencies relate to governments? Despite assumptions to the contrary, the work of rating agencies, in terms of their criteria and decision-making, is not regulated seriously anywhere in the developed world. Indeed, tight regulation would potentially destroy the key thing agencies have to sell: their independent opinion on market matters. However, some process by capital market regulatory agencies of "recognizing" rating agencies' activities is customary around the world.²⁷ This recognition is especially significant in the United States, where many states have laws governing the prudential behavior of public pension funds.²⁸ In these cases, the agencies' outputs are recognized as benchmarks limiting what bonds a pension fund can buy.

A central feature of United States and other countries' processes of governmen­tal recognition is regulators' reliance on wide market acceptance of a firm's rating. In turn, the agencies resist recurrent efforts to develop more invasive forms of regulation and hold up the public standard of market acceptance as the best test of their quality. They also oppose deeper incorporation of their ratings as benchmarks in law. Developing country governments often make ratings of domestic debt issues compulsory as a way of promoting the development of liquid, transparent capital markets.

Increasingly, ratings are key elements in transnational financial regulation. In 2001, the Bank for International Settlements proposed replacing established capital adequacy standards with a new system in which ratings play a significant role in esti­mating the risk exposure of banks.²⁹

Rating and Politics

Nuances of power and authority heighten the significance of rating. Rating agencies do possess, via rating downgrades, the capacity at times to coerce borrowers eager to obtain scarce funds. But relations between rating agencies and other institutions are more often about changing world views and influence than "power wielding." On the one hand, the influence of the rating agencies grows as new borrowers look to raise funds in lower-cost capital markets rather than borrow from banks in the tra­ditional way. In this environment, the number of agency branches is expanding, and the role of Moody's and Standard & Poor's is more significant: the agencies put a price on the policy choices of governments and corporations seeking funds.

On the other hand, many government administrations, particularly in the devel­oping world and Japan, have encouraged the formation of national bond rating agen­cies. These initiatives are intriguing. They suggest that the loss of government policy autonomy implied in the establishment of rating has not been imposed on govern­ments but is actually something states have sought, even promoted. Hence, a view of rating simply as a coercive force does not capture the whole story Consideration also must be given to where rating shapes, limits, and controls—often in connection with the generation of authority—rather than the brute application of power. Elab­orating this consideration is a key focus of this book.

Analytical Approach

In this book specific institutions and associated "micropractices" at the core of con­temporary capitalism are examined, in particular the "reconfiguring" effect these institutions and practices have on global economic and political life within sovereign states.³⁰ Natural science seeks to establish universal laws and considers specific events in terms of these laws. The objective is always generalization, and many social scientists have followed this path. Here, the purpose is similar to "process tracing," the historical development of interpretive frames actors use to understand the world.³¹

Specific events, institutions, and ways of thinking are associated with rating agen­cies. The focus on particular aspects of rating agencies—rather than on the positing of universal laws about agencies "in general"—means that the research design of this book is "realistic" and inductive. The design does not aspire to the "hypo- thetico-deductive mode of theory construction" that dominates much of social sci­ence.³² One way of viewing this book is as an exploration or probe that may help to create the basis for future hypothesis testing.

Substantively, this investigation is concerned with the veracity of different approaches, or general theoretical orientations, to motivation and action that are the subject of contemporary debate in the field of international political economy (IPE). These general theoretical orientations offer heuristics, in the form of relevant vari­ables and causal patterns that provide guidelines for research.

IPE has been dominated by rationalist approaches such as realism and liberalism, informed by economics, in which the heuristic is the struggle of rational actors with fixed preferences around scarce resources. This heuristic can be applied to any num­ber of problems as a guiding set of assumptions about what likely motivates an action." Here, this dominant rationalist lens is compared and contrasted with a very different general theoretical orientation.

This second approach draws on economic and organizational sociology and on the social sciences, other than economics. Rationalist approaches adopt the assump­tion that there is a one-to-one match between imputed material interests and social action. The constructivist approach can complement the instrumental cause-effect focus of rationalism. The heuristic focuses on processes through which the prefer­ences and subsequent strategies of actors (such as corporations and states) are socially constructed, varying over time and space, and defining the identity or nature of the actors in relation to others.³⁴The norms, identity, knowledge, and culture that comprise intersubjective structures—things held constant in rationalism—are among the things that constitute or regulate actors in this general theoretical orientation.³⁵

Both rationalism and constructivism are, as it is seen in subsequent chapters, essential for understanding bond rating agencies. The constructivist lens has, how­ever, so far been neglected in IPE, to our detriment.³⁶ In part, this book is an effort to correct that omission and to demonstrate the analytical contribution construc­tivist social science can make to IPE research.

International political economy started as a study of foreign economic policy, mainly of the United States and the European powers. These origins have led to IPE being dominated by the view that markets are very different from the typical insti­tutionalized manifestations of politics, like political parties and government bodies such as houses of representatives. Unlike most economists, many IPE specialists have been interested in the interaction of the economic and political spheres (under­stood as different motivations), which scholars with diverse approaches have thought were neglected.

Only through an analysis of this interaction could an understanding of interna­tional economic relations be formed, one that included many more variables than those economists have focused on. This area of analysis has contributed much to our understanding of the developing global order since World War II, especially of the creation and decay of the Bretton Woods regime. But global markets have developed and states have changed in form and behavior during the three decades since the end of that regime. Consequently, the strict separation of IPE subject matter, into a "states" category on the one hand and "markets" on the other, has become problem­atic. Increasingly, IPE thinkers have been concerned with intermediary institutions that are neither states nor markets but interact with both.³⁷ Some scholars have also looked at the economic sphere, to reappraise inherited notions of what markets are and how they work.³⁸

Economic sociology offers an alternative theoretical source for analytical insight.³⁹ The prime benefit it offers in abstract terms is to ground the agentcentric understanding (of states, of companies, of individuals) implicit in traditional IPE in a structure emerging from social relations. Waltz and international relations Neorealism offered a sense of structure. But that structure did not encompass market rela­tions and tended to minimize the role of actors other than states, even if the formal account of the approach gave space to other agents.⁴⁰

By contrast with the Neorealist vision of an anarchy of self-regarding units, the notion of "embeddedness" Granovetter identified—a key concept in economic soci­ology—sought to link institutions to the social relations in which they existed.⁴¹ In this understanding, economic life was not separate from society like a free-standing machine but was linked to historical and cultural circumstances and, therefore, vari­able over time and space.⁴² However, despite embeddedness, economic and institu­tional sociology has produced "evidence of global cultural homogenization."⁴³ This process of change is linked to pervasive myths or mental frameworks, which legiti­mate specific organizational forms (and negate others).

Mental or intersubjective frameworks are just as consequential as other social structures. As W. I. Thomas noted in 1928, "If men [sic] define situations as real, they are real in their consequences." Thomas claimed that people respond not just to objective things, like mountains and automobile accidents, "but also, and often mainly," to their collective attribution of meaning to the situation. As Coser points out, if people think witches are real, "such beliefs have tangible consequences."⁴⁴

The importance of mental frameworks is reflected within institutions. Meyer and Rowan argue that organizations and how they are structured reflect not the efficient undertaking of their function but the myths or mental frameworks that depict a pub­lic story about the organization.⁴¹ Internal rules and organizational forms within institutions reflect "the prescriptions of myths." These rules and organizational forms demonstrate that the organization is acting "in a proper and adequate man­ner." By conforming to the myth, the organization protects itself from interroga­tion. The key process is identifying elements of the myth and then reconfiguring the organization around them. Organizations, Meyer and Rowan suggest, typically face dilemmas between the prescriptions of these elements and their internal, shared sense of what they are really supposed to be about, and also between diverse com­peting myths held by different parts of society, such as government, interest groups, and market associations.⁴⁶

Professional judgment and analysis—and public expectations about its develop­ment and standards—is a key, societally legitimated rationalized element of the rat­ing agencies' mental framework. One conception of how this framework can be understood in its wider social context is through what Peter Haas and his fellow con­tributors have called epistemic communities.⁴⁷ Haas defines epistemic communities as "networks of knowledge-based experts" that address complex, seemingly techni­cal problems. The "recognized expertise and competence" of these professionals give them an authoritative claim to offering good advice, and their control of expert­ise is "an important dimension of power."

Haas suggests four features of epistemic communities: a shared set of normative and principled beliefs, shared causal beliefs, shared notions of validity in the area of expertise, and a "common policy enterprise" connected to enhancing human wel­fare. Epistemic communities neither guess nor produce data but interpret phenom­ena. The major role of the communities lies in ostensibly "less politically motivated cases," where they introduce a range of policy alternatives.⁴⁸ The communities dif­fer from the concept of profession in that they share normative commitments but such commitments may develop within professions (for example, the subset of econ­omists concerned with economic inequalities).

A normative element also distinguishes epistemic communities from other con­cepts such as policy entrepreneur.⁴⁹Haas argues that the communities do not behave as rational choice or principal-agent theory would predict because of the central role attributed to their beliefs. Epistemic communities are important in themselves because they "convey new patterns of reasoning" to policymakers and "encourage them to pursue new paths of policymaking," with unpredictable outcomes.⁵⁰

The concept of epistemic communities is relevant to this book's focus on patterns of reasoning, on the politics of technical expertise, and on the power that emanates from knowledge. However, this book parts company with epistemic communities over the key concept of normative beliefs. A subset of raters may share a conscious commitment to such beliefs, but this commitment is a defining element of epistemic communities. The notion of epistemic communities may be useful to the analysis of particular elements within the rating world to be examined in future work. An alter­native concept—embedded knowledge networks—is elaborated below.

Embedded Knowledge Networks

Embedded knowledge networks are analytical and judgmental systems that, in prin­ciple, remain at arms length from market transactions. "Embedded" does not mean that the networks are locked in and, thus, simply resistant to change. "Embedded" should not convey the idea of inertia, path dependency, or vested interests. Instead, it is supposed to suggest that actors view embedded knowledge networks as endoge­nous rather than exogenous to financial globalization. The networks are, therefore, generally considered legitimate rather than imposed entities by market participants.

How the networks construct and reinforce this collective understanding of them­selves is of great interest. Where institutions that are embedded knowledge networks in one society attempt to transplant themselves into others, they risk losing their embedded knowledge network status, unless they recognize the necessity of getting the market actors in these other places to recognize their endogeneity. To return to the discussion of myth and mental frameworks, rating agencies must adapt them­selves to public expectations of what they should be doing, as they expand from their American home base. Achieving endogeneity and, hence, legitimacy has been easier in some places than others for the major U.S. bond rating agencies.

The role of knowledge in investment decision-making is at the heart of embed­ded knowledge network activity. Market actors are overwhelmed with data about prices, business activity, and political risk. A typical form of knowledge output is some sort of recommendation, ranking, or rating, which ostensibly condenses these forms of knowledge. This knowledge output becomes a benchmark around which market players subsequently organize their affairs. Market actors can and do depart from the benchmarks, but these still set the standard for the work of other actors, providing a measure of market success or failure. In this way, embedded knowledge network outputs play a crucial role in constructing markets in a context of less-than- perfect information and considerable uncertainty about the future.

Rating agencies, acting as embedded knowledge networks, can be thought to adjust the "ground rules" inside international capital markets, thereby shaping the internal organization and behavior of institutions seeking funds. The agencies' views on what is acceptable shape the actions of those seeking their positive response. This anticipation effect or structural power is reflected in capital market participants' understanding of the agencies' views and expectations. In turn, this understanding acts as a base point from which business and policy initiatives are developed. The coordination effect of rating agencies therefore narrows the expectations of credi­tors and debtors to a well-understood or transparent set of norms, shared among all parties. Thus, the agencies do not just constrain the capital markets but actually provide significant pressures on market participants, contributing to their internal constitution.

Counterfactual Method

How might rationalist and constructivist analytical lenses be deployed in this sub­stantive discussion of rating institutions? Since the objective is to understand the implications of the particular rather than establish general laws, we need a suitable method of thinking through the implications of rating. For the type of cases described in this book, counterfactual analysis is an appropriate approach.⁵¹ In coun­terfactual analysis, the factor or variable thought most likely to be causal is subse­quently excluded from an alternative scenario the researcher constructs.⁵² Given this modification of what Weber terms the "causal components," we have to think through whether, in these changed conditions, the "same effect" would be expected empirically.⁵³ If, in the imaginative construct established, the supposition is that the effect would probably be different, we have likely isolated an adequate cause in the initial scenario and can feel confident about the analysis. But, as Weber cautions, causal significance of this sort always suggests a range of degree of certainty about causation.⁵⁴

One objection to counterfactual scenarios is, as Ferguson notes, the notion that "there is no limit to the number which we can consider." But the reality is quite dif­ferent. "In practice," suggests Ferguson, "there is no real point in asking most of the possible counterfactual questions" that can be imagined. Plausibility is key, as in all analysis. We are interested in what happened or could have happened, not what could not have happened. Our focus should be on "possibilities which seemed prob­able." Accordingly, there is a plausible set of counterfactuals, not an infinite number of alternatives for any situation. Even if we grant that this plausible set is always open to critique, by requiring us to rethink our arguments, the posing of counterfactuals is, as Ferguson suggests, a useful "antidote to determinism."⁵⁶

In the substantive chapters of this book, a rationalist account of rating agency effects is constrasted with a constructivist one inspired by economic sociology. The purpose is to demonstrate the utility of a constructivist-economic sociology analy­sis of rating agencies and, thus, of IPE problems more generally. Since constructivist accounts are not always better than rationalist ones, the working assumption is that the constructivist-economic sociology heuristic complements the rationalist account. In some cases, the most plausible explanation may be rationalist rather than constructivist.

Central and Supporting Arguments

Economists have been keenly interested in the question whether bond ratings actu­ally add new information to markets and thus affect market behavior. The central argument of this book concerns the intersubjective effect of rating, that is, how rat­ing affects the social context in which corporate and government policy plans are made. Specific attention is given to the power and authority of the agencies, and the implications of rating for private and public life.

Rating agencies are not the neutral, technical, detached, objective arbitrators they are assumed to be among people who see them as merely transmitting market views to investors. Capital markets (and other markets) are actually organized, coor­dinated, or "made" by processes of information gathering and judgment forming the rating agencies exemplify. These processes reflect particular ways of thinking and reject or exclude other ways.⁵⁷ The judgments produced acquire the status of understood facts in the markets—even when analysis shows they are at times faulty—because of the authoritative status market participants and societies attrib­ute to the agencies. These particular ways of thinking, which are hegemonic in the Western world and which the agencies enforce increasingly internationally, are referred to here as the mental framework of rating orthodoxy.

Most broadly, it is argued, the work of the agencies integrates further elements of economic and political organization around the world, pushing these toward a prevailing institutional pattern. In this emerging order, norms are increasingly shared, and policy converges around characteristically American "best practice."⁵⁸ American ideas have become the most important transnational ones.⁵⁹ Rating agency judgments contribute to this process, as does the work of other institutions like the IMF and World Bank.⁶⁰

Three supporting or mid-range arguments about the increasing importance of rating agency judgments are developed.

Supporting Argument 1: Investment

The first argument is concerned with investment, a central feature of any modern society that produces an economic surplus. Many economists, in the tradition of Hayek, assume that investment happens automatically if certain basic conditions hold.⁶¹ But investment may also be understood as an implicitly coordinated social process. Investment has its own history and particular constraints. It is therefore necessary to understand the context in which investment choices are made.

In current circumstances, the increasing importance of capital markets alters the basis on which investment is undertaken. As banks are displaced as key investment sources, gatekeeper power is concentrated in the hands of the small number of rat­ing agencies. Rating judgments are more important today and this trend will con­tinue into the future, because less investment capital in the form of loans is being allocated by banks. This change in the character of investment has significant con­sequences for corporations and governments seeking access to resources. Rating has become a key means of transmitting the policy orthodoxy of managerial best prac­tice. Much more of the world is now open to the consequences of rating judgments than was the case during the Cold War.

Centralization of investment judgment is the essential element of the first mid- range argument. This argument is supported by evidence from the relationship between corporate ratings and the cost of debt, Michael Milken's activities, and the rating of the automobile industry. Municipal rating adds further evidence of gate­keeping. The spread of the U.S. agencies and the emergence of local agencies in new markets also supports this first mid-range argument.

Supporting Argument II: Knowledge

The second argument is about knowledge. Knowledge is usually thought of as some­thing that transcends particular situations or times. In fact, certain forms of knowl­edge are more typical of some eras and places than others. Like the investment process, knowledge is a social creation, an arena in which particular understandings of the world compete for control over what is accepted as a basis for action and pol­icy. Politically, the key thing about knowledge is the moment when an idea changes from being an individual idiosyncratic view to one widely or intersubjectively held and collectively consequential. Rating judgments are not objective.

A specific form of knowledge at the heart of the rating phenomenon has conse­quences for what we think of as legitimate knowledge elsewhere in the policy process. The knowledge form that dominates rating tends to be analytical, to focus on how things do or should function in our world, in a cause-effect fashion. What this analytical form excludes is an explanation of origins: how institutions develop and also their potential for future transformation. Rating reinforces knowledge based on the assumption of a fundamentally unchanging world, one in which eco­nomic markets, for example, are thought to perform the same function today as "always."

Where did this specific form of knowledge originate? Sorel suggested the static, unhistorical way of thinking about knowledge is a technique linked to monetary accumulation.⁶² It eschews reflection and puts a premium on instrumental under­standing in the here and now. Capitalism is premised on such a knowledge form. But a static form of knowledge, under the changing conditions created by financial glob­alization, makes the capacity to anticipate the events of September 11, 2001, for example, inconceivable. Bond rating certainly did not create the static knowledge form, but rating agencies are transmitting and reinforcing this type of knowledge globally—with consequences for public and private policy around the world and, therefore, the daily lives of billions.

In the corporate world, the growth of the rating advisory industry and the rating of telecommunication firms support these claims. Problems with quality of life vari­ables provide evidence from municipal rating. The creation of Japanese rating agen­cies, discussed in chapter 6, also supports the claims.

Supporting Argument III: Governance

Ironically, perhaps, rating forces change in how we govern our lives because it spreads the static, instrumental form of knowledge, thus challenging established ways of thinking and acting. Governance is about how institutions or processes are organized in hierarehies and how these structures shape our lives as citizens and consumers. Sometimes, the existence of these governance structures is obvious, such as in the case of representative democracy. Other governance structures are quite diffuse. They operate in society much as operating systems do in computers, beneath the surface of things. The third mid-range claim made in this book is that established, historically derived norms and practices regarding governance are chal­lenged by the judgments of rating agencies: the views of appropriate constitutional arrangements and corporate governance approaches that the agencies promote are often derived from U.S. experience. Rating agencies did not invent these governance structures but act as interpreters, advocates, and enforcers of them around the world. When put in place, these patterns shape the nature of working life and the limits of democracy, making the former more insecure and competitive and the lat­ter less inclusive and meaningful.

In chapter 4, the problems in Japanese banks support this mid-range argument. New York City's financial problems provide evidence from the municipal world. Additional evidence from controversies over sovereign ratings, for both rich coun­tries such as Japan, Australia, and Canada, and developing countries, is given in chapter 6.

Plan of the Book

The three supporting arguments are developed in the chapters where the implica­tions of rating processes in particular contexts are investigated. First, however, in chapters 2 and 3, the book looks at the agencies: their internal organization, impor­tant features of their processes for creating ratings, and the relation between rating and regulation. Next to be examined is the power and authority of rating agencies, which underpin the mid-range issues. The key question asked is from what is rating power derived, and what are its limits? The context is development of the arguments about investment, knowledge, and governance to be investigated in subsequent chapters.

In chapters 4, 5, and 6, the discussion concerns corporate rating, municipal rat­ing, the sovereign rating of national governments, and the growth of rating agencies outside the United States. These accounts are organized in terms of the arguments about investment, knowledge, and governance that are developed in chapter 3.

This book focuses on more than one level of analysis, such as sovereign states. The politics of rating pervades the world order, requiring that we consider the effects on municipal government and private corporations as well as national states. In chapter 7, recent rating "failures" are explored. Why did these failures happen, what marks them as failures, and to what degree have they undermined rating authority? Just how resilient are the reputational assets the agencies possess?

Ironically, the financial crises of the 1990s and of the early years of the new mil­lennium may have actually enhanced the power of rating agencies: capital market financing has come to be seen as less risky than traditional bank lending, especially in emerging markets. Underlying market trends may be rescuing the rating agencies from their critics, even as the increasing importance of rating motivates further criticism.

The concluding chapter examines the significance of the discussion, in particu­lar the degree to which authority and power take on new forms in globalized condi­tions and how the agencies affect people in their everyday lives. What are some ways of responding effectively to the heightened role of the new masters of capital?