Rating Sovereigns

As dark clouds gather on the horizon of the global economy in the third year of the pandemic—with debt stocks swollen, interest costs rising, and growth undermined by energy insecurity and war—policy makers and pundits are anxiously watching sovereign credit ratings as the harbingers of the storm. What will it take to save Italy from the fatal downgrade to junk status that could bring down not just the fourth largest economy in Europe, but the entire eurozone with it? Could much of the developing world be a few downgrades away from devastating debt crises? Such concerns take history as their guide. In the late 1990s, a series of rapid-fire sovereign downgrades triggered the Asian financial crisis. In the late 2000s and early 2010s, sovereign downgrades played similar havoc with European countries—pushing Greece into default, and forcing eurozone authorities to take desperate measures to save Cyprus, Ireland, Italy, Portugal, and Spain from the same fate. In both instances, sovereign ratings were blamed not only for having failed to predict the debt servicing difficulties of the countries in trouble, but also for exacerbating the crises via belated and panicked downgrades.¹ Sovereign ratings exercise spectacular—and often disastrous—power in times of crisis.

At the same time, sovereign ratings also exert subtler, but no less problematic influence over the fate of countries in “normal times.” The judgment that sovereign ratings pass about a country’s creditworthiness consistently affects the interest rate that the country has to pay on its debt. Lower ratings increase the burden of interest costs on the budget, shrinking the amount of money available to provide public services, address social needs, manage the economy or fulfill any other important political, social and economic objectives. The larger the outstanding debt, the greater the budgetary impact of ratings. Given the high and growing indebtedness of governments around the world in the past decades, even a couple of dozen basis points of increase in interest costs associated with adverse rating changes can make a tangible effect on the budget. Furthermore, sovereign ratings affect the financing costs of all economic actors across the domestic economy, thereby indirectly influencing the functioning of equity markets, growth, unemployment and competitiveness.² Therefore, governments face very strong incentives to try to stay in the good graces of the rating agencies. 

Indeed, rating agencies have been called the “new superpowers” of our globalized world, whose influence over governments’ financial elbow room rivals that of the best-known wielders of financial power, such as the International Monetary Fund or the World Bank. And just as the concern about international financial institutions is that they make funding conditional on policy choices that countries would not make on their own, the worry about credit-rating agencies is that they interfere with the sovereignty of democratically elected governments by tying favorable funding conditions to specific policy choices. Indeed, scholarly research has shown that the Big Three regularly comment on the politics and policy choices of the countries they rate, ³ and they assign lower ratings to countries with center-left governments ⁴ and large welfare commitments. ⁵ The fact that unelected, unappointed, for-profit commercial organizations can put a price tag on electoral and policy choices in democracies around the world should give us pause.

Sovereign ratings continue to exercise immense power over the fate of governments—in good times and bad—despite vigorous attempts in the wake of the last crisis to “break the power” of the “Big Three” credit rating agencies—Fitch, Moody’s and Standard and Poor’s—in Europe and the US alike. Incensed by what they claimed were unfair,  aggressive and harmful downgrades in the midst of financial, fiscal and economic troubles that the Big Three themselves were partially responsible for, policy makers committed to sidelining ratings from public regulatory systems, and promised to create new, better (public) rating institutions. US authorities brought billion-dollar lawsuits against the Big Three. Congress created extensive regulatory oversight over the credit-rating business, and eliminated the public use of credit ratings under the Dodd-Frank Wall Street Reform and Consumer Protection Act. In Europe, the Parliament and Council created the European Securities and Market Authority with the explicit purpose of regulating the credit rating industry as one of its main responsibilities. Yet, more than ten years later, the Big Three are just as influential (and profitable!) as ever, and once more the public braces itself for a potential new wave of sovereign debt crises triggered by downgrades.

The resilience of sovereign ratings in general and the Big Three in particular is all the more remarkable because—quite apart from regulatory action—investors themselves had strong incentives to break with ratings and the Big Three, given the grave losses they suffered in successive crises due to rating failures. The fact that markets continue to rely on the ratings of the Big Three, instead of trusting their own risk assessment, experimenting with alternative risk metrics, or availing of the services of other rating agencies has profoundly puzzled expert observers. The staying power of the Big Three is truly striking. 

A common language

Understanding why the Big Three became such influential, and apparently unassailable, gatekeepers of sovereign-debt markets starts with debunking a common (intuitive but misleading) conception about the role that ratings play in financial markets today. Ratings—with their well-known alphanumeric code—denote categories of credit risk, ranging from virtually riskless ‘AAA’ all the way to default-prone ‘C’. As indicators of risk, ratings are commonly construed as a decision-support tool for investors: a due diligence of investment options to identify and highlight risks that investors might otherwise be unaware of. This conception of ratings is reflected, for example, in the lawsuits and legislative changes that sought to make rating agencies liable for the losses suffered by investors as bond prices collapsed in the wake of successive rating failures, implying that drastic rating changes exposed fraudulent, negligent or incompetent handling of the due diligence task that the Big Three took upon themselves. ⁶

Interpreting ratings as a due diligence mechanism is intuitive but anachronistic. Historically, rating agencies did start out as fact-finding and -analyzing enterprises. Emerging in the late 19th and early 20th centuries in the United States, they helped investors assess the bonds of railroad companies and provided businessmen with basic information about little-known clients they contemplated extending commercial credit to.⁷ But the compilation and analysis of difficult-to-access information declined in utility in the 1970s, with the advent of the information age and the ascendancy of institutional investors in bond markets. Armed with practically limitless (often privileged) access to information, modern computing power and armies of highly trained analysts and portfolio managers, contemporary institutional investors no longer need to rely on rating agencies for information, insight or judgment. 

Yet, rather than make them redundant, the rise of the institutional investor has made credit ratings an indispensable component of contemporary financial markets. Ratings became a crucial coordination mechanism to facilitate the interactions of institutional investors with one another and with their stakeholders. Instead of unearthing novel information, ratings now serve as commonly accepted, third-party indicators of credit risk to enable institutional investors, their clients and regulators to negotiate about risk, enter transactions, and manage relationships characterized by asymmetric information. 

Since institutional investors invest money on behalf of others, they may be inclined to take higher risk than would be acceptable to their clients in the hope of realizing higher returns. Managing this moral hazard is of central concern to clients and to regulators, but devising private or public rules to regulate the risk-taking of institutional investors is predicated upon a common understanding of how risky each credit instrument is. Since clients and regulators cannot possibly keep track of the riskiness of innumerable bonds from across the globe in the portfolios of institutional investors, portfolio mandates and official regulations depend on ratings as commonly trusted third-party indicators of credit risk to define limits on risk-taking. Without an independent measure of credit risk, institutional investment on behalf of millions of investors would be unfeasible. Ratings serve to mitigate information asymmetries between institutional investors, clients, and regulators through independent risk estimates.

In dealings among institutional investors, ratings serve as a shorthand for defining minimum quality standards for collateral used to secure against counterparty risk in the myriad transactions (mostly repos) that make up a large part of the volume of activity in modern financial markets. Rather than mitigating information asymmetry, in this case ratings allow for efficient transactions without having to negotiate (and repeatedly renegotiate, as circumstances change) standards for acceptable collateral in each specific deal. Without a common standard for denoting risk, transactions among institutional investors would be too cumbersome to be practical.

What ratings do in both of these settings is—as David Beers, Standard and Poor’s Global Head of Sovereign and International Public Finance once put it—provide a “common language of credit risk”⁸ that actors enmeshed in a complex web of financial market transactions can use to negotiate and regulate their relationships. By supplying a common language, ratings play a crucial role in supporting the relationships and transactions that constitute contemporary financial markets. They play an infrastructural function comparable in significance to that of SWIFT, the interbank payment system. Whereas SWIFT physically transfers messages about transactions, ratings provide the vocabulary for defining vital terms of transactions and relationships. Sovereign ratings—and particularly sovereign ratings on the upper end of the rating scale—play a particularly important part in this vocabulary, because highly-rated sovereign bonds make up the bulk of the so-called safe assets that are most commonly used to safeguard client-investor relationships and inter-investor transactions. Without a common language for what constitutes safe enough assets to reassure clients, regulators and counterparties, these transactions and relationships—i.e. modern financial markets—could not function.

Empowered by convention and failure

Why do the Big Three retain a monopoly over this medium of communication? Although several dozen credit-rating agencies are registered across the US and Europe, the Big Three continue to overwhelmingly dominate the rating market, with a practically 100 percent share in sovereign ratings.⁹ (A curious feature of the rating market is that it is characterized by a joint monopoly of Fitch, Moody’s and Standard and Poor’s—with Moody’s and Standard and Poor’s both having a 100 percent of the market and Fitch holding a somewhat lower share. That is because the overwhelming majority of collateral standards and private or public regulatory documents call for at least two ratings and specifically require ratings by Fitch, Moody’s and Standard and Poor’s.)¹⁰ In the wake of the last crisis, new contenders threw their hats into the ring with some fanfare, but their initiatives petered out before they could properly set up their operations, while existing competitors continued to be constrained to narrow niches of the rating market.

The Big Three gained first mover advantage in the market for a global lingua franca of credit risk between the mid-1970s and the early 2000s, when they were the only “nationally recognized statistical rating organizations” in the United States.¹¹ Obliged to use the Big Three for public regulatory purposes, institutional investors fell back on the same method of risk certification in their own private portfolio mandates and contracts, too. Thus, the ratings of the Big Three became entrenched in both public and private use in the United States just as modern financial structures were born. Given the dominance of American financial markets and investors within the international financial system, the practice of using the ratings of the Big Three as commonly accepted indicators of risk was exported as globalization progressed. The world settled on the convention that when market actors need to discuss credit risk, they universally speak the language provided by the Big Three. This convention goes a long way towards explaining why no competitors organically grew into serious challengers even after the Securities and Exchange Commission opened up the status of “nationally recognized statistical rating organizations” to new applicants, and the European Securities and Market Authority registered several dozen new rating agencies. 

Convention alone cannot explain why the Big Three retained their monopoly over the global language of risk, after the language they provided repeatedly proved so fragile. “AAA” is supposed to mean practically zero credit risk, whereas “BBB” reflects tangible, if moderate, doubts about payment capacity. If a AAA-rated sovereign can become BBB-rated within a relatively short period of time (like Ireland or Spain did in the European debt crisis), “AAA-rated” no longer denotes the kind of safety that market actors meant to enshrine in their contracts and portfolio regulations. When ratings drastically change, market actors are forced to make hasty changes to their portfolios to adhere to the stipulations of the contracts and regulations that bind them. Such forced adjustments have the potential not only to impose losses on individual investors but also to destabilize entire markets by triggering simultaneous adjustments across innumerable portfolios and placing enormous pressure on the prices of the affected bonds. 

In light of the deleterious effects of failing ratings, any competitor able to offer more reliably accurate indicators of credit risk should have the potential to supplant the Big Three. But reliably accurate indicators of credit risk—especially sovereign credit risk—are a mirage. Credit relations are characterized by uncertainty, not risk: it is not possible to accurately account for all possible contingencies that might affect the future ability and willingness of debtors to service their debt over the lifetime of bonds and assign “realistic” probabilities to all conceivable outcomes. This is especially true for sovereign debtors whose debt servicing capacity is determined by the complex interaction of political, economic, and fiscal factors. Absent a crystal ball, sovereign ratings (or any other forms of sovereign credit analysis) are no more than educated guesses about the future credit standing of countries, liable to be proven wrong from time to time by adverse surprises. 

Furthermore, even if analysts compensate for the limits of their foresight by preemptively lowering ratings to reflect threats to debt servicing capacity from conceivable negative surprises, they cannot insure ratings against market panics. Whereas even the most dramatic political, economic or fiscal blows rarely cause a country with previously strong debt servicing capacity to renege on its debt commitments, a “sudden stop” of funding, caused by panic, might do that. “Sudden stops” make it impossible for countries to roll over their expiring debt at an interest rate that they can afford to pay, forcing them to default even if they otherwise have good credit standing. Maintaining investment-grade rating on a sovereign that defaults is the ultimate disgrace that can befall a rating agency. Therefore, at any signs of panic brewing, ratings have to precipitously fall (towards speculative grade) to reflect the growing possibility that the given country is careening into default—irrespective of what the country’s economic, political and fiscal fundamentals warrant. But, of course, falling ratings fuel further sell offs—as investors try to comply with margin calls on collateral and rating requirements in regulated portfolios—triggering further downgrades. Once panic starts, ratings fail.

Indeed, sovereign ratings have so far always failed by falling victim to panic. The Irish example is illustrative of this dynamic. Even though the shock that hit Ireland’s public finances was tremendous—by bailing out domestic banks, the government increased the country’s debt by almost a hundred percent of the GDP—the country’s debt servicing capacity was not fundamentally undermined. Once the dust settled after the crisis, Irish sovereign ratings settled on A+. This was not the coveted AAA that Ireland used to hold, but a decent investment-grade rating. However, during the crisis, as investor panic drove yields on Irish debt through the roof, there was a very real possibility that Ireland would lose its access to market funding altogether, triggering a rapid succession of downgrades, plunging Ireland from AAA in 2009 into non-investment grade territory by mid-2011, when markets were finally calmed by a bailout package from the troika of the International Monetary Fund, the European Commission and the European Central Bank.¹² Whereas some downward adjustment in the longer term was clearly warranted by the shock to public finances, the mayhem of the 10-notch drop within two years (a manifest rating failure) was driven by market panic incommensurate with the original shock that triggered it, as evidenced by the five-notch correction after the crisis. Other country cases in the Asian financial crisis and the European debt crisis illustrate the exact same pattern.¹³

Given the universality of this dynamic in times of crisis, there is little reason to hope that similar dynamics will not unfold in the future, or that alternative indicators of credit risk would not be vulnerable to this mechanism. Failures of credit risk indicators are baked into the markets. Consequently, even if the Big Three commit mistakes in their credit-risk analysis (which they certainly do), any improvement on their performance is ultimately bound to be marginal given the inevitability of failure of any indicators of credit risk in the face of bad news and market panic. The measurement of credit risk is an essentially illusory undertaking, predestining anyone that attempts it to recurrent failure. This fact explains not only the petering out of private initiatives for supplanting the Big Three, but also the manifest reluctance of public regulators to get involved in the production of risk indicators.¹⁴

The futility of attempts to produce reliable estimates of credit risk does not undermine the vital need of contemporary financial markets for independent, authoritative, third-party indicators of credit risk, to be used in negotiating the relationships of institutional investors with their clients, regulators, and one another. What it does is militate strongly against destabilizing the current entrenched convention around the Big Three as providers of the crucial common language of risk for the sake of experimenting with alternatives. Reluctance to rock the boat in the absence of better options helps to account for the timidity of authorities in enforcing regulations that would have undermined the willingness or ability of the Big Three to continue to provide their services. The most conspicuous example of this was the reversal of the decision under the Dodd-Frank Act: to make rating agencies legally liable for their opinions. The legislative change was promptly revoked when the Big Three threatened not to authorize the use of their ratings in prospectuses and debt registration statements, causing severe dislocations in the market.¹⁵ Other provisions of the Dodd-Frank reforms—like the tasking of the Securities and Exchange Commission with either changing the business model of rating agencies or creating a board that randomly assigns credit rating assignments to one of the nine nationally recognized statistical rating organizations—were quietly shelved without any attempt to implement them. Similarly, in Europe, regulations never went beyond technical details like requiring a set rating calendar or rules for disclosure. Regulatory forbearance underscores that, for all their faults and failures, the ratings of the Big Three are the only game in town. 

Reflexivity

The tenacity of the Big Three is full of paradoxes. The firms derive immense power and spectacular profits from having joint monopoly on an impossible task. The fact that they are bound to repeatedly fail at that task—with grave consequences for financial markets and entire national economies—neither diminishes the demand for their services, nor threatens their monopoly. In fact, inevitable failure further cements their position as pivotal actors in financial markets whose authoritative judgment on risk governs portfolios and contracts around the globe. Nevertheless, for all their authority, the Big Three are also hostages to markets given the immense vulnerability of their ratings to sudden adverse changes in market sentiment. When markets respond to unexpected bad news with signs of panic, foreshadowing the possibility of a “sudden stop,” ratings have to tumble to reflect impending disaster. But exactly by doing that, they unleash a meltdown of bond prices, fulfilling the prophecy of disaster. Caught in a predictable, yet unavoidable, vicious cycle, ratings drive themselves and markets into failure—becoming both the victims and the culprits of crises. 

While intriguing in its own right, the logic of credit ratings also highlights crucial vulnerabilities in contemporary global financial architecture. Beyond the fact that the transactions that make up the bulk of financial market activity around the globe are governed by a shaky illusion (that credit risk can be measured), the logic of ratings also calls attention to the heightened danger of intensified market reflexivity.¹⁶ Financial markets have always been complex systems of interconnected decisions in which investors try to anticipate what the market at large might do, generating potentially destabilizing movements. However, in contemporary global financial architecture, the proliferation of relationships and transactions that require a common language of risk has amplified the coordination of destabilizing developments around the globe. Investors were always watching the market, but there used to be some room for differences in perceptions that would dampen herd effects. With a vast share of portfolios legally forced to follow changes in ratings en masse, ratings became the focal point of reflexivity overpowering any stabilizing effects of potential differences in opinions. The more one contemplates ratings, the more reasons emerge to eliminate them altogether. Doing so without dismantling the broader global financial architecture that encases it, however, seems doubtful.