Rating agencies – don’t blame the messenger!
Don’t get me wrong. Undeniably, the big international rating agencies pose a problem to world financial markets and policy makers. Their systemic importance, oligopolistic structure and scope of influence call for close scrutiny, questioning of motives and cautious interpretation of results. On the other hand, the opaqueness and complexity of markets, and the growing tendency to publicly attack Moody’s, S&P and Fitch out of anger and desperation is an invitation to policy makers and stakeholders to pursue own interests and in loudly joining the critics to distract from the root of the problem.
With increasing vulnerability of the world financial system, and the growing number and severeness of financial crises in recent years, criticism of rating agencies became ever harsher. It culminated when on two successive days on July 4th and 5th Standard & Poor’s first thwarted a French debt rollover plan for Greece and then Moody’s in neglecting economic progress downgraded Portugal’s debt to junk, lowering its rating by four notches from Baa1 to Ba2. Critics condemned the latter as an “act of vandalism”, a “punch in the stomach”, and called for breaking the oligopoly of rating agencies, banning them from rating countries and even “dissolving” them. On the opposite side, for instance, Ambrose Evans-Pritchard in the Telegraph titled “Europe declares war on rating agencies” calling the attacks “sinister repression” of the agencies.
In the heat of the debate economic and political interests collide and arguments become increasingly blurred. With outrage and uncertainty fuelled by markets testing one domino after the other, and spreads reaching ever new highs, there is a danger that in blaming the messenger one of the major achievements of modern financial markets gets severely damaged without solving the underlying problems.
From railroads to residential mortgages
The rating business dates back to the early days of industrialization in the United States when the railroad industry began to expand across the continent and into undeveloped territories and raised capital through the bond market. In 1860 Henri Varnum established Poor’s Publishing, which later became Standard and Poor’s. In 1868 he published the Manual of the Railroads in the United States containing operating and financial statistics and other information on the business conditions of the newly emerging borrowers. Moody’s Investor Services was established in New York in 1900 and, in 1909, became the first to issue credit ratings in the United States. Fitch Publishing Company, the predecessor of the third of today’s Big Three, was founded in 1913, again in New York. In 1924, the firm introduced the now familiar rating scale from AAA to D. In 1997, Fitch merged with IBCA, London. Today it is owned by a French holding company, FIMALAC, and has dual headquarters in New York and London, with the latter serving as a counterargument to an alleged anti-European bias of the Big Three.
Historically, rating agencies generated their revenues from subscription fees from investors for ratings on the creditworthiness of issuers and related research. Since the 1970s, they have been increasingly paid by the issuers themselves who regard ratings as an instrument to enhance investor confidence. The resulting inherent conflict of interest is one point of criticism. Over the past decade, the nature of the rating business changed even more as ratings for structured finance products now account for a remarkable share of revenues. The implication to the integrity of the rating process is a second point of criticism: As business evolved “from the rating of dynamic and on-going enterprises to static and defined-lived structured assets” (Mason and Rosner (2007), and with the subprime crisis in the US unfolding, doubts arose on the rigors of the agencies’ structured finance rating models and on their historic claims of being publishers (which in the US are protected by the First Amendment’s protections of the freedom of the press).
The agencies’ role is favoured by legislation and financial regulation which “enshrine credit-rating agencies as gatekeepers to the capital markets” (Douglas Elliott). For instance, investments of insurance companies and pension funds are based on ratings with the rules requiring immediate portfolio adjustments after downgrades. As a consequence, the often cited self-fulfilling prophecies of market reactions after downgrading are partly rooted in this quasi automatism. Stress tests are based on ratings. Basel II made it near mandatory for banks to use the ratings of the Big Three when calculating capital ratios. The concept allows banks to choose between a standardised approach to measure credit risks and an internal ratings based (IRB) approach allowing them to use their own internal models for risk assessment. Under the standardised approach external ratings by a recognized rating agency serve to measure the risk sensitivity of corporate, sovereign and interbank exposures. For unrated corporate and sovereign exposures risk weights are 100% (table 1).
Source: Jackson (2002)
Ratings are expert opinions about the creditworthiness of sovereigns, institutions and financial instruments. In this, they are one determinant of investment decisions – but not the only one. For example, in bond markets, other important influences are liquidity or price volatility. They explain why market prices for bonds with the same rating may differ considerably (Katz et al. 2009).
The rating process consists of two components: quantitative assessments of credit risk and expert judgments of a ratings committee. The fallacies became apparent with the Enron and WorldCom scandals in the early 2000s. Both firms had investment-grade ratings until just before their collapse. However, more recent events indicate that alternative inhouse models based on value-at-risk analyses which are characterised by high complexity, partly unrealistic assumptions and a crude combination of market valuations and synthetic prices are fraught with pitfalls, too. These days, many observers consider the choice between agency ratings and internal models a choice between Scylla and Charybdis.
Methodologies differ from agency to agency which is one explanation why ratings may differ and why, for example, up- or downgrading by one agency is not necessarily followed by the others (table 2 shows differences in methodologies for bank ratings). For instance, Fitch was the first to engage in explicit assessment of systemic risk. On the other hand, the firm has drawn some criticism for the use of sophisticated econometrics and neural networks in cases of limited samples of sovereign nations and sovereign defaults. In 2007, Moody’s introduced a new bank rating methodology ahead of the financial crisis systematically taking into account external support available to banks. S&P is currently revising its methodology substantially in reaction to the crisis which is expected – among other things – to alter the geographical distribution of potential rating actions with Asian banks tending to be upgraded and European banks downgraded. The latter can be expected to reinforce existing pressures on the agencies.
Table 2 Source: Packer and Tarashev (2011)
Beside the Big Three there are an estimated 130 to 150 credit rating agencies in the world. Many of them focus on special sectors, products or regions. Only the Big Three are recognized worldwide. They account for 95% of global ratings and over 90% of revenues. These days, in some markets, risks related to rating agency power are aggravated by a similar concentration of other actors. One example is structured finance. To cite a report by the French securities regulator from Rosner and Mason (2007):
“12 banks account for more than 70 percent of European deals, and three rating agencies cover the entire market (two of them accounting for 80 percent). … [in the French market] three legal firms account for more than 60 percent of the legal structuring work in the CDO market, and three others account for more than 50 percent of volume in the RMBS market.”
It is this “closed-shop” scenario which raises suspicions calling for strict rules and surveillance. On the other hand, the Big Three are few, and big, but they rarely act in a synchronized manner as, for example is observed in oil cartels and other oligopolies:
In a recent BIS study, Packer and Tarashev found considerable disagreements among rating agencies. Only in 8% of cases the Big Three assigned the same rating. For 33% of banks ratings differed by two notches or more. These results stand in contrast to the widespread view of an oligopoly where firms collude to influence markets, but in face of a growing number of countries on the brink of default and misery they probably do not suffice to resolve public concerns.
Too big to rate?
Growing uneasiness about power concentration is one reason why users become more and more prepared to reduce existing dependencies and allow more competition in this field. Another reason appears to be increasing political pressure as agencies refuse to accomodate political strategies. When Moody’s and S&P resisted the smooth bailout of French and German financial institutions in the current crisis, justified or not, this was the proverbial straw that broke the camel’s back.
Currently, more and more rating agencies are getting involved into the official rating business. The ECB includes the ratings of a fourth agency, the Canadian Dominion Bond Rating Service (DBRS). DBRS is also officially approved by the German BaFin – in addition to the Japan Credit Rating Agency Ltd and the German Euler Hermes Rating GmbH (a company of the Allianz Group). In 2011, three new German rating agencies were added: the Creditreform Rating AG, the Feri EuroRating Services AG and the PSR Rating GmbH. In Switzerland, beside the Big Three and DBRS FINMA recognised the Swiss Fedafin. After an examination process by BaFin and the Committee of European Securities Regulators (CESR) in November 2010 Euler Hermes became the first European registered credit rating agency. It was followed by the Japan Credit Rating Agency and Feri EuroRating Services in January and April 2011 respectively. These three are recognized as External Credit Assessment Institutes (ECAI), eligible to provide a basis for capital requirement calculations under Basel II European-wide.
Competition is also emerging in other parts of the world. Dagong Global Credit Rating Co., Ltd. from China caught the headlines when in July 2010 it published its first ever sovereign risk assessment giving its own government a higher debt rating than the US, UK and Japan – and the Netherlands and Germany a lower rating than the Big Three (table 3). Source: Alloway (2010)
The Dagong example demonstrates one aspect which is easily overlooked in the heat of the current debate: More competition among rating agencies, and a greater number, does not necessarily mean more favourable ratings. Exchanging a “US bias” against a “European” or an “Asian” one would aggravate the problem rather than contribute to the solution. In the end, surely only an Italian rating agency could truly judge Italian creditworthiness? And, given German dominance in the eurozone, shouldn’t the business rather be assigned to a German rating agency?
The answer is no. Even regarded from outside, there is no denying that the eurozone is in big trouble, that the most vulnerable countries at the periphery pose a considerable systemic risk, which in the current situation is not necessarily grounded in economic fundamentals, and that the big and comparably healthy core members refuse a sustainable solution. The dilemma has a name: In game theory the prevailing conflict between individual and collective rationality is known as the tragedy of commons. Ratings that do not reflect this situation are not worth the paper they are written on, and choosing a more “accomodating” rating agency only postpones necessary revisions and adjustments as markets expect, and again and again will try to test, a “political” solution.
As I wrote elsewhere, the financial industry is a global one – actually, it has long been THE example of a global industry – and assessing global firms with headquarters in different world places, and ever-changing business environments, requires a global view and a familiarity with networks and systems worldwide. Rating and rated must meet on an equal footing. In this respect, the Big Three have clear first-mover and size advantages.
In a globalized world, in many sectors and businesses size is an indispensible prerequisite for economic success: Exploration projects need great financial strength and persistence before investments bear fruit. In manufacturing scale economies determine international competitiveness. In international banking size allows to take the risks related to being a major player in world markets, to provide the financial infrastructure needed for global trade and investment and – as current events demonstrate – to survive the wide swings in market activity in times of crisis.
In the international ratings business it is a bit of all: High fix cost calls for the exploitation of scale economies, the global nature of clients’ activities requires a worldwide net of analysts, and for rating complex strategies and activities comprehensive methodologies are needed.
The advantage to reduce the dependence from the Big Three comes at a price. The tendency to increase the number of rating agencies, in particular when accompanied by introducing a national or regional element, bears considerable dangers:
Firstly it risks further destabilising the international system in increasing the number of voices and spectrum of opinions. Currently, one of the eurozone’s big problems is the cacophony of messages coming from Brussels and Frankfurt, from individual member country politicians, central banks, analysts and scholars. This will become even louder with the number of agencies increasing.
Secondly, asking for more “opinions” regarding the creditworthiness of countries and banks will reduce rather than increase transparency and make decisions for the uninformed investor or policy maker even more difficult.
Recently, an unfortunate tendency emerged among politicians and outside observers to tell the public what is right or wrong in this field. Furthermore, jumping on the bandwagon academic studies provide a “scientific” rationale for criticism: For instance, estimating relationships between sovereign debt ratings and macroeconomic and structural variables and then stating that downgrades have been unjustified ignores the very concept of systemic risk on which recent ratings on Portugal and other periphery countries were based thereby unduly reinforcing a general impression of agencies’ failure or wrongdoing.
Rating agencies are accused to neglect economic progress when downgrading countries. If the experts foresee that economic progress will play a minor role in market decisions, and other factors determining cost and access to finance will dominate, it is their very task to signal this result. Otherwise investors’ confidence in them were no longer justified. Further, rating agencies are blamed to accelerate trends. Of course, they do! If they do their job, even if no recommendation is intended, their results inevitably influence investor decisions and markets. The big challenge is to assure that they do their job well and that their results reflect existing risks and not special interests.
Size matters in this business and restricting size would mean restricting investors’ access to indispensable information and judgment of complex strategies, products and relations. The times when ratings were used to increase financial market efficiency by sparing investors the cost of gathering and processing information are long gone. These days, without the expertise and competence of rating agencies, their economic strength and wide networks, investors (and regulators, policy makers and analysts) are cut off from vital knowledge.
The emergence and strengthening of smaller competitors, and the growing tendency of investors to reinforce inhouse research, are an improvement in that they widen the pool of expertise. Nevertheless, the world of finance cannot do without a benchmark. Making this benchmark more reliable, and its functioning more transparent, calls for joint efforts to redefine rules and rethink structures – not for a lumberjack.
—–Additional introductory material can be found in Reszat (2005), Appendix D.
From → Markets