研究痛批穆迪嫌贫爱富

    标普（Standard & Poor）近期下调美国主权信用评级后，各界再次开始热议评级机构是否值得信赖：它们是否能够公平地评估各类债券？一份新调查可能火上浇油：印第安纳大学（Indiana University）、美利坚大学（American University）和莱斯大学（Rice）的教授发现，过去三十年里，三大评级机构中至少有一家——穆迪投资者服务（Moody's Investors Service）给私营部门债券的信用评分相对来说高于政府债券，有“注水”之嫌。

    更为致命的结论是：为评级机构带来更多收益的资产所获评级比那些无利可图的证券更为有利。“某种资产给评级机构带来的收入越多，评级乐观主义（宽大或注水）也就越明显，”研究报告的执笔者们写道，“证据强有力地表明，结构性产品的评级比公司债券要宽大得多（乐观），而市政或主权债券的评级显然更为严厉（更加悲观）。”

    尽管标普、穆迪和惠誉（Fitch）使用不同评级体系来衡量各种类型债券的信用，但所有资产类别都适用其自定标准。也就是说，当穆迪将一种债务担保证券、一种公司债券和一种主权债券均评为AA级，那就意味着穆迪认为这三种债券的信用程度是相同的。

    可是，乔•巴莱斯特里诺称，专业投资者长期以来对这些评级的共通性一直持怀疑态度。他效力于管理着3,550亿美元资产的投资管理公司Federated Investors，担任首席固定收益策略师，他指出：“各类资产之间存在不平衡或差别待遇，市政债券是最突出的例子。”

    几十年来，穆迪给市政债券适用了一套单独的标尺，使这些债券的评级颇有被低估之嫌。该机构还称，经验丰富的投资者理解市政债券的违约可能性要比评级相同的公司债券低。

    政府官员对此很不认同。2008年，康涅狄格州总检察长理查德•布卢门撒尔起诉了三大评级机构，援引一些内部研究，称这三家公司明知市政债券的违约率远低于企业债券。加利福尼亚州财政部长比尔•洛克耶则发起一场运动，试图迫使评级机构修改评级体系。

    最后，穆迪和惠誉（后者也为市政债券适用了一种单独的标准）作了让步。2010年，这两家机构调整了市政债券评级体系，使其与公司债券的评级标准更一致，引发数千种政府债券的评级上调。标普则坚持其统一评级标准并不歧视市政债券。

    不过，上述研究的结论不仅涵盖了市政债券。“这一事实没有变，主权债券相比公司债券而言遭到了不公待遇，而公司债券又不如结构性产品那么受优，”报告作者之一杰斯•科尔纳贾如是说。（尽管该研究只使用了穆迪的数据，但科尔纳贾认为其结论很可能也适用于标普和惠誉，因为三家机构的评级结果通常是相似的。）

    科尔纳贾及报告的其他作者分析了穆迪在1980-2010年间给出的信用评级，然后比较获得同等评级的不同种类资产的违约率，他们发现了巨大的不一致性。例如，被评为“A”级的资产中，只有0.49%的市政债券违约，主权债券从未违约，而获得同种评级的公司债券中有1.83%违约，金融债券有4.92%违约，结构性产品的违约率更是高达27.2%。

    The recent downgrade of U.S. debt by Standard & Poor's reignited the debate over whether ratings agencies can be trusted to fairly assess different types of debt. A new study could add fuel to the fire: Professors at Indiana University, American University and Rice found that, over the last 30 years, at least one of the big three agencies, Moody's Investors Service, inflated the credit scores of private debt relative to public bonds.

    Even more damning is their conclusion that assets that generated greater proceeds for the ratings agency were rated more leniently than less lucrative securities. "Ratings optimism (leniency or inflation) increases in the revenue generation by asset class," wrote the authors. "The evidence overwhelmingly suggests that while ratings of structured products were significantly more generous (optimistic) than those assigned to corporate issues, those assigned to municipals and sovereign issuers were significantly less generous (more pessimistic)."

    Although S&P, Moody's and Fitch use different letter systems to gauge the health of various types of bonds, each applies its own individual criteria across asset categories. So when Moody's says a collateralized debt obligation, a corporate bond and a sovereign issue are all rated AA, the implication is that the three types of bonds are equally healthy.

    But professional investors have long been dubious of the ratings' commonality, according to Joe Balestrino, the chief fixed income strategist at Federated Investors, an investment management firm with $355 billion in assets. "There are inconsistencies or discrepancies from one asset class to another," he says. "Munis are the most glaring example."

    For decades, Moody's used a separate scale for municipal debt, which gave the bonds the appearance of being underrated. The agency argued that sophisticated investors understood that municipal issues were less likely to default than similarly rated corporate issues.

    Public officials disagreed. In 2008, Conn. Attorney General Richard Blumenthal sued the big three ratings agencies, citing several internal studies that suggested the companies were aware that municipal bonds were defaulting at a much lower rate than private issues. Bill Lockyer, the treasurer of the state of California, led a campaign to force the agencies to modify their scales.

    Eventually, Moody's -- and Fitch Ratings, which also used different criteria for munis -- caved. In 2010, both agencies recalibrated their municipal ratings systems so that they better corresponded to their corporate rating scales, which resulted in upgrades for thousands of public bonds. S&P maintains that its general ratings scale does not discriminate against municipal debt.

    But the study's findings go beyond municipal bonds. "The fact remains that sovereigns are getting the shaft relative to corporates, and corporates are getting the shaft relative to structured products," says Jess Cornaggia, one of the authors of the paper. (Though the study only used data from Moody's, Cornaggia says the results likely apply to S&P and Fitch, which tend to produce similar ratings).

    Cornaggia and his co-authors looked at the credit ratings Moody's assigned between 1980 and 2010, then compared the frequency at which different assets that received the same letter grade defaulted. They found large disparities. For example, while just 0.49% of municipal bonds and 0% of sovereign bonds that received an "A" rating defaulted, the rate was 1.83% for analogous corporate bonds, 4.92% for financial bonds and 27.2% for structured products.