A Better Way to Rate Bonds

"Sunlight is the best disinfectant." – Justice Lewis Brandeis

A day before Lehman went bankrupt, they carried an A rating from Moody's on their debt. Likewise, AIG was rated AAA only 24 hours before they begged for their first multi-billion dollar tranche of bailout cash. Worse yet, the rescue bills have not created a new way of improving risk assessment for bonds, but rather have put a heavier reliance on the oligopoly of existing bond raters – Moody's, S&P, and DBRS.

Many articles have called for increased transparency in financial markets. A recent Wall Street Journal article by L. Gorgon Crovitz points out that FDR wisely opted for transparency rather than overburdening regulation. Crovitz also notes that XBRL, the more flexible and intelligent data standard for sharing financial information, might help make things more open. S&P itself has recently published a white paper suggesting three different models: issuer pay, subscriber pay, and government utility. The much-maligned SEC has always been torn between standardizing financial reporting to enable easy comparison across firms and allowing local complexity so that each firm can best disclose the unique challenges and financial situation they face. In the main, the SEC has correctly opted for specific openness over spurious standardization. XBRL can advance customized disclosure while allowing for easier comparability because digital formats allow for that kind of efficient customization. Put another way, XBRL can do for financial statements what HTML has done for pictures and text – allow local creativity, with global sharing.

But I think there is a better way – we can create a market in ratings.

Large institutional investors are often restricted to buying "investment grade" (often BBB or better) debt to fulfill their fiduciary responsibility to protect the assets under their management. Many can buy only A grade and up. In any event, the ratings for debt significantly affect the number of people who can invest in that product. In addition, the ratings agencies have a conflict of interest to "upgrade" the debt rating because they only get paid if the company issuing the obligation decides to use their rating system. So it's in the interest of S&P to "outbid" Moody's quality rating for the same cash flow stream. In addition, the rating companies are under no obligation to share the criteria by which they construct ratings as it is their proprietary intellectual property.

Two young geniuses, Toby Segaran and Jesper Andersen, have pointed out how odd this set of restrictions and incentives is and what harm it has brought to risk assessment in our debt markets. As a solution, they have suggested that we need two vital new "facilities" so that everyone can have easy access to much more information useful to rating bonds at a fraction of the effort and cost. They want to work with companies to help this happen, and to do it as a service for those companies that do not. First, they argue for a utility that makes corporate data much more accessible by accelerating the use of XBRL to report and publish financial data. Second, they believe we should have an "open market" in ratings, where individuals or firms can publish their debt ratings, and a leaderboard in which we collectively assess which ratings do the best job over time of predicting the outcomes of different debt instruments. They envision that, over time, the best ratings will win and subsequently become the new authorities. Those firms can then decide how to monetize their fame – through advertising, subscriptions, advice, or whatever model they want.

For instance, the movie rental company Netflix is offering a million dollar prize to anyone who can make a 10% improvement in its movie recommendation engine. Its leaderboard shows the progress of competing teams against that goal. Segaran and Andersen imagine a similar leaderboard for those who rate debt, and its performance. Such an approach would foster a competitive market in opinions on the quality of debt. To jump-start the market, we could also borrow Neflix's approach of offering a prize.

Over time, these ratings could become just as authoritative as the ones we have today and better at determining which bonds are truly "investment grade" and which are not. This would bring tremendous transparency and liquidity to the debt market and the ability for institutions to make better investment decisions.

This idea of taking a bureaucratic process like debt rating and making it into a market process is at the heart of capitalism itself. It is time to take this old corner of stuffiness and make it market ready; doing so could help us all assess risk more accurately and more efficiently. It won't get rid of the bond rating agencies – but it will make them compete for their privileged position. And as the grandson of immigrants, I like the idea of a meritocracy better than one where companies are born to their rank.

Also at HBS Publishing at http://tinyurl.com/cq5xjn

Wolframalpha vs. Google = Power vs. Simplicity?

Why Twitter is the Duct Tape of Marketing and Why Every Firm Needs to Know How to Use It