Ratings, an information business, have been linked to the growth of the largest capital market in the world since the 1920s. By 1995, the reach and power of rating agencies was such that Thomas Friedman commented in the New York Times, “you could almost say that 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 leveling it with bombs; Moody’s can destroy a country by downgrading its bonds.”
After The Analysis of Structured Securities came out in 2003, I shopped around the idea for a book that explained how ratings on structured securities would inevitably destroy the credit ratings business as we know it. Destroy—because the premise of structured ratings is statistical, and our Nationally Recognized Statistical Rating Organizations (NRSROs) compile data, they do not make statistical judgments. None of my colleagues at Moody’s had PhDs in statistics.
I would call the book Wizards of Odds. Our publisher, Oxford, was receptive, but mainstream publishers were not, for the patently obvious reason that structured finance and rating agencies were not, and never would be, front page news. Unfortunately, the truth is not always obvious, let alone marketable.
To wit: the SEC’s subprime mortgage crisis has nothing to do with mortgages. Yes, there is a credit crisis in the mortgage sector. It involves governments, pension funds, home owners and the banking system directly. However the SEC’s “subprime mortgage” problem is more fundamental. Its roots predate the mortgage bubble. So does rating agency malfeasance (however unwitting at the time) as anyone involved in the netherworld of a securitization-related litigation can attest.
The SEC’s “subprime mortgage” problem is that the crisis has exposed a very serious information problem in the capital markets, and a perverse truth is revealed: financial firms make less money from information than from exploiting loopholes in our fragmented, deeply flawed capital market information systems. The SEC is the guardian of these fragmented, flawed systems, so it is very difficult for the SEC to take any action at all.
But act it must. Capital markets are trading markets. The entire system of disclosure requirements from the SEC is based on static data. Static data are entirely ill-suited to trading. And, FASB 157 compels us to act as if we believe that prices generated from useless or inappropriate data are meaningful for valuation and reporting purposes. So, we are stuck in a wheel of self-interested, commercially viable disinformation.
To be fair, the SEC was nothing less than visionary in mandating static pool data disclosure for the Reg AB filing requirements. As measures of total risk, static pool data are superior to the accounting “information” used to compute average risk measures and hence easily manipulated.
But Reg AB does not go far enough. Structured securities are composed portfolios, they are not corporations. And data is not information. Information—ratings, FICO scores, Sharpe ratios—involves abstraction and synthesis. Financial information is intrinsically statistical. Hence, the seed of self-destruction argument.
The Sharpe ratio on structured securities backed by homogeneous asset pools (ABS, RMBS) will naturally transition to a higher or lower number. Normally, it will improve. That’s a loophole: good for the investor, bad for the seller. If you repackage the risk you add leverage, but the rating is likely to stay the same because nobody asked for a re-rating at the point of repackaging. That’s another loophole: good for the intermediary, neutral for the seller, potentially devastating to the investor. The basis risk between cash and synthetic CDOs is large but invisible since few people know how to model the cash. That’s yet another loophole of the same origin. It helps the banks pare risk and at the same time, teaches the hedge funds a lesson in humility. In the end, the rating agencies were not so biased as it seems. By failing to re-rate structured securities, they gave a little something to every one. And we are all losers for it.
Thus I fail to see how 17 CFR Parts 240 and 249b, Proposed Rules for Nationally Recognized Statistical Rating Organizations, solves what it purports to address. Surely the SEC knows, a real solution would dictate a massive overhaul of at least two information systems that we have relied on for information for over half a century. One is ratings. The other is GAAP. Neither is adequate for measuring risk in a continuous framework.
Our generation borrows more than it builds. Our institutional memory does not reach back more than a business cycle. “Technical expertise” in finance, accounting and taxation is the ability to exploit systemic loopholes: our experts would be challenged to propose real fixes, let alone integrate our patchwork information systems into the gold standard of efficiency that they would need for our capital markets to resume their intended function, of allocating scarce resources to worthy projects.
But if we were to start somewhere let us, as consumers, start by demanding access to good financial information—even if that only means that we learn first to discern it for ourselves.