The idea behind rating agencies is a noble one. But the business model and operations are flawed, as analyzed and criticized heavily since 2008. I don't think this critique is complete. But here I will not dwell in these areas. Rather I'll focus on one area that I think merits more scrutiny, namely the fundamental flaw of the discrete, grade-based rating model.
Chaos theory has long discovered that the non-linear positive feedback loop is an essential ingredient for many types of instabilities in system dynamics. Discretization is an easy way to create the non-linearity, a convenient trick for creating chaos out of otherwise stable systems. The difference between 0.9999 and 1.0001 may be immaterial; but if you discretize the former into 0 and the latter 1, the difference is amplified 5,000 times.
This is in essence what the grade-based rating system does. The fact that creating instability may not have been the intention is irrelevant. And we've seen the artificially created positive feedback loop in play over and over, especially since 2008. Change in the underlying credit risk of an entity is inherently jumpy, driven by news, which is inherently discrete. But as if this instability is not exciting enough, the rating agencies come in and dramatically enhance the positive feedback loop by artificially bucketing the otherwise loosely continuous quantity of credit risk into a few arbitrarily designated grades. The destabilizing effect of this disastrous design is only made more fantastic by the unfortunate government sanction of these chaos generators and various requirements on who can hold what based on such ratings.
In some sense, the disastrous design of credit rating system is at the very core of the 2008 crisis. Rating downgrades, coupled with regulatory requirements, trigger mandatory sales of distressed assets, accelerating the death spiral. Without such artificial instability, the crisis would probably have been a tiny bit less severe and more manageable. Note that this point is independent of whether the rating agencies are ahead of or behind the curve, or whether they have done their job properly by grossly undervaluing correlation risk and turning a blind eye toward overt distribution manipulation such as barbelling of CDOs.
The grade-based rating system is a product of backward times when the cash bond market was obscure and illiquid. One positive effect of CDS is that it has made credit risk much more transparent and liquid. Despite being widely vilified since 08, the CDS market is superior to the rating system in every aspect, starting from the mathematics level. With the properly reformed and regulated CDS market trading credit risk, independent research digging beneath the surface, and the blogosphere disseminating info (and mainstream media occasionally pitching in), there's no longer a place for rating agencies, which only add negative value by exacerbating systemic instability.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.