August 12, 2013
Although rating agencies have been the subject of much criticism, including much from yours truly, (here for instance) there is no clearly superior replacement for the existing business model. Even worse, there is not even much theoretical work on alternatives. Thus, it is exciting to see that Becker and Opp have posted a new paper, Replacing Ratings, that at least considers a plausible alternative.
Their paper examines “a unique change in how capital requirements are assigned to insurance holdings of mortgage-backed securities. The change replaced credit ratings with regulator-paid risk assessments by Pimco and BlackRock.” (1) But their analysis did not “find evidence for more accurate inputs to regulation.” (3, emphasis removed) Indeed, their “empirical analysis reveals that the old system was better able to discriminate between risks. As a result, the old system based on ratings not only provided higher levels of capital, but also ensured that capital was more appropriately related to risks.” (3-4)
By the end of their analysis, they believe that “the new system only recognizes current (expected) losses, but does not provide any buffer against possible future losses. Our results are consistent with regulatory changes being largely driven by industry interests.” (21)
They find the new system is worse than the old system and that the new system benefits the industry. So why should we care about this research at all? For at least three reasons:
- it identified a change in the insurance industry that has implications way beyond that industry;
- it compared how two different MBS evaluation systems performed; and
- it identified the drawbacks of the new system.
This is how we begin to build a body of knowledge about “viable alternatives to ratings.” (2) But, of course, there is much more work to be done.