Fixing the corporate credit ratings system via crowdsourcing

Credit ratings are already crowdsourced by the stock market. This article is actually talking about crowdsourcing credit rating _models_, which is actually quite different and may be a good idea. I don't buy all the hype in the article though..

>Even while Moody’s gave Lehman a clean bill of health, the alternative model showed the company was basically DOA. Andersen: The Lehman score was the lowest score we’ve ever seen.

In fact, I can't stand this type of backtesting analysis on credit risk. The reason that the model works and gives a low rating score is likely that it places a high value on the factors that caused the current recession, factors that are not likely to be the cause of the next major correction.

The real genius credit rating models are the ones that are predictive, rather than reactionary

The Big Three credit rating agencies (Moody's, S&P and Fitch) have been ripe for disruption for ages. Something's flagrantly broken when Lehman carries an A2 rating right up to its bankruptcy filing and AIG's troubles go unnoticed.

Maybe the transparency of open reporting standards (XBRL) and the insights of crowdsourcing (FreeRisk) can be combined to fix the system.

I wonder how easily their approach could be gamed by companies trying to improve their credit rating; presumably, this would be more difficult if there were several places that the 'crowdsourcing' was taking place?

Also, not to be too facetious, but the guy in the picture looks pretty smug and self-satisfied, which is a little offputting.

According to the article, freerisk.com is going to be a open data source for financial data taken from SEC filings. This is great for all types of fundamental analysis. I wonder why they're focusing so much on credit risk.

The use of a market system, like Inkling, would be an interesting approach to corporate credit ratings.