Why do the credit rating agencies move the sovereign debt market?
Markets play a simple yet crucial role: they price (or 'rate') assets, so as to ensure societal resources are allocated efficiently.
Markets do this better than any other institution yet devised because people need to put their money where their mouth is; it is not enough to say 'I think Greece will default' - if you are going to have an influence on the price, you have to be willing to take the risk of losing money if your opinion turns out to be wrong, and you would only do that if you have confidence in your information and analysis. That way, the price of different assets, including sovereign debt, is determined by the people with the best information (or, as is the case when it comes to extremely deep and liquid markets such as sovereign debt, the best ability to process the information that is freely available to all).
So, it is a serious perversion of this basic principle when the credit ratings agencies spend 60,000 dollars a year (that's the annual wage of a junior analyst) on analyzing Greek debt while investing exactly $0 on it, yet the effect they have on the price of that debt is equivalent to their controlling billions in funds. What the hell does it mean to 'rate debt'? Isn't this what the market is supposed to be doing by setting the price?
This state of affairs is so striking that I have to repeat this again. Markets work because people put their money on the line: ratings agencies don't. Furthermore, information on sovereign debt is abundant and the agencies have no informational advantage whatsoever (not to mention an appalling ratings record). Giving dodgy ratings to obscure CDOs is one thing, but rating sovereign debt should have had no effect on its price - information is aplenty.
I'm trying to work out a model as to why real market participants (you know, the ones with money on the line) don't completely discount the agencies' credit ratings of sovereign debt. Where I've got to so far implies that ratings agencies can only have a destructive effect, leading the market away from the efficient price: since the ratings do not reveal any new information (as I said, info on sovereign debt is abundant), the only reason they can move the price of debt is because they give an opportunity to smart money to profit at the expense of dumb money - money that is too stupid to rate sovereign debt independently, or is restricted from investing in debt below a particular rating.
I repeat: this is seriously messed up. Do the credit agencies have access to superior information on Greek debt compared to everyone (indeed anyone) else? No. Does the $60,000/ year S&P team analyzing Greek debt have such amazingly better insights than the collective might of analysts of hundreds of funds and banks across the globe? Hell no. Then why in the name of all that is folly does what S&P say affects the price of sovereign debt to such an extent?
And let me repeat a crucial point in case it wasn't made clear (a lot of repetition in this post, but this is what happens when your blogger starts getting a distinct feeling that either he has gone crazy or the rest of the world has). Don't try arguing that even if the credit ratings agencies don't have their money on the line they have their reputation to worry about - in case you just arrived on our beautiful planet from somewhere in the outer universe, they seriously messed up with CDOs with no effect whatsoever on their credibility and their ability to make money (there is a word for companies that can provide a crappy service and stay in business: they are called monopolies).
All in all, credit rating agencies have a destructive influence. They facilitate speculation at the expense of market efficiency, creating or reinforcing herd effects where market participants try to outfox the 'bigger idiots' rather than aim for the efficient price, seriously distorting the essential role of the market in accurately pricing assets. They risk nothing when pontificating on the value of different assets, and they add nothing real to the stock of available information - just a focal point for speculation. Furthermore, they are so embedded in the system - with funds having restrictions in their constitution restricting what they can invest in depending on its credit rating - that it's difficult to see a way forward.
We can start by pointing out the absurdity of the situation.
AIG was rated "AAA" at the time it went bust (had to be bailed out by the US government).
"I'm trying to work out a model as to why real market participants... don't completely discount the agencies' credit ratings"
Model 1. Could one model be around funds letting their investors know what they're investing in. Lots of funds are set up to only invest in bonds above a certain credit rating and once a holding is down-rated they have to sell their entire holding. This might plausibly explain the impacts of the rating agencies on bond prices. It also can make sense for investors as a way of attempting to get around moral hazard problems with their fund managers. If the investor wants a really low risk asset allocation, they can either leave it to their fund manager to achieve that (knowing he might sometimes stick with a slightly higher risk asset to attempt to outperform other funds) or you can link to a factor exogenous to the fund, which creates a role for a credit agency.
Model 2. Fund managers love to herd because their contracts aren't set up right (moral hazard again), and this gives them an incentive to follow the agencies. If a fund manager makes a decision different to the rest and it works out there's limited upside, if it fails then they have a big chance of losing their job for being a maverick. I've seen it written by some 80s fund managers that everyone held IBM whether they thought it would do well or not because if it did well and you weren't holding you got fired but if it did badly and everyone was holding then you were OK.
That said, I still hate rating agencies!
A framework in which the value of a debt issue depends not only on fundamentals but also the ability of the company or government to ‘play the game’ would be useful. For government debt, part of ‘playing the game’ is signalling a commitment, no matter how empty, to a low-overhead civil service. Much of the equity market works this way. Financial news agencies, television and print, like to spout their boilerplate and have companies spout boilerplate back to them that fits into their boilerplate, and if the company doesn’t play their silly little finance game they’re punished.
"What the hell does it mean to 'rate debt'? Isn't this what the market is supposed to be doing by setting the price?"
If the return on an investment follows a distribution of some sort, then the market price represents the expected value of that distribution. Unless investors are completely risk-neutral, some kind of estimate of the variability of returns can be useful and meaningful completely independent of the expected value.
In effect, the market price gives the mean and the rating agency tries to tell you the variance.
If a price plunges when a rating decreases, that means that the awareness of increased variance has caused market participants to seriously re-assess the way they've estimated expected value.
Do you know of any journal articles or press releases discussing the unwarranted credibility of agencies' (or just S&P) credit ratings?