While you might say the ratings agencies are behind the curve, the ratings are an integral part of the financial system.. and some would claim too integral.
To be sure, I'm not suggesting that ratings should be eliminated. I do believe that the practices deployed by the ratings agencies need dramatic improvement and the assigned ratings need to be far more reliable. That will mean a lot more work, as the simple approach of largely inputting data (which is not generic) into models and then running with the results has had a dismal track record when subjected to empirical data. Fixing things will require far more active human judgment and analysis, really digging into data, working to understand the context in which countries/companies function, drawing upon relevant historical analogs, and providing probabilistic scenarios and milestones along each reasonably possible path, among other things. Such fixes won't limit error, but the quality, utility, and reliability of information should be markedly improved. As noted earlier, if the empirical research suggests that even simple data such as current account data provides a much better picture of risk than the current ratings system, then payments that are provided to ratings agencies are furnishing negative value.
Markets react in panic when these agencies downgrade a country or company despite that the markets knowing nothing has actually changed. Spain is an almost classic example. Fitch downgraded Spain from AAA to AA+ a month or so ago and the markets went into a panic mode. When the news organisations started to interview the Fitch people, then suddenly the markets noticed that the downgrading also meant that Spain went from downward outlook too stable outlook.. something they had missed at the start. Then the markets recovered over the next day. And for the record AA+ is still one hell of a good rating as the Fitch rep. said. So if the ratings agencies are behind the curve, then the markets should have factored in a downgrade for the most part.. but they have clearly not since a downgrade can start of massive selling and panic. And the real problem with Fitch's downgrade and the markets is that S&P downgraded Spain a few weeks earlier so there is no excuse what so ever for the markets not to have factored in such a downgrade.
Market prices reflect a combination of fundamentals and psychology. Sometimes the bias is toward one end and sometimes the other. Markets can be reasonably rational. They can also grow irrational. Hence, markets can and do go to excess, sometimes for extended periods of time. The theory that markets are enormous "discounting mechanisms" (quite accurately reflect all that is known about a company or country) is incorrect. Human foresight--be it individual or collective--is very limited, so by definition no human institution or construct can avoid the limitations of human foresight. Indeed, if markets were very good discounting mechanisms, short- and long-term volatility would be far lower than it is, because markets would have a good understanding of where things stand and they would not be continually caught by surprise by major economic developments, not to mention non-economic ones. Of course, there are also unforeseeable events, too, and those add a further limitation on use of markets to discount the future. In short, markets offer one piece of guidance, but one still needs to dig deeper to determine whether such guidance is reasonable given the situation or it is not.
And in the case of Greece I would claim the ratings agencies along with the speculators and naked short selling pushed Greece over the edge. Greece did not even get time to sort out its house before the markets turned on Greece and drove it to the brink and the ratings agencies had a huge part in that since each time they downgraded Greece for whatever reason, they forced up the yield on Greek bonds, which in turn meant they had to downgrade Greece yet again. A snowball effect. And when you have 3 major agencies then you get snowball effect fast if the markets keep reacting in panic mode for each downgrading or warning and it all happens with in a few weeks or months and there is no freaking way any country or company can fix a problem in that short a time.
While the sudden ratings downgrades amplified the adverse developments concerning Greek debt, I believe Greece should have been at junk status well before the debt crisis erupted. Greece has a long history of debt default. To assume that the fairly recent lack of such difficulty was a "new normal" so to speak was absurd, especially when one considered the magnitude of Greece's fiscal deficits, its enormous current account deficits, that 99% of its debt was held by foreigners, and the long-term fiscal challenges confronting the country.
That is why I say dissolve all rating agencies and form a single independent agency to value large assets. The present system clearly does not work and is easily manipulated by the big players as we saw with Lehman Brothers and the sub prime crisis.
I don't know whether a new ratings agency or multiple new ones would be more effective (I'd probably lean toward the latter, as competition can improve industry performance). But there is little doubt that the present approach, bad understanding of risk, lack of due diligence due to overreliance on models that prove badly flawed time and again, conflicts of interest, lack of regular reporting/transparency concerning effectiveness of assigned ratings, among other factors, does not work. Unfortunately, it appears that U.S. financial regulatory reform will not produce the dramatic overhaul that is needed in that business.