See e.g. these physics preprints on the arXiv about the housing bubble in Las Vegas, British islands, and the U.S. The last preprint from mid 2005 has identified a bubble and predicted the turning point for the bubble in mid 2006. Pretty good.A clever speculation became somewhat more important for the ultimate profits and losses than the actual value of houses or work. Speculators - a category that suddenly included a lot of ordinary people - were buying houses for borrowed money because they were almost certain that the price would keep on increasing. The price has been doing the same thing for years and one could use mathematical induction to predict the rest. They were so certain because they saw other people who were also certain. This feeling feeds itself. Such a synergetical psychology is an example of a positive feedback.
Because of some fundamental limitations constraining many physical quantities, such an escallation eventually has to stop. The prices are still not decreasing much but the sales certainly are. Equally importantly, banks and people no longer want to lend money. Some subjects depend on the loans and can't pay. Their insolvency influences other subjects and the potential problems can start to spread to other segments of the economy. That's another positive feedback mechanism in the system whose consequences are mostly negative.
Most of the "money" in the whole system are various debts of various degrees and this makes the whole system too interconnected. I feel that this is generally a bad thing.
Meanwhile, central banks have tools that can influence the situation. While many subjects act irrationally or in strange ways that have only become rational because of some previous irrational acts of others ;-), the central banks can do many things that compensate the behavior of these other subjects - it usually ends up as an overcompensation - but that also has some side effects. If you look, there really exists no deeply rational reason why the market should begin to act wildly right now - the global economy is in a great shape - but it does.
Is the herd market mentality a good thing?
Well, I don't think that it is such a good thing. It is ultimately a consequence of individuals' being unable to judge the external world independently. It follows from a lack of rationality and a weak desire to evaluate the actual risk and the actual value. It arises because other players' psychology becomes more important than the external world which is a bad thing. A psychological disease is behind most economic crises and behind excessive volatility. But I don't think that central financial institutions should be doing too much about it.
Instead, what I think that the central bankers and financial authorities should be doing is to offer their decent ideas about the actual value of certain products, especially financial products (but also houses) - a value that they consider fair and sustainable. They should say what they consider overpriced and what they consider undervalued. Expressing this opinion would already have a stabilizing effect on the markets. If the central bankers were saying twice a month that houses in XY are 20% overvalued, I guess that it could be enough.
We rarely hear central bankers to say such things. Surely, they respect the market. But the market is controlled by imperfect people. And those people who are elected to be central bankers or something like that are supposed to have similar - in fact, much better - abilities to estimate the value and the risk than the average market participants. Of course that a central banker may be wrong and the market can eventually prove that he is wrong. The leaders of the Soviet bloc were surely wrong almost about everything they have ever said (and done!) about the economy. But if a central banker in the current world understands the economy more than others, and he should, it should simply be an unlikely scenario.
Because of the feeling that liquidity is suddenly absent, the Fed has added 24 billion USD to bank reserves. The European Central Bank was more radical. During the last two days, the ECB has loaned more than 155 billon EUR i.e. 210 billion USD to banks at a bargain 4% rate. The Fed's action is clearly negligible in comparison. The difference is an order of magnitude.
Is that a good thing for the ECB to do? Well, I am not sure. It is hard to imagine that this event won't accelerate inflation. More importantly, as Gene argues in the comments, such a help prevents the erring greedy players from being punished and this experience will reduce the self-correcting ability of the market in the future. Also, the Fed is now expected by the markets to cut the rates in September. Is that a sensible expectation? Well, I think that it mostly depends on further injections of cash into the economies that may take place in the following weeks as well as on the ultimate consequences of the recent injections. If they pump too much money into the system, it is pretty clear that the liquidity problem may be solved but other risks such as inflation may become more real. That would make rate cuts very unnatural. On the contrary, they might be forced to raise the rates which is what they should do anyway. Not only the bubbles but also the large trade deficit more or less arises from the loose fiscal policy of the Fed.
Are overreactions good?
Just like the markets usually overshoot, central banks may overshoot, too. They may overreact. But I think it is very wrong for them to overreact. Such an overreaction is the counterpart of taxes that are higher than 100%. It is an event that makes the psychology of some government officials more important for the outcomes than the rest of the economy. Such things shouldn't occur because they diminish the relation between the price of products and the actual value or risks involved - a relation that is important for the markets to work well.
It is probably a good idea for them to pump some money to the economy right now. But if they pump a modest amount such as Fed's 24 billion USD, they are already helping the system. If they're sane, they know that it is impossible and undesirable to try to solve 100% of everyone's problems. Overhelping the system with ECB's 210 billion USD or more could turn out to be counterproductive, more irrational, and more unstable than the bubbles and wiggles created by regular market participants who have caused the situation in the first place.
And that's the memo.