The Dow, S&P500, and Nasdaq, fell 3.3-3.9% Tuesday, inducing calls for more regulation to prevent such "irrational" price moves. For example, Steven Pearlstein wrote: Years of dirt-cheap debt have spawned bubbles across a wide range of asset classes,… It’s been a wonderful ride … sooner or later it had to end. … It’s a confidence game that has little to do with the underlying economic fundamentals … make no mistake: This is herd behavior, as irrational on the way down as it was on the way up.
Robin Hanson: I believe you to have a very wry sense of humor, as any rational actor should have seen the comedy behind such a regulatory proposal. The very concept of a free market is the information obtained by buyers and sellers in their discovery of the market clearing price. It is the outcome of price discovery that contains such information, i.e. quantities demanded and supplied across the time horizon. Such a regulatory scheme to prevent price volatility presumes foreknowledge of that information. In fact, it presumes an alternative price discovery system, one that will set the "right" price because it "knows" the right price. It also presumes a straight line for prices, as under the foreknowledge of such price information incumbent in the regulatory scheme, no volatility would be acceptable, as any price volatility would be the result of an exchange made away from the "known" price.
It is unfortunate that some of the comments made missed your humor, as Pearlstein's suggestion is based on the notion (backed by no evidence) that Tuesday's volatility was unreasonable and unwarranted. That Pearlstein suggests regulators could know what volatility is reasonable and customary, in a manner he doesn't specify, is not only foolish, but demonstrates his naivety of, and likely hostility to, free markets.
And such a concept as offered by Pearlstein is more of the "equal outcome" mantra fostered by the Left, i.e. price discovery is discrimination, and we can't have any discrimination.
If markets behave irrationally, it's because they are influenced by humans, who also behave irrationally. The problem is, any proposed regulatory body would also be controlled by humans, and thus would also behave irrationally.
Hal, in a non-accountable regulatory regime to reduce volatility, there is also the possibility of rare very costly regulator misjudgements. The difference is that we can't very easily see such mistakes, to evaluate how well the regulation works.
To perhaps state the obvious, if it's profitable to curb market volatility, you don't necessarily need a government program to do so. Investors have incentives to take positions and profit from excess or inappropriate volatility, thereby reducing or eliminating it.
Seems like the big problem with making this kind of program self-funding, and one reason why people don't do it voluntarily, is what Keynes described. This plan can work pretty well most of the time and generate profits, but eventually it will fail catastrophically. You keep buying as the market keeps dropping, until you run out of cash, the market drops further, and you're wiped out. (Or vice versa for market bubbles, where you're taking a short position.) Even a hundred billion dollars won't last forever, not in a global market.
Spencer, how does Congress deal with a regulator who refuses to do their assigned regulation task? Fire them of course. I don't see why regulators need all inside info to just reduce volatility.
You may have a good idea here on one condition. The regulator must have advance access to all the"insider" information that moves stocks. The most common problem regulators deal with is insider information. So if you can figure out how to provide the regulators with this insider informationI'll support your proposal.
But how can you deal with the efficient market problem? If the regulators simply index the funds provided them over the long run they will beat essentially every portfolio manager. So if congress gives the regulator a sum they could simply index and live nicely off the dividend income without ever having to do anything -- about what the typical owner of inherited wealth does.
Peter, if the regulators accept the premise that by inducing trading halts you will reduce error by traders, can't they impose that on themselves (say, if they always had to use a computer program to trade their inventory of money and stocks that would automatically start refusing commands when the news/market has a certain volatility) and thereby make better trading decisions and beat the market?
Stuart, Alex, Peter, I did not propose this regulator scheme to achieve all possible regulatory goals, but only to achieve the goal of reducing "excess volatility," which is the problem Pearlstein complained about.
While there are probably many proposed regulations for which your reaction is appropriate, I think there are some possible regulations for which your analysis is misleading. In particular the word "push" has connotations that tend to obscure some possible effects of regulations on prices.I suggest that rapid market moves tend to cause traders to act less rationally than normal. The speed with which some trading decisions need to be made is partly a function of the speed with which competing traders are making their decisions, and in times when news changes unusually rapidly these competitive pressures cause traders to devote less thought to any particular decision. (By news, I mostly mean financial price information rather than English-language style news).It's possible that temporary trading halts can cause traders to have more time to think about their decisions and come to wiser choices as a result. If so, that would alter prices in a way that doesn't imply that regulators could make money by better trading, since the regulators would be subject to the same competitive pressures that cause other traders to act less rationally than is optimal.I'm not at all confident that rules causing trading halts are wise, since they probably have some harmful effects that I don't know how to measure. And I have no reason to think that government regulators would do a better job of creating such rules than stock exchanges would.There are some rumors that part of the problems with Tuesday's trading resulted from rules that temporarily suspend some types of automated trading during big market moves, which redirected some trading to human specialists who may have been overwhelmed by the volume. These rules seem to have been the result of one conspicuous case (the 1987 crash) where automated trading caused as least as much harm as more direct human trading, but this seems more like what military strategists call "fighting the last war" rather than a reaction to a persistent pattern of problems.
I'm not sure about the usefulness of giving some "regulators" a big pile of cash and letting them loose... but I am sure that trying to regulate the stock market to achieve certain price levels or market activity levels is foolish. The regulators would need to know where the stock market "should" be and would need to know how to make it get there.
Unfortunately, regulators are all human beings and would be chosen by politicians from either their cronies/donors or stock traders.
Unbiased, competent market control almost always looks good on paper... but is usually impossible to achieve. As was said above, regulators should work to ensure that the market is as efficient as possible by preventing fraud and enforcing some basic rules.
Strawman. I'm not aware of any regulators in the world who try to "rationalise" stock market prices. This isn't the role of regulators. Their role is instead to police certain rules, usually ones dealing with various kinds of informational inequality.
I suggested a regulatory regime wherein we could find out whether or not regulators can improve prices or not.
Ah, I think I see your point now. I personnally feel that the type of stock micromanagement you describe is foolish (the only roles of regulators should be to prevent big cascading crashes and insider dealing).
But if we want to prove it, your regime may not be enough. If the "push" regulators lose all their money, there could be three reasons:1) There is no rational price for stocks that the regulators can guess2) There is, but bringing the market to the rational price is not good for the market3) There is, and bringing it to the rational price is good for the market, but not profitable for the regulators who have to buck market trends to do so
We should find a way of dealing with that third situation.
Is there any possibility of setting up a basket of stocks (say half the stocks in the market) and get the "push"-regulators to intervene only on those? Then, at the end of the year, the regulators announce which stocks they were working with, and randomly choose new ones for the coming year.
Then after a run of five or six years, compare the regulated versus unregulated behaviours of the stocks (taking into account the money regulators spent, of course).
This feels like a completly impractical idea, but some variant of it may be workable, on a small scale (once we solve the issue that stock prices are not independent from each other); then we would know if the interventions of "push" regulators work, and, if so, whether they are worth it.
Stuart, I did not mean to claim that most stock price movements are rational, or that regulators always fail. I suggested a regulatory regime wherein we could find out whether or not regulators can improve prices or not.
Knack, yes of course there is a lot of wild randomness. I just suggested that speculators play a part in making prices more rational than they would otherwise be.
Robin, I'm not sure I understand what you mean when you say that speculators rationalize stock prices. Keynes, a probablist, said famously that markets can remain irrational longer than you can remain solvent. Speculators act as the extension of as well as the correction to that irrational phase.
Look at Mandelbrot. His explanations of markets and wild randomness conform to my experience more than what they tried to shoe-horn into me at business school about EMH etc. But how widely discussed are his views?
Twice I've pressed a button on my computer keyboard and the consequence was a 10% move in a global market. You'd probably say that the market was responding to new information. I don't see it that way, particularly based on what was reported by the "rational" participants both before and after each incident.
This idea of rational markets, dare I suggest, is a bias worth overcoming.
Sorry to say it Robin, but this article looks a bit biased (or maybe it just seems that way to me). It seems to imply that most stock-market movements are rational, that they are always economically benign, that there is a "long run" that stocks tend to, and that regulators always fail.
I've read economists and met investement bankers who disagree with all those statements.
Now, I'm not an economist myself, so not qualified to take a position on these issues (though, very superficially, the fact that prices change so drastically and rapidly seems to imply that they are not driven solely by market fundamentals). But if there are a variety of views on this, I feel you should at least address them, and say why they aren't valid in this case.
Robin Hanson: I believe you to have a very wry sense of humor, as any rational actor should have seen the comedy behind such a regulatory proposal. The very concept of a free market is the information obtained by buyers and sellers in their discovery of the market clearing price. It is the outcome of price discovery that contains such information, i.e. quantities demanded and supplied across the time horizon. Such a regulatory scheme to prevent price volatility presumes foreknowledge of that information. In fact, it presumes an alternative price discovery system, one that will set the "right" price because it "knows" the right price. It also presumes a straight line for prices, as under the foreknowledge of such price information incumbent in the regulatory scheme, no volatility would be acceptable, as any price volatility would be the result of an exchange made away from the "known" price.
It is unfortunate that some of the comments made missed your humor, as Pearlstein's suggestion is based on the notion (backed by no evidence) that Tuesday's volatility was unreasonable and unwarranted. That Pearlstein suggests regulators could know what volatility is reasonable and customary, in a manner he doesn't specify, is not only foolish, but demonstrates his naivety of, and likely hostility to, free markets.
And such a concept as offered by Pearlstein is more of the "equal outcome" mantra fostered by the Left, i.e. price discovery is discrimination, and we can't have any discrimination.
If markets behave irrationally, it's because they are influenced by humans, who also behave irrationally. The problem is, any proposed regulatory body would also be controlled by humans, and thus would also behave irrationally.
I rest my case.-Matt
Hal, in a non-accountable regulatory regime to reduce volatility, there is also the possibility of rare very costly regulator misjudgements. The difference is that we can't very easily see such mistakes, to evaluate how well the regulation works.
To perhaps state the obvious, if it's profitable to curb market volatility, you don't necessarily need a government program to do so. Investors have incentives to take positions and profit from excess or inappropriate volatility, thereby reducing or eliminating it.
Seems like the big problem with making this kind of program self-funding, and one reason why people don't do it voluntarily, is what Keynes described. This plan can work pretty well most of the time and generate profits, but eventually it will fail catastrophically. You keep buying as the market keeps dropping, until you run out of cash, the market drops further, and you're wiped out. (Or vice versa for market bubbles, where you're taking a short position.) Even a hundred billion dollars won't last forever, not in a global market.
Spencer, how does Congress deal with a regulator who refuses to do their assigned regulation task? Fire them of course. I don't see why regulators need all inside info to just reduce volatility.
You may have a good idea here on one condition. The regulator must have advance access to all the"insider" information that moves stocks. The most common problem regulators deal with is insider information. So if you can figure out how to provide the regulators with this insider informationI'll support your proposal.
But how can you deal with the efficient market problem? If the regulators simply index the funds provided them over the long run they will beat essentially every portfolio manager. So if congress gives the regulator a sum they could simply index and live nicely off the dividend income without ever having to do anything -- about what the typical owner of inherited wealth does.
Peter, if the regulators accept the premise that by inducing trading halts you will reduce error by traders, can't they impose that on themselves (say, if they always had to use a computer program to trade their inventory of money and stocks that would automatically start refusing commands when the news/market has a certain volatility) and thereby make better trading decisions and beat the market?
Stuart, Alex, Peter, I did not propose this regulator scheme to achieve all possible regulatory goals, but only to achieve the goal of reducing "excess volatility," which is the problem Pearlstein complained about.
While there are probably many proposed regulations for which your reaction is appropriate, I think there are some possible regulations for which your analysis is misleading. In particular the word "push" has connotations that tend to obscure some possible effects of regulations on prices.I suggest that rapid market moves tend to cause traders to act less rationally than normal. The speed with which some trading decisions need to be made is partly a function of the speed with which competing traders are making their decisions, and in times when news changes unusually rapidly these competitive pressures cause traders to devote less thought to any particular decision. (By news, I mostly mean financial price information rather than English-language style news).It's possible that temporary trading halts can cause traders to have more time to think about their decisions and come to wiser choices as a result. If so, that would alter prices in a way that doesn't imply that regulators could make money by better trading, since the regulators would be subject to the same competitive pressures that cause other traders to act less rationally than is optimal.I'm not at all confident that rules causing trading halts are wise, since they probably have some harmful effects that I don't know how to measure. And I have no reason to think that government regulators would do a better job of creating such rules than stock exchanges would.There are some rumors that part of the problems with Tuesday's trading resulted from rules that temporarily suspend some types of automated trading during big market moves, which redirected some trading to human specialists who may have been overwhelmed by the volume. These rules seem to have been the result of one conspicuous case (the 1987 crash) where automated trading caused as least as much harm as more direct human trading, but this seems more like what military strategists call "fighting the last war" rather than a reaction to a persistent pattern of problems.
I'm not sure about the usefulness of giving some "regulators" a big pile of cash and letting them loose... but I am sure that trying to regulate the stock market to achieve certain price levels or market activity levels is foolish. The regulators would need to know where the stock market "should" be and would need to know how to make it get there.
Unfortunately, regulators are all human beings and would be chosen by politicians from either their cronies/donors or stock traders.
Unbiased, competent market control almost always looks good on paper... but is usually impossible to achieve. As was said above, regulators should work to ensure that the market is as efficient as possible by preventing fraud and enforcing some basic rules.
EI
Strawman. I'm not aware of any regulators in the world who try to "rationalise" stock market prices. This isn't the role of regulators. Their role is instead to police certain rules, usually ones dealing with various kinds of informational inequality.
I suggested a regulatory regime wherein we could find out whether or not regulators can improve prices or not.
Ah, I think I see your point now. I personnally feel that the type of stock micromanagement you describe is foolish (the only roles of regulators should be to prevent big cascading crashes and insider dealing).
But if we want to prove it, your regime may not be enough. If the "push" regulators lose all their money, there could be three reasons:1) There is no rational price for stocks that the regulators can guess2) There is, but bringing the market to the rational price is not good for the market3) There is, and bringing it to the rational price is good for the market, but not profitable for the regulators who have to buck market trends to do so
We should find a way of dealing with that third situation.
Is there any possibility of setting up a basket of stocks (say half the stocks in the market) and get the "push"-regulators to intervene only on those? Then, at the end of the year, the regulators announce which stocks they were working with, and randomly choose new ones for the coming year.
Then after a run of five or six years, compare the regulated versus unregulated behaviours of the stocks (taking into account the money regulators spent, of course).
This feels like a completly impractical idea, but some variant of it may be workable, on a small scale (once we solve the issue that stock prices are not independent from each other); then we would know if the interventions of "push" regulators work, and, if so, whether they are worth it.
Stuart, I did not mean to claim that most stock price movements are rational, or that regulators always fail. I suggested a regulatory regime wherein we could find out whether or not regulators can improve prices or not.
Knack, yes of course there is a lot of wild randomness. I just suggested that speculators play a part in making prices more rational than they would otherwise be.
Robin, I'm not sure I understand what you mean when you say that speculators rationalize stock prices. Keynes, a probablist, said famously that markets can remain irrational longer than you can remain solvent. Speculators act as the extension of as well as the correction to that irrational phase.
Look at Mandelbrot. His explanations of markets and wild randomness conform to my experience more than what they tried to shoe-horn into me at business school about EMH etc. But how widely discussed are his views?
Twice I've pressed a button on my computer keyboard and the consequence was a 10% move in a global market. You'd probably say that the market was responding to new information. I don't see it that way, particularly based on what was reported by the "rational" participants both before and after each incident.
This idea of rational markets, dare I suggest, is a bias worth overcoming.
Sorry to say it Robin, but this article looks a bit biased (or maybe it just seems that way to me). It seems to imply that most stock-market movements are rational, that they are always economically benign, that there is a "long run" that stocks tend to, and that regulators always fail.
I've read economists and met investement bankers who disagree with all those statements.
Now, I'm not an economist myself, so not qualified to take a position on these issues (though, very superficially, the fact that prices change so drastically and rapidly seems to imply that they are not driven solely by market fundamentals). But if there are a variety of views on this, I feel you should at least address them, and say why they aren't valid in this case.
If the opposite of pro is con, then the opposite of progress must be congress. . . ;-)