> > If prices really are going up for a period of time
> > solely on expectation that someone else will always be willing to pay
> > prices even more unjustified by business fundamentals than the price the
> > previous buyer paid, then it would be possible to predict that the
> > overbid stocks will inevitably move downward by a large amount.
>
> Yes, but:
> 1) Speculators often do not realize the market is a bubble.
If so, this amounts to an argument against, not for, the efficiency of the
markets.
Prices are
> based on expectations of future demand, and that demand depends on future
> profits and other factors. For example, why is Amazon stock selling at a
> positive price, when it has been a "river of no returns"? Many share
> owners expect the company to make a profit in the future, or that the
> company will be sold at a higher price. That may never happen, but nobody
> knows for sure. People can expect higher future profits, and ex post this
> will seem unrealistic, but ex ante, nobody knows. It's not really greed,
> but hope, that creates momentum.
Hope as well as greed can reduce market efficiency. Hope and greed can
drive either investments based on careful, rational analysis, or wild-eyed
speculation. How realistic was that hope when Amazon was selling for well
over $200 a share with zero earnings in a low-margin business while other
well-financed companies preparing their own internet launch were selling for
many times less and a modest P/E ratio? You still sound like you're arguing
that any investment made with the hope of a future return (and that is all
investments) contributes to an efficient market. As long as all investors
hope to make money on their investments, this hope itself makes the markets
efficient. Well, no, that can't be right. As the contrarians say, "The
crowd is always wrong." (Of course, if the markets really are as efficient
as some people seem to think, then the contrarians are always wrong. My
point is just that good intentions--the intent to make a profit--do not make
a market automatically efficient, or efficient by definition.)
>
> 2) Many traders follow the trend. Trends can continue for a long time.
So
> even though traders know the trend will peak out, nobody knows when, so
> traders keep buying. The problem is that the trend is not a straight
line,
> but jagged, so nobody knows when a downturn is the major one or a
temporary
> one.
That's why trend traders trade with stops.
>
> Markets can be efficient all the way up the trend and down the trend,
> because at each time and price, traders are using the available
> information.
Well, no. Most trend traders are pure or nearly pure technical traders.
They willfully ignore most of the available information and concentrate
exclusively or mainly on two factors--price and money management techniques.
You can make money this way--in fact, a lot of money--but it does not
contribute to market efficiency. In fact, very many trend traders in a
stock or commodity ought to accentuate distortions in market pricing.
>The problem is that nobody knows the future. Also, financial
> markets are not completely efficient, as knowledge has a cost, so it is
> sometimes possible to profit from knowledge even if it is publicly
> available.
Yes, they are "not completely efficient," although many efficient market
theorists seem willing to admit this only when backed into a corner. Are
they inefficient enough to make more money than a buy-and-hold method over a
long period of time? According to many random walk enthusiasts, after
accounting for transaction costs, the answer is no, and they cite their own
computerized backtests for evidence. On the other hand, many professional
traders who have made fortunes from in-and-out trading will say yes. The
question continues to be debated.
>
> > If that is what you mean by an "efficient market" then to say that
> securities markets are efficient becomes a tautology rather than a
theory.<
>
> If markets were not efficient, prices would not reflect yields and
discount
> rates and competition. But they do tend to do so. So they are efficient,
> though not perfectly so.
>
> For example, if I want to buy an ounce of gold, I am not likely to find
> someone who will sell it to me at $200, and few would pay $400. The fact
> that most are buying and selling at a prevailing market price at some
> moment implies efficiency about the transmittal of information, about
> competition, and profit maximization. If the gold market were
inefficient,
> we would see gold trading at widely different prices at the same time.
Not necessarily. You have already conceded that markets are inefficient to
some degree, even though an asset will not trade at widely different prices
at the same time. The question is, how inefficient is the market? Have any
economists attempted to quantify this? I know of one swing trader, relying
mainly on stock options to maximize leverage, who has more than doubled his
trading account every year for the last six or seven years or so, although
his performance has dropped off considerably since the bear market began in
2000 (although still doubling). Random walk theorists would say this is
either impossible or blind, dumb luck. I suspect the degree of inefficiency
in the markets is underestimated, although I also agree it takes either
quite a bit of work or quite a bit of sophistication--more than the typical
mutual fund manager possesses--to take advantage of it.
~Alypius Skinner
