No Such Thing as a Bad Investment.

MarketWatch, a Dow Jones-owned website that should know better, published an article a little while ago called “The world's worst investors”, chock full of terrible, terrible advice.

Which doesn't bother me. Lots of people who should know better give terrible advice. What bothers me is the smug way he says it.

Brett Arends, the author, tells the story of Mark Villa and Lucie White, two doctors in Houston who lost a lot of money getting out of the stock market decline of 2008 and who are now getting back to investing. Arends virtually detonates in an air-burst of condescension and misinformation. His basic complaint is, as he says, Mom and Pop, as he calls them, “have consistently bought high and sold low.”

Which is demonstrably true, but then he follows it up by saying, “You could have made astonishing super-normal profits over the past thirty years just by selling stocks when Mom and Pop were buying, and buying stocks when Mom and Pop were selling.”

That statement is arrant nonsense. If it were true, Arends could have done it himself and made a fortune. He'd be too busy blowing his nose with hundred-dollar bills to waste time writing misleading financial columns.

The basic intuition is exactly what he said: if it's a bad idea to buy a particular stock, or stocks in general, at a particular time, it's necessarily a good time to sell. If Drs. Villa and White, let's call them V&W, actually were so dismayingly amazing at picking losers, they would equally amazing at picking winners — they're the same stocks, if you do the opposite and sell when V&W buy, and vice versa.

It would call into question why investment companies spend money on experts and computers and research when they could just do the exact opposite of whatever V&W are doing.

Perhaps Arends has a mental image of amateur investors like Villa and White as a panicky herd of animals, like cows or sheep, that stampede in some direction, while a few savvy professional — I'm sure Arends is casting himself in this role — pick off the strays like wily wolves.

But numerically, it just isn't so. In the open range of the stock market, the wolves outnumber the sheep. Most trading is done by huge institutions: mutual funds, pensions, large investors. In a liquid market, it's exactly as impossible to make a spectacularly bad investment as to make a spectacularly good one. If a thousand people are selling a particular stock, and nine-hundred ninety nine people plus you are buying that stock, how likely is it that one side or the other is making a huge mistake?

Someone — it is supposed to be John Maynard Keynes, but it wasn't, it was a Forbes analyst named Gary Shilling — wrote “Markets can remain irrational a lot longer than you and I can remain solvent.” Which would mean that even if there were something in Arends's fantasy about panicking amateurs capsizing the market, there's no way to benefit from that knowledge. If the price of a stock were driven irrationally low, it wouldn't be enough to buy that stock; you have to buy it near the bottom and then hold it until it recovers — something that might very well never happen.

Underlying all this is the Efficient Market Hypothesis, which says that prices on traded assets (like stock) already reflect all publicly available information. No amount of research is going to help you find you a good deal; no amount of stupidity is going get you a bad one. Arends calls the EMH “a cult”.

The interesting thing about the EMH isn't that it's wrong, although it is, a little. The interesting thing isn't even that it's self-contradictory, although it is, a lot. The really interesting thing about EMH is that it depends on being self-contradictory in order to be useful.

It's easier to see why the EMH is wrong if you think about why it's right. Imagine a market where the price of stock was completely detached from its value. All prices are set at random.

In such a world, it would certainly behoove even the dimmest investor to do some research. If Google is selling for $1 a share (here in the real world, it's selling for 800), go and snap that up. A market that started out completely inefficient would quickly be thronged with investors, bidding up the $1 Google stock to $100 or $700, something reasonable, something “efficient” as economists say.

But you can see what happens. As more and more people do more and more research and investing, the bargains dry up, and there's less return on doing the hard work of research and the risky work of investing.

Likewise imagine a perfectly efficient market. Every stock traded for exactly the correct price (which, to be technical, is the risk-adjusted net present value of the dividend stream). In a perfectly efficient market, there's absolutely no point in picking one stock over another, any more than you pick one dollar bill over another when getting cash from a teller. They're all the same to you.

But why would a perfectly efficient market stay efficient? Nobody's bothering to do any research — because there's no money in research.

Imagine you had a vacant lot choked with weeds. To get rid of the weeds, you brought in a small flock of goats. The goats can, of course, make short work of the weeds but if they eat all the weeds, they will themselves starve. For the goat/weed ecology to work, there needs to be enough weeds to keep the goats alive to eat the weeds.

In my (okay, rather tortured) analogy, “weeds” are inefficient, inaccurate prices: great stocks and terrible stocks. The “goats” are investors, they are only in the market because they're looking for weeds to eat, but by eating them, they get rid of them. Eventually, for a given size lot, there are just enough goats to eat the weeds down to a stubble.

Eventually, all the EMH guarantees is that the market will be “mostly” efficient, not “perfectly” efficient — in fact, it guarantees the market won't be perfectly efficient. There has to be enough inefficiency to reward people for doing research and modeling, rooting out that inefficiency, feeding themselves on the inefficiency.

Is the market going to reward you for doing that work? No, probably not. There are people with decades of experience and scads of computers and brains the size of beach-balls working on this problem 24/7, so they are probably going to do a better job than you. You might get lucky, of course, but you might get unlucky. It's like a lottery where most people win a little.

And Arends, for all his smugness, certainly knows this. He writes, “In late 2008 and early 2009 I was tasked by The Wall Street Journal with finding good value stocks for people to buy. And my job was very easy.”

Note that Arends doesn't mention investing any of his own money in those easily found bargains. He seems to be still working for a living.

 Guest Author Michael Lorton is a San Francisco-based programmer and writer. He writes the daily blog Blue Apsara about culture, economics, travel, law, and the occasional giant, carnivorous lizard.

Financial Data image courtesy of Shutterstock 

blog comments powered by Disqus

The Featured Five