Scott Sumner  

It's not a beauty pageant

Thank You for Correcting Me, M... The Mellow Heuristic...

Over at TheMoneyIllusion I received the following comment:

Macro trading is not about getting it right, it is about getting what others think is right. And that can be almost anything ...
That reminded me of Keynes's famous claim that successful investing was like guessing the winner of a beauty pageant. You don't try to pick the prettiest woman, but rather the woman that you think others will find pretty.

That's a clever line, but I don't think it's correct. I believe markets are roughly efficient, which means that it's very difficult for people to forecast (risk-adjusted) excess returns. But let's say I'm wrong; how would you go about doing so? One option would be to try to forecast the future type of mass hysteria that the public will succumb to. The other option would be to identify current irrationality in the investment community, which will eventually fade away, as new information makes it clear that their forecasts were unrealistic. I believe it would be easier to notice in 1999 and 2000 that tech stocks were too high due to "irrational exuberance," than to forecast back in 1994 that there would soon be a bout of irrational exuberance.

If you look at successful investors like Warren Buffett, they generally don't try to buy stocks that they believe will be overvalued in the future, rather they try to buy stocks that they believe are currently undervalued. Those are not easy to spot, but the alternative approach is even more difficult.

If we return to the beauty pageant example, Keynes's analogy doesn't really make any sense. For example, to make money betting on the winner of a beauty pageant, you'd need to do much more than predict which lady other bettors will find attractive, you'd have to predict how their views of relative beauty would evolve after the pageant. That's because the starting odds on each woman winning would already reflect the public's perception of their relative attractiveness. You'd need to predict how that public appraisal would change after the talent and swimsuit competitions. I suppose that's possible, but it's more about predicting "fundamental values" of beauty.

If we insist on a beauty pageant analogy, here's the one I'd use. Suppose the average beauty pageant gambler bets on the basis of the stock photos provided by the pageant to the press. That's the "public information." If you were Warren Buffett, you'd want to send detectives around observing the women in their daily lives, to get a better overall appraisal of their beauty and poise than you'd get from a single stock photo. (Or their ability to talk convincingly about the need for world peace.) That's what I mean by "fundamental" beauty. You don't want to forecast what others think, that's already embedded in the price, you want to forecast the TRUTH, which is what others will come to think once more information is available (in this case the pageant itself, in the world of stocks that would be how profits play out over an extended period of time.)

In both cases you are basically trying to forecast what people will learn over time. Of course my analogy is far from perfect, as even after a pageant you could argue that perceptions of beauty are more subjective than corporate profitability.

PS. In a future post I'll explain what's wrong with claiming, "the market can stay irrational longer than you can stay solvent."

Comments and Sharing

COMMENTS (14 to date)
Airman Spry Shark writes:

I often wonder if Keynes was mis-transcribed (since verbal scare quotes are easy to miss), and he actually said "Markets can remain 'irrational' longer than you can remain solvent."

bill writes:

Great post. And I really really look forward to that future post on markets, irrationality and solvency. I've used that phrase so often. Maybe (!) I'll have to stop. LOL

Kevin Erdmann writes:

It's ironic that most individual traders seem to believe that they are trading against an irrational market price.

My favorite update of the Keynes line is:

"Markets can remain solvent longer than you can remain irrational."

From JC Parets (All Star Charts), via Joshua Brown at via Jeff Miller at "A Dash of Insight". Although, I don't know if he's the original source.

I think it would be a huge step forward in finance if people instead thought of speculative trades either as regulatory arbitrage or as an arbitrage between their individual discount rate and the discount rate of the marginal investor. In some ways, your commenter was correct, inasmuch as the discount rate applied to different cash flows reflects unobservable and even inscrutable investor preferences. There are plenty of ways to capture individually useful tactical gains without irrational investors on the other side. The presence of a large number of irrational traders is not evidence of an irrational market.

Chuck writes:

The commenter is talking about trading, not investing. Even if other stock buyers never bid up the shares of Buffet's companies, he can get a return from dividends. Traders don't wait around for dividends.

James G writes:

libfree writes:

I don't know if it will work for me forever but I've had good luck buying the stock everyone suddenly hates. Think BP after the oil spill. Big stock hit but it wasn't something that people would remember 12 months later. That's somewhat similar.

Jose Romeu Robazzi writes:

Prof. Sumner
Great discussion. If I am not wrong, many people have commented on the idea of Macro investing/trading as a speculative game, as opposed to a investing activity. Two that I can remember of are Seth Klarman and George Soros. Seth Klarman, a well known fundamentalist, bottom up investor, thought that since macro fundamentals are well known to almost any investor in a very short time period, there is no point in doing research, like the one you suggested in your pageant analogy, since the likelihood that some info still not known to all will be found is very small. In his words, what matters is what other investors think, and that is how it becomes a speculation game. Soros goes even further, the says "if prices are moving, invest early and investigate later", which means in practical terms the same thing. Since these are two of the greatest investors of our time, there must be some truth to what they say ... My own view is that since return distributions are not normal, virtually all asset return have leptokurtotic distributions, there is always some way to exploit them... Long and variable leads! expectations matter a lot, if you get the time frame right!!

Peter writes:

Chuck, you benefit from dividends even if you only hold your stock for a second. Otherwise, you'd be able to make huge safe returns by buying just before dividends and selling right after.

The only important thing for dividends is how much time in total you hold stocks. So if you only hold stocks for 1% of the time then you only get 1% of the dividends. If you hold stocks 50% of the time but only hold each "trade" for a second you still get 50% of the dividends.

JJ writes:


Chuck, you benefit from dividends even if you only hold your stock for a second. Otherwise, you'd be able to make huge safe returns by buying just before dividends and selling right after. "

Isn't that already priced into the stock. I don't think people are stupid enough to sell the stock right before the dividend for below market value.

Michael Moran writes:

My wife had an aunt who was a manager of a mutual fund, and one time ranked in Barrons as one of best mutual fund managers. One of her comments was she was not that good at figuring out which "stories" would succeed, but she was good at figuring out which "stories" Wall Street would buy.
Investing is very complex, and many different successful methods. Not sure a generalization such as the the one of the beauty pageant can be proved or disproved.
Also, when valuing stocks, everyone measures against peers? So REITs valued verses other REITs, energy companies against other energy companies. Generally valuation given to group is accepted. Same as Beauty Pageant, you accept valuation metric of "judges" and only try to find bargains within metric.

anon writes:
That's because the starting odds on each woman winning would already reflect the public's perception of their relative attractiveness.

Huh... that was exactly Keynes' point - that prices are reflective of public perception (and public perceptions of others' perceptions, etc.) rather than fundamentals.

The point isn't to figure out how the market's perception will change and ride on that wave. The point is that fundamentals (your personal view of a contestant's "beauty") are often divorced from market prices (the group's view of a contestant's "beauty"). You know what happens when market prices and fundamentals diverge? Excess returns!

People always forget that, even if a company remains undervalued forever, the fact that you are getting dividends greater than what you should be, given the risk, means you outperform.

Scott Sumner writes:

I am traveling so just a few brief comments.

Jose, I don't think we can assume that a successful investor necessarily knows why he or she was successful.

Anon, My point was that eventually the truth (about beauty and stock valuation) will be revealed.

Don't try to predict what others think, just try to figure out fundamental value, and compare to current price.

TracKing writes:

Rational pricing always seems like an interesting argument. It's very compelling in many ways. Do you think public company CEOs ever think their companies are overvalued? There are certainly some signals that this happens occasionally.

I heard an interview where Elon Musk referred to his company's valuation by the public market as "generous". This was back when TSLA climbed from 50 to 100 and he used that opportunity to raise additional capital (while also buying a big chunk himself). The stock has since doubled again.

Over what time period do prices need to be efficient to count? I think this is complicated and you can kind of handwave around any argument. Why did BOND, an exchange traded fund drop and remain down for a day (even compared to the underlying assets) when Bill Gross announced he was leaving PIMCO. People make markets efficient and they have all types of informational issues. It's just complicated. I think at best we can say markets are "pretty good" and possibly even "good enough".

Jose Romeu Robazzi writes:

"I don't think we can assume that a successful investor necessarily knows why he or she was successful."

Agree. People want to think their skills allowed them to make the "right" investment decisions. When it comes to fundamentals I could not agree more.

But a speculator does not have to "know" anything, it is possible to design trading rules that take advantage of return distributions that show "a structure" (e.g. are not normal distributions). If I would summarize this philosophy, it would go on this lines:

1. Look at what prices are doing, not what the "should be doing"

2. believe the fat tails. Extreme movements are much more frequent than most people think.

3. Keep your risk under control (expected and unexpected)

4. Stay liquid

5. Design strategies that potentially will reach "large numbers" of events reasonably fast

Comments for this entry have been closed
Return to top