« Mad Men | Main | Assorted links »
What is the aggregate cost of trying to beat the stock market?
Investors collectively spend around $100 billion a year trying to beat the stock market. That’s the finding of a rigorous effort to measure the total costs of Americans’ efforts to surpass the returns they would have received by simply holding a stock index fund. The huge price tag helps explain why beating a buy-and-hold strategy is so difficult.
Here is much more, and from Felix Salmon. You can think of this sum as the amount it costs to keep markets relatively efficient, a source of societal fraud and rent-seeking, a Nash equilibrium mixed strategy (no one tries to beat the market on every margin), a donation to the broader social good, or most properly all of the above. Interfluidity adds comment.
Posted by Tyler Cowen on March 10, 2008 at 01:10 PM in Economics | Permalink
Comments
My biggest problem with those who claim buy and hold as a default option is that such a policy is awfully hard to maintain in practice, noting that people are emotional humans, not machines willing to follow the advice of obscure professors.
How many people bought into the notion of buy and hold, and following that advice they held the Nasdaq index as part of their portfolio, right up until late 2002 or early 2003, when they instead cried uncle and dumped their entire position and then bought houses as an investment instead? Care to guess when they capitulate with their housing investments?
Posted by: happyjuggler0 at Mar 10, 2008 1:52:14 PM
isn't this just signaling? Buying index funds signals that you are average or worse at investing and you don't aspire to anything more. Picking stocks and buying more active funds signals that you are a confident risk-taker. You might lose money, but you'll impress many more people at parties.
Posted by: DK at Mar 10, 2008 1:58:59 PM
"You can think of this sum as the amount it costs to keep markets relatively efficient, a source of societal fraud and rent-seeking, a Nash equilibrium mixed strategy (no one tries to beat the market on every margin), a donation to the broader social good, or most properly all of the above."
Actually I think it's more properly thought of as part of the entertainment industry. This is not a frivolous point. The more closely you look at how most investment advisors behave, the more economically accurate this classification seems.
Posted by: ZF at Mar 10, 2008 2:16:19 PM
I'm confused.
If no one, anywhere, ever tried to beat the market, who would buy (or sell) the stocks that index funds are tied to? And if there were no buyers/sellers, how would their stock prices rise?
Posted by: Patrick Minton at Mar 10, 2008 2:31:08 PM
You can get Buffett's letters for free, and if they don't make sense to you, stop trying to beat the market.
Posted by: Andrew at Mar 10, 2008 2:40:07 PM
Of course, approximately half of the people (or, half of the money) who attempt to beat the stock market are successful.
Money spent to money managers who handle Harvard's endowment is very well spent.
The real question is how much was the aggregate cost paid by people who FAILED to beat the stock market. Assumedly, a large number -- but is it more than half of $100 billion?
One wonders whether any money managers are willing to take their fee in "% of the amount I beat the market over a five year period."
Posted by: Rich B. at Mar 10, 2008 2:42:45 PM
Patrick,
That's a good question, one these efficient market folks never seem to address. But it's also not that important. The only stock pickers that matter are the ones who are right. We'd be better off if all the other noisemakers in the market quit trying and tried to do better at their day jobs. If the marginal imbecile quit trying to beat the market, the people who know they are good wouldn't stop.
Posted by: Andrew at Mar 10, 2008 2:43:53 PM
If the marginal imbecile quit trying to beat the market, the people who know they are good wouldn't stop.
But this is where the equilibrium comes from. If you can beat the market, you can make a lot of money. But once the market becomes too efficent, you can't make enough money doing it for it to be worth your while. If things actually went like this, you would see cycles. Actually, the mechanisms are assuredly far more obscure. But isn't the marginal imbecile's participation necesary for others to beat the market? Or to put it another way, is it possible to have only people who buy and hold and only people who beat the market buying and selling? Do the people who buy and hold then slightly underperform?
Posted by: mpowell at Mar 10, 2008 2:50:58 PM
If no one, anywhere, ever tried to beat the market, who would buy (or sell) the stocks that index funds are tied to? And if there were no buyers/sellers, how would their stock prices rise?
Assumedly, retirees and pensioners would sell their index funds (and the underlying stocks) to current investors who are dollar cost averaging every month. Stock prices would be determined by the underlying values of the assets during these regular transactions.
Posted by: Rich B. at Mar 10, 2008 2:59:53 PM
Or to put it another way, is it possible to have only people who buy and hold and only people who beat the market buying and selling? Do the people who buy and hold then slightly underperform?
Yes and yes. If all noise traders disappeared, speculators and liquidity suppliers (market makers, contrarian traders etc.) would earn returns at passive investors' expense. This means that there's no "tragedy of the commons": Steve Waldman seems to be incorrect.
Posted by: guest at Mar 10, 2008 3:52:25 PM
While I personally agree with the sentiment, throwing up absolute numbers about how much investors collectively spend per year isn't effective. Why stop at annual waste? For dramatic effect, why not list the global costs in the entire history of the market?
Would it not make more sense to provide relative measures? Why not compare these numbers to the total money spent on investing, or the money spent to research and invest in index funds? Maybe even divide this number by the number of investors so that we can have a per capita number that we can benchmark against, say, annual income?
Shiran
Posted by: Shiran Pasternak at Mar 10, 2008 4:01:27 PM
@Rich B.:
"One wonders whether any money managers are willing to take their fee in '% of the amount I beat the market over a five year period.'"
An excellent description of carried interest in a private equity fund structured to use both clawbacks and a preferred return.
Posted by: at Mar 10, 2008 4:34:20 PM
"You can think of this sum as the amount it costs to keep markets relatively efficient, a source of societal fraud and rent-seeking, a Nash equilibrium mixed strategy (no one tries to beat the market on every margin), a donation to the broader social good, or most properly all of the above."
Yeah--thank you, stock pickers. I've been free-riding on you ever since Vanguard started selling index funds to small investors, and I'm grateful. One additional benefit: you can listen to news people explaining that the cause of any downturn after a good day is "profit taking." So all this activity has bought us a moron detector, too.
Posted by: Alan Gunn at Mar 10, 2008 6:27:42 PM
"Yeah--thank you, stock pickers. I've been free-riding on you ever since Vanguard started selling index funds"
Eactly what I was going to say, almost to the word.
Posted by: Slocum at Mar 10, 2008 7:28:11 PM
Patrick, those who believe in an efficient stock market and indexing do acknowledge that active traders
are necessary to maintain the market's efficiency. Currently a small percentage of shares are held by index funds, so this is not a problem. Nor probably will it ever be, given human nature.
Posted by: Scott F. at Mar 10, 2008 7:53:24 PM
One problem I have with index funds is that the most popular ones are easily manipulated. I still remember when AOL was added to the S&P 500. Smart traders made millions of dollars buying AOL on the announcement and then selling it to the index funds on the market close the day before the stock was actually added. After seeing that, I'll never buy an index fund.
Of course, if you really buy into the idea that you can't beat the market, then you don't have to buy an index fund. Just randomly pick 5 or 10 stocks and buy and hold, avoid the manipulation, and avoid the management fees. At $7/trade and 10 stocks, holding for 10 years, you only need $7368.42 before you can match that 0.19% management fee, and if you don't have even *that* much in your retirement account you can live with a little less diversification anyway.
Personally I do my research and buy the stocks I think are going to perform the best. In my IRAs I buy one stock every year after making my $4000 contribution (round trip commission of 0.35%). And then I hold unless I come up with a really compelling reason not to.
Posted by: Anthony at Mar 10, 2008 9:39:51 PM
I think ZF's argument this be seen as part of the entertainment industry is a good one, at least for retail investors. It should be clear that they are utterly outclassed in terms of transaction costs and information by institutional players, making high frequency trading of stocks (and even worse, trading options), a wealthy person's lottery. The only way I can see these people adding information and overall market value is by having genuine inside information like news of an impending drug trial failure.
Posted by: jonm at Mar 10, 2008 10:17:34 PM
The flipside of believing that you can't beat the market is believing that the market can't beat you. In that case, you'd be happily buying Pets.com and Nortel at the dizzying heights of the dot-com era, or selling your house for a tulip bulb, because that's what a by-definition rational and efficient market said they were worth.
The efficient market hypothesis presumes that all investors are equally rational and well-informed. Neither really holds true. Behavioral economics is putting paid to the notion that humans are rational, and in fact recent research in datacrunching market numbers has found some decidedly non-random-walk effects: the unreasonable effectiveness of momentum investing, for instance. And the modern Internet, with its huge selection of highly specialized blogs and other resources, can give a decisive advantage to those who have the time and ability to datamine it to filter useful information from the noise. This means information asymmetries are larger now; it's a very different world from when everyone read the same Wall Street Journal and Fortune articles. If you were reading well-regarded economics blogs, for instance, you knew about the recession in December already, but that story flew under the radar of the mainstream media until January.
Finally, passive index-fund investing has really worked well in the last 25 years, but there have been long periods when it has worked very poorly indeed: 1930 to the early 1950s, 1965 to 1982, 1989 to today and counting for the Japanese stock market, 2000 to today and counting for the NASDAQ Composite index. It is very likely that we are again entering such an era, and the recent period was more of an anomaly than the norm.
Posted by: at Mar 11, 2008 1:32:10 AM
Evidently we are choosing to right off any value to the economy based on stock picking; when in fact the nature of the profession is to choose which companies to allocate capital to. Who can we lend to and what rate to make it worthwhile. As markets are to an extent now breaking down their is real value to this. Without stock pickers making markets efficient than who chooses not to give capital to bad companies. Ideally, this would prevent dot.com bubbles where bad companies get cash.
Posted by: sean at Mar 11, 2008 5:31:44 AM
People who buy index funds are on average buying overvalued stocks relative to some theoretical fair value: once a stock is slightly overvalued, passive buyers push it further into overvalued space. A successful active investor should be able to take advantage of this by "arbitraging" overvalued stocks back to fair value. If such a scenario held in practice, we would have passive investors that lose a little bit vs benchmark all the time, and this "lost value" would go to active investors. This would hold also if there were no losers among the latter.
Posted by: CRP at Mar 11, 2008 7:39:24 AM
"The efficient market hypothesis presumes that all investors are equally rational and well-informed."
No it doesn't. It presumes that *some* people are perfectly rational and well-informed, and that those people are willing to enter into highly leveraged transactions of boundless magnitude until the market price is driven up or down to the fair value.
It's still a somewhat ridiculous assumption, at least if you take it literally down to the penny, but not *quite* as ridiculous as your assertion.
Posted by: Anthony at Mar 11, 2008 8:28:13 AM
I'm not adding anything that the above comments have overlooked, but I offer a journal article and a pop piece to elaborate on some of the above points:
* The simplest and best model I have seen, which solves the "what if everyone believed in efficient stock prices?" paradox, is the approach taken by Grossman and Stiglitz (e.g. here). Grossman's book The Informational Role of Prices is a wonderful collection on this stuff, by the way. E.g. he deals with futures prices in just the way you'd want from a neoclassical model. In other words, this is a rare occasion when I think the formal model really does what it's supposed to--it chooses the real-world cause and models it simply, rather than focusing on something that in reality isn't that relevant and elevating it to supreme importance in the model.
* I spell out the Hayekian implications of stock speculation in this pop piece.
Posted by: Bob Murphy at Mar 11, 2008 10:44:04 AM
Someone should write (after discriminating research) a substantial report on the History of Intermediation (in the U.S., with comparisons to other commercial areas).
The "elimination" or displacement of the function of what used to be called "middlemen" is an ongoing process and succeeds to the extent the market places are "open," and not constrained by external intrusions (culture, tradition, inertia, governments...).
However, that same degree of openess creates opportunities for some to "make them a job," partially based on specialization of labor, divisions of efforts; as evidenced by the rise of those who devised and marketed various derivative finacial instruments, established mutual funds, REITS, often creating value from efficiencies, which are succeeded by excessive intermediation (beyond viable needs).
Those are honorable callings, but like so many other callings, they are not always answered by persons who can remain honorable.
Posted by: R. Richard Schweitzer at Mar 11, 2008 10:57:26 AM
"carried interest in a private equity fund structured ... clawbacks and a preferred return"
If you're actually investing under this assumption, I hate to burst your bubble, but you're wrong. To the extent that the hurdle is close to the market return, you ignore non-management (transaction, board, financing, etc.) fees, ignore the time value of the management fee stream, ignore the (dis)incentive effect of large, guaranteed, management fee income, ignore leverage (risk), and finally throw out the option value of carry, maybe you're right.
Bottom line ... only people who've never spent any serious time actually investing in PE funds would assume that GPs are compensated only for the amount by which they outperform the market.
Posted by: Mike G at Mar 12, 2008 9:22:57 AM
Another item the EMT folks never bring up is governance. If everyone bought indexes, there would be no shareholders with concentrated interest enough to offset management. We, in fact, don't have to go to the theoretical extreme because this is already a problem.
Posted by: Andrew at Mar 12, 2008 2:34:59 PM
