When Measures Become Targets: How Index Investing Changes Indexes

in·dex noun : an indicator, sign, or measure of something.

In Vietnam, under French colonial rule, there was a rat problem. To solve the rat infestation, the French offered a bounty on rats, which could be collected by delivering a rat’s tail as proof of murder. Many bounties were paid out, but the rat problem didn’t improve. Officials soon noticed rats running around without tails–people were cutting off the tails and releasing the rats to breed, so as to increase the pool of potential bounty revenue for themselves.

The same thing happened in Colonial India[i]: a bounty was offered on cobras because they were attacking people, which caused people to breed cobras for more bounties, and ultimately resulted in a higher cobra population when the bounty system was abandoned and the breeders released their now worthless snakes.

What you choose as a measure of performance—rat tails or dead snakes in these examples—is what you end up getting. Tell a portfolio manager that they will get paid on Sharpe Ratio, and you can bet that their Sharpe will improve, relative to other measures of investing success. This can become a problem.

Goodhart’s law says “when a measure becomes a target, it ceases to be a good measure.” Rat’s tails and dead cobras were measures of progress against vermin overpopulation in Vietnam and India. But those measures then became the targets of enterprising citizens, and in the process ceased to be good measures. In America, home ownership was a measure of quality of life and happiness, so it became a target, and we know how that ended. This is becoming a bigger and bigger problem in the stock market.

Stock indexes were originally designed as reference points—measurements of market success. But now they are also the targets for the most popular investing strategies in the world, for both pure indexers and closet indexers. As of 2009 when the below chart ends, roughly half of mutual fund assets are indexed or closet indexed[ii]. It’s grown a lot since then.

petajisto chart

To save ourselves as investors from some trouble down the road, we should start asking ourselves where Goodhart’s law applies in the market. Nearly free access to markets is a wonderful thing. Bogle should get a Nobel prize. But management fees are one thing, valuations another. I’d rather pay 100 bps for an S&P 500 index fund trading at 12 times normalized earnings than 5 basis points for the same market trading at 25 times earnings, which it does as of March 2016.

If indexes are now primary targets, are they still good measures of stock market performance? This issue shows up all over the place. Let’s look at an example in the world of value investing.

Price to Book: Measure, Then Target

Price to Book has been the most common measure of valuation. It was the factor originally used in Fama and French’s seminal 1993 paper (and a driving factor for the nearly $400 billion asset manager with which Fama is associated). It is also the factor that is used to define value in the most prominent style indexes: the Russell Growth and Value indexes.

Here is the cumulative excess return of the cheapest stocks by a few different measures of value between 1963 and 1993, when the paper was published, and then, below that, the same cumulative excess for the factors between 1993 and 2015, and finally between just 2007 and 2015.

value early value middle value late

As I’ve written elsewhere, how you define value has a huge impact on returns (up to 100% in the last ten years if you’d just swapped out sales for book value).

Price to book was a key measure, then it became the target around which hundreds of billions in assets built value portfolios and indexes, and along the way has decoupled from other major value factors. Everything else—sales, earnings, EBITDA, free cash flow, total yield—has been a better indicator of value since book to price was first held out as the defining value factor.

Interestingly, if price-to-book value goes fully out of favor (don’t think we are there yet), it may finally make price-to-book a good measure of value again! Watch for big names to change their definition of value from price-to-book to something else.

Measure or Target

The return of the total market will always be a relevant and accurate measure of stock market success and performance: it is, after all, reflective of the total return earned in aggregate by all investors. But I am not concerned with the aggregate, which I cannot influence. I am concerned about you and me!

Active management is on its heels, and it should be. Since the crisis, it’s lost gross and net of fees because 1) its more expensive and costs compound (which is all Bogle ever really said) and 2) large/mega cap have been doing very well, helping indexes relative to non-cap weighted active managers. But cost conscious active management can work very well for long periods of time, as it did in the early to mid-2000s following a valuation bubble in the very mega cap stocks that dominated the S&P 500 and other cap-weighted indexes. In almost every meeting these days, I hear “we are (or are moving) passive in large cap.” It is ubiquitous. The primary measure is now the primary target.

Here is Seth Klarman writing (a touch early!) in 1991 about indexing more broadly:

“I believe that indexing will turn out to be just another Wall Street fad. When it passes, the prices of securities included in popular indexes will almost certainly decline relative to those that have been excluded. More significantly, as Barron’s has pointed out, “a self-reinforcing feedback loop has been created, where the success of indexing has bolstered the performance of the index itself, which, in turn promotes more indexing.”  When the market trend reverses, matching the market will not seem so attractive, the selling will then adversely affect the performance of the indexers and further exacerbate the rush for the exits.”

Indexes are affected and changed by asset flows into strategies which target those indexes. This is true when hundreds of billions of dollars seek low price-to-book strategies, and for the entire market as a whole. Fund flows affect everything, especially when those flows have T after them. To this point, this has worked in one direction, for the most part. Portfolios with decent valuations, by historical standards, are only achievable today through portfolios that are very different from the S&P 500. Costs are very important. But there are management fees and there are valuation multiples. At the very least, I urge you to consider both.


Thanks to Alex Rubalcava for introducing me to Goodhart’s law.


[i] (apocryphally)

[ii] http://www.cfapubs.org/doi/pdf/10.2469/faj.v69.n4.7