Don’t Look Back

When the Michelin brothers Andre and Edouard began their travel guide in 1900 as a marketing tool for their tire company, there were fewer than 3,000 cars in France. The roads were a primitive, local network, often unmarked, and there were no gas stations – drivers had to carry a metal container into a pharmacy to purchase fuel.

The Michelin Guide helped ease the uncertainties of automobile travel and in 1926 was expanded to include fine-dining restaurant reviews with the now-famous star ratings: one star for “a very good restaurant in its category,” two for “excellent cooking, worth a detour,” and three stars for “exceptional cuisine, worth a special journey.”

How Michelin awards its stars remains somewhat of a mystery, but it is reportedly a painstaking process. Its inspectors, often trained chefs, may visit a restaurant numerous times over half a year or more to evaluate the quality of ingredients, “mastery of flavour and cooking techniques, the personality of the chef in his cuisine, value for money and consistency between visits.”1

Afterwards, according to Michelin, there are debates and sometimes “heated discussions” before agreement is reached on how many stars a restaurant should receive. A three-star restaurant must create a meal that “is an emotional experience that is engraved in one’s memory for many years to come.”

The world of investing awards stars, too, most notably the five-star rankings assigned to mutual funds by the data-analytics firm Morningstar. However, the evaluation process is far more mundane: Morningstar checks a fund’s historical performance (returns & volatility) over the past three, five and sometimes ten years, comparing the results with the fund’s peers. The best historical performance gets the most stars.

The problem is, an investment’s historical performance tells us nothing about how it will perform going forward. To illustrate, consider the track records of three investment strategies offered by two competing fund companies that we’ll call Roadrunner and Coyote.

The strategies target large U.S. growth stocks, developed international stocks, and emerging-market stocks. Here are the ten-year performance results for each company’s funds:


As the chart shows, an investor would have done far better in Roadrunner’s funds than in Coyote’s funds. What’s noteworthy, though, is that each company’s funds are identical in every respect except one: the date they started.

Coyote launched its funds in January 2008; Roadrunner launched eleven months later in December 2008. Within a span of ten years, eleven months doesn’t seem like much. However, the result was that the Coyote funds had to suffer through the entire 2008 market crash, causing its historical record to fall off a cliff compared to Roadrunner, which fortuitously sidestepped most of the declines.

In reality, the fund strategies above are the historical returns for three market indices – the S&P 500 Index, the MSCI EAFE Index, and the MSCI Emerging Market Index – starting eleven months apart. As we can see, where you start and stop the period for evaluation with each index has an enormous effect on the performance results you see.

This phenomenon of “endpoint sensitivity” affects all types of data series, but is especially problematic for investors trying to make investment decisions based on how their portfolios have been doing recently. Depending on where you stop the clock, certain asset classes will be up or down, in both absolute terms or relative to each other over the period you’re evaluating. However the performance during that period tells you virtually nothing about what will happen next with each of the assets.

At best, the historical performance data is a starting point for inquiry, raising questions about what drives the source of return for any investment. The answers to those questions will just as likely come from subjective, qualitative assessments not unlike the ones that the Michelin inspectors use to assess the likelihood of consistency in a restaurant’s offerings.

This is why a lot of smart investors ignore price-movement data altogether. As the Chief Investment Officer of Vanderbilt’s endowment fund recently noted:

What does a track record tell you, really? It’s a measure of past performance, but generally not statistically significant. Our goal is to assess the probability that a manager will generate strong results in the future, not to hope that what happened in the past will persist. We make every attempt to be objective in our analysis, but we make little effort to perform quantitative assessments.2

It’s difficult, though, to persuade an investor that past performance doesn’t matter. We don’t yet have any data about the future, so what else is there? Experience matters in so many other areas of life, so why not here?

Investment marketers are more than happy to support these beliefs and even pay Morningstar for the right to advertise a fund’s star ratings. The ads often display the stars more prominently on the page than the name of the fund itself. Regulatory disclaimers about “past performance, yada, yada, yada” are buried in the fine print at the bottom.

Morningstar itself emphasizes that its star ratings aren’t meant to be predictive, but acknowledges that “[m]illions of people trust Morningstar Inc. to help them decide where to put their money.”3 Meanwhile, it continues to tally the last three, five and ten years of historical data and award its stars to the Roadrunners over the Coyotes, knowing it’s likely that they’ll switch places in the coming years.

If Morningstar really wanted to help investors understand that the historical data should be used more for asking questions than answering them, there is a simple solution it could implement: change its Stars to Question Marks. A fund company willing to acknowledge how little its recent performance matters and to display those question marks next its name would definitely be worthy of further exploration.

1 History of the Michelin Guide [].
2 “Why Starting a Hedge Fund Is Like Opening a Restaurant,” SumZero Headlines, December 7, 2018.
3 “The Morningstar Mirage,” The Wall Street Journal, October 25, 2017.