III. Index Investing: Some Positives and an Inherent Flaw

Key take-aways

  1. There are clear benefits to passive investing—low fees most importantly, but also greater steadfastness, as cause to blame and switch fund managers falls away.
  2. Passive investing has largely been practiced as index investing, and index investing is by its nature trend-following, and a de facto price momentum strategy.
  3. Trend-following strategies participate in and contribute to herding behaviors. Such behaviors, by neglecting price discipline, dampen long-term returns.
  4. A strategy of investing in stocks with low valuation multiples, on the other hand, provides higher returns without taking any additional risk.
  5. Index investing is likely to grow in popularity, resulting in more excess return available to investors adhering to a valuation discipline.

Discussing active versus passive investing has become akin to bringing up politics or religion at the dinner table.  People have passionate views on the topic, in the event they care at all.  And once a person has a firm opinion on the topic, however he has come to hold that opinion, it is nearly impossible to change his mind.  The arguments for and against have become incredibly intellectualized—elaborate, nuanced and brain-cramping.  So, I am going to try my best to keep this discussion simple.  And I am not going to try to change anyone’s mind.  I am going to stick to some common-sense observations.  If you have not developed a hardened point of view on this topic, I hope this is helpful.

Active and Passive Investing Defined

Let’s begin with some basic explanations of what active and passive investing entail.  Both are approaches to selecting securities and putting them together in a portfolio for purpose of investment.  There are now active and passive funds for many different types of investments (or asset classes), but for writing simplicity I will focus here on stock funds.

A clarifying note on terminology: mutual funds vs exchange-traded funds (ETFs)

Discussions on active versus passive investment approaches often are confusing due to misuse of terminology; specifically, active versus passive gets mixed up with the distinction between mutual funds and exchange-traded funds (ETFs).  As we will discuss here, passive and active investing are two different approaches to selecting stocks.  Mutual funds and ETFs, on the other hand, are two different types of pooled investment vehicles (pooled meaning many individual investors buy shares in a fund, thereby “pooling” their money, so the fund can then more effectively invest in a diversified portfolio of stocks for the benefit of those investors).  The first ETF was introduced only in the early 1990s, and they differ from mutual funds in certain administrative and procedural ways.  Most notably, investors can buy or sell ETFs throughout the day, just as they would an individual stock, whereas mutual funds are purchased and sold based on end-of-day pricing.  Today, mutual funds include both actively-managed and passively-managed portfolios.  The overwhelming majority of ETFs are passively-managed, though there are some actively-managed ETFs.

In building portfolios, actively-managed funds seek to find stocks that will provide the highest risk-adjusted return.  They employ a wide variety of philosophies, strategies and processes that they believe help accomplish this objective.  Selection criteria can be both quantitative and qualitative.  They expend considerable effort and resources toward this task, as they believe it adds value in the form of higher returns.

The basic premise of passive investing is that stock markets are “efficient.”(1)  Which is to say, stocks are always fairly priced in the sense they reflect all information known to investors.  As a result, it is impossible to beat the market over time.  To test this premise, academic studies have looked at the distribution (or cross section) of returns achieved by active managers and concluded it is not significantly different (statistically speaking) from the distribution of returns of portfolios generated by randomly selecting stocks.  They further assert that even if certain funds exceed the market over time with consistency, this is still a matter of chance.  (To use an analogy to explain this, with thousands of investors each given 20 coin tosses, a lucky few will toss say 17 or more heads).  And in any case, it is difficult for investors to identify these “winning” managers in advance.  Given stock selection is thus a fool’s errand, an investor should not waste her time picking stocks or her money paying someone else to pick stocks for her.  Rather, stocks should be selected and portfolios constructed in as low cost a manner as possible, so as to minimize the drag of this cost on returns.

Benefits of Passive Investing

Low Fees

Low fees have accordingly been a distinguishing feature of passive funds.  The best example comes from the longest-running passive fund in the U.S., Vanguard 500 (originally named First Index Trust), which has returned a compound annualized 11.18% from 9/30/1976 through 12/31/2017, whereas the return for the S&P 500 over the same period was 11.28%.(2)  This difference is admirably small.  Low fees are an unambiguously good thing for fund investors.  As passive funds have accounted for an increasing share of assets in stock funds overall, the average expense ratio of stock funds on a dollar-weighted basis has fallen from 0.99% to 0.63% (99 cents per $100 invested to 63 cents per $100 invested), from 2000 to 2016.  And competition from passive funds has contributed to fees for active funds steadily declining over time.  Over this same period, fees at active funds fell from 1.06% to 0.82%.  Fees at active funds remain considerably higher than those for passive funds, which in 2016 had an average dollar-weighted expense ratio of just 0.09%.(3)  Fees at active managers will likely continue to drop.

Passive Investing Has Been Practiced as Index Investing

95442243-square Interestingly, given the statistical foundation of their core premise, as passive funds actually came into being for offer to the public, their managers did not select stocks by placing every stock ticker into a bowl and randomly pulling out enough to create a diversified portfolio.(4) Rather, they sought to replicate one of the many indices already published by index providers as a measure or proxy of the stock market.  We can think of three reasons they chose to do so: (1) this eliminates stocks that are so small, they would not work in a fund as it grew bigger, though this could have been dealt with by only placing tickers of stocks with a minimum size into the bowl; (2) the indices had a balance of companies in different industries, while picking stocks out of a bowl could result by chance in portfolios heavily skewed to one industry, undermining diversification; (3) while portfolios of randomly-selected stocks would statistically track to the average market returns over time, the portfolio returns could diverge from the market in any given year and most certainly would.

This third reason is probably the most compelling.  Returns that varied from existing measures of the market—the widely-quoted stock indices—would likely unnerve investors and undermine their confidence, especially for what was then a new approach; while tracking these indices guarantees their returns are very close to these measures each and every year.  This is a ready-made advantage, certainly from a marketing perspective, versus active funds whose returns naturally vary from the indices year-to-year, just as a randomly generated portfolio would.  To this day, most passive funds continue to track indices (though they could now throw away the bowl and use computers to randomly generate tickers).

So, passive funds are today in practice index funds.  I will use this term going forward, as it more accurately reflects these funds’ approach.  So, what’s an index?  An index is a sampling of stocks—some broader than others in terms of the number of individual companies in the sampling—as determined by an established index provider/publisher.  The index provider formulates, publishes and adheres to rules-based criteria and guidelines in constructing its index.  It applies these criteria at regular intervals, so the stocks in the index change over time.  The original objective in doing this was to provide some measure of the market as a whole or a segment of it over time.  Sorry to rehash what indices are and how they work, but I think it is important to realize that with an index fund, stocks are being selected.  But they are being selected, not by the fund manager, but by a third party, the index provider according to its methodology.  Well-known index providers include Standard & Poor’s (S&P), Russell and MSCI.  Different providers use different methodologies.  For example, Russell indices are based on strict quantitative rules, most importantly market cap cut-offs.  S&P’s process is a bit more subjective; for the S&P 500, a committee decides which companies to include each year based on eight primary criteria.

Elimination of Skill Uncertainty

In the last section, I described two investor behaviors that drive performance chasing—picking winners and sharing blame—based on funds’ recent returns.  Because index funds simply mimic a known benchmark, there is no longer cause to blame its managers, certainly not for a lapse in skill.  Eliminating the “blame game” trigger should help diminish performance chasing and enhance constancy, which we consider a second merit of index investing.

Evidence supports this intuition.  In the first graph in our section on Chasing Performance, you may have noticed that while active funds delivered a higher annualized return in the study period (9.73%) versus passive funds (9.66%), the return actually experienced by active fund investors was lower (6.85%) versus those in index funds (6.95%).  So, the return shortfall investors experienced relative to their funds was “only” 2.77% per year for index funds versus 2.88% for active funds.  Unfortunately, the shortfall for index funds is still significant.  Untimely switching remains, as the impulse to pick winners continues with index funds—certain funds will do better than others in different market environments, based on how the indices they mimic are composed.  This has been exacerbated by a proliferation of different types of index funds, including ones that track certain industries.  Most importantly, the overall problem of poor market timing through market cycles remains.

Inherent Flaw of Index Investing

This leads us to an inherent flaw in index investing.  By virtue of their design, most index funds are trend-following.  That is because most indices—and all of the most well-known ones—are market-capitalization weighted.  This means that the percentage of any stock in the index reflects the size of that company as measured by market value.  For example, at the end of 2017, the largest company in the S&P 500, Apple Inc, had a 3.61% weight in the index, while the smallest, Under Armour, had a 0.01% weight.  Given each index was originally designed to be a broad measure of the market, this makes intuitive sense.  Bigger companies should have a bigger weight, as overall they are by definition a larger weight in the overall market portfolio.  As market participants change their assessment of what companies are worth, i.e. their stock prices and thus market values change, their weight in the market portfolio and the index likewise change.

523020777 But when an index itself becomes an investment strategy, a potential problem arises.  Stock prices of companies in the index move by virtue of being in the index, as the index as a whole is bought and sold.  When an index fund tracking the S&P 500 is bought, this adds to the demand pushing up the prices of these 500 stocks.  Among the 500 stocks, the biggest companies are purchased in greater size by virtue of their weight in the index.  This price appreciation may attract additional buying from investors enticed by this strong recent performance, which in turn pushes prices up further.  And so it goes. When market-capitalization weighted indices become investment strategies per se, they are de facto trend-following, price-momentum investment strategies.  Trend following strategies participate in and contribute to herding behaviors, especially in the self-fulfilling prophecy stage.  Such behaviors, by neglecting price discipline, dampen long-term returns.

This trend-following aspect of index investing is not a problem when the investment dollars in index funds is relatively small.  The price momentum effects of these funds are swamped by the price discovery effects of active investors.  For a long time, this was the case.  The first index fund, Vanguard 500 mentioned earlier, was launched and offered to the public in the United States in August 1976.  From this humble beginning, index funds have steadily grown and in fact have exploded over the last 20 years.   Looking exclusively at funds that invest in U.S. stocks, at the end of 2016, there were 291 index mutual funds and a further 700 exchange-traded funds, according to data from the Investment Company Institute.  Of the $7.9 trillion in U.S. domestic stock funds, 59% was in active mutual funds, 22% in index mutual funds and 19% in ETFs.

In terms of stock price impact, trading volumes are more important than assets under management.  Index fund proponents rightly point out that the stocks held in index portfolios turn over less than those in actively-managed portfolios.  But trading volumes are significantly impacted by flows of investment dollars into and out of funds, as well.  As shares in funds are purchased or redeemed, their managers have to buy or sell the stocks in the portfolio.  Fund flows in recent years have been dominated by index funds, both index mutual funds and ETFs, as shown in this chart.

Constancy Investors analysis. Source data: “2017 Investment Company Fact Book: A Review of Trends and Activities in the U.S. Investment Company Industry,” 57th edition, published by Investment Company Institute (ICI), May 2017.

Over the 10 years ending December 31, 2016, nearly $1.5 trillion has flowed out of active U.S. stock funds (on a net basis, i.e. purchases minus redemptions), while over $1.3 trillion has flowed into index funds (both mutual funds and ETFs).  Any open-minded observer should be curious as to the impact, on both individual stock prices and the stock market as a whole, of such a massive shift in how people are investing in stocks.  As index funds continue to grow, the trading resulting from index fund purchases and redemptions will become more impactful on stock prices.

Now we come to the final and most important part of this discussion.  Stay with us.

Benefit of Valuation Discipline

Study after study has shown that adhering to certain disciplines for selecting stocks have consistently delivered returns over the long term that are higher than those of the broad market, as measured by indices.  As you might have guessed, this fact has become mired in the debate about whether markets are efficient and the merits of active and passive investing.  Critics of the efficient market idea argue that such studies challenge the basic premise, because efficient markets shouldn’t allow for such predictable excess returns.  Efficient market supporters, on the other hand, have simply incorporated these findings into their framework.  For example, later in his career, the author of the efficient market hypothesis, Eugene Fama, acknowledged the observation that two measurable attributes of stocks (again, “factors” in the parlance) have supported higher returns—market capitalization and valuation.  More specifically, stocks of smaller companies and stocks trading at lower valuation multiples have delivered higher returns.  Mr. Fama therefore pragmatically incorporated these factors (along with a measure of a stock’s volatility relative to the market, called beta) into his model for calculating expected return.  Very simply, if a company was smaller or priced at a lower valuation multiple (or had a higher beta), the higher its expected return.  In his revised framework, after taking these factors into account, he finds the pricing of stocks as efficient.

Here’s my position—who cares about the intricacies of this debate?  You shouldn’t.  As pragmatists, all you should care about is generating the highest returns possible, within a risk framework in keeping with your time frame.  Small cap stocks and value stocks have higher returns—that’s all we care about, period.  As illustrated in the section Taking Advantage of Time Frame under the “Time Frame and Mindset” tab above, the higher returns of small cap stocks can benefit a portfolio, but given small cap returns are more volatile, you need to have sufficient time frame.  Note this can be accomplished with index funds mimicking small cap indices, like the Russell 2000, the Russell 2500 or the S&P 600.  But, much more importantly, a discipline of investing in stocks with low valuation multiples—and which can be applied to stocks of both small cap and large cap companies—provides higher returns without taking any additional risk.

In the last section on Chasing Performance we looked at a valuation factor, specifically the ratio of stock price to earnings per share over the last twelve months (P/E) for companies in the Russell 1000, a broad market index.  Let’s look again at P/E ratios, but now for stocks in the S&P 500, given this is the index most tracked by index funds.(5)  I also lengthen the study period to year end 1992, as far back as our source has the relevant data, but only rebalancing at the first of each year (versus monthly previously).  I specifically look at the annual growth of $1,000 invested in the S&P 500 versus a portfolio of those S&P 500 stocks with the cheapest P/E ratios (bottom 20%, after excluding companies with negative earnings), measured and rebalanced at the start of each year.  For comparability, I market-cap weight the low P/E portfolio, consistent with the S&P 500 methodology.

Constancy Investors analysis: Growth of $1 invested in stocks of companies in the S&P 500 Index that are in the lowest (cheapest) 20% on Price/Last-Twelve-Months Earnings (LTM P/E), rebalanced annually, versus growth of $1 invested in the S&P 500 Index. Data source: Bloomberg.

As you can see in the graph above, consistent with every other study on the topic, this analysis shows the higher returns resulting from a valuation discipline, low P/E ratios in this case.  To quantify, the compound average annual return for the S&P 500 was 9.7%, while that for the bottom quintile P/E (i.e. lowest or cheapest 20%) was 11.7%.  Given the power of compounding, while the $1,000 invested in the S&P 500 index grew to $10,080, the same investment in the low P/E portfolio grew to $15,815 over the 25-year study period.

Passive funds as they mostly exist today, as index funds, do not participate in the excess return consistently realized by low valuation strategies.  The indices most heavily tracked by index funds do not take into account valuation multiples in stock selection.  Index funds thus rely on other investors to price stocks efficiently.  Their own stock selection criteria (by market capitalization) result in a de facto trend-following, price-momentum strategy. 

Index investing has been a step in the right direction for investors.  It has focused investors on important advantages—less of a drag on returns from fees, potentially greater constancy by eliminating blame.  But index investors must also understand and acknowledge they are forgoing returns by investing in strategies that go along with and in fact contribute to the trend-following behaviors that amplify cycles.     

It is fitting to conclude this section as we began.  On the topic of passive versus active investing, we are not going to change anyone’s mind.  Based on the long-term trend, their current cost advantage and the marketing muscle now available to the large fund companies committed to indexing, we expect that index funds will continue to grow in popularity and become an increasing share of all stock funds.  With more investor dollars employed in index funds, which are by their nature trend-following, I would expect this to add to the performance-chasing dynamic that has always operated in investment cycles.  If I am correct, this means more excess returns will be increasingly available to those that invest according to the proven disciplines, valuation above all.

In the next section, A Sensible Approach, I elaborate the common sense reasons why the valuation factor—valuation discipline, as I prefer to call it—works.  And under the Invest tab above, I specifically explain how these disciplines can be systematically applied in investment portfolios.

(1)The efficient market hypothesis (EMH) is an idea that was formalized by Eugene Fama in 1965 and has been much studied and developed in academic finance ever since.  The reasoning goes that because market participants are trying to generate returns, they gather information to give them an edge in buying and selling stocks.  In so doing, prices become efficient in that they embed all known information.  Stock prices therefore move randomly, as new information arises.  A corollary of the EMH is that all the effort to gather information undermines the original motivation for that effort, which is to say it becomes impossible to earn above-market returns over time (adjusted for how much risk is taken, with the volatility concept of beta as the measure of risk).

(2)The index return is in fact purely theoretical.  An index fund’s returns will be less than the return of the index it is tracking.  This is due to (1) the fees that the index fund charges to cover the administrative costs of running the fund and its profit, (2) trading commissions it pays to brokers in buying and selling stocks, and (3) “friction costs” which results from moving the prices of the stocks around at the time of purchase or sale.  Note these are “buy and hold” total returns for both, including the assumption that dividends are automatically reinvested back into the Vanguard 500 and S&P 500, respectively.  Dollar-weighted returns (i.e. the returns investors actually experienced based on when they actually had money in the index or the fund) likely fall short, due to investors’ tendency to time purchases and sales poorly, as discussed in the Chasing Performance section.

(3)The source for this data on expense ratios is “2016 Investment Company Fact Book: A Review of Trends and Activities in the U.S. Investment Company Industry,” 56th edition, published by Investment Company Institute (ICI), May 2016.  Expense ratios include portfolio management fees and administrative costs (fund administration, daily accounting and pricing, shareholder services, distribution costs and other costs).  The expense ratios cited are averages calculated on an asset-weighted basis; simple averages are substantially higher (about 0.60% higher).  This is because most investors fortunately put more money in funds with lower costs.  This is true for active funds and even more so for passive funds.  In turn, larger funds can spread their costs, some of which are fixed, over a larger asset base.

(4)For a passive fund to truly replicate an entire market, for example the entire U.S. stock market, a fund would have to buy a bit of the many thousands of publicly-traded stocks in the U.S.  To do so would be impractical and administratively challenging.  Thus, the need and practice of using a representative sampling.

(5)With respect to U.S. domestic stock mutual funds, 47% of index funds track the S&P 500 index.  Source: “2016 Investment Company Fact Book: A Review of Trends and Activities in the U.S. Investment Company Industry,” 56th edition, published by Investment Company Institute (ICI), May 2016, p. 219.  An equivalent breakdown for the S&P is not available for U.S. domestic stock ETFs.