A Comparison of Passively Managed, Small-Value Funds

The result of the differences in fund construction rules means that three passively managed funds in the very same asset class can produce three very different outcomes. Since all passively managed funds are simply following their construction rules, assuming they are efficient at minimizing trading costs, they are all accomplishing their specific objectives.

The following table, with data from Morningstar as of January 31, 2020, shows various value metrics for three passively managed, small-value funds from three different fund families: the index fund of Vanguard and the structured funds of Dimensional and Bridgeway. (Full disclosure: My firm, Buckingham Strategic Wealth, recommends Dimensional and Bridgeway funds in constructing client portfolios.) The table provides the weighted average market capitalization to show the relative exposures to the size premium, and four different value metrics – price-to-earnings(P/E), price-to-book (P/B), price-to-sales (P/S), and price-to-cash flow (P/CF) – to show the relative exposure to the significant premium provided by value stocks.

VISVX has a much larger average market cap, more than twice that of DFSVX and more than four times that of BOSVX. In addition, it has significantly higher valuations relative to each metric than DFSVX and BOSVX. DFSVX also has more than twice the market cap of BOSVX and has higher P/E, P/S and P/CF valuations than BOSVX.

With the data and concepts in mind, let’s now take a look at how the funds have performed – did we get what we expected? Since the first full year for Bridgeway’s fund was 2012, we will examine the returns for the last eight calendar years (2012–19) and year-to-date (March 3, 2020). To see if we got what we expected, we will also look at the returns of Vanguard’s 500 Index Fund (VFINX). The following returns data is from Morningstar.

Vanguard’s small-value fund includes about 14% of real estate investment trusts (REITS), while Dimensional’s and Bridgeway’s funds do not because REITs are treated by those firms as separate asset classes. That can create/explain some of the differences in performance.

In 2012, VFINX returned 15.8% and VISVX returned 18.6%. Since small value outperformed, we should again expect to see DFSVX outperform VISVX – and that’s what happened, as DFSVX returned 21.7%. Bridgeway’s BOSVX returned 17.7%. Given its greater exposure to the size and value factors, we should have expected BOSVX to outperform. The likely explanation is that with a new fund, the risk of tracking error is greater as the fund builds its positions.

In 2013, VFINX returned 32.2% and VISVX returned 36.4%. Since small value outperformed, we should again expect to see DFSVX outperform VISVX – and that is what happened, as DFSVX returned 42.4%. And BOSVX outperformed both, returning 44.6%, just what we should have expected. Vanguard’s performance was hurt by the performance of REITS – Vanguard’s Real Estate Index Fund Admiral Shares (VGSLX) returned just 2.4%.

In 2014, VFINX returned 13.5% and VISVX returned 10.4%. Since small value underperformed, we should expect to see DFSVX underperform VISVX – and that is what happened, as DFSVX returned 3.5%. And BOSVX underperformed both, returning 0.8%, just what we should have expected. VISVX also benefited from the strong performance of REITS – VGSLX returned 30.3%.

In 2015, VFINX returned 1.3% and VISVX lost 4.8%. Since small value underperformed, we should expect to see DFSVX underperform VISVX – and that is what happened, as DFSVX lost 7.8%. BOSVX also underperformed VISVX, losing 6.6%, though it unexpectedly slightly outperformed DFSVX. VISVX again benefited from the relative performance of REITS, as VGSLX returned 2.4%.

In 2016, VFINX returned 11.8% and VISVX returned 24.7%. Since small value outperformed, we should expect to see DFSVX outperform VISVX – and that is what happened, as DFSVX returned 28.3%. And as we should have expected, BOSVX outperformed both, returning 34.5%. The performance of VISVX was negatively impacted by the performance of REITS, as VGSLX returned just 8.5%.

In 2017, VFINX returned 21.8% and VISVX returned 11.7%. Since small value underperformed, we should expect to see DFSVX underperform VISVX – and that is what happened, as DFSVX returned 7.2%. And BOSVX underperformed both, returning 6.0%, just what we should have expected. REITs hurt VISVX, as VGSLX returned 4.9%.

In 2018, VFINX returned -4.58% and VISVX returned -12.3%. Since small value underperformed, we should expect to see DFSVX underperform VISVX – and that is what happened, as DFSVX returned -15.1%. And BOSVX underperformed both, returning -17.2%, just what we should have expected. VISVX benefited from the relative performance of REITS, as VGSLX lost just 6.0%.

In 2019, VFINX returned 31.3% and VISVX returned 22.6%. Since small value underperformed, we should expect to see DFSVX underperform VISVX – and that is what happened, as DFSVX returned 18.1%. And BOSVX underperformed both, returning 13.9%, just what we should have expected. VISVX benefited from the relative performance of REITS, as VGSLX returned 28.9%.

Through March 3, 2020, VFINX lost 6.8% and VISVX lost 12.5%. Since small value underperformed, we should expect to see DFSVX underperform VISVX – and that is what happened, as DFSVX lost 17.2%. And BOSVX underperformed both, losing 19.4%, just what we should have expected. VISVX again benefited from the relative performance of REITS, as VGSLX lost just 2.1%.

The exposure to the factors of size and value explain the relative performance of the three funds – there were just two outliers in the data, the underperformance of BOSVX in 2012 and its outperformance of DFSVX in 2015 (in both cases the likely explanation is random tracking error).

Takeaways

There are a couple of important takeaways. The first is that all three passively managed funds were all doing their jobs well. The differences in performance weren’t explained by good or bad management. Instead, they were explained by the fund’s structure – how they’re designed, and the “laws of style purity.” When an asset class does well, you should expect the fund with the most exposure to the factors explaining its outperformance to have the highest return. And when an asset class does poorly, you should expect the fund with the most exposure to the factors will underperform.

The second is that choosing a fund should be based on how much exposure you want to the factors, and also a fund’s expense ratio. When making the decision, you want to be sure you weigh both. It might be that the fund with a higher expense ratio is the better choice, as it might have more exposure to the factors that determine returns and carry premiums. In other words, it is not just cost, but cost per unit of expected return (and risk), that matters.

Another more important point to cover is the psychological bias known as “recency.”

Recency bias

Recency bias causes investors to allow recent returns to dominate decision-making while ignoring long-term evidence. Over the most recent decade, small value stocks have underperformed the large and more growth-oriented stocks that make up the S&P 500. As we have seen, that underperformance explains the relative performance of the three small value funds, with VISVX producing the strongest returns.

I’ve learned that one of the greatest problems preventing investors from achieving their financial goals is that when it comes to judging the performance of an investment strategy, they believe that three years is a long time, five years is a very long time, and 10 years is an eternity. Even supposedly more sophisticated institutional investors – those who employ highly paid consultants – typically hire and fire managers based on the last three years’ performance. On the other hand, financial economists know that when it comes to investment returns, 10 years can be nothing more than “noise,” a random outcome. For example, we have had three periods of at least 13 years over which the S&P 500 underperformed riskless one-month Treasury bills:

• 17 years: 1966-82
• 15 years: 1929-43
• 13 years: 2000-12

Financial economists know that all risky assets go through long periods of poor performance, and it requires discipline to stay the course. Such long periods should not cause you to abandon your belief that riskier assets should have higher expected (but not guaranteed) returns. The three periods of 13 years or longer of negative equity premiums demonstrate that this is just as true of large stocks as it is of small value stocks. The long-term evidence is that small value stocks have outperformed the S&P 500 Index. According to Ken French’s data library, from 1927 through 2019, the Fama-French small value research index returned 14.5%, outperforming the S&P 500 Index’s return of 10.2%, by 4.3% per annum. The outperformance is intuitive, as small value stocks are much riskier. For example, since 1927, while the standard deviation of the S&P 500 has been around 19%, it’s been about 28% for small value stocks. Thus, we should not abandon our belief that small value stocks should outperform in the future, nor should we abandon our belief that DFSVX and BOSVX should have higher expected returns than VISVX because they have more exposure to the small and value factors.

Recency bias helps explain why investing is the only endeavor in which, when something gets cheaper, investors want to buy less. The result of that bias has been that the performance of small value stocks over the past decade has led many to abandon them. On the other hand, successful investors have learned to buy stocks like they buy socks – when they are on sale! And the underperformance of small value stocks over the past three years has resulted in their being the cheapest (relative to the market) they have been since the Great Depression, the only exception being the end of the tech bubble era.

Morningstar ratings

Despite the fact that Morningstar’s ratings have less predictive value than simply ranking by expense ratios, many people rely on them when choosing funds (likely because they are unaware of the evidence). Even Morningstar’s Vice President of Research John Rekenthaler stated: “There’s actually not much difference between mid-ranked funds and top-rated ones. Three-star, four-star and five-star funds have been found to perform pretty much alike.” Which is just what the study, “Morningstar Ratings and Mutual Fund Performance,” found.

One problem with Morningstar’s rating system is that it overweights recent performance, placing more weight on three-year performance than on five- and 10-year performance, and ignoring longer-term performance. Relying on such ratings has systematically resulted in the underperformance of pension plans. In addition, short-term performance should have no relevance when it comes to passively managed funds whose performance is based on that of the asset class, not the manager. And as we know, all risky assets go through long periods of poor performance.

Despite the fact that all three small value funds were doing exactly what they were designed to do, Morningstar currently gives VISVX a five-star rating, DFSVX a three-star rating, and BOSVX just a two-star rating. That’s because of the recency bias in the ranking – one that you should ignore.

Summary

The returns of the three small-value funds we examined are well explained by their exposure to common factors (market beta, size, value and quality). And while VISVX has the lowest expense ratio, it’s cost per unit of exposure that matters, not just cost. When choosing a fund, you should also consider the fund’s trading strategy (does it act as price-taker, or does it trade patiently), its fund construction rules (such as the use of the aforementioned momentum screens), and its exposure to the common factors that explain returns.

The following table, with data from Portfolio Visualizer, provides the loadings (amount of exposure) for the three funds to various factors. The data covers the period September 2011 (inception of BOSVX) through December 2019. The value exposure is measured by the price-to-book ratio. BOSVX has the highest exposure to the size and value factors as well as the quality factor (because of its use of multiple value screens, including not just P/B but also P/CF and P/E and P/S).

As you consider your investments in small value, make sure you are not being influenced either by recency bias or the related dreaded condition of “relativism” – comparing the performance of your portfolio to that of a popular index reported on daily by the media. These conditions can cause you to stray from a well-thought-out plan of diversifying your sources of risk, which leads to what I consider the greatest anomaly in investing. While investors idolize Warren Buffet, they tend to not follow his advice: Avoid market timing, but if you can’t, then be fearful when others are greedy and greedy when others are fearful.

To help you avoid such mistakes, the table below presents the value metrics of three small-value funds and compares them to that of the S&P 500 Index. Data is from Morningstar as of January 2020.

At the end of 1994, shortly after the publication of the famous Fama-French research on the size and value factors, the P/E ratio of the S&P 500 was 14.5. At the end of January 2020, it was almost 28% more expensive. On the other hand, the P/E ratio of small value stocks was 11.7 at the end of 1994, about the current average P/E of the three small value funds, and in the case of DFSVX and BOSVX, the current P/Es are lower. In other words, the spread in valuations between value and growth stocks have widened dramatically.

Since there is no evidence of investor ability to forecast such changes in valuations, the question investors should be asking is: What tends to happen after such periods of changing valuation spreads? Avantis’ research team provided the answer by examining the performance differences over the five years following periods of above- and below-average valuation spreads. This should be of particular importance because the valuation spread between value and growth stocks is currently at a historically extreme high level, the 97th percentile in the U.S. And it is even more extreme in international markets.

1980-2019

Subsequent Five-Year Return Difference Between Value and Growth (%)

Data from 12/31/1979-12/31/2019. Periods greater than one year have been annualized. Past performance is no guarantee of future results. Stocks are represented by Russell size and valuation indices. Source: Avantis Investors, Morningstar, Russell.

Periods following above-median valuation spreads experienced much better future relative performance for value stocks versus growth stocks. The premium was almost seven times as large (0.6 versus 4.0) after periods of above-median valuation spreads than after periods of below-median valuation spreads. In addition, over all periods the performance gap is much wider (4.5 times as wide) in small than in large stocks (0.8 versus 3.6). This shows that there is information in the spread in valuations between growth and value stocks. For example, in terms of the cash flow-to-price (price-to-book) metric, Avantis found that when the spread was in the 25th percentile, over the year value underperformed growth by an annual average of 3% (6%). However, over the next 10 years value still outperformed growth by an annual average of 3% (2%). Note the importance of taking a long-term perspective. When the spread was at the 75th percentile, over the next year value outperformed by an annual average of 12% (14%), and over the 10 years it outperformed by an annual average of 7% (7%).

This certainly doesn’t guarantee the future outperformance of value stocks; there are no guarantees when investing in risk assets. Investing is, however, about putting the odds in your favor. And here the evidence is clear – the odds favor value over growth by more than almost any other period we have experienced, with the only recent exception being the bursting of the bubble in growth stocks in March 2000. The following table shows you what happened when that bubble burst.

2000-2007

Thanks to the research team at Dimensional, we can also examine the performance of the value premium (as well as the market beta and size premiums) after periods when they underperformed for 10 years. The data covers the period June 1927 through December 2018 and shows the premium over the succeeding 10 years.

Once again, we see the importance of discipline and patience. Those investors who abandoned a value strategy after 10 years of a negative premium missed out on an annualized premium of more than 8% over the following 10 years. The results are very similar to those for the market beta premium. The important message is that 10 years for risk assets is likely to be nothing more than noise, and therefore discipline, adhering to your well-thought-out plan, is the key to investment success. Those who abandon their strategies get the worst of all worlds. They feel the pain of the losses, and then when they abandon the plan, they don’t reap the benefits.

Since the goal of investing is to buy cheap and sell expensive, ask yourself if you should consider reducing or even eliminating your exposure to small-value stocks, or adding to it, perhaps simply by rebalancing and adhering to your long-term plan. As you consider the decision, here’s one more data point to remember. Over the five-year and two-month period from January 1995 through February 2000, VFINX outperformed DFSVX by a cumulative 88%. By the end of 2001, less than two years later, DFSVX’s total return over the full period 1995-2001 had surpassed that of VFINX, 190% versus 180%. That’s how fast return trends can reverse.

Larry Swedroe is the chief research officer for Buckingham Strategic Wealth and Buckingham Strategic Partners.

Important Disclosures:

Information from sources deemed reliable, but its accuracy cannot be guaranteed. Performance is historical and does not guarantee future results. Indices are not available for direct investment. Their performance does not reflect the expenses associated with the management of an actual portfolio nor do indices represent results of actual trading. Total return includes reinvestment of dividends and capital gains.

Read more articles by Larry Swedroe