The Re-Death of Value, or Déjà Vu All Over?

The underperformance of value stocks over the past 10 years has received much attention from the financial media and led at least some investors to conclude that value investing is dead. From 2009 through March 2019, while the S&P 500 Index returned 14.2 percent per annum (total cumulative return of 290 percent), the Fama-French large value and small value research indexes returned 11.3 percent per annum (total cumulative return of 200 percent) and 13.0 percent (total cumulative return of 249 percent), respectively. (Fama-French data is from Ken French’s website.) Many observers believe that the premium has disappeared because the publication of the research demonstrating the value premium led to overcrowding.

Before you jump to the conclusion that value investing is dead, it’s worth going to our trusty videotape to review the historical evidence. I’ll begin with a review of the cover story of the August 13, 1979, issue of Businessweek (BW), “The Death of Equities.” The following are some of the important, need-to-know “insights” BW provided.

• At least 7 million shareholders had defected from the stock market since 1970.
• Equities, more than ever, are the province of giant institutional investors.
• Pension money could now be invested in not only stocks and bonds but also in real estate, commodity returns, and even gold and diamonds.
• Institutions could withdraw billions from both the stock and bond markets.

These insights led Robert Salomon, general partner of the prestigious investment banking firm Salomon Brothers, to conclude that “We are running the risk of immobilizing a substantial portion of the world’s wealth in someone’s stamp collection.”

BW continued with this observation:

Before inflation took hold in the late 1960s, the total return on stocks had averaged 9 percent a year for more than forty years, while AAA bonds rarely paid more than 4 percent. Today the situation has reversed, with bonds yielding up to 11 percent and stocks averaging a return of less than 3 percent throughout the decade.

This led BW to conclude:

This death of equity cannot be seen as something a stock market rally will check. Only the elderly who have not understood the changes in the nation’s financial markets, or who are unable to adjust to them, are sticking with stocks.

Based on the above, BW issued its famous warning:

• For better or for worse, then, the U.S. economy probably has to regard the death of equities as a near-permanent condition.

We now know that the S&P went on to return about 11.5 percent per annum, defying prognostications by so-called experts.

The case of the “Death of Equities” is a perfect analog for the current skepticism around certain factor strategies, and especially value stocks. While there has been a significant value premium over the long term, the premium has been time varying, just as it has been with equities in general—and all risk assets (if that wasn’t the case, there would be no risk and no premium). As you will see, there is really nothing unusual about value’s underperformance over a decade; it’s just that investors either have poor memories and/or they don’t know their investment history. In fact, the underperformance of value (mainly driven by the outperformance of the FANG stocks—Facebook, Amazon, Netflix and Google) is very similar to its underperformance in the latter part of the 1990s. (All Fama-French table data is from Ken French’s website.)

Analyzing the 1994-1999 Period

During this period (which followed on the heels of the publication of the famous Fama-French research demonstrating the existence of the size and value premium), large value stocks underperformed the S&P 500 by 108 percent, and small value stocks underperformed by 84 percent. As mentioned in the first paragraph, the recent cumulative underperformance for large value has been 90 percent and for small value 41 percent. There’s another important similarity between the two periods (2009-April 2019 and 1994-1999).

Returns to Companies with Negative Cash Flow and Negative Earnings

In the vast majority of years, companies with negative cash flow produce negative returns for investors. For example, data from AQR Capital Management over the 24-year period, 1995-2018, show that there were just five years when they produced positive returns, and in only two of those were the returns above a few percent. Those two years were 1999 and 2018. In 1999 companies with negative cash flows returned 19 percent, and in 2018 they returned 6 percent. Similarly, companies with negative earnings produced negative returns in all but seven of those years, and in just two of them were the returns above a nominal percentage. In 1999, companies with negative earnings returned 27 percent, and in 2018 they returned 8 percent. In both cases, these abnormal/unexpected outcomes contributed to the negative value premium. However, over the long term, companies with these traits do not make for higher-returning investments. In fact, the AQR data shows that, over the 24-year period ending in 2018, they have produced negative returns—about -3 percent for companies with negative earnings and about -5 percent for companies with negative cash flow.

It’s important to note that, while we or any researcher can do these types of diagnostics ex-post, they don’t provide an ability to predict when this happens ex-ante. The same argument applies to the equity premium—when the stock market tanks, we can ex-post talk about why, but it’s not clear that there is an ability to predict it in advance. That is why it is called risk. And risk has to be unpredictable.

Been There, Done That

The evidence demonstrates that the decade-long underperformance of value stocks is not unusual. It’s worth noting that during this period of the late 1990s, when technology stocks dominated (similar to the recent experience), legendary investor Warren Buffett was thought to have “lost his touch,” as he dramatically underperformed.⁽¹⁾ However, Buffett did not lose faith in his belief in value investing during that period. And I believe it is safe to say he hasn’t lost confidence today either.   

What many investors fail to understand is that such periods of outperformance typically are caused by dramatic changes in valuations, with the underperformance of value stocks causing them to become much cheaper relative to growth stocks, which eventually has led to the restoration of the value premium. As the table below demonstrates, that’s exactly what happened over the eight years after the growth bubble burst in 2000—large value and small value outperformed by 60 percent and 218 percent, respectively.

Analyzing the 2000-2007 Period

In reviewing the above tables, some might question their shorter-term nature relative to the now decade-long underperformance of value. To address that issue, let’s again go to our trusty videotape for a history lesson. 

How Long Is Long Enough?

If the decade-long underperformance of value has convinced you that the value premium is gone, would the much longer 17-year period of a negative equity premium have convinced you (as it did many, including Salomon Brothers and Businessweek) that the equity premium was dead?

 Analyzing the 1966-1982 Period

Over this 17-year period, the S&P 500 Index not only underperformed totally riskless one-month Treasury bills but also underperformed large value stocks by 368 percent and small value stocks by 1117 percent! It’s also worth noting that the S&P 500 Index underperformed totally riskless one-month Treasury bills over the 13 years 2000-12 and the 15 years 1929-43. Note that the three periods total 45 years, or half of the 90 calendar years since 1929! And, in each of these periods, value stocks provided diversification benefits, outperforming the S&P 500.

Further Evidence

Eugene Fama and Ken French examined the volatility of the market beta, size, and value premiums in their study “Volatility Lessons,” published in the Third Quarter 2018 issue of the Financial Analysts Journal.⁽²⁾ The study covered the period from July 1963 through December 2016. Fama and French noted that, while most of the news about equity premium distributions for longer return horizons is good, there is bad news. They used the realized monthly returns from the period their study covered to construct long-horizon simulation returns and found that for the three- and five-year periods that are often the focus of professional investors, negative equity premiums occur in 29 percent of three-year periods and 23 percent of five-year periods of simulation runs. Even for 10- and 20-year periods, negative premiums occur in 16 and 8 percent of simulation runs, respectively.

Fama and French found similar results for the size and value premiums and concluded that this is simply the nature of risk—if you want to earn the expected (the mean of the distribution of potential outcomes) premiums, you must accept the fact that you will experience losses, no matter how long your horizon. Said another way, if you can’t stand the heat, get out of the kitchen.

They concluded:

The high volatility of stock returns is common knowledge, but many professional investors seem unaware of its implications. Negative equity premiums and negative premiums of value and small stock returns relative to market are commonplace for three- to five-year periods, and they are far from rare for ten-year periods. Given this uncertainty, investors who will abandon equities or tilts toward value or small stocks in the face of three, five, or even ten years of disappointing returns may be wise to avoid these strategies in the first place.

Clearly, the returns to all risk assets are not only time-varying, but all risk assets can be expected to experience long periods of underperformance. That means that discipline is the key to successful investing. If you don’t know your history and don’t have faith, long periods of underperformance will lead to the abandonment of a strategy, typically right before returns revert to their long-term mean performance. A key ingredient for having faith is understanding why a premium should exist and what might cause it to underperform. 

Source of the Value Premium

One of the great debates in finance is whether the source of the value premium is risk-based or a behavioral anomaly. In our book, “Your Complete Guide to Factor-Based Investing,” Andrew Berkin and I present the evidence showing that there are good arguments on both sides. Thus, it’s likely the answer isn’t black or white. For example, we show that the academic research demonstrates that value firms not only have poorer earnings and profitability compared to growth firms, but their greater historical leverage and more irreversible capital increases their risks in times of financial distress. Stocks that do poorly in bad times should command large risk premiums. Thus, investors demand a higher return on value stocks than on growth stocks as compensation for higher vulnerability due to financial distress. That said, the literature also provides us with several behavioral explanations for the value premium. One such explanation is that investors are systematically too optimistic in their expectations for the performance of growth companies and too pessimistic in their expectations for value companies. Ultimately, prices correct when expectations are not met. My ETF.com post of November 29, 2017, also includes a detailed discussion.

Before concluding, let’s turn to consider a possible explanation for why value has underperformed over the past decade.

Explaining Value’s Underperformance

One candidate for explaining the poor performance of value stocks is the difference in duration of value and growth stocks. Growth stocks, with more of their earnings in the distant future, are longer duration stocks than value companies (which have more of their cash flows “front-loaded”). With that in mind, consider what the economic environment has been like since the onset in 2007 of the Great Financial Crisis:

• Economic growth has been much weaker than after any other economic recovery, with not a single year so far with as much as 3 percent growth in GNP.
• In addition to weak economic growth, all the surprises on the inflation front have been below expectations.
• The combination of weak economic growth and lower than expected inflation (a global phenomenon) has led to a sharp decline in interest rates, both nominal and real.
• Falling interest rates benefit longer duration stocks more than shorter duration stocks. 

As supporting evidence, consider the findings from the study “Does the Stock Market React to Unexpected Inflation Differently Across the Business Cycle?” Chao Wei, the author, attempted to answer the following questions:

• Does the stock market react to unexpected inflation differently across the business cycle?
• Returns of what types of stocks respond more negatively to unexpected inflation?
• Are the responses asymmetric across the business cycle?

Following is a summary of his findings:

• There is strong evidence that nominal equity returns of firms with lower book-to-market (BtM) ratios (growth stocks) are more negatively correlated with unexpected inflation. Growth stocks are longer duration stocks, and thus an increase in the stock risk premium impacts them more than it does value stocks. In addition, value companies typically are characterized by having more leverage, and inflation reduces the real cost of fixed rate debt.
• The lower the BtM ratio (the more “growthy” the stock), the stronger the negative correlation between the nominal equity returns and unexpected inflation at the time of recession.
• There are significant cyclical patterns in the responses of the excess market return to unexpected inflation. The stock risk premium responds much more negatively to unexpected inflation during contractions than expansions. The results are statistically significant.

The 1970s provide a good example of the relationship between unexpected inflation and the value premium. Inflation in 1970 rose 5.5 percent. By 1979 the inflation rate had increased to 13.3 percent. While the value premium (the annual average return of value stocks minus the annual average return of growth stocks) has been 4.7 percent since 1927 and the inflation rate averaged 3.1 percent, in the 1970s when inflation averaged 7.4 percent, the value premium averaged 8.4 percent.

For those interested in learning more about the correlations of inflation and growth and value stocks, I suggest you read Bill Bernstein’s piece on the subject.

We now turn to addressing the question of whether the value trade has become overcrowded, beginning with a review of the findings from two studies that sought the answer to this question.

Has the Value Trade Become Overcrowded?

The December 2018 study “Characteristics of Mutual Fund Portfolios: Where Are the Value Funds?” by Martin Lettau, Sydney Ludvigson and Paulo Manoel, provides a comprehensive analysis of portfolios of active mutual funds, exchange-traded funds (ETFs) and hedge funds through the lens of risk (anomaly) factors such as size, value and momentum. Among the questions they try to answer are: To what extent do active fund managers exploit these factor premia? If there are limits to arbitrage, do active funds contribute to the existence of these anomalies, or do they overweight underpriced stocks?

Among their important findings was that neither mutual funds nor ETFs systematically tilt their portfolios toward profitable factors, such as high BtM ratios, high momentum, small size, high profitability and low investment growth. In fact, for some factors, mutual funds target the low-return leg of long/short factor portfolios rather than the high-return leg. This bias is especially strong for BtM ratios. In fact, they found that there are virtually no high-BtM funds in the sample, while there are many low-BtM “growth” funds. For example, only seven out of 2,657 funds in their sample have a BtM score in the fourth quintile or above.

Supporting evidence comes from David Blitz, who demonstrated in his February 2017 paper, “Are Exchange-Traded Funds Harvesting Factor Premiums?” that, while some ETFs are specifically designed for harvesting factor premiums, other ETFs implicitly go against these factors.⁽³⁾

Specifically, Blitz found:

From a factor-investing perspective, smart-beta ETFs tend to provide the right factor exposures, while conventional ETFs tend to be on the other side of the trade with the wrong factor exposures. In other words, these two groups of investors are essentially betting against each other.

We can also address the issue of overcrowding by examining the spreads in valuations of growth and value stocks. If overcrowding has occurred, we should see a dramatic narrowing in valuations. 

Have Spreads Narrowed?

In 1994, right after the publication of the famous Fama-French research on the size and value premiums, the price-to-book (P/B) ratio of U.S. large growth stocks was 2.1 times that of large value stocks, and their price-to-earnings (P/E) spread was 1.5. (Data is from Dimensional.) If there was overcrowding, we should see narrowing of the spreads.

Using the Vanguard Russell 1000 Growth ETF (VONG) and the Vanguard Russell 1000 Value ETF (VONV), we find that as of April 30, 2019, the P/B of VONG was 6.3 and the P/B of VONV was 1.9. The spread had widened from 2.1 to 3.3. The P/E of VONG was 22.2 and the P/E of VONV was 14.8. Here we find the spread was unchanged at 1.5. At least here, we see no evidence that cash flows have caused the value premium to narrow.

AQR provides us with further evidence demonstrating that the value trade has not become overcrowded.

Valuation Spreads 

AQR has data going back to 1984. Using a composite value measure that includes P/B, P/E, price-to-forecasted earnings, and sales-to-enterprise value, they found that as of March 31, 2019, the valuation spread between U.S. value stocks and growth stocks was trading at the 85th percentile (only 15 percent of the time was value trading cheaper relative to growth), developed international value stocks were trading at the 91st percentile, and emerging market value stocks were trading at their cheapest relative level (100th percentile).  To achieve industry neutrality, value spreads were constructed by comparing the value metrics within each industry and then aggregated to construct a portfolio. Note that industry neutrality presents a more “conservative” picture, as an unconstrained measure would put the U.S. valuation spread at the 97th percentile.

The bottom line is, it appears that despite what many investors believe, a massive net inflow into value stocks relative to growth stocks has not occurred. Importantly, the research also shows that the valuation spread does contain information on the future value premium.

Valuations Matter

Adam Zaremba and Mehmet Umutlu, authors of the March 2019 study “Strategies Can Be Expensive Too! The Value Spread and Asset Allocation in Global Equity Markets” examined whether the value spread (the difference in valuation ratios between the long and the short sides of the trade) was useful for forecasting returns on quantitative equity strategies for country selection. To test this, they examined a sample of 120 country-level equity strategies replicated within 72 stock markets for the years 1996 through 2017. They found: “The breadth of the value spread can predict the future returns in the cross-section. We show that equity strategies with a wide value spread markedly outperform strategies with a narrow value spread. In other words, if you wonder which strategy might produce decent payoffs in the future, pay attention to the value spread.”

Their findings are consistent with prior research. For example, the June 2018 study “Value Timing: Risk and Return Across Asset Classes,” by Fahiz Baba Yara, Martijn Boons and Andrea Tamoni, also found that valuation spreads provide information. The authors demonstrated that “Returns to value strategies in individual equities, commodities, currencies, global government bonds and stock indexes are predictable by the value spread. … In all asset classes, a standard deviation increase in the value spread predicts an increase in expected value return in the same order of magnitude (or more) as the unconditional value premium.”

Jim Davis’ 2007 study “Does Predicting the Value Premium Earn Abnormal Returns?” also found that BtM ratio spreads contain information regarding future returns. However, he also found that style-timing rules did not generate high average returns because the signals are “too noisy”—they don’t provide enough information to offer a profitable timing signal. Confirming Davis’ findings, AQR’s research into style-timing rules led them to conclude that they increase turnover and trading costs, making them even less effective for implementation.

For value investors, the above findings are good news, as the relatively poor performance of value stocks in the U.S. over the past decade has led to a widening of the P/B and P/E spreads between value and growth stocks, restoring them to about the same level (or wider) than they were when Fama and French published their famous study, “The Cross-Section of Expected Stock Returns.”

Summary

Investors face a choice. They can either own a traditional 60 percent market-like equity/40 percent bond portfolio, which has most of its risk (as much as 90 percent or more, depending on the maturity and quality of the bond holdings) concentrated in the single factor of market beta, or they can choose to diversify across as many unique sources of risk and return that they can identify that meet all their established criteria. The first path is the comfortable one in the sense that your portfolio will not cause any tracking error regret—you won’t be underperforming popular benchmarks that are reported on a daily basis by the financial media. On the other hand, that strategy will likely be highly uncomfortable during periods like 1973-74, 2000-02 and 2008, when the single factor (market beta) that dominates their portfolio’s risk suffers from severe bear markets (and also over long periods, such as the three 13-year or longer periods discussed above).

Failing conventionally is always easier than failing unconventionally (misery loves company). And, while based on the historical mean and historical volatility that the market beta premium should be expected to be negative about 9 percent of the time over 10-year periods, we should expect more frequent failures in the future because current valuations are much higher than historical valuations. Thus, we should expect a smaller premium. According to data from Ken French’s website, from 1927 through 2018 the market beta premium was 8.5 percent. Most financial economists expect it to be much smaller going forward, perhaps half as much. A smaller premium with the same volatility means greater odds of negative returns for U.S. stocks.

The second path—diversification—is more likely to lead to successfully achieving goals. However, it does mean having to live with the fact that your portfolio will perform very differently than traditional portfolios, creating the risk of tracking error regret. In that sense, diversification is not a free lunch. It means living through uncomfortable periods, even long ones. And during periods of failure, it means failing unconventionally, which is much harder to deal with.

Given that you must accept that, whichever path you choose, you will have to live through uncomfortable times, it seems logical that you should choose the path that gives you the highest odds of achieving your goals. And that is choosing the more efficient portfolio—the more diversified one—and saying, “I don’t give a damn about tracking error regret because relativism [how you performed relative to some popular index] has no place in investing.”  

And finally, investors can use the knowledge about the difference in exposure to inflation risk of value and growth stocks when making asset allocation decisions. Investors who are more prone to the risks of unexpected inflation (such as retirees) may wish to have lower exposure to more growth-oriented stocks (with their longer duration). Similarly, investors who hold portfolios that are more growth-oriented may wish to reduce their exposure to inflation risk either by reducing the maturity of their nominal bond holdings, increasing their allocation to Treasury inflation-protected securities (TIPS), or adding a value tilt.