Swedroe: Finding The Source Of Value

There has been long and great debate as to the source of the value premium. There is a belief among many academics that the value premium is actually an anomaly (in contradiction to the efficient markets hypothesis) and the result of persistent pricing errors made by investors.

For example, those from the “behavioral school” of finance believe investors naively extrapolate past growth when evaluating a company. Thus, they overreact to that information, resulting in a situation where growth companies are persistently overpriced and value companies are persistently underpriced.

Behavioralists also find that investors often conflate familiarity with safety. Because they tend to be more familiar with popular growth stocks, those stocks tend to be overvalued.

In any case, the argument rages on among financial economists over whether the value premium is risk-based or behavioral-based, with supporting evidence on both sides. And it is certainly possible the issue isn’t either black or white, with both the mispricing and risk explanations playing a role in the premium.

Recent Supporting Research

Hui Guo, Chaojiang Wu and Yan Yu contribute to the literature on the source of the value premium with their study “Time-Varying Beta and the Value Premium,” which appears in the August 2017 issue of the Journal of Financial and Quantitative Analysis. Their study covers the period July 1963 through December 2012. Following is a summary of their findings:

• There’s a strong countercyclical variation in the value premium’s market beta. Macroeconomic variables are informative in estimating the conditional (that is, conditioned upon the state of the economy) capital asset pricing model.
• Consistent with risk-based explanations for the value premium, its market beta increases sharply during business recessions and decreases during business expansions—value stocks become riskier (safer) in bad (good) times, helping explain the value premium. Specifically, there was a sharp spike in the value premium’s conditional beta in each of seven business recessions over the period July 1963 to December 2012. In addition, the conditional beta tends to decrease during business expansions.
• Value’s conditional market beta correlates positively with unemployment. As unemployment rises (falls), the beta of value stocks increases (decreases), increasing (decreasing) their risk.
• Value’s conditional market beta correlates negatively with industrial production. As industrial production falls (rises), the beta of value stocks rises (falls), increasing (decreasing) their risk.
• Disaster risk accounts for a substantial portion of the value premium due to higher investment adjustment costs (value companies tend to have more irreversible capital).
• The conditional beta correlates negatively with the procyclical inflation rate and price-to-earnings ratio—value stocks become more (less) risky in bad (good) times.
• The results were robust to various periods, including from 1949 through 2012 (the longest period for which unemployment data was available) and also from 1927 through 2012.

Guo, Wu and Yu concluded: “Our novel evidence of a strong correlation of the conditional beta with key aggregate economic activity gauges suggests that investment-based models remain a viable explanation of the value premium.”

Their findings are consistent with other research that has found a risk-based explanation for the value premium. I’ll review some of that literature, beginning with the 1998 paper “Risk and Return of Value Stocks.”

Risk-Based Explanations

The authors, Nai-fu Chen and Feng Zhang, make the case that value stocks contain a distress (risk) factor. They examined three intuitive measures of distress present in value companies: cutting dividends by at least 25%, a high ratio of debt to equity, and a high standard deviation of earnings.

Chen and Zhang found that the three measures all captured the returns information (produced high correlations) contained in portfolios as ranked by book-to-market (BtM) ratio. When these metrics were present, returns were greater. Because all three measures have simple, intuitive risk interpretations by virtue of their association with firms in distress, the authors state that it is not surprising the risk factors they studied were both highly correlated to each other and with BtM rankings.

Their conclusion was that value stocks are cheap because they tend to be firms in distress, with high leverage, and that face substantial earnings risk. They therefore offer higher returns due to the greater risks facing value investors.

Next, we’ll look at the 2005 study by Lu Zhang, “The Value Premium.” He concluded that the value premium could be explained by the asymmetric risk of value stocks. Value stocks are much riskier than growth stocks in bad economic times and only moderately less risky than growth stocks in good times. Zhang explains that the asymmetric risk of value companies exists because value stocks are typically companies with unproductive capital. Asymmetric risk is important for the following reasons:

• Investment is irreversible. Once production capacity is put in place, it is very hard to reduce. Value companies carry more nonproductive capacity than growth companies.
• In periods of low economic activity, companies with nonproductive capacity (value firms) suffer a greater negative impact in earnings because the burden of nonproductive capacity increases, and they find it more difficult than growth companies to adjust capacity.
• In periods of high economic activity, the previously nonproductive assets of value companies become productive while growth companies find it harder to increase capacity.
• In good times, capital stock is easily expanded. In bad times, adjusting capital levels is an extremely difficult task—especially so for value companies.

When these facts are combined with a high aversion to risk (especially when that risk can be expected to appear in difficult economic times and investors’ employment prospects are more likely to be in jeopardy), the result is a large and persistent value premium. This is consistent with the results of Motohiro Yogo’s 2006 study, “A Consumption-Based Explanation of Expected Stock Returns.”

Yogo found that value stocks deliver low returns during recessions, when the marginal utility of consumption is highest. In other words, the returns of value stocks are more procyclical than growth stocks. Thus, investors must be rewarded with high expected returns to hold these risky stocks.

Value & Distress

We’ll next examine a 2005 study, “Is the Book-to-Market Ratio a Measure of Risk?” The authors, Robert Peterkort and James Nielsen, developed a leverage-based approach to investigate the BtM effect. Because leverage is risky, it isn’t surprising that they found a positive relationship between higher stock returns and market leverage.

Their regression results showed that, when compared with market leverage, the BtM ratio added a small amount of explanatory power for stock returns. They wrote that they believe the incremental explanatory power of the BtM ratio is due to additional information about the riskiness of the firm’s assets. Thus, they concluded the BtM effect is mostly a leverage (risk) effect.

Additionally, when the authors considered only companies that they called “all-equity” firms (those with minimal amounts of mostly current debt, as opposed to long-term liabilities), there was no BtM effect at all.

If the value premium is an anomaly, the mispricing should show up in all high-BtM firms, not just those with high leverage. These findings are consistent with those of Ralitsa Petkova’s 2006 study, “Do the Fama-French Factors Proxy for Innovations in Predictive Variables?”

Petkova found that value companies tend to be firms under distress, with high leverage and high uncertainty of cash flow. Therefore, shocks to the default spread (the spread between bonds of higher-rated credit and lower-rated credit) explain the cross section of returns, and is consistent with value being a measure of distress risk.

What’s more, growth stocks are high-duration assets (much of their value comes from expected future growth), making them similar to long-term bonds. Value stocks, on the other hand, are low-duration assets, making them more similar to short-term bonds.

Term-Spread Shocks

Thus, shocks to the term spread (the difference between short-term bonds and long-term bonds) also explain the cross section of returns, and is further consistent with value being a measure of distress risk.

Next, we’ll examine the 2014 study “Value Premium and Default Risk,” which covered the period 1927 through 2011. The authors, Mohammed Elgammal and David McMillan, found there was a “positive relationship between default risk and the value premium for both large and small firms together with a leverage effect.”

They concluded: “The results show a positive association between the default premium and the value premium accompanied with evidence for a leverage effect on the value premium. This lends support to the risk-based explanation for the source of value premium. That is, where the default premium captures systematic risk in the macroeconomy and that the value premium is associated with rational decision making on the part of investors. Value stocks characterized by poor performance, earnings and profitability compared with growth stocks are more vulnerable to the risk of default and lead the investors to require a higher return on value stocks as leverage increases.”

These papers demonstrate the link between value stocks and financial distress at the asset level. Value stocks are not simply great bargains waiting to be scooped up as free money. Rather, they are cheap for a reason—a reason related to their riskiness.

Influence Of Economic Activity

Finally, we have the 2009 study “The Value Premium and Economic Activity: Long-Run Evidence from the United States.” To test the validity of the risk explanation, the authors, Angela Black, Bin Mao and David McMillan, examined the relationship between the value premium and macroeconomic variables, such as industrial production, inflation, money supply and interest rates. Their study covered the period 1959 through 2005. Following is a summary of their findings.

First, in times of economic expansion, when industrial production rises, value stocks become less risky relative to growth stocks. Thus, the prices of value stocks increase more than the prices of growth stocks. The result is that the spread between the high-BtM and low-BtM stocks narrows, and the value premium declines. In bad times, value stocks become riskier relative to growth stocks. The result is that the prices of value stocks decrease faster than growth stocks, and the value premium increases (a sign of increased risk).

Therefore, the relationship between the value premium and industrial production is negative. This certainly was the case during the most recent recession, which lasted from December 2007 through June 2009, when the value premium was -0.44% per month.

Second, a similar, negative relationship exists between the value premium and the money supply. Following an increase in money supply, stock prices increase. The prices of value stocks tend to increase more than growth stocks, and the value premium shrinks. When money supply decreases, stock prices decrease, with the prices of value stocks decreasing more than growth stocks, and the value premium rises.

Third, there is a positive relationship between the value premium and interest rates. As long-run interest rates rise, stocks become less attractive than bonds and stock prices decrease, with the prices of value stocks decreasing faster than the prices of growth stocks. That leads to an increasing value premium. When interest rates fall, the prices of value stocks increase faster than the prices of growth stocks. This leads to a decreasing value premium.

Overall, Black, Mao and McMillan found that value stocks are more sensitive to bad economic news, whereas growth stocks are more sensitive to good economic conditions. They concluded that the value premium is largely based on fundamental risk factors within the economy and arises through macroeconomic risk.

Summary

The bottom line is that there are simple and logical risk-based explanations for the existence of a value premium. And I’ll add one more intuitive explanation: Value stocks have been more volatile than the market. From July 1926 through June 2017, the annual standard deviation of the Fama-French Large Cap Index was 24.8%. The annual standard deviation of the Fama-French Large Cap Value Research Index was much higher, at 41.4%.

We see the same pattern in small stocks. The annual standard deviation of the Fama-French Small Cap Index was 43.3%. It was 50.1% for the Fama-French Small Cap Value Research Index.

In summary, the academic research demonstrates that value firms have poorer earnings and profitability compared to growth firms. And value firms’ greater leverage increases their risks in times of financial distress. Stocks that do poorly in bad times should command large premiums. Thus, investors demand a higher return on value stocks than on growth stocks as compensation for their higher vulnerability due to financial distress.

That said, there are good behavioral explanations for the premium as well. That is why the debate rages on, and why I believe that, while the value premium isn’t a free lunch (there are good risk-based explanations), it just might be a free stop at the dessert tray.

Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.