Glamour Can Distract Investors

There’s very strong historical evidence to support the existence of a value premium in equity markets. While there’s no dispute over the existence of the value premium (value stocks have provided an annual average return 5% higher than growth stocks over the long term), there is much debate over the cause of the difference in returns.

In one camp are financial economists who argue that the value premium is a risk story—value stocks are riskier than growth stocks. As a result, investors must be compensated in the form of higher expected returns. Others argue for an alternative explanation—there’s a persistent overvaluation of the earnings prospects for growth stocks. And what’s more, there’s a wealth of evidence on both sides.

The Allure Of Growth

Keith Anderson and Tomasz Zastawniak—authors of the October 2015 study “Glamour, Value and Anchoring on the Changing P/E”—contribute to the literature supporting the persistent overvaluation of the earnings prospects for growth stocks.

Their working hypothesis was that the differing experiences of glamour and value investors can be explained by the well-documented behavior of anchoring.

Anchoring is a form of cognitive bias in which people place an inordinate amount of importance on certain values or attributes, which then act as a reference point, and the influence of subsequent data is weighted to support their initial assessment. For example, some investors will tend to hang on to a losing investment because they’re waiting for it to break even, anchoring their investment’s present value to the value it once had.

Anchoring is such a powerful force that, even in experiments when subjects could plainly see the anchor and that it could not possibly be any sort of guide to an answer, the bias continued to play a role.

For example, in a famous experiment, Daniel Kahneman and Amos Tversky asked subjects to spin a roulette wheel rigged to stop at 10 or 65, and then asked them to estimate the percentage of African nations in the United Nations. Subjects who saw a result of 10 guessed 25% while those shown 65 guessed 45%.

Anchoring On The P/E Ratio

Anderson and Zastawniak hypothesized that investors may anchor on the price-to-earnings (P/E) ratio of a stock when they initially invest in it.

They write: “Given an observed high P/E of 25 [investors] may think (consciously or not), ‘Thousands of investors, some of whom are better informed than I am, already are paying $25 for each $1 of current earnings. This must be a valuable, high growth company to justify that.’”

The authors then posit that such investors “fail to adjust their future expectations sufficiently according to mean reversion.” Thus, having now bought the stock, investors “expect the P/E to change slowly, if at all. As time goes on, the P/E decile changes, and different prospects for returns attach to each decile. If there is a differential drift in the P/E and hence returns between value and glamour stocks that are not expected by investors, this could account for why glamour investors end up disappointed.”

Anderson and Zastawniak’s study covered the period 1983 through 2010. When ranking stocks and forming P/E deciles, they divided positive P/E companies into deciles and then added five more deciles to include the firms with losses (34.1% of the company/year data points represent losses), producing a total of 15 “bins.” They then used these 15 bins, as the loss-producing companies represented about one-third of the total.

They paired each year’s decile number with the decile to which the company moved in the following year. If no earnings were recorded the following year, decile “00” was coded, as the company went into administration, was taken over or otherwise ceased to be quoted. There are thus 15 bins in year 1 and 16 bins in year two, resulting in 240 (or 15×16) possible transitions from one year to the next. Finally, they calculated the equally weighted return from each of these 240 possible transitions.

Study Results

Following is a summary of Anderson and Zastawniak’s findings:

• For all deciles, the most likely decile for a company to appear in the following year is the same decile as the current year.
• A glamour company has a 34% chance of becoming a loss-maker next year, compared with just 25% for a value company.
• Extreme loss-makers (decile 1) and value companies are most likely to remain in the same decile next year, at 32% and 34%, respectively. The deciles between them are much more likely to move, with the probability of remaining in the same decile only 15-20%.
• Extreme loss-makers find it particularly difficult to break out of the spiral. They have only a one-in-six chance of turning a profit in the next year, compared with a 27% chance of being delisted.
• Unsurprisingly, the worst loss-makers (decile 1) that either delist or stay in the first decile lose money for their investors. However, surprisingly, companies only have to make their losses less severe and excellent returns can be expected—an average gain of 131% for loss-makers that move from decile 1 to decile 5 (the least severe loss-makers).
• Glamour stocks that remain in the same decile provide three times the rate of return of value stocks that remain in the same decile (36% versus 12%). This could help explain the preference for glamour stocks. However, few companies stay in the same decile in the following year, and glamour stocks have a greater tendency to move deciles—roughly five-sixths of the time—and then give very poor returns. In addition, if glamour stocks start making losses, they suffer greatly. Glamour stocks that move from decile 6 to decile 1 lose on average 41%. On the other hand, value shares are much more likely to remain in or near the value P/E decile.
• Glamour investors appear to be underestimating the tendency of their preferred stocks to change decile, by at least 18%, and possibly much more. Additionally, the average change in P/E is large.
• Smaller companies, particularly small high-growth companies, are much more likely to move from the glamour deciles to the value deciles than large companies. These are exactly the sort of stocks that are most likely to be dominated by retail investors who are more prone to behavioral biases. I would add here as well that they are also exactly the type of stocks most difficult for arbitrageurs to short, thereby correcting mispricings.
• NYSE stocks are 28% more likely than Nasdaq stocks to move into the higher deciles next year, while Nasdaq stocks are 42% more likely than NYSE stocks to become loss-makers or delist.
• Shares in the extreme-loser decile have respectable mean returns, but the median returns are very poor and the standard deviation of returns is double what it is for the glamour and value deciles. Any good returns from an extreme-loser stock depend on the firm becoming profitable, or at least stopping such heavy losses, but each year there is an almost 60% chance of the firm either ceasing to be quoted or remaining a decile 1 stock.
• There was a 7.5% average annual difference in returns between decile 6 (glamour) and decile 15 (value).
• The returns for glamour investors are relatively poor whatever their time horizon. However, value investors can expect superior returns if they hold for two to three years (not just for one year) as recommended by Benjamin Graham. The returns to value stocks, while just 5% for the first year, were 21% for the second year and 15% for year three. After that, returns decline to little more than the returns on glamour shares.
• The value decile’s standard deviation of returns is only slightly greater than that of the glamour decile, and not enough to account for the higher returns.

Conclusions

When considered together, the authors write, the findings support their thesis that glamour investors anchor on the high P/E value for growth shares while ignoring the high likelihood that the P/E ratio will change in the future.
Later this week, we’ll take a closer look at a possible explanation Anderson and Zastawniak offer to account for investors’ preference for growth stocks, as well as some other evidence in the academic literature that supports their findings.

This commentary originally appeared March 28 on ETF.com

By clicking on any of the links above, you acknowledge that they are solely for your convenience, and do not necessarily imply any affiliations, sponsorships, endorsements or representations whatsoever by us regarding third-party Web sites. We are not responsible for the content, availability or privacy policies of these sites, and shall not be responsible or liable for any information, opinions, advice, products or services available on or through them.

The opinions expressed by featured authors are their own and may not accurately reflect those of the BAM ALLIANCE. This article is for general information only and is not intended to serve as specific financial, accounting or tax advice.

© 2016, The BAM ALLIANCE