…once the investor pays a substantial amount for the growth factor, he is inevitably assuming certain kinds of risk; viz., that the growth will be less than he anticipates, that over the long pull he will have paid too much for what he gets, that for a considerable period the market will value the stock less optimistically than he does.
– Benjamin Graham
Given the dismal economic outlook and resurgence of the virus, we remain somewhat skeptical at the resiliency of the equity markets, which recouped most of the first quarter declines. We reviewed client portfolios last year, raising liquidity where needed, and will do so again this quarter. The S&P 500, buoyed by strong performance from the large tech stocks that dominate the index, ended the quarter down a little more than 3% while the broader MSCI All Country World Index returned -6.3% year-to-date. The dominance of the ‘FAANGM’ stocks (Facebook, Amazon.com, Apple, Netflix, Google & Microsoft) becomes apparent when one considers that the average stock in the S&P 500 returned -10.8%. Despite the second quarter rebound, the ‘stay at home’ divide still separates this year’s winning and losing stocks. The over $5 trillion of monetary and fiscal stimulus that flooded the market since March succeeded in its intended effect of lowering discount rates and inflating asset prices. Questions remain about the longer term impact of the pandemic on corporate profitability. The market appears to be discounting a fast recovery, which creates a potential for disappointment and further volatility. For a historical example, the market suffered a short downturn in response to 9/11 and the recession that occurred in late 2001, but suffered a much larger decline in late 2002 as corporate earnings continued to languish.
Is value investing dead?
The long-term outperformance of value investing remains impressive. For US stocks, owning the cheapest 30% of the market and rebalancing annually resulted in an annualized return of 12.1% from 1926 through April 20201. Owning the most expensive 30% returned 9.2%. This nearly three percentage point return advantage translated to an eleven-fold difference in the ending wealth of a value portfolio compared to a growth portfolio. While the value portfolio in this example exhibited volatility approximately 30% greater than the broad market, the outperformance was sufficient to swing all the various risk-adjusted return metrics in value’s favor. The advantage for value stocks remains unchanged after excluding the 1929 Crash and Great Depression. From January 1946 through April of this year, value outperformed growth by about the same amount. Value appears to be a universal phenomenon in markets – using the same metrics, value outperforms growth by similar or greater margins for non-US stocks. This commentary will focus on the US, but the same observations apply to
international stocks as well.
Over the past few decades, a considerable amount of academic research has explored why value investing worked so well. The typical rationale given by value managers is that investors overpay for widely loved stocks that have demonstrated strong past performance and undervalue companies that are out of favor. Substantive academic research supports this simplistic explanation. Nobel Prize winner Eugene Fama, the author of the Efficient Market Hypothesis, authored a paper in 20072 that quantified the gains to value investing from value stocks getting re-rated to core or growth stocks and the losses suffered by growth investors from growth stocks falling out of favor and becoming core or value stocks. Fama termed this phenomenon migration, and found it accounted for almost all of the outperformance of value stocks relative to growth stocks.
While value’s long-term outperformance remains, we are currently in the midst of the longest documented period of underperformance for value investing. Since the end of the last period of value outperformance in 2007, the 10.5% return of the Russell 1000 Growth Index was over twice the 5.2% return of the Russell 1000 Value. Value had a few moments of outperformance, particularly the recovery in 2009 from the post-Financial Crisis lows, along with a short burst in 2016, but overall it was a miserable decade for the strategy.
This underperformance leads many observers to speculate that value investing is dead. The narrative of technology disrupting old economy businesses is most often cited and makes a compelling story, as it relates to most people’s personal experience. However, technological disruption is a constant of American capitalism and growth has outperformed in markets, such as Europe, that lack big tech companies like Microsoft or Amazon.com. Furthermore, disruptors can themselves get disrupted, such as the collapse of Blackberry and Palm in the 2000s.
As value investing is widely known, another possible explanation is that sophisticated investors arbitraged away the return premium, creating a more efficient market without this seemingly easy way to obtain excess returns. This explanation fails to account for the fact that record amounts of capital flowed out of active value managers into passively managed index funds and ETFs. The discount of value stocks relative to growth stocks also widened during this period, a phenomenon one would not expect if too much money was chasing the strategy.
So why has value performed so miserably? The answer is likely a combination of factors. One issue is that migration has declined. A study by Baruch Lev and Anup Srivastava3 at NYU and the University of Calgary, respectively, found that during the 2007-2018 period, value stocks remained value stocks for 3.3 years on average before revaluing upward, 32% longer than the 1989-2006 period. The longevity of growth stocks also increased by a similar amount. They also found that the frequency of increases of 10% or more for value stocks and declines of 10% or more for growth stocks also decreased during this period. Why did migration decline? Certainly the continued disruption of retail, advertising and IT services by companies such as Amazon, Alphabet (Google) and Microsoft played a role. The continued underperformance of banks, a significant component of value portfolios both in and outside the US also contributed. Lev and Srivastava note that most value stocks fell either in the highly regulated banking, insurance and utilities sectors or in low-return, asset intensive businesses, such as retail and transportation.
Another factor is changes in the relative valuation between growth and value stocks. All else equal, value will out (under) perform when the valuation discount to growth stocks narrows (widens). Research Affiliates noted in a paper published last quarter that the entire underperformance of value over the 2007-2020 period can be explained by the relative change in the average price / book ratio4 between growth and value stocks – i.e., value stocks got cheaper relative to growth stocks over the past twelve years. However, as the next paragraph details, price / book is a flawed measure. Investments in intangibles and share buybacks account for part of the increase in the difference in price / book ratios. Using other valuation metrics, the change in average valuation contributes to the underperformance, but cannot explain all of it.
The third factor relates to flawed measurements of value. Financial accounting standards were developed during an age when hard assets – factories, railroads, real estate, etc. – were the sources of corporate wealth. Accounting standards do not consider investments into these hard assets as expenses deducted from income. If Ford Motor Company spends 10 cents of every dollar reinvesting in upgrading its factories, this expenditure does not detract from its earnings and the investments become an asset on the company balance sheet with a small fraction of that expenditure deducted against future income as depreciation. However, leading modern companies invest primarily in intangible assets – research & development, brands and business processes. These expenditures are treated as business expenses – fully deducted from earnings and they do not create an asset on a company’s balance sheet. Additionally, the share buybacks favored by today’s leading companies directly reduce accounting book value. Compare Amazon.com to Ford. Amazon.com has shown barely any profits as it has reinvested nearly all its earnings into research & development. Prior to COVID-19, Ford Motor Company’s stock traded at about its accounting book value with a price / earnings multiple of around 8x. Amazon.com, on the other hand, currently trades at 21 times book value and 130 times earnings. Amazon.com certainly commands a premium valuation and may be overvalued, but it is ridiculous to suppose that the fair value of the company is more than 95% below its current value, which one would believe if book value was a reliable metric. As for Amazon’s astronomical P/E ratio, adding back research & development expenses to earnings – treating it as an investment analogous to Ford investing in factories – reduces this metric to a more reasonable 29 times earnings. Lev and Srivastava attempt to quantify this by recalculating book value by recasting accounting to treat R&D the same as investments in hard assets. They found that 40-60% of value and growth stocks changed classification during the 2008-2018 period after applying this adjustment. While this did not erase the underperformance of value, it narrowed the disparity with growth stocks considerably.
We have been aware of the issue with conventional valuation metrics and avoided traditional value strategies. The Research Affiliates Fundamental Index (RAFI), which we have used since 2006, offers a good solution to the problems with conventional value investing. This strategy avoids the simplistic reliance on a single valuation metric, rather it invests in every stock within its capitalization range but with weights based on robust metrics of company size that, unlike the market-cap weighting of major indexes, are not dependent upon the price of a security. The strategy uses sales, cash flow and dividends to weight stocks within its universe, periodically rebalancing to take gains from stocks that have appreciated and rebalance into stocks that have declined relative to their fundamentals. Not only have the RAFI strategies outperformed comparable value indexes during the 2008-2020 period, the US version outperformed the S&P 500 from 2008 through mid-2019, after losing ground over the past three years as tech stocks dominated the market.
Growth stock valuations are beginning to create risks for investors. A paper by Research Affiliates5, the creators of the Fundamental Index strategy, compared the valuations of growth stocks with their subsequent earnings, which they termed clairvoyant value. They found that while the earnings of growth stocks did indeed grow faster than the market, they did not grow as fast as their valuations implied, resulting in an eventual ‘crash’ when growth expectations returned to more realistic levels. Coca-Cola’s run as a growth stock in the 1990s illustrates this. The stock appreciated nearly 10x between 1990 and 1998, as the company streamlined operations and penetrated new markets in the developing world. Coke’s earnings grew by around 20% annually from 1990-1998. The company began the decade trading at around 20 times expected earning and ended the decade trading around 50 times. However, growth began to slow in the late 90s and the stock lost half its value between 1998 and 2003, only reaching a new high in 2014. Coke’s poor performance over the past 20 years offset its great run in the 1990s, leaving the stock’s performance behind the S&P 500 since 1991. While Coke remained a profitable, quality company throughout this whole period, its aggressive valuation in the late 1990s
set investors up to fail.
While the valuations of the FAANGM stocks remain comfortably below the extremes of 1999-2000, another cohort of stocks now exceeds even those lofty valuations. Based on Price / Sales, a valuation metric that can encompass companies that have not (yet) turned a profit, the top 10% of US stocks is now more expensive than it was back in March 2000, the peak of the Tech Bubble. Ranking US stocks on this metric, the top 10% at June 30 sported Price / Sales ratio above 12.4, and the top 5% trade at more than 28 times sales. Just over three years ago, at the end of 2016, the comparable numbers were 7.7 and 11.0. For a company that has not yet turned a profit, trading at 20 times sales requires both continued extraordinary revenue growth and future high profit margins. Microsoft, perhaps the most consistently profitable company in the S&P 500, has a net profit margin of 33% and trades around 10 time sales. A company trading at 20 times sales would have to double its revenue and obtain Microsoft-level profit margins before it could begin to justify this valuation. Of the FAANGM stocks, Facebook sported the highest historical Price/Sales ratio, reaching a peak of 18.4 times in 2014 before the company’s ad revenue really took off. Now, 5% of the market is priced over 50% above Facebook’s highest valuation. A notable stock in this cohort is Zoom, a star of pandemic work and social life, which trades at an astronomical 89 times sales. As investors learned during the 2000-2002 period, buying good companies at astronomical valuations equates to a bad investment, and buying mediocre ones at these valuations becomes a disaster.
While the US technology giants are great companies and have dominated markets for the past several years, it makes sense to remain diversified. The valuations of these stocks require aggressive growth in order to continue to outperform the broad market. As advertising companies, Facebook and Google depend on the strength of the overall advertising market. These companies also face significant regulatory pressure, particularly from the European Union. Microsoft and Amazon.com compete for cloud computing (Amazon Web Services, not online retail, is the primary driver of the company’s earnings). While client portfolios benefited tremendously from the performance of these companies, current valuations will lead to sharp declines if and when the market decides current levels of growth become unsustainable. With some basic assumptions, a company priced to deliver 10% earnings growth could see its stock price fall by nearly 25% if the market revises that expectation to 7%. Value stocks hold the advantage of low expectations, with a modest improvement in outlook able to deliver good upside performance. With growth stocks historically expensive relative to value stocks, it would take just a small combination of growth stock disappointment and /or positive value stock surprise for value to stage a comeback. However, with the virus driving the market, the current outperformance of growth could easily continue for several years, so we would not recommend an overweight to value. With the US lagging far behind other developed countries in controlling the coronavirus, this also is not a time to abandon international diversification. Finally, while Fed action reduced the income available from safe liquidity to record lows, investors should still maintain an adequate allocation to support several years of liquidity needs.
1 Ken French: Portfolios formed on Book-to-Market
2 Eugene Fama and Ken French: Migration https://papers.ssrn.com/sol3/papers.cfm?abstract_id=926556
3 Lev and Srivastava, Explaining the Recent Failure of Value Investing
4 Research Affiliates: https://www.researchaffiliates.com/en_us/publications/articles/reports-of-values-death-may-be-
5 Research Affiliates: Clairvoyant Value https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1263127##