by Charles Roth, Thornburg Investment Management
If the efficient markets hypothesis is questionable at the stock level, it’s more so at the index level. As the number of indices skyrockets, the number of companies declines, and passive flows grow, active price discovery becomes more and more valuable.
What does the exponential proliferation of stock indices amid the decline in publicly listed companies, coupled with the surge of flows to passive investments, mean for market efficiency? The question is gaining in relevance for fund investors of all types, from those making top-down allocation decisions in passive products to those looking to active managers in search of benchmark-beating returns.
It also includes those plunking ever-more money into “smart beta” ETFs, which purport to identify the sources of excess returns, based on the historical contributions of key “factors” traditionally identified by active managers. Also known as “styles,” they mainly include momentum (earnings or share price growth), quality (high return on equity or assets), value (low share price-to-earnings or book value), market cap (small is beautiful), or lower volatility. Multifactor ETFs that combine these styles in a single wrapper are of more recent vintage.
While active managers run backward-looking quantitative screens to filter for stocks exhibiting one or more of these characteristics, they also engage in forward-looking, qualitative analysis of business models, balance sheets, earnings, company investment cycles, management quality and valuation metrics to forecast future cash flows, per-share earnings and dividend potential. If they’re right more often than not about the more limited number of researched stocks in their portfolios, that is, if the contributions from their winners outweigh the detractions from their losers, they generate excess returns for their shareholders. Passive products, by contrast, seek to diversify away single-stock risk with a mass of holdings in a bet on general economic growth, or, if they’re thematic, attempt to time market cycles in sectors or investing styles.
Rather than populating portfolios with highly vetted, individually selected stocks on backward- and forward-looking fundamental analysis, passive investors look to index-level movements to drive market returns. Systematic, or common, drivers affecting portfolio returns may include “expectations about monetary policy, inflation and other macroeconomic factors,” as a March report from the Bank of International Settlements put it.1 “They do not devote resources to seeking out and using security-specific information relevant for valuing individual securities. In effect, they free-ride on the efforts of active investors in this regard. Hence, an increase in the share of passive portfolios might reduce the amount of information embedded in prices and contribute to pricing inefficiency and the misallocation of capital.”
As passive’s share of investment funds increases, perhaps the price discovery in which active managers engage—at essence, the margin of safety assessment of upside potential and downside risks in individual stocks—becomes more valuable.
The efficient markets hypothesis stipulates that because share prices reflect all publicly available information, it’s virtually impossible to beat the market return. But even economist Paul Samuelson, who in the early 1970s derided most active mutual fund managers as financial “gun-slingers” incapable of consistently beating market returns, distinguished between the hypothesis at the micro and macro levels. Known as “Samuelson’s Dictum,” market efficiency was thought to be greater at the individual stock level than at the index level. Why? Because market participants could more easily spot a growing divergence in a company’s share price from its business fundamentals, exploit it and thereby correct it. At the index level, by contrast, individual stock stories are aggregated, and their fundamentals—both positive and negative—are averaged away, leaving index valuations much harder to interpret and gauge.
Price-to-Dividend’s Predictive Power Is Singular
In a 2006 academic paper entitled Samuelson’s Dictum and the Stock Market, Jeeman Jung and Robert Shiller found that price-to-dividend ratios of individual firms “have had some significant predictive power for subsequent growth rates in real dividend,” but when they’re lumped into an index, their predictive power dissipates or produces a false signal.2
Thornburg portfolio manager Ben Kirby has found much the same in practice. “Top-down, index-level market timing and valuation mean reversion is tough,” Kirby notes. “We can more effectively analyze the timing, duration, and probability of cashflows for a specific company than we could for an index. The bottom-up signal is so much clearer and so is more actionable compared to the index-level mean reversion argument.”
For example, in September 2015, Barron’s ran an article entitled, "Rising Interest Rates Will Boost Value Stocks," citing an academic who pointed out that value stocks have historically outperformed growth stocks during periods when the U.S. Federal Reserve tightened monetary policy.3 Moreover, “value has outperformed (growth) over the long haul, as a number of well-known studies…have observed” Since then, the Fed has hiked its Federal funds rate a cumulative 175 basis points.
As the table below shows, investors in BlackRock’s iShares factor ETFs who three years ago had overweighted value, which has subsequently lagged, and underweighted momentum, which is growth-stock heavy, would have painfully underperformed the MSCI USA Index. But the SIZE ETF also underperformed this index in the longer periods. While the Quality Factor ETF performed in line, the iShares MSCI Multifactor ETF , which combines the main investment styles, has materially underperformed in the period.
Timing mean-reversion bets on single-factor ETFs are difficult to pull off. When’s the market top? When’s the bottom? But what accounts for the underperformance of the iShares MSCI Multifactor ETF? Its aggregation of factors should enable it to act “as an alternative to a traditional long-term U.S. stock holding when seeking outperformance,” according to BlackRock.
Overvalued Index Darlings
“There is a very strong relation between the number of indices a stock is a part of and its price-to-book ratio: the more popular a stock is with index providers, the more expensive it is,” wrote Vincent Delaurd, INTL FCStone’s global macro strategist, in a recent report.4 “The relation is also visible, albeit less so, on other valuation metrics, such as the price-to-earnings, price-to-sales, and price-to-dividend ratio,” he added. According to Deluard, U.S. stocks “in more than 200 indices are 2.5 times more expensive than those in less than 75 indices.” Initially, the most favored stocks with index funds should outperform “because they benefit from this extra demand," he explained. “But after a while, index darlings should become so overvalued that their forward returns will eventually disappoint.”
The surge of investor flows into passive vehicles, including smart beta ETFs, has sparked debate about whether the flows are distorting valuations. “'It is aggressively oversold right now,' " Rob Arnott, one of the industry’s biggest proponents, was recently quoted in the Financial Times as saying.5 “'It shouldn’t be seen as a panacea. Factors can become materially expensive, and performance will mean revert. Some factors have been performing better precisely because they’ve been getting inflows.'” The same article noted that BlackRock estimates $1.9 trillion of assets in dedicated factor strategies and predicts the number to hit $3.4 trillion by 2022.
At the end of July, global ETFs and exchange-traded products of all types numbered nearly 7,500, counted more than 14,427 listings and held assets of $5.12 trillion, up from $774 billion a decade ago, according to ETFGI. Add in passive index fund assets, and the total rises to roughly $8 trillion, or 20% of aggregate investment fund assets by June 2017, up from just 8% the decade before.6
Yet as passive continues draw to assets, the number of publicly traded companies globally has declined to just more than 43,000 at the end of 2017, from nearly 45,000 in 2014. In the U.S., which sports 41% of global stock market capitalization, the number of listed companies plummeted by nearly half to 4,336 last year from a peak of 8,090 in 1996, according to the World Bank. And the number of indices globally? In January, the Index Industry Association counted 3.29 million, 95% of which are equity focused. “Top-down,” index-level analysis and benchmarking are skyrocketing.
Growing market inefficiency at the index level should over time make it easier for active managers at the micro level, especially because some trends may not mean revert. As technology develops, consumer preferences change, and regulatory environments shift, conventional portfolios can become vulnerable to structural displacements, as we recently discussed. “Winner-take-all” tech business models, for example, are radically changing the landscape for traditional media, telecommunications and cable companies, among other industries disrupted in one way or another. Rather than index-level mean reversion, fundamentals-based, flexible, forward-looking active investing can make tactical asset allocation more effective in delivering idiosyncratic returns.
Perhaps we’re seeing the early stages of passive-driven, index inefficiency playing out in Japan, where the share of passive vehicles ran around two-thirds of total equity investment fund assets in June 2017. The Bank of Japan, the country’s monetary authority, controls more than 60% of Japanese equity ETFs as part of its “quantitative easing” program. Then there’s also the massive Government Pension Investment Fund, whose allocation to passive equity funds surpassed 80% in 2016.7
By the end of 2017, 85%, 66% and 55% of Japanese large-cap actively managed funds outperformed their benchmarks in the one-, three- and five-year periods, respectively, according to S&P Dow Jones Indices’ Spiva report. Small- and mid-cap actively managed funds also outperformed in all three periods. The average returns of active funds in all three periods for both categories beat their respective benches in both equal-weighted and asset-weighted terms.
Of course, if passive investors are seeking the market return, investors in active funds are looking for managers who beat both the market return and the average of their active peers over the long run. Given the dispersion in returns of actively managed funds, their shareholders surely aren’t seeking the category’s “average” performance. They should seek funds with above-average long-term records, reflecting high conviction, high-active share, benchmark-agnostic portfolios.
Bottom-up research and relative value comparisons of individual securities can generate uncorrelated returns when assembled into a portfolio that considers both their individual risk/reward propositions and discrete contributions to broader portfolio diversification. Contrary to conventional wisdom, focused portfolios can exhibit lower correlations. In our experience over more than three decades, less correlated investments enhanced asset allocation by reducing volatility and improving a portfolio’s potential for strong, risk-adjusted returns across changing markets and unpredictable economic cycles. Much comes down to the efficiency of active price discovery at the stock level, amid a rising passive tide, index proliferation and the decline in listed companies.
1. The Implications of Passive Investing for Securities Markets-Bank for International Settlements (BIS), March 2018
2. Samuelson’s Dictum and the Stock Market-By Jeeman Jung and Robert Shiller, Yale University, 2006
3. "Rising Interest Rates Will Boost Value Stocks" –Barron’s, 9/12/2015
4. "The Index Bubble: Why It Is Time for an Anti-ETF ETF"-By Vincent Delaurd, INTL FCStone. 8/14/2018
5. "Can factor investing kill off the hedge fund?" –Financial Times, 7/22/2018
Copyright © Thornburg Investment Management