After the extraordinary performance of growth and momentum stocks in 2017, many investors are asking why bother with value stocks? In a late-stage global bull market exhibiting global synchronized gross domestic product (GDP) growth, why should value – often associated with early-stage economic recovery
– succeed? We argue that now is precisely the time to emphasize value, especially with growth stocks trading at extreme premiums to value stocks and dispersion in market multiples well above long-term averages. History has also indicated that maintaining a value discipline, particularly when valuations are high and interest rates have reached a floor and turned upward, has produced stronger relative returns than a growth orientation over full market cycles.
The mere existence of economic cycles and corresponding fluctuations in interest rates constrain investor enthusiasm for equities. Stock prices do not completely disconnect from underlying fundamentals since active managers, private equity funds, and strategic buyers have ultimately taken advantage of mispricing and closed the valuation gap. In the long run, a value style bias has been rewarded. As illustrated in Exhibit 1, the MSCI World Value Index has “clobbered” the MSCI World Growth Index, outperforming by 2.1% annually since 1975 (the inception of these indices). Remarkably, this superior value return has exhibited lower volatility (14.4% annualized) than the growth return (15.3% annualized volatility). We observe a narrower value advantage (but still outperforming by 0.3% annually) if we view this in an All-World (including Emerging Markets) context, though the MSCI Emerging Markets Value and Growth indices only extend back to 1997.
While it is clear that value has outperformed growth over the long run, most investors do not have the luxury of a 40+ year investment horizon. However, a value investor has not require one. Exhibit 2 shows the performance of the MSCI World Value Index versus the MSCI World Index over rolling 5-year periods since 1975. 72% of the observations show positive relative performance of the MSCI World Value Index. Due to the construction methodology of the Value and Growth indices, the chart below would look nearly identical if we judge Value’s performance versus Growth (rather than the MSCI World itself).
Admittedly, the MSCI Value and Growth indices are not perfect – they fail to completely isolate their respective namesake style effects. This is because any index constructed around one dimension may bring with it many other risks and biases such as country, currency, sector, and style allocation effects. We can utilize the proprietary Causeway Risk Lens to disaggregate these effects and uncover which active risks drive the greatest differences between the MSCI World Value Index and the baseline MSCI World Index. Our Risk Lens currently estimates annual tracking error of the MSCI World Value Index at approximately 2%. Style exposure differences (overweight to value and underweight to momentum and growth) drive 57% of this active risk, sector differences (underweight to information technology and overweight to financials) drive 29%, with the remaining portion explained by differences in stock-specific risks. MSCI matches the country/currency weights between its Value/Growth and baseline indices, so these are not factors.
Even after controlling for these effects, however, we find the same long-term outperformance of value over growth. To arrive at the return streams for “pure value” and “pure growth,” we utilize the style returns from Causeway’s multi-factor risk model, which disaggregates cross-sectional returns into those attributable to the broader market and those attributable to specific countries, sectors, currencies, and styles. After isolating value and growth in Exhibit 3, the total returns fall from those in Exhibit 1 since we have now stripped out the market’s returns, but the cumulative effect has been very similar – over twice the return by investing in value versus growth (though growth briefly surpassed value during the height of the late 1990s/early 2000s tech bubble). And this outperformance is just since 1991 (the start of our risk modeling) versus 1975 in Exhibit 1.
Continue reading here.
Also check out:
