Abandoning Diversification
Does the S&P 500 deserve its premium valuation over small, international, and value stocks?
By: Daniel Rasmussen
We are believers in the benefits of international diversification. We are convinced by the long-term evidence that value stocks should earn a premium over time. And we think there’s a greater opportunity for alpha generation in small and micro caps compared to mid and large caps.
But those beliefs about how to invest over the long term have not been short-term winners. On the contrary, the winning trade over the last decade has been to go long US technology stocks or to put the bulk of your money in the passive indices they dominate (the S&P 500 and the Nasdaq).
There are two ways to interpret the clash between long-term evidence and short-term results.
The first is the idea that there’s been a paradigm shift: that the US economy is simply a far better engine than Europe or Asia, that software is eating the world, that traditional industries are in decline, and that we’ve entered a winner-takes-all world where large caps should dominate small caps.
The second is the idea that markets are cyclical, that what works one decade rarely works the next decade, and that the recent results were largely attributable to changes in valuations and investor preferences that have little predictive power for the next decades’ fundamentals.
It's hard to prove the first interpretation, though there’s certainly a plethora of anecdotal evidence. But the way to evaluate the second interpretation would be to look first at the extent to which changes in valuation multiples have explained stock returns rather than changes in fundamentals.
Let’s start with international stocks versus US stocks. The below chart shows the discount on enterprise value-to-sales ratio of international stocks (as measured by the S&P International 700 index) versus US stocks (as measured by the S&P 500 index).
Figure 1: International Discounts on EV/Sales vs US Stocks
Source: Capital IQ
Over the past decade, international stocks went from trading at about a 15% discount to US stocks on EV/sales to a 50% discount, making quite a dramatic case for the cyclical theory.
We covered the widening in the small-cap discount last week, so we’ll focus in today’s research on the discount for value stocks in the US and internationally. The below chart compares the EV/EBITDA ratio for the cheapest 30% of stocks in each market to that of the S&P 500 index.
Figure 2: International Value Discount Relative to the S&P 500
Source: Capital IQ. NOTE: Value stocks defined as the cheapest 30% of all stocks with a minimum market capitalization of $25M.
Value stocks globally “cheapened” by 20 percentage points or more over the past decade relative to the S&P 500. US, developed Europe, and EM value stocks are now trading at a 50% discount to the S&P 500, while Japan is trading at a staggering 70% discount. This is ignoring the size effect, which should further widen the discount.
The valuation multiples of international stocks, small-cap stocks, and value stocks have become significantly cheaper relative to US stocks over the past decade, with discounts in some cases widening more than 35 percentage points.
The natural next question, though, is to what extent this steep discount is justified by fundamentals. Three of the most common rationalizations are differences in return on assets, margins, and growth rates.
Let’s start with return on assets. The below chart shows the difference between the return on assets of the S&P 500 and international stocks (as measured by the S&P International 700), small-cap stocks (as measured by the S&P 600), and value stocks (as measured by the S&P 500 Pure Value) over the past 13 years.
Figure 3: ROA of International, Value, & Small-Cap Stocks Minus the S&P 500
Source: Capital IQ
The chart shows that, while international stocks, value stocks, and small-cap stocks have lower ROAs on average than US stocks, the differences are small and there’s been relatively little change over time.
We next look at margins. The below chart shows the difference between the EBITDA margins of the S&P 500 and international, small-cap stocks, and value stocks (as measured by the same indices as in Figure 3 above).
Figure 4: EBITDA Margins of International, Value, & Small-Cap Stocks Minus the S&P 500
Source: Capital IQ
International stocks, value stocks, and small-cap stocks have lower margins than S&P 500 stocks but are relatively unchanged from a decade ago, albeit with high volatility for value stocks.
When looking at return on assets or margins, we don’t see the type of sharp drop that we see in the valuation multiples. Nothing about ROA or margins explains the sharp drop in valuations for international, value, and small-cap stocks relative to the S&P 500.
But what about growth? The below chart compares the trailing three-year revenue CAGR of international stocks, small-cap stocks, and value stocks relative to the S&P 500 over the past decade.
Figure 5: Growth Rates of International, Value, & Small-Cap Stocks Minus the S&P 500
Source: Capital IQ
Small caps have consistently grown revenues faster than large caps, which makes their discount puzzling on this metric. Value stocks have swung from growing faster to growing slower to growing faster to growing slower, and we suspect this volatility will continue, making the current valuation gap for value stocks seem difficult to justify.
The case for a fundamental gap is best for international stocks. International stocks have consistently grown slower than US stocks by about 5% per year over the past decade. This seems perhaps the closest to a fundamental explanation for the lagging relative returns of value and international stocks, if we believe that growth is predictable and persistent and that US companies will continue to have a significantly higher growth rate than international stocks.
The question of whether US or international stocks will have faster revenue growth over the next decade is insoluble. But we can ask a more basic question: are growth rates persistent and predictable? We will be publishing a major study on that topic this fall, one that we hope will help contextualize this question. And if, as we plan to argue, growth rates are neither persistent nor predictable, then extrapolation of the international growth lag into the future might be creating a big and exploitable expectation error for investors.
Perhaps at the time when many investors are abandoning diversification in favor of putting all their money in the S&P 500, it’s time to reconsider—and make the bet—that valuation gaps that have driven wide relative return differences might be strongly mean reverting.