Is it time for U.S. Small Caps to shine? If you pick them right

Re-examining our underweight of U.S. Small Caps

Given recent volatility in foreign equity markets and somewhat disappointing performance relative to U.S. stocks year-to-date, we examined our underweight of U.S. small caps to fund our overweight of foreign developed-market (DM) and especially emerging-market (EM) stocks. We are not underweight U.S. Small Caps to fund an overweight on U.S. Large Caps, as our U.S. Large-Cap allocation is also underweight vs. our long-term allocation; because we prefer foreign over domestic stocks, especially since late 2017. We are constructive and overweight global equities vs. fixed income.

That said, we regularly compare U.S. Small to U.S. Large Caps as part of our global-equity allocation strategy, which is the focus of this note. Our analysis shows that very demanding valuations or high secular earnings-per-share (EPS) growth expectations persist at U.S. Small Caps vs. U.S. Large Caps, despite recurring disappointments through this cycle in their ability to outpace U.S. Large-Cap EPS growth. Our analysis of 1Q S&P 500 vs. Russell 2000 EPS results suggests that there aren't yet signs of Small-Cap EPS growth materially outpacing large-cap EPS growth in 2018 or beyond, despite Small Caps being greater beneficiaries of the corporate-tax-rate cuts. Thus, we still prefer U.S. Large over U.S. Small Caps and continue to prefer foreign stocks vs. the S&P 500.

Russell 2000 1Q EPS growth (ex. Health Care) similar to S&P 500, but bigger tax benefit

The headline 1Q year-over-year EPS growth rate for the Russell 2000 (R2000) is very strong at 39.8%. However, the R2000 Health-Care sector had large and similar losses in both 1Q17 & 1Q18, which boosts the index's overall EPS growth in a very distorting way. Excluding Health Care, R2000 EPS growth is 26.6% year-over-year, which is similar to 27.6% at S&P ex. Health Care. Yet R2000 had a bigger tax-cut benefit given most of its revenue is generated domestically. We estimate a ~20% tax-benefit boost to R2000 EPS vs. 10% at most for the S&P 500. This means 1Q18 EPS growth from operations is ~7% at R2000 vs. 15%+ at S&P 500. If we exclude all R2000 companies with negative EPS in 1Q18 and a year ago, then R2000 1Q18 EPS growth is 17.7%, which is slower than the 25.7% overall at S&P. We do see R2000 EPS growth accelerating from its 2015-17 slump, but it appears to have less operating leverage to the improving economy vs. large caps.

Making sense of Russell 2000 EPS growth requires sector-level examination

A couple R2000 sectors had impressive 1Q EPS growth, including Consumer Discretionary at 47% and Financials at 38%. This makes sense given the cyclical and domestic nature of these two. However, other R2000 sectors posted 1Q EPS growth that appears fully attributable to the tax cut. Such as Industrials at 21% and Technology at 19%. Defensive domestic sectors like Staples and Utilities posted better EPS growth of 24% and 27%, respectively, but also mostly tax-cut-driven. We realize that 1Q U.S. GDP wasn't great and that Q2 should be stronger, but the underlying trend has been upward for several quarters now and yet this R2000 EPS growth remains unimpressive.

Russell 2000 PE on 2018E EPS is 26, 19.7 ex. Health Care & 17.6 ex. negative EPS

Consensus 2018E R2000 EPS growth is 50%+, but ~17% ex. negatives, vs. about 21% for S&P. R2000 trades at 26x 2018E EPS, we consider bottom-up 2018E R2000 EPS reasonably credible, but this price-to-earnings (PE) ratio is high even with negative EPS given it is nine years since the last recession. The R2000 PE is less demanding ex. Health Care and negative EPS stocks. The S&P PE on bottom-up 2018E EPS, to which we have now aligned our estimate, is 16.8x or 4% cheaper than R2000 ex negatives. To compare PEs, consider relative EPS growth + dividend yield of both 2019 and beyond.

Historically R2000 has the same EPS growth + dividend yield as S&P 500

Since 1995, the average EPS growth + dividend yield is the same for R2000 and S&P 500, both 8.4%. R2000 has higher EPS growth (6.9% vs. 6.5%) yet lower dividend yield (1.5% vs. 1.9%) than S&P 500. On this basis, investors were equally rewarded for long-term ownership of R2000 and S&P 500. But the R2000 has much greater earnings cyclicality, so that's why we think it is justifiable that R2000 ex negatives has historically been trading at a small discount to the S&P 500 on average.

PE of Russell 2000 ex. negatives is 18.9 on annualized 1Q18 EPS, S&P is 17.4

If we annualize 1Q18 EPS, R2000 trades at 35.8x such annualized recent-quarter actual EPS, but 18.9x ex. negative EPS stocks; this is still a 9% premium to S&P at 17.4x annualized 1Q18 EPS. Since 1978, R2000 ex. negative trailing EPS stocks trades at 4% discount to the S&P on average. Is the current higher-than-average valuation premium at R2000 justified by the earnings outlook? If we look beyond 2018, consensus 2019E R2000 EPS growth is 28% and 10% for the S&P 500. We consider the 2019E bottom-up R2000 EPS highly doubtful, as this far out tends to be far too high. We think the Russell is likely to generate 5-10% EPS growth and the S&P 5-8% in 2019.

How to invest in small caps? Carefully and selectively

demanding PE and continued slow EPS growth, excluding the one-time tax cut benefit, at the Russell 2000 overall amidst an aging cycle calls for selective U.S. Small-Cap investing in our view. Careful consideration of valuations, current and long-term earnings, and even takeover potential, are important to navigate in this environment. For instance, although Small-Cap Health Care was unprofitable in past years and losses are expected again in 2018, most of the year-to-date Russell outperformance comes from Health Care owing to a mergers-and-acquisitions wave in Biotech.