Index Capital Insights Issue #5

This issue of Index Capital Insights is part 2 of a 3-part series on small cap stocks. This month, I’ll discuss a few important distinctions between prominent US small cap benchmarks and ETFs.

Let’s begin with a survey of the landscape. The second quarter is an annual milestone as we approach the Russell reconstitution. This year it is anticipated that several of the largest R2000 members will graduate to the Russell 1000, including Super Micro Computer (SMCI), Microstrategy (MSTR), and others. As I noted recently, highlighting the importance of benchmark selection, investors holding both an S&P 500 index fund and a Russell 2000 index fund are facing unnecessary turnover as SMCI was added to the 500 at S&P’s March rebalance and will be sold from R2000 funds at Russell’s reconstitution.

A simple screen of broad (not growth or value) small cap ETF’s from ETF.com shows the largest funds by AUM. Of the top 10, 7 are purely passive funds tracking one of the leading US small cap benchmarks. The other 3 lean more towards the active space, to varying degrees. Table 1a covers the passive benchmark ETF’s.

Table 1a, Leading Small Cap Benchmark ETF’s

Note Table 1a and 1b omit index mutual funds and other index trackers. Taking them into account would undoubtedly, and markedly, increase the indexed capacity utilization (AUM/Index Cap) of the Russell 2000. According to FTSE Russell, the 2023 Russell reconstitution amounted to about $134 billion traded across US Exchanges[1]. However, the S&P SmallCap 600 is the leader in small cap ETF AUM – likely due to IJR’s placement in iShares’ Core ETF lineup and the 600’s prior record of outperforming R2000.

Table 1b covers the active ETF’s. CALF and FNDA purport to be index funds, but the indexes they track are highly bespoke, relatively narrow, rules-based cuts of small cap universes. Such customized indexes are often designed with a single use case in mind, such as serving as underlying benchmark for an ETF or structured product. Benchmarking the benchmarks in these situations may be challenging, as the indexes themselves become an integral part of the perceived alpha. DFAS is a truly active fund, which has the virtue of not muddying its “activeness” with its benchmark.

Table 1b, Leading Small Cap Active ETF’s

Among leading small cap benchmarks, index methodologies vary greatly. Table 1a shows that one result is the range of total market capitalization represented in each index. The S&P SmallCap 600 is significantly narrower than other leading small company benchmarks, followed by the Russell 2000 and DJ Total Stock Market Small Cap which both select a specific number of securities in size rank order. The CRSP US Small Cap Index selects securities using a pre-defined range of total market capitalization rather than a target constituent count and is the broadest of major small cap benchmarks. An advantage of CRSP’s approach is the index responds naturally to changing skew in the underlying stock market capitalization distribution. For example, fewer stocks are included in the large cap size benchmark, and more in the small cap benchmark, as more total stock market capital is “consumed” by the largest companies.

Another significant variation resulting from methodology differences is sector exposure. Chart 1 highlights the 2 leading benchmark ETFs, IJR (SP600) and IWM (R2000), and shows their comparative exposures across the 11 GICS economic sectors.

Chart 1

No doubt the picture will change somewhat after the Russell reconstitution becomes effective in July, but some key differences in sector exposure have roots in methodological differences other than the timing of constituent changes. For example, the S&P SmallCap 600 is perennially underweight Health Care because many small biotech firms operate with losses, and S&P’s methodology requires positive EPS to be eligible for the 600.

There is a lot more detail that could be covered about index methodologies, but I’ll close with a look at each ETF’s historical loading on the Fama-French small company factor (SMB). After all, if you are going to the trouble of vetting and implementing an explicit small cap allocation, the small company premium is what you are shooting for.

Table 2, Benchmark ETF Fama-French 3 Factor Model regression coefficients / t-stats

The factor regression shows interesting results. Over the sample period, IWM (R2000) had the highest small company factor loading with the most statistically significant t-stat followed by IJR (SP600), SCHA (DJ TSM SC), and then VB (CRSP SC). All the ETFs also generally had statistically significant loadings on value, but IJR’s was the highest coefficient with the highest t-stat. Given the S&P index rule requiring positive earnings, these results seem to make sense intuitively. The count-based index ETFs (IWM and SCHA) offer straightforward small cap factor exposure, while the S&P index is a narrower small cap index with a value tilt. The CRSP index tracker, VB, provides small cap exposure but in a less pure manner given the breadth of its market coverage.

For an investor, given a desire to explicitly allocate to US small cap stocks, choices must be made along a continuum of trade-offs. A sound approach (among many I’m sure) is to first select the benchmark family you want to build your portfolio around. Select a benchmark family that most closely aligns with your indexing and investing philosophy. I prefer a core position in a total market index fund combined with a modest small cap allocation to a fund tracking the small cap index from the same index provider, resulting in controlled duplication of the small cap names in the indexed portfolio. Another reasonable choice might be to split the small cap allocation into 2 sleeves - directing a portion to an index fund and the other to a truly active, yet diversified, fund like DFAS. In any case, as you make portfolio decisions remember that benchmark selection is an active process requiring time and attention.

[1] See 2024 Russell US Indexes reconstitution: summary of preliminary changes, May 24, 2024.