The Myth of a Declining Size Premium


The so-called size premium, where stocks of smaller companies perform better than larger rivals over time, is again being questioned.

In the past five years, the small-cap Russell 2000 Index had gained an average annualized 12.8%, as of mid-June, according to Morningstar. By contrast, the blue-chip S&P 500 Index had returned 13.7%.

But such an uptick has been highly uneven. It also comes as longer-term average annualized data show small caps still coming out ahead.

"Any subset of the data, such as five years' worth of data, is bound to contain significant errors in its attempt to describe the risk and return of an index," wrote Mark Hebner in the 2007 edition of his classic book on index funds.

Given evidence uncovered by academic luminaries such as Eugene Fama and Kenneth French showing a long-term premium for owning small-cap stocks, we've decided to update a chart that appeared in earlier versions of "Index Funds: The 12-Step Recovery Program for Active Investors."

The refreshed graphic (see below) captures total cumulative returns of small cap stocks. But instead of simply comparing large caps to small caps, our analysis takes a more granular view. A wealth of research by Fama and French, among others, finds that the size premium exists even within small-cap stocks. 

In the chart below, we've replicated IFA's research process from 2007 by breaking the US market into 10 distinct categories, or deciles, according to market-cap size and the S&P 500 for comparison.

As depicted here, the "largest" decile of stocks is pictured on the far-left side and labeled as the "first" decile. At the other end of the spectrum is the "10th" decile, or the smallest market cap.

Notice the wide swings that can take place during shorter timeframes. Then again, we've framed such slices of market history against a much broader data set (1927-2017). Against such a comprehensive 90-year backdrop, the importance of exposing a portfolio to the size factor becomes quite evident.

History of the size effect

This is consistent with a wave of academic research going back to 1981 when Northwestern University's Rolf Banz formally identified that a size premium existed. It's worth noting, however, that his original paper points out that the effect "is also not very stable through time." Banz concludes that his analysis shows "substantial differences in the magnitude of the coefficient of the size factor."

So how can investors construct robust strategic portfolios that recognize a size premium does exist, but can be quite elastic?

Exposure to small caps can vary within different funds and indexes. For example, the Vanguard Small Cap Value Index fund is listed by Morningstar with more than twice the average market-cap size of the DFA U.S Small Cap Value Portfolio: $3.8 billion vs. $1.7 billion.

Such a stark contrast in two seemingly similar small-cap funds can make a huge difference in terms of capturing the full brunt of the size premium.

In markets where size might become less definitive in terms of returns, research also indicates that designer indexes slanting away from growthier names with less profitability can greatly enhance small-cap performance. (See table below.)

Size Factor Overview (1975-2017)
(Top 3,400)
Low Prof
Small Caps
ex Small
Low Prof
Annualized Compound Return 12.58 12.31 14.53 5.11 15.76
Annualized Standard Deviation 15.13 14.9 19.59 28.49 18.68
Annualized Premium (vs Large Caps) -- -- 2.22 -- 3.44
Annualized Tracking Error -- -- 9.93 -- 9.29
Average Monthly Return 1.09 1.07 1.3 0.76 1.37
T-stat of average return
difference vs large caps
-- -- 1.85 -- 2.62
Average Number of Names 3,383 906 2,477 382 2,095
Source: Dimensional Fund Advisors.

In a new research report by Dimensional Fund Advisors, analysts Kaitlin Simpson and Wei Dai find domestic small caps as a group produced an annual compound return of 14.53% from 1975 through 2017.

Meanwhile, small-cap growth stocks with relatively low profitability metrics generated 5.11%. Also, annual standard deviation, a measure of volatility, was appreciably higher for this corner of the small-cap market.

Perhaps more revealing is that in benchmarking all styles of small caps, the study's authors found a major performance boost from investing in smaller companies while avoiding growthier names with low profitability.

On an annual compound return basis over the period studied, gains shot up to 15.76%. At the same time, overall portfolio risk (as measured by standard deviation) fell by almost a full percentage point.

The upshot: such a tilt bumped the size premium against portfolios heavy in large-cap stocks from 2.22% to 3.44% on an annualized basis. It also had a strongly positive impact on the t-statistic, which measures the probability that outperformance of small caps versus large caps could have occurred from chance alone.

The new DFA study, which looked at the top 3,400 stocks in the U.S. market, found a T-stat of average return difference between small caps and large caps of 1.85.

In general, a positive alpha with a T-stat greater than 2 indicates a 2.5% or lower probability that the excess returns are due to luck.

Minus smaller growth stocks with low profitability, however, the T-stat spurts to 2.62. This gives us further evidence that passively managed portfolios seizing on both the size premium and profitability factors can generate higher returns, net of fees, over time for investors.