Considerations for the “Forgotten Size Segment”: Mid Caps

While many allocate specifically to large caps and/or small caps, mid-caps tend to be something of a forgotten size segment. Our research team conducted an under-the-hood analysis of smart beta indexes to detail their underlying factor exposure and explain factor return drivers in performance comparisons. The relative tilt of a portfolio or index to small-cap stocks and value stocks can have a large impact on performance,¹ and one of the interesting takeaways from our factor analysis was just how different the size indexes and exposure factors were for the various mid-cap indexes. In this blog post, I will first show the size factor analysis across mid-cap indexes and then provide a framework that can serve as a strategic baseline for considering how much of a portfolio could be allocated to mid-caps.   The Mid-Cap Size Factor Spectrum: WisdomTree at “Book-Ends” For definitions of indexes in the chart, please visit our glossary. • Fundamentally-Weighted “Book-Ends”: It’s interesting that the WisdomTree MidCap Dividend Index (WTMDI) is the largest mid-cap Index shown, whereas the WisdomTree MidCap Earnings Index (WTMEI) is the smallest. Size factors of 0.20 to 0.52 represent a wide variation in size exposure, in our opinion, and therefore very different potential risk and return characteristics. Even within the market capitalization-weighted peers, the S&P MidCap 400 Index’s small-cap tilt was more than 50% greater with its size loading factor of 0.43, compared to 0.28 for the Russell Midcap Index. • “Book-Ends” Offer a Size Exposure that Is Materially Different from the Large and Small Caps: While the S&P 500 (large cap) and the Russell 2000 (small cap) have size factors at opposite ends of the spectrum (at -0.15 and 0.85, respectively), the mid-cap indexes occupy a notably different niche, with factors that range from 0.2 to 0.52. This suggests that mid-caps potentially offer a different return experience than their large- and small-cap peers. • Blends Worth Considering: Based on these figures, a blend of approximately 75% WTMDI and 25% WTMEI would match the size factor of the Russell Midcap Index, whereas a blend of approximately 72% WTMEI and 28% WTMDI would match the size factor of the S&P MidCap 400 Index. These are very different portfolios, which helps us emphasize that benchmark selection is very important when considering mid-cap equities in the U.S. Having discussed the unique ways of accessing mid-caps and how different the size factor exposure can be across index families, let’s turn to the larger question of how much of a total portfolio should be allocated to mid-caps. One framework to answer this question begins with the Russell 3000 Index as of December 31, 2013, and cuts the exposure as follows. Approximate weights: • Large caps: Over $10 billion in market capitalization (78.4% weight) • Mid-caps: Between $2 billion and $10 billion in market capitalization (16.4% weight) • Small caps: Below $2 billion in market capitalization (5.2% weight) We ran a factor analysis that combines this portfolio allocation to WisdomTree Large-, Mid- and SmallCap Indexes in the same proportions as large, mid- and small caps appear in the Russell 3000 Index. The resulting factor exposures are listed in the table below.   Building Exposure to the Market Capitalization Size Spectrum • Dividend Blend: Simply put, the major attributes of the dividend blend, compared to the Russell 3000 Index, are 1) lower beta, 2) exposure to larger companies, and 3) a greater tilt toward the value style that is largely similar to the Russell 3000 Value Index. • Earnings Blend: The major attributes of the earnings blend, compared the Russell 3000 Index, are 1) slightly lower beta, 2) very similar size, and 3) very similar “core” allocation with a value factor similar to the Russell 3000 Index (core). The crucial point is that all of this was accomplished with nearly 100 basis points of incremental average annual performance.   Conclusion In our research on smart beta, we used a factor model analysis to describe the exposure inherent in WisdomTree’s Index methodology. We believe a critical part of evaluating the smart beta approach is taking a look under the hood of the strategies. We hope this research and blog series helped frame why and how WisdomTree’s U.S. Index family achieved the results it did over the real-time period analyzed.   ¹Source: Eugene F. Fama and Kenneth R. French, “Size, Value, and Momentum in International Stock Returns,” 6/21/11.