UIT Managers Exhibit Poor Stock Selection Skill
Unit investment trusts (UITs) are SEC-regulated investment vehicles in which a portfolio of securities is selected by a sponsor and deposited into the trust. Although the number of UITs has fallen from approximately 12,000 in 1990 to about 5,000 by the end of 2014, assets held in UITs have grown steadily since the financial crisis, increasing from about $20 billion at the end of 2008 to more than $100 billion by 2015. (These numbers exclude the eight ETFs that are structured as unit investment trusts.)
Unlike mutual funds, equity UITs invest in fixed portfolios of stocks for a predetermined period of time. Generally, equity UITs terminate within two years of their launch date (although some are longer), with many of them terminating at one year. At the end of this specified term, holdings are liquidated and investors receive the proceeds from their investment. A benefit of UITs is that, unlike mutual funds, they don’t have to hold cash to meet redemption needs. The lack of a “cash drag” provides UITs with an advantage. And their lack of turnover keeps total costs down.
Importantly, the fixed portfolio (buy and hold to termination date) of UITs provides a unique sample to specifically measure stock selection skill (isolating it from the other part of active management, market timing).
Kimberly Luchtenberg, author of the January 2017 paper “REIT Unit Investment Trusts and Fund Manager Skill,” examined the stock selection skills of fund managers using a sample of real estate investment trust (REIT) UITs. As Luchtenberg hypothesizes, a benefit of limiting the study to REIT UITs, instead of the entire UIT universe, is that, because “the REIT population is much smaller than the entire population of stocks, it would be reasonable to suspect that informed managers have the ability to become more skilled at selecting REITs than they are at selecting stocks in general.”
Her study covered the period May 2009 through July 2015 and 37 REIT UITs, each holding 25 REITs using an equal weighting. Because managers must sell assets at the end of the term, and market impact costs can occur, returns nine days from the UIT’s termination date are excluded (of course, investors still must bear those costs, but this is separate from the issue of stock selection skill).
Luchtenberg used the Fama-French three-factor model (beta, size and value) and the Carhart four-factor model (adding momentum), with the data for high minus low (HML), small minus big (SML) and momentum (MOM) coming from Kenneth French’s website. Luchtenberg found that REIT UITs deliver negative three-factor and four-factor alphas, though the results were not statistically significant.
We can also observe whether active management is likely to add value for REIT mutual funds. For the 15-year period ending in 2016, the Vanguard REIT Index Fund (VGSLX) posted a Morningstar percentile ranking of 39, meaning it outperformed 61% of the surviving REIT funds. However, Morningstar’s ranking contains survivorship bias. Over that 15-year period, 18 of the 50 REIT funds that existed at the start of the period were no longer in the database—just 32 survived.
If we adjust for survivorship bias by making the assumption that the nonsurvivors had below-benchmark returns, then VGSLX’s ranking changes to the 25th percentile. It outperformed 75% of actively managed REIT funds. In other words, active management is just as much of a loser’s game in REITs as it is elsewhere—it’s a game that, while it’s possible to win, the odds of doing so are so poor, that it’s not prudent to try.
Further Evidence On UITs
George Comer and Javier Rodriguez, authors of the May 2015 paper “Stock Selection Skill, Manager Flexibility, and Performance: Evidence from Unit Investment Trusts,” examined the performance of 1,487 diversified UITs over the period 2004 through 2013. The authors excluded sector UITs, such as the REIT UITs that Luchtenberg studied.
In analyzing the performance of UITs, Comer and Rodriguez compare their returns to four benchmarks. The first is the single-factor (beta) CAPM model. The second is the Carhart four-factor model (beta, size, value and momentum). Because the UITs held an average of about 12% in international equities, the authors used a third model, adding an international index (the MSCI World ex U.S. Index) to the Carhart model. Their fourth model was a trust-specific benchmark, such as a Dow Jones sector index. Following is a summary of their findings:
- Regardless of the model used, UITs generated significant negative alphas—between -2.5% and -2.8%.
- More than 65% of the UITs generated negative alphas.
- The number of trusts with negative and significant alphas is four times as large as those with positive and significant alphas. Note that, given expense ratios of about 0.23%, expenses don’t explain the negative alphas.
- None of the trust series were able to generate significantly positive alphas at the 5% level of significance.
- Poor performance was consistent across asset classes, as the average alphas were all negative.
- The 2008 financial crisis was not a main driver of the negative alphas.
- UITs significantly underperformed mutual funds, suggesting that restricting flexibility and maintaining full investment in the market doesn’t result in better risk-adjusted performance.
- Observable trust characteristics, such as trust size and expenses, weren’t able to explain the poor performance.
- There was no evidence that their inability to trade was the main driver of the results, as UIT performance slightly improved during the life of the trust.
- Poor performance does persist within trust families. Focusing on the Carhart model results, the average primary trust alpha was -2.8% and the secondary trust alpha was -2.9%. Both of these estimates are statistically significant at the 1% level.
Comer and Rodriguez concluded their tests suggest that the negative alphas of UITs reflect poor stock-selection skill. This is an interesting finding, one that’s very different from the evidence on actively managed mutual funds. Studies on individual investors have found they also exhibit poor stock selection skills, but this isn’t true for mutual funds.
For example, a study by Russ Wermers, “Mutual Fund Performance: An Empirical Decomposition into Stock-Picking Talent, Style, Transactions Costs, and Expenses,” which was published in the August 2000 issue of the Journal of Finance, found that, on a risk-adjusted basis, the stocks’ active managers outperformed their benchmark by 0.7% per year. Unfortunately for investors, stock-picking alpha wasn’t close enough to generate alpha after expenses, as total costs, including the cost of “cash drag,” exceeded 2%.
The evidence makes clear that UITs are another in a long list of products that are meant to be sold (because of the high sales charges embedded in them), but never bought. The bottom line is that, whether we’re looking at UITs or actively managed mutual funds, investors are more likely to achieve their goals by limiting their investments to passively managed vehicles.
This commentary originally appeared February 27 on ETF.com
By clicking on any of the links above, you acknowledge that they are solely for your convenience, and do not necessarily imply any affiliations, sponsorships, endorsements or representations whatsoever by us regarding third-party Web sites. We are not responsible for the content, availability or privacy policies of these sites, and shall not be responsible or liable for any information, opinions, advice, products or services available on or through them.
The opinions expressed by featured authors are their own and may not accurately reflect those of the BAM ALLIANCE. This article is for general information only and is not intended to serve as specific financial, accounting or tax advice.
© 2017, The BAM ALLIANCE