Actively trading passive vehicles

Since 1998, I have taught a course on investments at the Schulich School of Business, York University. I have noticed an increasing trend of ETF usage amongst my students, be it in the stock market simulation game in the course, or in their personal portfolios. This should not come as a surprise. After all, we would expect millennials to embrace new technology and products more readily, and to have an appreciation for the benefit of low costs and transparency.

That said, it does not mean that by investing in predominantly passive ETFs (passive ETFs track a well-defined index of securities), my students are necessarily taking more of a buy-and-hold approach. As a recent publication¹ by Fogertey and Boroujerdi of Goldman Sachs highlights, in the institutional space, active managers are among the largest users of ETFs. And John Bogle, founder of the Vanguard Group, has spoken many times about the “temptation to trade” that ETFs create.

My PhD student and co-author, Markus Broman (soon to join the Whitman School of Management at Syracuse University), and I conducted an empirical study² to examine the issue of short-term trading in ETFs. By short-term, we are not referring to high frequency traders, who are also active ETF users, but to institutional money managers whose quarterly holdings data we have access to through the Securities Exchange Commission (SEC) 13-F filings. By implication, then, the shortest holding period of an ETF in our study is a quarter, or three months. We show that for passive U.S. equity ETFs, the average holding period was 2.8 quarters between 2006 and 2012. Less than three-quarters clearly do not qualify as buy-and-hold. Of course, we do not expect all institutional investors to buy and hold; many of them are active portfolio managers. One way to benchmark 2.8 quarters is to compare it to the average holding period of the individual stocks that make up the ETF’s underlying index. We estimate the average value-weighted holding period to be two quarters longer for the same period. Even SPY, which tracks the S&P 500 Index and is the world’s largest ETF, has an average holding period of only 3.3 quarters, compared to 5.8 quarters on average for the S&P 500 constituents.

In the study, we also investigate whether ETF liquidity, when compared to its underlying securities, is what attracts and facilitates shorter-term trading. We measure liquidity using three secondary market measures (proportional quoted spreads, turnover and price impact), and a primary market measure unique to ETFs (creation/redemption activity). Indeed, we find that ETFs that are generally more liquid are held by an institutional investor for a shorter duration, and they tend to attract investors that are classified, a priori, as having shorter investment horizons³.

As the title of this article suggests, it seems paradoxical that a passive vehicle would be used in a shorter-term, active strategy. There are several strategic reasons for institutional investors to use ETFs this way, such as quick exposure and rotation to style, region or sector; hedging; transition and cash management, to name a few. However, casual observations suggest that some retail investors, in a move away from professionally managed funds with higher fees and toward self-directed portfolios, are also in favour of actively managing passive ETFs. In this case, let’s hope they do their research, diversify and benchmark performance properly.