The Long/Short ETF Mousetrap: CSM A Winner

September 15, 2014

  • An overall beta of 1 to the market (meaning it doesn’t just play the beta manipulation game)
  • Any “bet” on a stock can only be +/- 40 bps of its weight in the S&P 500

The minus sign there is critical. It means a bunch of stocks that have a tiny weight in the S&P 500 actually get negative weights in the 130/30 index. That’s where the 130/30 comes in—you raise cash from the short selling, and you use that cash to buy more of what you like, to get your overall exposure back up to a beta of 1.

In terms of implementation, the actual ETF—CSM—takes a hybrid approach, gaining most of its short exposure through swaps, and getting its long exposure through a combination of plain old purchases of shares and swaps.

As these types of strategies go, it’s actually been pretty effective, and not terribly expensive. With an expense ratio of 45 bps, CSM is hardly insane, although the median tracking difference would suggest the true cost is a bit more like 80 bps.


The tracking is in fact so tight around that 80 bps, it makes me fairly certain what you’re seeing is the cost of the swap positions. The 30-50 bps strikes me as about right for that, so keep that in mind when assessing the “true cost” here.

And what do you get for that? Well, it turns out, you get alpha—actual statistically significant, risk-adjusted outperformance, at least over the last year.


We often get accused of not leaving room for active management in our analytics methodology. Well, this is how you make room for active management. You actually have to deliver on a better mousetrap.

CSM is holding very tight to its promise of beta 1, and beating the market. It does this all while keeping surprisingly tight sector exposure too. Note that our segment benchmark here isn’t the S&P 500; rather, it’s the 281-stock MSCI USA Large Cap Index. Measured against the S&P 500, these sector skews are even less dramatic:


Find your next ETF