By Corey Hoffstein, Newfound Research
The Importance of Long/Short Portfolios
Anybody who has read our commentaries for some time has likely found that we have a strong preference for simple models. Newfound’s Justin Sibears, for example, has a knack for turning just about everything into a conversation about coin flips and their associated probabilities. I, on the other hand, tend to lean towards more hand-waving, philosophical arguments (e.g. The Frustrating Law of Active Management[1] or that every strategy is comprised of a systematic and an idiosyncratic component[2]).
While not necessarily 100% accurate, the power of simplifying mental models is that it allows us to explore concepts to their – sometimes absurd – logical conclusion.
One such model that we use frequently is that the difference between any two portfolios can be expressed as a dollar-neutral long/short portfolio. For us, it’s long/short portfolios all the way down.
This may sound like philosophical gibberish, but let’s consider a simple example.
You currently hold Portfolio A, which is 100% invested in the S&P 500 Index. You are thinking about taking that money and investing it entirely into Portfolio B, which is 100% invested in the Barclay’s U.S. Aggregate Bond Index. How can you think through the implications of such a change?
One way of thinking through such changes is that recognizing that there is some transformation that takes us from Portfolio A to portfolio B, i.e. Portfolio A + X = Portfolio B.
We can simply solve for X by taking the difference between Portfolio B and Portfolio A. In this case, that difference would be a portfolio that is 100% long the Barclay’s U.S. Aggregate Bond Index and 100% short the S&P 500 Index.
Thus, instead of saying, “we’re going to hold Portfolio B,” we can simply say, “we’re going to continue to hold Portfolio A, but now overlay this dollar-neutral long/short portfolio.”
This may seem like an unnecessary complication at first, until we realize that any differences between Portfolio A and B are entirely captured by X. Focusing exclusively on the properties of X allows us to isolate and explore the impact of these changes on our portfolio and allows us to generalize to cases where we hold allocation to X that are different than 100%.
Re-Thinking Fees with Long/Short Portfolios
Perhaps most relevant, today, is the use of this framework in the context of fees.
To explore, let’s consider the topic in the form of an example. The iShares S&P 500 Value ETF (IVE) costs 0.18%, while the iShares S&P 500 ETF (IVV) is offered at 0.04%. Is it worth paying that extra 0.14%?
Or, put another way, does IVE stand a chance to make up the fee gap?
Using the long/short framework, one way of thinking about IVE is that IVE = IVV + X, where X is the long/short portfolio of active bets.
But are those active bets worth an extra 0.14%?
First, we have to ask, “how much of the 0.18% fee is actually going towards IVV and how much is going towards X?” We can answer this by using a concept called active share, which explicitly measures how much of IVE is made up of IVV and how much it is made up of X.
Active share can be easily explained with an example.[3] Consider having a portfolio that is 50% stocks and 50% bonds, and you want to transition it to a portfolio that is 60% stocks and 40% bonds.
In essence, your second portfolio is equal to your first plus a portfolio that is 10% long stocks and 10% short bonds. Or, equivalently, we can think of the second portfolio as equal to the first plus a 10% position in a portfolio that is 100% long stocks and 100% short bonds.
Through this second lens, that 10% number is our active share.
Returning to our main example, IVE has a reported active share of 42% against the S&P 500[4].