Before continuing, you need to have a solid grasp of alpha and beta – otherwise the rest of this article will make no sense.
Back in part three, we developed a method for trading the S&P 500 for use in the long term speculative account. In other words, we’ve figured out how to trade beta. That’s good, but one of the goals of this speculation method is to limit beta exposure to +- 20% of the account value. The purpose of this limit is to mitigate account damage in unexpected crash situations (when long) or boom situations (when short). Since I’ve allocated 20% of the portfolio to beta trend following (which could thus produce betas between +-20%), ideally everything else in the portfolio should have a beta of zero. That’s a difficult requirement for stocks, however, because they essentially all have a positive beta, and in most cases that beta constitutes at least half of the stock’s movement. What we need is something that acts like a stock, but with no beta component.
The solution is something called a synthetic instrument or spread. You’re in essence creating a new financial instrument by combining two (or more) existing ones in a specific way. I’m being a bit vague or general here because there are numerous ways to construct a synthetic instrument – it all depends on what you want to do. In this case what we want to do is create a beta-neutral stock spread – in other words, just the alpha component of a stock.
I already gave an example of how to do this in the alpha and beta article. You simple compute the regression slope coefficient, or look it up on a website like google finance. It will be labeled just “Beta”. Let’s use Intel stock (INTC) as an example. Google gives Intel’s beta slope as 1.08. So to create the beta neutral spread, we would do this: for every $1000 of INTC we had, we would short $1080 of SPY (a S&P500 index fund). The result would be the alpha component of the INTC, with the beta subtracted out. Alternately, if you wanted to short INTC would would have to buy corresponding amounts of SPY to cancel the beta.
What we’ve just done is something called hedging. When you buy a financial instrument, you’re generally exposing yourself to multiple risk/reward influences simultaneously. In this case, you are exposed both to Intel’s performance as a company and to the broader market. Sometimes you want that. But sometimes only want exposure to one of those influences. Hedging is the process of adding other securities to your portfolio that as best as possible cancel out the unwanted influences, leaving just what you want. In this example we want Intel’s performance as a company, but not the broader market. So we buy INTC and short the broader market. The ratio in dollars between the two is the beta coefficient.
This hedging process, in a slightly different form, is what gave “hedge funds” their name. The original concept was that a hedge fund would be long some stocks and short others, thus hedging out most of the beta. They weren’t as precise about achieving zero beta as I’m being because, back in 1950 or so when the term was coined, instruments like SPY or ES that you would use to do the hedge didn’t exist. So they just paired a long stock position with a short stock position. The idea was the same, we’re just higher tech now. Eventually the concept of a hedge fund drifted and became any limited partnership trading complex positions. These days most hedge funds are not notably beta hedged. But the original hedging idea had a lot of merit, especially in uncertain markets, and is an important inspiration for this portfolio.
So we’ve created this synthetic instrument based on INTC stock. It should be obvious that we could do the same to essentially any stock, and get an entire population of stock-like spreads that are independent of the S&P average. This is a good thing, because that’s exactly the sort of instrument we want for the speculative alternatives portfolio. The problem now is one of narrowing our scope – since we can create a beta neutral spread out of any stock, which ones do we use? And how to we profitably trade those synthetic instruments? That I’m going to have to delay for next week, when I promise we’ll finally figure out the majority of the positions in the portfolio no matter how long an article it takes.
Continued in part 5