Managing assets via "alpha": an expert in asset management explores a special technique—"Alpha Transfer"—that amounts to a form of telecommuting for excess returns. Here's how it works

Risk & Insurance, June, 2002 by Andrew P. Hofer

I work in New York City, but live in the suburbs. New York City's job market offers tremendous opportunities for people with financial and managerial skills, so I am pursuing my career here. On the other hand, I have young children, and the schools and environment in my hometown of Princeton, N.J., are perfect for them. The result is I spend more than 15 hours a week commuting.

As investment managers, we often face similar dilemmas. Our objective is to identify mispriced securities, but we often find these pricing anomalies in parts of the market we cannot conveniently access or within other markets entirely. What if we could "telecommute" to those other markets and bring the excess return ("alpha" in our jargon) back to our own part of the market without taking excessive risk? It is entirely possible, and money managers are increasingly doing it today. We call it "Portable Alpha" or "Alpha Transfer."

One of the reasons the S&P 500 index has been a difficult benchmark to beat (at least with a long portfolio of S&P 500 stocks) is that it is no small task to have an information advantage in this market. Large capitalization U.S. stocks are the most widely researched and scrutinized in the world. Furthermore, these companies manage their information disclosure carefully (for regulatory as well as management reasons), so all analysts are generally working with the same information. It is, therefore, extremely hard to generate fresh insights on individual stocks in the S&P 500. It is a very efficient market in that respect.

From a pure asset class perspective, however, the S&P 500 is a highly desirable asset. Large capitalization U.S. equities have an enviable track record of providing both high real returns and deep liquidity.

Consistent excess returns and attractive nominal returns do not necessarily come from the same place. Chart one on the top right of page 56 compares annualized returns for various asset classes over the last 10 years with the ability of the median manager to generate consistent excess returns on that asset class, as measured by the information ratio.

Since the information ratio indicates the level of excess return divided by the volatility incurred relative to the benchmark, it is an indicator of the potential persistency of alpha generation. As you can see, information ratios are higher for fixed income categories than equity categories and are particularly high for short maturity, high-quality credit exposures.

Wouldn't it be nice if the average S&P 500 manager could outperform their benchmark as consistently as the average fixed income liquidity manager? The idea behind alpha transfer is just that; to graft the highest consistency of alpha to the asset class that most closely matches the investor's desired risk and return profile.

How Is It Done?

The simplest example of alpha transfer is equitization. There is a very deep and liquid market in futures contracts on the S&P 500. Using 10 percent to 20 percent of the portfolio as margin, the investor can establish a synthetic long position in the index. The remaining 80 percent or so of portfolio assets can then be used to generate alpha in another asset class. If that alpha can be generated in a market neutral fashion (such as with equal-weighted long and short positions), alpha transfer is complete. The portfolio will receive at least 80 percent of the alpha of a fully invested portfolio plus the index return on the S&P 500.

This "smart bonds/dumb stocks" technique underlies various "enhanced equity index" products. Enhanced equity index managers take advantage of the fact that that S&P 500 futures contracts are constantly arbitraged to a level where the implied return on the underlying funds is about LIBOR (London Interbank Offered Rate, the interest rate at which London market banks borrow from other banks). In this case, the manager takes the synthetic stock position and attempts to exceed LIBOR with the investable portfolio. The margin by which the manager exceeds LIBOR can be exceeded translates to the excess return of the fund over the S&P 500. These strategies have consistently beaten the S&P 500 by more than 0.5 percent per annum over many reporting periods, placing them among the top large capitalization equity funds. This process of linking a non-equity asset class alpha to the equity markets is called "equitization." Many hedge fund managers already offer "equitized" and "market neutral" versions of their products.

Total return swaps can also deliver a synthetic position in another asset class. For instance, after establishing a portfolio in the high-alpha asset class, the investor can go to the over-the-counter market and swap the return stream for the alpha class (represented by an index) for that of a more desirable asset class (also represented by an index or basket of securities). The drawbacks of this technique are that swap rates between asset classes move around substantially based on a number of factors, including market expectations, liquidity, dealer inventory hedging requirements and other demand and supply anomalies. For instance, during 2001, we saw indicated prices on a total return swap on the high yield bond market fluctuate by up to 300 basis points. While swaps are therefore an unreliable platform on which to build an alpha transfer product, managers can use them opportunistically to improve returns in the alpha transfer process.