The most dramatic change in investment management in the 1990s was the application of financial engineering to enhance portfolio return and control risk. The derivatives revolution of the prior decade provided the tools. The obsession with index benchmarks created the demand. Now, as we start a new decade, a style known as portable alpha increasingly will be seen as the most prominent expression of this trend.
The basic principle behind portable alpha is deceptively simple. The tradition part is easy. Identify a way to add value against some standard benchmark; that is, create alpha. Second, transport that alpha onto the target benchmark by using a swap contract or a combination of futures. That's technology. The net effect is the target index return enhanced by the investment manager's ability to create alpha in any market.
An example will show the benefits and suggest the risks of this approach. Start with a $100 million investment strategy designed to beat the one-month London interbank offered rate with a modest degree of risk. If LIBOR is running at 5% annually, the strategy's target return might be 6%, or an average alpha of 1 percentage point a year.
The next step is to transport the alpha onto a target index, let's say the Standard & Poor's 500 stock index. To do this, one might enter into a $100 million notional value "total return" swap. This is where the investment manager agrees to pay LIBOR on $100 million notional value to the swap counterparty in return for receiving the capital appreciation and dividends on a hypothetical $100 million S&P 500 index portfolio. (An equivalent way to transport the alpha is by using a combination of short eurodollar and long S&P 500 futures, but the specific details of this trade add little to understanding the basic approach.)
Looking forward one year, suppose the investment strategy earned 6.5%, while LIBOR averaged 5% and the S&P 500 increased 15%, including dividends. In this example, the alpha generated was 1.5 percentage points. The investment manager owes the swap counterparty $5 million (5% LIBOR on $100 million) but the swap dealer owes the manager $15 million for the S&P 500 return. These sums are netted and the dealer would pay $10 million net to fulfill the obligations of the swap.
The total outcome for the investment is the sum of the returns. The manager earned $6.5 million on the original investment and received $10 million from the swap dealer, for a total return of $16.5 million, or 16.5%. Note this just equals the S&P 500 total return plus the realized alpha.
What happens if the S&P declines? Starting with the same original investment and swap, suppose the total return of the S&P 500 had been -10%. The investment manager still would owe $5 million for the LIBOR payment, but now the swap counterparty would "owe" a negative $10 million, which means the investment manager pays again. The total net transfer is $15 million from the manager to the swap dealer.
The bottom line on this scenario is the manager earned $6.5 million on the original investment while paying out $15 million on the swap. The investment in total lost $8.5 million, or -8.5%, which once again is 150 basis points better than the S&P 500. In fact, under all market scenarios, the final outcome is the target benchmark return plus the alpha.
This principle can be used quite broadly. Alpha can be generated in many different areas and transported onto virtually any index. The limiting factor is the availability of derivatives to carry out the alpha transfer.
If you haven't yet asked what can go wrong with all this, now is the time to do so. The above examples were somewhat stylized to stress the basic theory. There are a few practical wrinkles to consider.
First there are costs. Derivatives transactions are not free and the number of basis points negotiated with the swap dealer will have the effect of reducing the alpha. The good news here is that while not free, a simple derivatives transaction like that described above is fairly cheap and doesn't affect returns too much. If, however, you are interested in transporting a target alpha onto a relatively obscure index (e.g. a small-cap value index) be prepared to pay more.
Take reasonable risk
Second, and most important, is the ability to generate alpha with reasonable risks. Some people might desire a modest sized net alpha and expect it to be present continuously. Others might be willing to accept greater volatility in exchange for a higher target alpha. In the examples above, if the original investment had failed to earn the 5% LIBOR rate after costs, the alpha would have been negative and the total return would have fallen short of the S&P 500.
Third, an alpha shift strategy involves derivatives and counterparty risk if those derivatives are over the counter. The good news here is that these risks are no more complicated or severe than those assumed when implementing a foreign currency hedging program. Implementing a portable alpha strategy requires one simple decision to begin. Do I build or buy? A fund with great skill in adding value in its fixed-income portfolio might be tempted to shift that alpha onto a domestic equity portfolio by using futures or a swap directly. It is important, however, that the appropriate skills and risk controls are in place before proceeding. The 1990s are littered with sad examples of corporations and plan sponsors that acted without adequate preparation or controls to unfortunate results.
What to look for
For most institutional investors buying a portable alpha product is probably preferable to building. In shopping around, one should be alert to important differences across products. Do the available products target the right benchmark? Today the S&P 500 dominates the landscape. In the next few years, more choices no doubt will be available.
What is the source of the alpha? The largest existing funds find alpha in the fixed-income market and shift it to equities. These funds tend to be more challenged to produce a positive alpha when interest rates rise. Commonfund's portable alpha product, Index Plus, is structured to generate an alpha against LIBOR using relative low risk absolute return strategies. If successful, there should be low correlation between its alpha and movements in either stocks or bonds.
Finally, be prepared for periods of underperformance. Alpha generation involves active strategies and invariably there will be such times. On average, however, sound portable alpha strategies should add value net of costs.
Whatever route one takes, it is important to understand and accept the goals and risks of the alpha engine.
Portable alpha strategies employ the best financial engineering tools available to investment managers to shape returns and control risk. Today the applications are mostly limited to enhancing returns of major U.S. indexes like the S&P 500. Tomorrow there will be a wide range of options to investors, and what is now novel will seem quite routine. For those investors who want to control their benchmark risk, these strategies are among the best opportunities to beat the street.