Recent changes to accounting standards, such as the introduction of FRS 17 in the UK, have helped develop the mindset in investing and product structuring known as liability-driven investment (LDI), where a fund is managed with respect to its true liabilities, or proxies of them, rather than against a peer average or technical market benchmark.

Meanwhile, the downward trend in the real yield of long-maturity bonds had led to a situation in which many pension funds find themselves severely under-funded. Fund managers are increasingly being forced to seek excess returns while operating within tightened risk budgets.

These two concerns, the need for higher returns and the more careful monitoring of risk have whetted the appetite for more innovative investment ideas. The key to maximising returns while constrained by a small budget of risk lies in identifying a large number of uncorrelated investment strategies. However, given the difficulty of identifying low-risk diversifying investments under current market conditions, the advent of a relatively new class of strategies, known as portable alpha, is timely.

The fundamental law of active management

The most commonly used measure of a portfolio manager’s performance is the information ratio. Grinold’s ‘Fundamental Law of Active Management’ states that the information ratio can be decomposed into two factors, which he describes as the information coefficient (basically a measure of the skill of the manager) and the breadth of the portfolio (essentially, the number of uncorrelated strategies pursued by the manager).Taking the manager’s skill to be constant, risk-adjusted returns increase with the number of independent bets taken by the manager. Thomas (‘Engineering an Alpha Engine’) examines the relationship between information ratio and breadth, demonstrating, under certain assumptions, how an increase in the number of strategies from 1 to 20 can change a portfolio manager’s information ratio from a lowly 0.2 to a stellar 0.9. Thus, the need for diversification is paramount.

Traditionally, managers have used asset classes such as hedge funds, private equity, commodities and real estate to add diversification to their portfolios. For example, hedge funds were formerly viewed as pure alpha strategies, leveraging on manager skill to achieve high returns uncorrelated to market performance. There is evidence, however, that as the hedge fund sector has become increasingly heavily populated, the opportunities for non-directional returns have partially disappeared. Consequently, hedge fund performance in recent years is more and more driven by beta and their diversifying power is shrinking.

Furthermore, other classes are perceived to be particularly risky investments at the moment. In a climate of rising interest rates worldwide there are indications that real estate prices are approaching a downturn. On the other hand there is uncertainty over whether commodity prices can sustain the growth levels of recent years.

The need for new uncorrelated strategies is undeniable. The greater proliferation and understanding of financial derivatives has opened up new opportunities for returns that are a function of the manager’s skill or access to information. For example, credit default swaps can be used to implement simple non-directional strategies. This can be achieved by selling protection on a number of “safe” corporate names (using available credit research) while simultaneously buying protection on a number of the most risky ones. The deal can be constructed to be duration neutral and spread duration neutral. In this way, the manager has arrived at a strategy that is immune to general market movements and is dependent only on the quality of available research.

Portable alpha as an independent bet

Portable alpha strategies are similar in principle to the CDS strategy described above in that they use derivatives to remove the beta component from a deal designed to isolate a single aspect of the investment process – in this case the manager skill.

The strategy involves the use of swaps to separate the alpha and beta components of a portfolio, removing the asset manager’s performance from its benchmark and “porting” it onto another benchmark. For example, by combining a long position in an actively managed government bond fund with a short position in a future on the fund’s benchmark we can construct a portfolio consisting only of the active return of the fund. The alpha can now be treated as a brand new asset class, suitable for inclusion in, for example, a pension fund portfolio.

The key to understanding the success of this asset class lies in the insight that the alpha component of a portfolio’s performance is, by definition, uncorrelated with the performance of the benchmark.

For example, if, for illustrative purposes, we look at the performance of an actively managed US equity fund with a benchmark of the S&P 500, over a period from December 2004 to September 2006, figure one (p32) of the fund (red line) against its benchmark (green line) shows that while the fund is highly correlated with the index, it also consistently outperforms the index thanks to the managers stock picking ability. The third line represents a combined long position in the fund with a short position in a future on the S&P 500, which is a commonly traded and highly liquid derivative.

Figure two charts the correlation of the fund with the index and the alpha with the index. While the correlation of the fund to the index is consistently close to 1.0, the “portable alpha” asset has a very low absolute correlation to the index, with an average close to -0.2.

Similar strategies may be implemented using, for example, a total return type portfolio or indeed any portfolio whose benchmark can be replicated using the array of liquid swaps currently traded, including interest rate, inflation and currency swaps. Indeed, a portfolio consisting of a basket of alpha strategies can be created, adding a multimanagement layer to the process. The figure three reflects the performance of a portfolio of five top performing US equity funds tracking the S&P 500, and of the alpha component of the portfolio. Again we can see that the resultant alpha strategy is uncorrelated with the S&P index. This type of strategy can be used to leverage on fund picking, as opposed to stock picking skills.

Since the search for alpha can be thought of as a zero-sum game – for every active manager who achieves returns in excess of the market average, there must be another who fails to beat their benchmark – a fund manager must be free to seek extra returns in any category where they are perceived to have an advantage, whether due to superior stock-picking (or fund-picking) skills or superior research.

In an LDI context, this means the manager is not restricted to seeking extra returns amongst those asset classes that most closely match a fund’s liabilities. For example, the returns from a total return or Libor+ strategy may be ported onto a portfolio designed to replicate a stream of pension fund liabilities, creating a solution that is immunised from interest rate and inflation risk, but still capable of targeting the extra return needed to amortise a deficit.

The growth of LDI coupled with the vast array of OTC derivatives now commonly traded has resulted in a much broader universe of solutions available to pension fund consultants. Using portable alpha techniques, pension funds are free to invest in any type of vehicle in any region, using swap and futures to remove alpha from its benchmark allowing consultants to leverage on the fund picking and multi-management skills of asset managers.

In association with Pioneer Investments.

_{Pioneer Investments is a trading name of the PGAM group of companies (the “Group”). The Group has over €217bn of total assets under management as at end of September 2006. We provide a wide range of investment solutions including mutual funds, alternative investments and structured products to clients that include institutions, corporations, intermediaries and private investors around the world. Institutional clients account for over €35bn.}

Printer friendly version