Using portable alpha to enhance a passive strategy

“Everyone likes a winner”, says Guy Fletcher, Head of Research and Client Solutions, Sanlam Investments

The debate between active and passive investing has long been topical. The passive premise is built on the benefits of lower management fees and, very often, associated lower turnover costs. Actual excess returns through active stock selection has proven to be difficult to sustain and the vast majority of active managers, locally and internationally – particularly if one takes survivorship bias into account – underperform typical benchmarks.

So, ‘go passive’ case closed?

The flipside of this argument is human emotion. Most investors dislike being consigned to the (albeit marginal) underperformance scrapheap when stories of excellence abound. Everyone likes a winner. And being associated with one (through your investment strategy) is next best to being one, even though it’s hard to find and often cyclical. So, you have to have some active in the mix, right?

But what about a third option? (and we’re not talking smart beta here – that’s for another chapter).

Instead of simply replicating a given benchmark as closely as possible, derivatives give us an added dimension to enhance returns, without using leverage. This is with the addition of portable alpha.

Let’s explain

Portable alpha involves seeking the market return (commonly referred to as beta) from one asset class and excess return (commonly referred to as alpha) from a different one, and then merging the two. And the best form of portable alpha is the one that is completely uncorrelated to the beta we want to enhance.

The one area of investing in which alpha has persisted is in the money market spectrum. If one observes the performance surveys over time, almost every money market manager has delivered in excess of their benchmark, implying that alpha is not in elusive supply in this part of the market (unlike the equity market). In comparison to returns on short-term deposits, the money market funds deliver returns some 70bp to 150bp higher. And the enhanced yield funds can be up to double this! To put this in perspective, if you were able to consistently achieve these sorts of returns above an equity benchmark after costs, you would be a very happy investor.

So, how do the money market guys do it? Well, they have the term structure of interest rates on their side (you get a higher yield for longer maturity). They use their understanding of interest rate cycles to good effect by choosing fixed vs floating rate notes, plus they can put aside some of their investments in higher-yielding credit instruments (under stringent controls), all of which adds up to better returns.

So how does the “porting” process work?

First, we replicate as much of the benchmark as possible with derivatives. In the case of a SWIX Top 40 index, there are futures contracts that are designed to expire at EXACTLY the same level as the index on specific dates. Prior to that date, the future will trade at a value that is different from the index. This is because, as a contract, it has different attributes to a normal investment. In other words:

• It requires only approximately 10% collateral against the exposure to the market that it purchases – this money must be placed with SAFEX (South African Futures Exchange) as a buffer against severe market movements and will earn a typical short-term deposit rate
• It must make good on the daily profit/loss of the contract (called margining) – if the contract owner is unable to do this, SAFEX will close out the position
• It receives no dividends (since it is not actually invested in the market, merely in a contract that derives its value form the market)
• The balance of the monies not required for collateral are assumed to be invested in cash

So the value of the contract is a function of four elements, namely:

• The current spot price of the underlying index (S)
• The expected future amount of dividends (d)
• The expected cash return (r)
• The term (in days) to expiry of the contract (t)

This can be written as:

Fair value = S * (1-d+r) ^t/365

Portable alpha essentially maximises the return for the third element (cash) – as we have seen above, money market managers pretty much all add positive value.

Diagrammatically, the process can be represented as follows

Most individuals will understand that markets can move aggressively (think of market crashes in 1969, 1987, 1998, 2002 and 2008) so one has to be diligent in managing these positions, particularly having additional capital available for extreme events. However, even with 20% put aside on an immediate call in addition to the initial margin at SAFEX (we have never had a negative market move of >15% in one or two days), we can still invest some 70% in enhancing activities.

Since we will contractually receive the exact return on the index, the additional return over the expected short-term deposit that we will receive from enhancing the cash is all “alpha” and is portable (moveable) onto the index return!

Let’s have a look at some charts (for illustrative purposes, we are only looking at the last 42 months):

1. SIM’s Active Income fund – the portable alpha engine of choiceThe chart above indicates the following information:
   1. SIM’s Active Income has delivered an average excess return over cash (represented by STFIND) of 151bp p.a. after costs since the end of 2013
   2. Best rolling 12 months has been 329bp, worst has been -10bp.
   3. Porting 70% of this onto a SWIX 40 index, implies that our best return would have been SWIX40 +230bp, and our worst would be SWIX40 -7bp.
2. Adding value to a pure index investmentThe cumulative impact of this portable alpha strategy is profound:
   1. a pure passive approach (using ETFs) would have delivered index -217bp since December 2013 (an implementation and management fee cost of some 0.62% p.a.)
   2. the portable alpha strategy indicated above would have delivered index +459bp (1.29% p.a.)
   3. this is a swing of 676bp or 1.88% p.a.
3. Positioning an index plus portable alpha strategy against active managers
   1. Most active managers use the larger SWIX index as their benchmark. Thus, only approximately 80% of this index is represented by the SWIX40 (or large cap component) analysed above.
   2. For ease of reference, we assume that the “tail” (or non-large cap element) can be replicated at a cost of 62bp p.a. (as per the ETF experience above)
   3. Hence the added value of the portable alpha element to the full SWIX will be closer to 110bp (80% * 151bp + 20% *-62bp).
   4. Below is a scatter plot indicating the results of all active managers within institutional surveys in the 3 years to the end of Apr 2017 (the middle of our period). The scatter plot uses the full SWIX benchmark and shows the annualised excess return of that manager vs his tracking error (risk). All managers are estimated to have a 45 bp p.a. management fee.
   5. The cross-hairs on the Passive plus Portable Alpha portfolio show that, during this period, only five funds delivered a higher return than our strategy (fully 43 delivered lower returns) and everyone did it with higher risk.

Conclusion

The concept of portable alpha has been around for decades, and is a highly effective strategy in delivering excess returns over benchmarks. It is particularly appropriate in higher yielding environments (such as South Africa) where money market instruments deliver vastly improved returns over cash on deposits at commensurately lower risk.