Last September, we noted that hedge funds in South Africa and several other off-the-beaten-path places seemed to be holding up okay through the August storm. Now with gold prices flirting with all time highs, South African managers (hedge and long-only) seem to be attracting a lot of renewed interest. The April edition of Institutional Investor magazine shines a light on South Africa. Next week, Terrapinn will be hosting an “Alpha Beta Summit” in Cape Town. And last month HedgeWeek published a special report on the country’s hedge fund industry. HedgeWeek observed in an article published alongside the report that since March 2004, the South African hedge fund index had grown by nearly 20% per annum (vs. 12% for the MSCI World).
But is it really alpha? To address this question for us, we welcome the following guest contribution from Helena Conradie of major South African money manager Sanlam Investment Management. Helena is the Head of Sanlam Investment Management’s equity quant boutique that manages over R21 billion. She is a CFA charterholder and has an MSc in Applied Mathematics Cum Laude from Stellenbosch University.
Special to AllAboutAlpha.com by: Helena Conradie, SIM Equity Quants
In just more than 18 months people all over the world will flock to South Africa to attend the world cup soccer event, paying generously to see amazing flair and display of talent. But would they consider South Africa as the location for amazing alpha?
At any given time there is a finite amount of alpha available for fund managers to hunt. And as we all know, it is “all about alpha!” The diversity of stock returns across all sectors (the cross-sectional volatility) is a good indication of the presence of alpha. So does the South African rainbow provide the alpha hunter with enough diversity?
The above chart illustrates the difference in cross-sectional volatilities for several developed countries, emerging markets and South Africa in particular. At first glance we can see that, globally, we are in a historically low alpha environment. It has become increasingly difficult to find those rare sources of alpha amongst stocks that currently show such herd behaviour.
But on closer examination we see that, as expected, emerging markets provide the investor with greater alpha opportunities across time than in the developed world. South Africa, a subset of emerging markets, provides less diversity than the emerging markets as a whole, but nevertheless more than developed markets.
As all managers will know, desperately hunting down the prey is only the first step taken in the alpha racethe next challenge lies in the efficient capturing of this sought-after alpha in the portfolio. Add a set of handcuffs to the hunt and you have a real challenge! Sector limits, strategic holdings, limits on active holdings are all constraints that hamper the fund manager’s ability to transfer his ultimate view to the portfolio. However, the handicap of the long-only constraint outweighs them all.
To illustrate the significance of the long-only constraint in the South-African environment, let’s construct three portfolios each with different active risk targets. Assume the manager of these three funds has a positive view on Anglos (AGL), Standard Bank (SBK) and Amplats (AMS), some of our largest stocks, and a negative view on Imperial (IPL) and Exxoro (EXX). This view should result in a positive active bet on AGL, SBK and AMS and a negative active bet on IPL and EXX. Right? Not necessarily!
For a low risk fund of 0.5% active risk such as the enhanced fund above, the manager is clearly not hand-cuffed at all. She can comfortably express her (low active risk) view in the portfolio and, as a result, 100% of her view is transferred to the fund.
However, when the same manager is required to take on more active risk, she runs into the detrimental effect of the long-only constraint. At an active risk of 5% such as the fund below, only 61% of her view can be replicated in the portfolio. Furthermore, the requirement for more aggression can produce an unintended reversal of views, in this case SBK, driven purely by the need to finance the increase in risk.
At 7.5% active risk (below) the handcuffs are well and truly on! Less than half of the manager’s investment view can be translated into the portfolio. Worse still, only seven out of the forty-stock universe can accommodate such aggression (marked with yellow stars below).
Like the manager above, the African alpha hunter needs to be “unleashed”. African alpha exists, but managers in this investment universe are often hand-cuffed by mandate constraints that prevent them from capturing alpha efficiently. Following the global trend, South Africa is turning to the flexibility of short-extension products as a possible solution to a hamstrung alpha hunter! The first official institutional 130/30 fund is about to be launched (by Sanlam Investment Management) and others will follow soon, no doubt.
What does it mean for the long-only breed of alpha hunters? The improved efficiency of a 1X0/X0 approach to portfolio construction can’t be ignored (see figure 4)! The transfer coefficient on the 5% tracking error can be improved from 60% (long-only) to 75% (110/10), to 85% (120/20) or to as high as 90% (130/30). No matter how much alpha the hunter can find, fund constraints, particularly at aggressive risk levels can get in the way.
Despite these obstacles, the chart above illustrates the relatively high efficiency achieved by managing enhanced index funds (active risk up to 2.5%) in the long-only space. This fund manager can take comfort in the knowledge that at least 80% of her view is transferred to the portfolio.
Still, there is therefore a strong case for a barbell investment approach in the traditional long-only space â€“ enhanced index at the lower levels of risk and 1X0/X0 at the more aggressive end of the spectrum – particularly in South Africa where markets aren’t quite as informationally efficient as in other parts of the world.
– H. Conradie, April 30, 2008
The opinions expressed in this guest posting are those of the author and not necessarily those of AllAboutAlpha.com.
Editor’s Note: This 2005 academic study concludes that active management does indeed outperform passive management in South Africa – but that, as usual, fees can eat up all of this outperformance. So the critical question raised by Conradie’s piece above is therefore “Does the 1X0/X0 strategy leverage alpha enough to overcome those fees?”