Alpha in Asian long short funds
Looking back over the years from 2007 to 2010 is a good “stress test” to see how Asian long short hedge funds managed during the last year of the bull market, the correction and subsequent recovery. We undertook this review on most of the region’s larger managers. We faced a few challenges measuring alpha given differing styles and country mixes. It is hard also to avoid a survival bias given a number of managers failed to survive till the end of 2010.
Our review used simple regression analysis to look at each respective fund manager’s beta to the relevant indices to get a measure of alpha. So essentially the Alpha = Manager Return - [(beta * Market Return) + ((1-beta) * risk free rate)]
We are aware that this has its limitation so as well as incorporating an alpha measure based on a static beta we also looked to assess the manager’s alpha using a 6month trailing beta and an alpha based on the actual reported net exposure.
The two shorter term measures of alpha using 6month trailing beta and actual reported net exposure were used to try and assess if managers were adding alpha by adjusting their net exposure. Our hypothesis was that if the alpha using the shorter term measures of beta were lower than that achieved using the static beta than we could assume the manager was adding alpha by adjusting net exposure. That is if the manager correctly captures market moves by adjusting net more of the return would be apportioned to the beta return and so the alpha should be lower than that achieved using a static beta. The actual reported net proved to be the least stable of these measures and does not provide that stable data.
Our calculation of alpha is after fees, which in the case of hedge funds, is the combination of management fees, performance fees and some direct fund costs. We have also tried to look at total return benchmarks were possible, though the fund will suffer some loss of withholding tax on dividends.
Below is a quick snapshot of the aggregated fund performance looking back over the last four years of manager performance (2007-2010).
|Ave. Static Beta||Ave. Alpha using static Beta||Ave. Alpha using Ave. 6m Trailing Beta|
|Asia Pacific Long/Short|
|MSCI AC TR Asia Pacific (AP)|
|MSCI AC TR Asia Pac ex-Japan (APxJ)|
|MSCI AC TR Japan USD|
|MSCI AC TR Japan Local|
|MSCI AC TR China Local|
|CSI 300 Index|
|MSCI TR India USD|
|MSCI TR India Local|
|MSCI TR Australia USD|
|MSCI TR Australia Local|
The table above shows the average alpha using a static beta and 6month rolling beta for the Asian Long/Short, Japan Long/Short, China Long/Short, India Long/Short and Australia Long/Short funds. This displays the alpha of the funds by geographic remit and dependant on which index we measured against.
Overall with the exception of India there is reasonable positive alpha despite the very challenging environment. There is also some indication, especially with the Pan Asian managers, that some value was added though de-risking and re-risking.
The Pan Asia Long/Short alpha has been a bit disappointing however some of the variable long bias managers came out looking strong. These tend to be single star manager funds with limited capacity. They tend to be tactical in running to cash and the alpha driver is limiting the negative returns that damage long term compounded returns. A handful of the more stable low net long bias managers also looked to have produced steady levels of alpha, whilst some of the more aggressive funds have struggled.
Pan Asian Long/Short also presents a challenge as many managers have Japan in their mandate, yet few fully weight it. Therefore, a source of alpha has been avoiding Japan. This shows as a higher alpha on the broad pan Asian index and a lower alpha if Japan is excluded.
As second problem, particularly in 2010 is the effect of currency and markets with limited shorts. The total return of Asia ex. Japan in USD was nearly 20% in 2010, yet the average Pan Asian Long/Short managers returned around 4.5%. Based on their average beta this looks like negative alpha. However, over half the return came from foreign exchange and the performance of the less “hedge-able” markets of Asean and India. Most managers try and protect an absolute dollar return and hedge foreign exchange. The markets with the best stock hedges were little changed.
The most notable observation from the table above is that the China and Australia managers came out looking like they delivered the most alpha. The most disappointing alpha appears to have been generated by the India Long/Short funds.
We continue to wonder why the China hedge fund alpha is so much higher than that witnessed in the broader Pan Asian funds or other single country funds. We do question if this is a function of the measurement of alpha and if the alpha witnessed to date is sustainable.
There is an argument that in China detailed on the ground research will be rewarded. We are in an environment that any visible sustainable growth stories tend to be fully valued fairly quickly. The Chinese consumer story is also well understood. Therefore, waiting for confirmation of a thesis in the financial data is often too late. There is real value in monitoring what is selling, pricing trends and changing business models. There are also serious risks in investing in China based only on a top down thesis. Corporate governance standards are highly variable and stocks really need to be thoroughly vetted. As a result on the ground research can add considerable value on both the long and short side.
Many China managers avoid the sectors which are subject to high government policy risk. This tends to keep them out of the financial, telecom, utility, energy and real estate sectors which account for the majority of the market capitalisation. Most have a strong bias to consumer, services, industrial, healthcare and technology. This has certainly been a very favourable bias over the last few years and a good source of alpha.
It was noticeable that the China funds managed by individuals who have only ever worked domestically in China have outperformed their colleagues with international experience. This has in part contributed to the out-performance of the China Long/Short funds versus peers. Versus the MSCI China index the locally trained fund managers generated alpha of +20.3% versus +9.7%for internationally trained fund managers and when we looked at the Alpha versus the CSI 300 index the differential is +20.4% compared to +10%.
It should be noted that despite the alpha generated by the solely domestic trained managers we have found it hard to establish why they have an edge. They tend to be from more trading backgrounds, though most now have significant fundamental research teams. Arguably they may have an edge in interpreting policy and may have better sources of market intelligence.
The lack of alpha in India was very poor. Clearly fund managers felt the market was less exposed in 2008 to the global issues which subsequently proved wrong. They also had to contend with changes in regulation with regard to shorting, with the Indian regulator trying to phase out P-notes. Hedging the currency also became prohibitively expensive due to the behaviour of the non deliverable forwards market.
Part of the issue is the high beta which effectively means performance fees were paid for the market return on 2007. Although the market regained the end 2007 levels by end 2010, most hedge fund managers did not. Most of the longer biased funds capitulated and missed some of the bounce. Only a few managers had a consistently executed hedging strategy throughout the period. This still raises a troubling question of why Indian alpha is so poor. Part may also be explained by the sensitivity of the market to foreign flows. This creates large slippage when foreigners enter or exit en masse.
It still remains highly questionable whether investors should bother with Indian hedge funds given the poor or negative levels of alpha witnessed historically. The situation has improved a bit with increasing activity in their single stock futures market and some more managers are trying to embrace a more tightly managed net exposure. Currently we see very little demand for managers in the space.
The Australian Long/Short managers appear to have added good alpha over the last four years however the biggest caveat is that many of these funds are probably not scalable and it would be interesting to see how sustainable the alpha would be with significantly more assets. The average beta is a little misleading as many are market neutral, yet there are a few very directional managers. Many of the managers have an active trading style and typically this is very hard to scale over a few hundred million. Therefore, they rarely reach a size that institutions can get involved. That said there are a couple notable funds that have done very well.
Japan at the headline level didn’t appear to add as much alpha however the alpha does come with significantly less beta as compared to the other single country funds. It is worth noting that there are a group of managers with in the analysis who have performed even better than the headline levels of alpha with even less beta despite having to contend with terrible market conditions for the last 4 years. It is hard to get excited about the broad direction of the Japanese market, but we see the market as a good diversifying alpha source.
What this data can’t capture is the changing face of the industry. There have been a growing number of launches from experienced individuals that previously working in large global funds or banks. Many of these managers run lower net exposures and have more experience on the short side. As a result we see the level of beta declining. Investors have also become more discerning on paying hedge fund fees for beta.
The first 7 months of 2011 remained challenging though we saw steady alpha. The average of the larger long short fund returned 3-4% on a market little changed. However, the damage in August looks severe as many of the former winning stocks have significant reversals. I fear many managers may find they have wiped out most of their alpha for the year. We remain convinced that Asia remains a stock pickers market and long short is a good way to express this. However, the path will not be easy in the shorter term.