Our discretionary overlay is the most distinctive feature of our multi-factor quant investment process. It is also the most difficult to explain. It comprises two components: our macro-overlay whereby we endeavour to position the portfolio so that it is better suited to the market environment, and our micro overlay which focusses on improving the predictive power of our investment process at a stock level.
Our decision to dial down momentum exposure and mitigate stock specific risk by getting more short exposure via index futures in 2020 and 2021 are examples of our macro-overlay. We have a strong track record of improving risk adjusted returns via these overrides.
Our micro discretionary overlay takes many forms and varies greatly on the alpha profile of the stock being analysed. In broad terms, we try to determine whether the stock is compatible with the mispricing opportunity we’re targeting. It is not fundamental analysis.
We have recently updated our main marketing presentation to better explain how this discretionary analysis works. In the Appendix, we have included some long and short stock examples which highlights some of the information we analyse on the main stock profile page of Conquest, our portfolio management system. If you would like us to send you these slides, please let us know.
In this month’s commentary, we will discuss the discretionary overlay vis-à-vis a stock which has been in the news a lot recently: Alibaba. We have not chosen Alibaba because it is an outsized position. We have over 1,000 stocks in the portfolio and our long position in Alibaba is relatively modest. We have selected Alibaba primarily because most people have some degree of familiarity with the company.
Alibaba satisfies our Net Cash Alpha, Growth at a Reasonable Price, and Long Value screens.
Our Net Cash Alpha screen identifies stocks which are extremely cheap on an enterprise value basis. There are a relatively large number of stocks in Asia with cash rich balance sheets, far more than in Europe and the United States.
For Net Cash Alpha stocks, our discretionary analysis focuses on whether the cash can be used to boost the shareholder yield. We typically discount net cash holdings for SOEs given the cash is unlikely to directly benefit shareholders. This doesn’t apply to Alibaba and in March the company increased its buyback from $10bn to $25bn. Historically, the company doesn’t have a great track record of following through with repurchases, but this time it has been different. Prior to upsizing its buyback program, it had already purchased $9.2bn of stock.
Our Growth at a Reasonable Price screen identifies stocks which ae relatively cheap compared to their growth profiles.
For this screen, we focus primarily on the stocks’ historic and forecast growth profiles. On average, stocks with strong forecast growth tend to underperform slightly. In other words, forecast growth tends to be a contrarian indicator. Although this sounds counterintuitive it makes sense when viewed through the lens of these stocks being priced for perfection and hence prone to undershoot lofty expectations.
During periods of investor euphoria – like we’ve witnessed recently – the potential overpricing of growth expectations is more acute and hence we’re more attuned to this potential mispricing anomaly.
Factors measuring historic growth, however, do tend to be positively correlated with stock returns, albeit only mildly. Given this, our analysis of growth stocks tends to focus on stocks with strong structural growth drivers – such as a large “moat” or barrier to entry for competitors – but which have growth forecasts which aren’t disproportionate to their realised growth.
Alibaba has an extremely strong growth profile. There are a few charts in Conquest that we review, but our preferred chart simply shows historical and forecast revenue on the left y-axis and the corresponding revenue yield on the y-axis for the last 10 years and the next 3 years.
Chart 1: Alibaba revenue growth profile
Source: Factset, OQFM
Historical revenue growth has a nice “exponential” profile whereas revenue forecasts are more modest. We also note that growth has not been fueled by debt; indeed, the company has been increasing its net cash position over the last few years.
Chart 2: Alibaba net cash profile
Source: Factset, OQFM
Cashflow growth has also been strong. All the key charts and other diagnostics we monitor in Conquest indicate that Alibaba has the type of growth characteristics we’re looking for.
Our Long Value screen identifies stocks which are very cheap based on robust valuation factors that exhibit strong predictive power over the medium to long term. This screen also includes hurdles to exclude stocks which rate badly based on quality and certainty factors.
For Long Value stocks, our discretionary analysis focuses on potential structural, regulatory, or other risks that could render the stock a “value trap”. Alibaba isn’t structurally challenged and, as we discussed, has an extremely strong growth profile.
On the topic of regulatory risks, the situation is less clear cut. Any discussion of Chinese regulatory risks often morphs into a political or ideological debate. We prefer to stick to the facts, focusing on past actions and statements of intent by key decision makers.
Alibaba has had to deal with numerous regulatory issues over the last couple of years. Ant’s listing was axed in 2021. Last year it was also fined for anticompetitive behaviour (forcing merchants into exclusive sales arrangements) and this year it’s likely to be fined for incompetence (a data leak from a misconfigured Alibaba Cloud server). These actions are consistent with US and European regulations. Both the FTC and EU are currently pursuing antitrust investigations focusing on the big tech companies.
Further, there are indications the Chinese government believes that investor angst over the tech crackdown and, more broadly, a move towards “common prosperity” has been overdone. In May, Liu He, a vice-premier and President Xi Jinping’s closest economic adviser, said China “must support the platform economy, and sustain the healthy development of the private economy”.
Recently, investors have also been fretting about the US delisting risk. On July 30, the SEC added Alibaba to a list of firms that face delisting from US exchanges. Alibaba’s share price immediately plunged.
Many commentators that play down the delisting risk maintain that it’s unlikely to happen. We don’t take on a view on this issue given it is partly driven by geopolitical concerns and political posturing. It may happen and given US stocks currently trade on higher multiples than any other developed market, it could have a short-term price impact.
However, our investment view is shaped by one overarching principle: share prices are ultimately determined and driven by fundamentals. Any investment methodology that isn’t based on this premise has to exploit very short term anomalies and typically fits within the realm of high frequency trading. We do have factors and some screens that focus on short-term earning and price momentum, but we ultimately rely on rational price setting.
Applying this principle to Alibaba means that whether or not it maintains a US listing is largely moot. If Alibaba’s share price is cheap relative to its cashflows and growth potential, it will outperform, regardless of this issue.
It is also important to note that Alibaba doesn’t need access to US capital. As we’ve discussed, it has a very strong balance sheet with a sizeable net cash position. Alibaba also has an extremely liquid and fungible listing in Hong Kong. This will soon become the “primary” listing, probably before the end of this year, making it eligible for southbound trading via Stock Connect.
Of course, we could be wrong! Our alpha screens exhibit strong out-of-sample backtest results. However, in the world of quant factor investing, strong backtest results equate to an after-cost success rate that only provide a slight edge to the investor. It’s akin to counting cards in Blackjack – back in the days when there weren’t so many decks in the chute; it only provides you with a slight edge over the house. Also, our micro discretionary overlay is designed to only slightly further skew the odds to our advantage.
This means we need a lot of breadth. It’s something we harp on about all the time, but we do it for a good reason: it’s incredibly important. We need breadth in terms of alpha drivers, rebalance trades and number of positions in the portfolio. This greatly lowers the bar for success. So long as we get it more right than wrong, we can generate very attractive risk adjusted returns. In contrast, high conviction funds need to make consistently good calls, lest they significantly underperform. And it’s very hard for investors to distinguish between luck and skill when analysing past returns.
We know which approach we prefer.
Comments