Our year in review includes an analysis of environmental risk factors, changes in market structure (and how we adapted to these changes), as well as notable stocks and investment strategies from 2024.
Lowlights
Our largest inflow occurred on our strongest performing day this year, which diluted the performance of our commingled fund. Chinese equity markets were closed in early October when Chinese equities were surging and, given the fund’s long H-share and short A-share positioning, this artificially increased fund return volatility. Another way to view this is that the fund’s absolute and risk-adjusted performance was actually stronger than the numbers suggest.
The Korean short-selling ban impeded our ability to generate alpha in this market. Historically, Korea has been a strong market for us, and we expect this to continue once the short-selling ban is lifted in March.
The fund performed relatively poorly in Australia. The blame for this lies with me. I’ve struggled to comprehend the continued strong outperformance of several growth stocks, which I believe are overpriced. I still believe in the concept that a good company can be a bad stock and my discretionary overrides in this market detracted alpha.
Highlights
The fund performed strongly in 2024. The fund only registered two down months: -0.47% and -0.74% in July and August, respectively. Downside volatility was exceptionally low and the fund’s Sortino ratio was 5.92. We have now generated strong positive returns in each of the last four calendar years. Considering the extreme rotations between value and growth, the broader shifts in risk-on and risk-off sentiment, and the significant volatility within our universe of Chinese stocks, this represents a remarkable achievement. We believe that few single-strategy funds have matched our high level of return consistency over this period.
We lured Lowell away from Point72. Lowell is a great hire and brings additional depth and expertise to our research team.
We achieved strong fundraising success. For various reasons - including my reluctance to travel and engage in marketing efforts and unwarranted investor concerns over simplistic measures of idiosyncratic risk - it took some time to gain traction with investors. Recently, however, investors have shown strong interest in our fund, and demand now exceeds our available capacity.
Operational improvements include trading with more brokers, more closely monitoring slippage, and using trading algorithms that optimize the trade-off between market impact and opportunity costs. Increasing our trading frequency with Morgan Stanley has also improved the level of internalization, and it has been rewarding to work with brokers who have supported us rather than use Goldman Sachs for all our trading. We intend to do more of this in 2025.
We made significant improvements to our rebalance process, focusing on adding more screens to facilitate more active trading. We believe this is particularly important in the current market environment, and we discuss the reasons for this below in the context of changes in market dynamics.
Market Dynamics and R&D Initiatives
I particularly enjoyed reading Cliff Asness’ recent research paper titled “The Less-Efficient Market Hypothesis.” I typically avoid finance research papers because, much like legal documents, they often explain simple concepts in the most obscure and complex ways. Asness’ paper, however, was an exception - it was extremely easy to read and deeply resonated with me.
Asness postulates that investors who “make bad decisions based on emotion, story, tastes that are not relevant for risk and return, and behavioral biases” are having a bigger impact on setting share prices. Potentially these investors are playing a more active role driving share prices due to the increased use of social media and the “gamification of trading”. Other investors who have historically profited from this behaviour by taking the other side of the “misguided positions” are now less active in the market, having transitioned to passive index investing.
Inefficient stock pricing - i.e., disconnects between stock prices and fundamental values - was prevalent in 2024 and will likely persist for the foreseeable future. While this may offer greater rewards for patient, long-term investors, it poses a significant challenge for a fund like ours, which strives to generate consistent alpha.
Our primary approach to addressing this challenge is by targeting short-term pricing distortions driven by liquidity flows. In 2024, we enhanced our investment process by adding additional rebalance screens to achieve this. This also resulted in a marked increase in trading frequency. We now typically execute more than 200 trades each day.
Accompanying this change, we also adjusted the distribution of position sizes by effectively limiting the tails of the distribution. Our largest long and short positions are now smaller than they used to be, helping to mitigate the risk posed by extreme stock movements that run contrary to stock fundamentals.
Fund Return Volatility
A recurring theme in 2024 was low fund return volatility. The fund’s return volatility remained below our 7% p.a. target in all but one month, and for many months, it was significantly below this level. This was largely driven by the market environment and adjustments to our investment process.
The market environment was characterised by relatively low cross-sectional return volatility and, more importantly, significant currency depreciation relative to our base currency (the US dollar). This latter factor consistently dampened portfolio gross exposure (i.e., leverage), making it more challenging to achieve our fund return volatility target.
Adjustments to our investment process primarily involve implementing tighter risk controls. When I launched the fund, I had relatively loose risk controls, which was a deliberate decision aimed at generating stronger headline performance
As an example of the relatively loose risk constraints, during the first couple of years, I struggled to find good short stock opportunities in Asia. Shorting Asian index futures was also problematic due to large index stock weights and sector biases. Consequently, to achieve market neutrality, I shorted S&P 500 and Nasdaq futures contracts, offsetting this with a net long exposure to Chinese equities. Needless to say, this did not work well; among other issues, it hindered our fundraising efforts. Why would an Asia-focused fund be shorting U.S. futures contracts?
After Zicai joined the team and built a robust performance attribution system, we discovered that this risk positioning significantly increased fund return volatility while materially detracting from the fund's performance. Fortunately, the alpha generated through our stock selection process was strong enough to overcome this performance handicap.
The solution was clear: we needed to impose tighter risk constraints, particularly regarding net country limits, and focus more on stock selection. To those who may be worried about the prospect of us ever shorting US futures again, we also modified our risk limits, effectively prohibiting us from shorting futures contracts outside of Asia.
More recently, we also adjusted the distribution of position sizes by limiting the tails.
All these changes make sense in terms of alpha generation. However, they also dampen fund return volatility and the potential to generate strong headline performance. To counter this, we should have been more aggressive increasing gross exposure, particularly in 2024. Fortunately, we have a very conservative capacity limit, allowing us to take up portfolio leverage without incurring significantly higher market impact costs.
It's important to acknowledge and learn from mistakes, such as having relatively loose risk constraints. However, it’s also good to balance this with recognition of good decisions. In addition to focusing solely on Asia, implementing a conservative capacity limit is an example of good decision-making.
Environmental Risk Factors
There are risk factors that consistently influence stock returns, such as country, sector, and size exposure. These factors are captured by generic risk models.
There are also risk factors specific to the market environment. These factors either aren’t captured by generic risk models or do not receive the weighting they deserve.
As part of our discretionary overlay, we integrate these risk factors into our portfolio construction process. We believe this provides us with a significant edge over quants using a systematic investment approach.
In 2024, we monitored and constrained the portfolio’s exposure to three significant risk factors: AI, yen sensitivity, and interest rate sensitivity.
The emergence of AI as a risk factor is discussed here: https://www.oqfundsmanagement.com/post/investor-sentiment-ai-as-a-risk-factor
In previous market commentaries, I discussed why we increased our analysis of aggregate yen sensitivity. Japanese equities are extremely currency-sensitive, more so than any other large equity market. The yen has also depreciated markedly over the last three years, and dollar-yen is currently approaching 160 - a level that, based on past interventions to support the currency, seems to be a line in the sand beyond which further depreciation is not tolerated. The yen is also seen as a safe-haven currency and, in the event of an extreme exogenous shock, may appreciate markedly.
Constraining sector exposure is not sufficient to address this risk factor. Focusing solely on yen betas is also inadequate. In addition to yen betas, we monitor the percentage of revenue generated overseas, particularly in the United States, given the U.S. dollar has been particularly strong.
Interest rate sensitivity has emerged as a key risk factor, given that central banks are adopting less hawkish stances and, with the exception of the BOJ, are either cutting rates or are likely to cut rates in 2025. Lower interest rates benefit high-duration growth stocks and bond proxies such as REITs and utilities. Interestingly, these stocks lie at opposite ends of the risk spectrum, necessitating a more nuanced approach than is typically required when addressing risk issues.
Interesting Stocks
With over 1,300 positions in the portfolio, there are always interesting stock stories. In this section, I highlight six notable stocks from 2024, one from each of the key markets in our stock universe.
Australia: Pro Medicus (PME AU) Pro Medicus has been one of the best-performing stocks in our universe since the fund was launched, rising more than tenfold over the last five years. In 2024, the share price increased by more than 170%. Unfortunately, we maintained a sizeable short position in this stock over the course of the year. There are many things to like about the company, particularly its strong growth profile. Typically, outperforming growth stocks have strong broker recommendations. Pro Medicus, however, has nearly as many sell recommendations as buy recommendations, with a consensus recommendation that is more than half a standard deviation below the Australian mean and the Australian Health Care sector mean. Many analysts cite the company’s extreme valuation, with its market cap approaching 100 times its revenue. Analysts have been busy upgrading their target prices, but this is normal when their target prices are significantly below the market price. There has also been a lot of insider selling, and it is one of the worst-rated stocks in Australia and our stock universe based on this metric.
Japan: AEON (8267JP) When I talk about the impact shareholder perks have on stock prices, AEON is my favourite example. It’s a stock with a tiny dividend yield of around 1% and an earnings yield that isn’t much higher (its one-year forward earnings yield is 1.6%). Sales per share are strong but this is countered by a low gross profit margin. As the largest supermarket chain in Japan, it also doesn’t offer much in the way of blue-sky growth potential. What it does offer, however, is the well-known AEON Card, or Omoiyari Card, which provides attractive discounts, rebates, and exclusive benefits. These perks need to be accounted for when analysing the stock’s shareholder yield. In 2024, AEON underperformed the Japanese market, rising just over 17% (versus 25% for the Topix). Its performance was dragged down by its Yen beta (it’s hurt by a weak currency), but it still managed to outperform its sector peers.
Hong Kong: China Life Insurance (2628 HK) One of the great mysteries to me is why some highly liquid H-shares trade at a significant discount to their A-share listing, even after the introduction of the Stock Connect program in November 2014. Both A-shares and H-shares have the same dividend entitlements and voting rights. In my opinion, the best example of this anomaly is China Life Insurance. As of the end of 2024, the H-share price was $14.68, while the A-share price in HK dollars was $41.92. Put differently, the H-share traded at a 65% discount to the A-share.
Singapore: Singapore Telecom (ST SP) Singapore Telecom is one of our largest long positions, and its share price rose over 31% in 2024. It owns stakes in several Asian telecommunications companies, most notably Bharti Airtel (BHART IN), which currently represents over 75% of Singapore Telecom’s enterprise value. Bharti Airtel rose almost 58% in 2024, significantly outpacing Singapore Telecom. In addition to its other shareholdings, the value of its underlying businesses - including the second-largest telecommunications company in Australia (Optus) and a growing regional data centre business - makes it, in our view, one of the most undervalued companies in Asia based on a sum-of-the-parts analysis.
Taiwan: TSMC (2330 TT) TSMC has a local listing (2330 TT) and an ADR listing in the United States (TSM US). For many years, the two securities traded close to parity (after adjusting for currency). In 2024, however, the share prices of the two listings diverged, with TSM US trading at a significant premium to 2330 TT, peaking at approximately 30% in February. It finished the year at approximately a 17% premium. Arbitraging this difference is not feasible given that special regulatory approval is required to convert 2330 TT into TSM US. The AI frenzy is particularly acute in the United States and appears to be the main driver of this pricing anomaly.
Korea: Posco Future M (003670 KS)
Writing about this stock is extremely frustrating, but I will do it anyway. It’s a good company, but I believe there’s a lot not to like about the stock. In particular, it has had negative valuation factor z-scores across our entire suite of factors since the fund was launched in 2020. In 2023, it resembled a meme stock, with Korean retail investors getting excited about its role in the electric vehicle supply chain, particularly in the production of battery materials. It was also one of the most shorted stocks by institutional investors, and when retail investors successfully lobbied to have short selling banned in October 2023, its share price spiked. We had a large short position in the stock, and it significantly detracted from the fund’s performance. At the time, I struggled to deal with the competing forces of poor quant factor scores and strong liquidity support. I ended up taking an each-way bet and covered half of our short position. This was unfortunate, as the stock performed extremely poorly in 2024, declining over 60%.
Interesting Investment Strategies
Buying discounted H-shares The A-H premium/discount primarily varies based on overseas investors' views on China. In recent years, overseas investors have been extremely bearish on China, which has manifested in a significant discount for H-shares. There have been instances, such as in late 2014, when H-shares, in aggregate, actually traded at a premium to A-shares. As I discussed in our December newsletter, there are various ways to measure the A-H premium/discount. One aspect we monitor is the discount or premium of the H-share that is most expensive relative to its A-share listing. Based on this metric, H-shares offered the best value proposition compared to A-shares in 2024 in many years.
Trading A-H pairs with a positive carry
This is one of my favourite trading strategies from 2024, which I documented in a comprehensive report published on our website (please note that this report is password protected). This strategy became less attractive towards the end of 2024 as borrowing costs increased for numerous A-shares. However, the following A-H pairs still offer appealing positive carry opportunities: China Petroleum & Chemical Corporation (600028 CH and 386 HK), China Everbright Bank (601818 CH and 6818 HK), and PetroChina (601857 CH and 857 HK).
Trading dual listings
To my way of thinking, significant price differences between dual-listed stocks undermine the efficient market hypothesis. Stock prices are determined by supply and demand rather than by fundamental valuations, and the levels of demand can vary significantly across different markets. In the previous section, I mentioned TSMC’s dual listing and the significant premium of its U.S. listing relative to its Taiwan listing. Another good example is Rio Tinto’s dual listing in Australia and London. The Australian listing finished the year at over a 23% premium to the London listing, nearing a multi-year high. Incredibly, there have been periods when the Australian listing traded at a discount.
Shorting Asian stocks (ex-China) with China exposure This has been one of my favourite strategies that has manifested since the recent selldown in Chinese equities. Investors have been selective in terms of expressing their bearish views on China. There are stocks listed in Asia that rely on Chinese demand and/or Chinese supply chains, which are richly priced. In contrast, many Chinese stocks listed in Hong Kong are priced at levels comparable to those during the Global Financial Crisis. In 2024, we were able to take advantage of some of these opportunities, particularly in September and early October when Chinese stocks listed in Hong Kong rose sharply.
Comments