top of page
Search

BOJ policy change and the end of zero-Covid

December was an interesting month in Asia for two key reasons:


1. The BOJ’s yield curve control policy change on December 20, allowing 10-year bond yields to fluctuate by plus or minus 50 basis points around its 0% target, instead of the previous band of plus or minus 25 basis points.


Following this announcement, the Yen appreciated by more than 4% relative to the US dollar.


Japanese equities are more sensitive to currency moves than any other major market. Exporters, which had been major beneficiaries of a weak Yen over the course of the year, declined sharply. The Topix Index declined by almost 2% in the afternoon session (the announcement was made during the lunch break).


2. The CCP’s zero-Covid exit, marking an end to lockdowns which were hindering economic growth.


When investors fully grasped the enormity (and permanency) of this change, stocks which are beneficiaries of China reopening - most notably tourism and leisure stocks – rallied.


Both events had one thing in common: they represented a major turning point in financial markets.


Quant factors typically perform poorly around such turning points. A good example is November 2020 when positive vaccine news resulted in a significant market reversal and the severe underperformance of quant factors, including value and momentum factors.

December 2022 was no exception with quant factors performing poorly in Asia. Chart 1 shows the rank IC for our composite quant score for the key Asian markets in our stock universe.


Chart 1: All Star rank IC December 2022

Source: OQFM


The data for Singapore is somewhat noisy given the small sample size. In contrast, the HK/China universe is large, and it is extremely rare to see such poor factor performance over the course of one month in a market with this level of breadth.


Digging deeper into the factor performance data for HK/China, it’s apparent that both value and momentum factors detracted significant alpha. Chart 2 shows the factor performance for our alpha composites.


Chart 2: HK/China composites rank IC December 2022

Source: OQFM

The poor performance of both value and momentum factors is rare. Typically, these factor families are negatively correlated given outperforming stocks tend to be relatively expensive and vice versa. The correlation does, however, vary over time. Chart 3 shows the correlation between our Value and Momentum composites over the last 15 years in Hong Kong.


Chart 3: HK Value and Momentum composites correlation

Source: OQFM

Recently there has been a significant spike in the factor correlations. This is something we monitor in our risk management system (Lidar). In keeping with our investment philosophy, we also supplement this empirical data with additional insights based on what’s driving market sentiment. We have been closely monitoring the risk posed by China reopening and, specifically, the risk of shorting underperforming stocks which are expensive due to artificially low earnings, cashflow and dividend data resulting from the lockdowns. This is an example of how we monitor risks which aren’t captured in risk models based on historically calibrated data and can control risks more effectively than systematic managers. It helped us a lot in November 2020 across all markets and in December 2022 in HK/China.


Focussing on Japan, the rank IC of our composite quant score was only marginally negative, despite the extreme market reversal following the change in the BOJ’s monetary stance. However, factor performance leading into this announcement was relatively robust and the sudden currency reversal impacted the performance of our Sentiment composite which detracted significant alpha in December (Chart 4).


Chart 4: Japan composites rank IC December 2022

Source: OQFM


Fortunately, we constrain our net Yen sensitivity as it is a key risk factor in Japan, and this reduced the performance impact following the BOJ’s announcement. Our monitoring of Yen betas also enabled us to run real-time screens to identify stocks where the intra-day return wasn’t proportionate to the level of currency sensitivity. There were numerous stocks with strong Yen sensitivity which only marginally underperformed, and numerous stocks with minimal Yen sensitivity which strongly underperformed. Historically, we have a strong track record of exploiting extreme market environments by screening for oversold and undersold stocks, the best example in Japan being in the aftermath of the earthquake and tsunami crisis in 2011. The cross-sectional share price moves on December 20 weren’t of the same magnitude, but there were still some attractive alpha opportunities which we were able to exploit.


Looking forward, we are encouraged by the opportunities in Asia, particularly in HK/China where we are still seeing some attractive mispricing anomalies. These anomalies have been around since the fund was launched, but in 2020 and early 2021 the monetary policy backdrop and excessive liquidity made them difficult to exploit. More recently, rising bond yields have resulted in more rational price setting in most markets. The key exception has been HK/China where investors have been fretting over “investability” issues. It is interesting to note that the rank IC for our composite quant score over the last 6 months has been positive in all key markets globally except for HK/China (Chart 5).

Chart 5: 6 month All Star composite December 2022 rank IC

Source: OQFM

Recently, many of the issues which have discouraged institutional inflows into Chinese equities have been addressed. We believe this will result in more rational price setting and provide a nice tailwind for our investment process in 2023. It’s a good time to be running a discretionary multi-factor quant process in Asia.


20 views0 comments

Recent Posts

See All

Investor sentiment & AI as a risk factor

“In the short run, the market is a voting machine but in the long run, it is a weighing machine.” This is a quote from Warren Buffett in 1987 who paraphrased a passage from Graham and Dodd’s seminal b

Comments


bottom of page