Associate Professor Mardy Chiah

Using threshold and piecewise regression models on monthly data (January 1969¿March 2021), we examine the asymmetric effects of inflation and the US ten-year Treasury bond interest rate on gold price returns, identifying two distinct regimes. Inter alia, it is found that the responsiveness of gold returns to changes in inflation and interest rate depends on the magnitude of monthly inflation itself. Given the recent multi-trillion-dollar stimulus packages, the US is likely to switch into a ¿high¿ inflation regime (as defined by our endogenously determined threshold), leading to gold once again becoming a potent hedge against inflation. The results support the view that when monthly inflation in the US exceeds 0.55%, gold exhibits significant responses to changes in both inflation and the ten-year Treasury interest rate. However, when inflation is moderate or low, gold remains somewhat non-responsive. We thus argue that such asymmetric and size-dependent responses are the main causes of the lack of consensus in the literature regarding the hedging capability of gold.

Akbas et al. (2021) demonstrate that a more intense daily ¿tug of war¿ between overnight noise traders and daytime arbitrageurs predicts higher future returns in the US market. We investigate whether the daily tug of war contains predictive information about future stock returns in China. Using the frequency of negative daytime reversals, we find no significant difference in the future returns of stocks with a high versus a low level of intensity in this tug of war. However, we find persistent positive overnight returns followed by daytime reversals of almost similar magnitudes once we decompose the future returns into their overnight and daytime components. Thus, positive returns of the overnight component and negative returns of the daytime component cancel out each other, resulting in no predictive relationship between the daily tug of war and future returns in China.

This paper examines the impact of energy price uncertainty on a range of value anomalies. We demonstrate that the value premium is substantially stronger in periods of heightened energy price uncertainty. Energy price uncertainty exerts an asymmetric effect on the value anomalies, whereby downside energy price uncertainty accentuates the return differences between value and growth stocks compared to upside energy uncertainty. These findings are consistent with the argument that value firms possess a larger amount of inflexible assets than growth firms. Therefore, they struggle more to adjust in periods of elevated energy price uncertainty. We also demonstrate that energy price uncertainty has predictive power on the value premium one-month ahead. Using the Feasible Generalized Least Squares predictive model, energy price uncertainty can help mean-variance investors to obtain a positive annual utility gain across the value anomalies for up to 16.71%.

We analyze firm characteristic variables that potentially explain the momentum effect. Using a decomposition procedure developed by Hou and Loh (2016), we find that variables shown to have relations with momentum do not necessarily explain the effect. Among the firm characteristic variables considered, variables related to the direction and trend of trading volume or share prices play an important role in explaining price momentum. The contribution is primarily dominated by aggregated signed trading volume. Our results are robust to subsample analysis and different momentum period definitions. This paper sheds light on the sources of momentum returns.

Utilizing a recently developed measure of investor sentiment based on a large sample of news photos (i.e. Photo Pessimism), this paper examines the relation between investor sentiment and stock market returns around the world. We find that investor sentiment proxied by Photo Pessimism is negatively related to contemporaneous returns in 37 international equity markets. Photo Pessimism positively predicts subsequent market returns and trading volume, reflecting temporary sentiment-induced mispricing and the subsequent correction. The effect of Photo Pessimism on global stock markets is stronger amongst smaller markets and in periods where risk aversion is elevated. This indicates that sentiment-induced mispricing is more pronounced in the presence of limits-to-arbitrage.

This paper examines the trading patterns of Australian retail investors during the COVID-19 pandemic in 2020. It addresses the possibility that the Australian stock market may be used as a substitute for gambling due to prolonged lockdowns during the COVID-19 pandemic. Using transaction-level data, we document that retail trading volume increases substantially during 2020, overtaking institutional trading. This trading behavior is consistent with the conjecture that the stock market offers an easily accessible platform as a gambling substitute. We further explore two additional motivations for gambling in the stock market. We find that investors treat stock trading as a fun and exciting activity especially during the second phase of lockdown from July to October 2020. Examining the profitability of retail trading, we show that stocks heavily bought by retail investors consistently generate negative returns. This result resembles the negative expected outcome of gambling loss and demonstrates the danger of trading in a volatile market environment.

This study evaluates the impact of investor mood and mood contagion on the cross-section of Australian stock returns. While we document a day-of-the-week effect in Australia, there are noticeable differences to the US (Birru, 2018). Australian stock returns on Tuesday are significantly lower than other weekdays. This ¿Tuesday Blues¿ effect is most prominent amongst speculative stocks which, being more difficult to value, are more likely to be influenced by investor mood. We further report that Australian speculative stock returns are more correlated with factors indicative of same-day domestic mood and lagged US mood. Exploiting the 14-hour difference in time zones, we find that Tuesday returns to Australian speculative stocks are lower following extremely negative Monday mood or high risk aversion in the US. Overall, our results are consistent with mood contagion across international markets, which contributes to the observed daily patterns in stock returns after recognizing differences in time zones.

The size effect, whereby small firms outperform large firms, is not only a pervasive phenomenon in financial markets but also an important pricing factor in the Fama and French models (1993; 2018). However, several studies document that the size effect in recent decades has disappeared in the US and other international markets. Hou and Van Dijk (2019) show that the disappearance of the size effect can be attributed to profitability shocks in the US. This paper documents a weak size premium on the Tokyo Stock Exchange in Japan over the 1983¿2014 period. Similar to the US, a strong size premium resurfaces after adjusting for the price impact of profitability shocks.

This study investigates oil price risk exposure of financial and non-financial industries around the world during the COVID¿19 pandemic. The empirical results show that oil supply industries benefit from positive shocks to oil price risk in general, whereas oil user industries and financial industries react negatively to positive oil price shocks. The COVID¿19 outbreak appears to moderate the oil price risk exposure of both financial and non-financial industries. This brings important implications in risk management of energy risk during the pandemic.

There is an ongoing debate about the economic implications of excessive bank risk-taking and profitability. We examine this issue from the perspective of bank shareholders. Contrary to evidence for non-financial stocks, we find that operating profitability is negatively related to risk-adjusted bank stock returns. This negative relationship can be attributed to the nature of the banking business, where profit and systematic risk are intrinsically linked, and the previously documented ¿betting against beta¿ anomaly. We further demonstrate that more profitable banks are riskier and therefore have greater demand from leverage-constrained investors, resulting in higher valuations and lower than expected subsequent returns. The negative relationship between profitability and risk-adjusted returns is increasing in bank scale, as moral hazard problems and the use of market-based activities accentuate the link between profit and systematic risk in large banks.

We compare the performance of the time-series (TS) and cross-sectional (CS) momentum strategies in the US and China. The CS strategies by default are zero net investment strategies, whereas the TS strategies take on a time-varying net long position in risky assets. In the US, we confirm that the performance of the TS and CS strategies are not substantially different, as it is explained by a time-varying investment in risky assets. However, we find positive and significant return differences between the TS and CS strategies in China, after adjusting the CS strategies for a time-varying investment in risky assets. Our results suggest that in China, the difference between the performance of the TS and CS strategies might not be only limited to a time-varying investment in risky assets.

Recently, Gerakos and Linnainmaa (2018) decompose book-to-market into past book-to-market and changes in market and book values of equity. They find that book-to-market is highly persistent over time, and its explanatory power on US equity returns disappears when changes in market value of equity are accounted for. We find that, in both the US and Australian equity markets, changes in book-to-market stem largely from changes in firm size. However, contrary to the findings in the US market, book-to-market remains a strong predictor of Australian equity returns in the decomposition analysis. The results suggest that the finding in Gerakos and Linnainmaa (2018) is sample specific to the US equity market.

This study examines the comovement between eight prominent Australian asset pricing anomalies and their corresponding US counterparts. It confirms the continued existence of these anomalies in Australia and finds that these anomalies do not co-move with their US counterparts. Given the conflicting findings in prior research on the integration or segmentation of the Australian and US equity markets, this study adds to the body of evidence supporting segmentation.

This paper examines the impact of COVID-19 on trading volume in stock markets around the world. We document a large spike in trading volume in 37 international equity markets. The surge in trading volume is found to be associated with the national culture and institutional environment of individual countries. In particular, investors tend to trade more heavily in societies characterized by a higher level of trust and individualism, as well as a lower level of uncertainty avoidance. Investors are also more willing to trade in wealthier nations, as well as those with stronger protection of legal rights, better governance systems, and greater gambling opportunities.

This paper examines the day-of-the-week effect in anomalies (Birru, 2018) in 24 international equity markets. In particular, the returns to the quality-minus-junk (QMJ) factor on different weekdays are studied. The QMJ factor resembles the difference between speculative and non-speculative stocks in Birru (2018). We find that QMJ generates a positive (negative) premium on Monday (Friday). The QMJ premium on Monday is 2.89 times higher than its daily average. Our findings are consistent with the notion that investors are more optimistic (pessimistic) about the future prospect of speculative stocks on Friday (Monday). As such, there is a stronger preference for non-speculative (speculative) stocks on Monday (Friday), leading to the day-of-the-week effect around the world.

Equity markets outside the US are generally dominated by small-sized stocks that are outside the investable universe of institutional investors and professional money managers. In this paper, we compare the performance of a range of competing factor models in pricing large Australian stocks. By doing so, we shed light on the mixed findings in prior studies and the issue of national and international pricing of assets. Using a comprehensive sample spanning a period of 35 years, we document that the Fama and French (2015) five-factor model is superior despite a few close matches with some of the competing models. As the sample expands from the top 300 to the top 500 stocks, the superiority of the five-factor model becomes more apparent. There is also evidence that profitability and investment factors help to explain the cross-section of stock returns. Finally, although large Australian stocks are integrated with the US market, domestic factors are more important drivers of expected returns in Australia.

Using a factor-pricing approach, this paper investigates the extent to which the factors in the Fama¿French five-factor model, including momentum, explain Australian equity returns. A comparison of the United States and Australia suggests common components in asset returns. All the factors examined are useful in pricing Australian equities, whereas the HML factor is redundant for the United States. The findings suggest that the Fama¿French five-factor model should be at least considered as a benchmark model in asset pricing research.

Recently, Fama and French () propose a five-factor model by adding profitability and investment factors to their three-factor model. This model outperforms the three-factor model previously proposed by Fama and French (). Using an extensive sample over the 1982¿2013 period, we investigate the performance of the five-factor model in pricing Australian equities. We find that the five-factor model is able to explain more asset pricing anomalies than a range of competing asset pricing models, which supports the superiority of the five-factor model. We also find that despite the results documented by Fama and French (), the book-to-market factor retains its explanatory power in the presence of the investment and profitability factors. Our results are robust to alternative factor definitions and the formation of test assets. The study provides a strong out-of-sample test of the model, adding to the comparative evidence across international equity markets.