Representative Publications (Date, newest)

Our earlier findings in Chia et al. (2020) show that Malaysian public listed firms should care about their stock liquidity because of the direct implications on firm value. But what liquidity-enhancing strategies should these Malaysian listed firm pursue? To offer recommendation, we empirically examine the untested proposal of Amihud and Mendelson (2000, 2008) that expanding shareholder base is an effective strategy to enhance stock liquidity.

This research is made possible because the “End of Year Shareholdings by Type of Investor” database assembled by Bursa Malaysia provides the required data of “number of shareholders” for each firm at the end of each calendar year. To justify the inclusion of this new variable of shareholder base in a liquidity model, we first provide the theoretical bases available in the existing literature. Our baseline and extensive robustness checks confirm a nonlinear relationship between shareholder base and liquidity. While more shareholders are associated with higher liquidity, the negative effect of wider spreads kicks in when shareholder base exceeds a threshold level, which we found is due to higher volatility induced by noise trading. More importantly, the economic effect of our new variable on liquidity is at least four times larger than the combined effects of all ownership variables. It thus lends support to the call for corporate managers to pursue shareholder-boosting strategies.

When exploring why liquidity declines at higher levels of shareholder base, we also consider the possibility of adverse selection costs imposed by informed trading. Even though such channel has been ruled out, one interesting insight from the empirical analysis with PIN is that none of the investor groups – local individuals, local institutions, local governments, local nominees, foreign individuals, foreign institutions and foreign nominees – are informed traders. Only the coefficient of blockholdings is significantly and positively associated with PIN, suggesting that the only informed investor group in Bursa Malaysia is blockholders. Hence, informed trading is determined by shareholding sizes and not investor types.

WoS Citation      Google Citation

After the macro work on aggregate liquidity in Liew et al. (2018, 2020) that has policy implications to stock exchange regulators, we turn our attention to firm-level analysis. Should Malaysian public listed firms care about their stock liquidity after the public listings? To provide an answer, we re-examine the relationship between liquidity and firm value using data for all non-financial firms traded on Bursa Malaysia over the sample period of 2000–2015.

Among the novelties of this paper, we examine the possible existence of threshold liquidity level, and explore three moderating variables that are unique to the local corporate landscape- political connections, foreign nominee ownership and ownership of local institutions which are largely state-backed. The key finding is that the firm value benefit only kicks in when liquidity exceeds the threshold level. Hence, the challenge is for Malaysian corporate managers to further boost liquidity so as to increase the market valuations of their listed firms, especially for public firms with political connections, higher foreign nominee ownership and higher foreign institutional ownership.

WoS Citation      Google Citation

Since foreign investors are found to consume liquidity on the Malaysian stock market in our earlier study (Liew et al., 2018), we conduct the first empirical analysis to evaluate the effectiveness of liquidity provision through proprietary day traders (PDTs). The key novelty, however, is the extension of liquidity to higher-order moments, in particularly our proposal of conditional liquidity skewness.  Using the  “Trading Participation by Category of Investors” database but at daily frequency, we find evidence that PDTs’ trade volume is associated with higher aggregate liquidity. However, such improved liquidity comes at a cost to investors, as proprietary day trading is found to be associated with higher conditional volatility and conditional skewness of closing percent quoted spreads. The former is due to the exchange-imposed immediacy for PDTs to close their open positions, whereas the latter can be attributed to the exclusive rights granted to PDTs to engage in intraday short selling.

WoS Citation      Google Citation

After the pioneering work of Lim et al. (2016), this sequel continues to scrutinize the role of foreign investors in Bursa Malaysia, shifting the focus to macro-level analysis that has direct policy implications. We construct weekly market liquidity measures by aggregating firm-level estimates using value-weighting scheme, with the selected closing percent quoted spread proven by Fong et al. (2017) to be the best proxy for Malaysian stocks. Our paper is also the first to utilize the “Trading Participation by Category of Investors” database assembled by Bursa Malaysia.

With these unique data, we undertake the first empirical analysis to determine the impact of foreign equity flows on aggregate liquidity of the Malaysian stock market. The research addresses the prevailing public anxiety on the recent massive withdrawals of foreign portfolio investment from the country. Our findings show that foreign investors act as informed traders who reduce aggregate liquidity of the local bourse. There is evidence that the gross inflows of foreign institutions destabilize the local stock market through their adverse impact on aggregate liquidity, confirming the negative press reports on foreign investors during periods of volatile foreign equity flows. However, liquidity condition in Bursa Malaysia remains resilient when foreign investors flee the market as there is ample liquidity provision from local government-backed institutions to cushion the negative impact arising from such capital flight, thus protecting the local market from liquidity dry-up.

From the perspective of asymmetric information models, the reduction of liquidity by foreign trading implies that foreign investors are informed traders in the Malaysian stock market. Their information advantage can be derived from either privileged access to private information or skilled analysis of public news. The first possibility has been ruled out by Chia et al. (2020b) unless they are blockholders, whereas the latter is confirmed by Lim et al. (2016) who find that foreign investors are elite processors of public news in Bursa Malaysia.

WoS Citation      Google Citation

This is the first paper in the extant empirical literature to examine the role of foreign investors in Bursa Malaysia, despite its incorporation for more than four decades. The absence of such research in the past is because complete ownership data are not reported in annual reports and commercial databases. We capitalize on the release of such data when Bursa Malaysia commercializes its database “End of Year Shareholdings by Type of Investor“.

Among the novel contributions, this paper shows the existence of a threshold level in foreign ownership, explores the underlying price efficiency channels, and determines the differential impact of foreign investor heterogeneity. A unique finding that we uncover is the important information role played by foreign nominees, who have been found to be elite processors of public market-wide and firm-specific news in the Malaysian stock market. However, such price discovery does not occur when foreign investors trade through direct accounts.

Our unique finding on foreign nominees complements the discovery of informed trading through the accounts of children in Finland by Berkman et al. (2014).

WoS Citation      Google Citation

Autocorrelation-based tests are popular because the presence of autocorrelations is theoretically grounded. This paper advocates two recently developed autocorrelation-based tests in which the lag order/holding period is selected automatically, and thus avoids the arbitrary selection that causes conflicting results. In the application, we show that the new tests should be employed in rolling framework to track the changing degree of efficiency. This is because even the most developed U.S. stock markets are not efficient all the time!

WoS Citation            Google Citation

The survey paper by Lim and Brooks (2011, JES) shows that statistical tests of the weak-form market efficiency focus on uncovering autocorrelations, nonlinear serial dependency and long memory. However, we argue in this paper that the most appropriate econometric approach is to examine whether the stock returns are Martingale Difference Sequence (MDS), implying no serial dependence, either linear or nonlinear, in the conditional mean of the return series. In our empirical application, we advocate the generalized spectral martingale test, and show that the test can pick up dependency structure that passes through variance ratio tests, while at the same time will not falsely proclaim market efficiency when there exists serial dependence in higher order conditional moment.

WoS Citation          Google Citation

This paper departs from trade openness’s traditional focus on economic growth to a relatively less-explored area, the informational efficiency of stock markets. There are at least two novel contributions to the existing literature. First, we measure market efficiency in relative rather than absolute form, using the absolute value of the variance ratio minus one with the optimal holding period been selected automatically. Second, we find that the positive relationship between trade openness and market efficiency is only significant when de facto measure is used, contradicting the existing theoretical model where openness is defined as the elimination of trade barriers. To rationalize our new finding, we offer an information-based explanation in which trade openness signals higher future firm profitability, and investors tend to react faster to information when there is less uncertainty about a firm’s future earnings or cash flows.

Our call to further determine the lack of association between financial liberalization and market efficiency has been picked up by Naghavi and Lau (2014).

WoS Citation        Google Citation

Nearly five decades after Eugene Fama proposes the efficient markets hypothesis (EMH), there is still no consensus on whether the stock markets are efficient or not. This paper offers a new perspective on the subject matter, arguing that informational efficiency varies over time (evolving market efficiency) and across stock markets (relative market efficiency). After assembling empirical evidence supporting the concepts of relative and evolving market efficiency, this paper argues in a favor of a new framework, the adaptive markets hypothesis (AMH), to replace the once dominant EMH.

The idea of our paper is nicely executed by Niemczak and Smith (2013).

WoS Citation        Google Citation

This paper operationalizes the adaptive markets hypothesis (AMH) by testing its two implications. We first follow previous studies to test AMH’s implication of time-varying return predictability, and reaffirm market efficiency is not a static characteristic. The novelty of our paper lies in providing the first empirical evidence on the often neglected implication of AMH, namely the degree of return predictability is dependent upon market conditions as manifested by market crashes, fundamental economic or political crises, economic bubbles and regulatory regimes. This latter implication requires measuring market efficiency in the relative sense, which we implement using automatic variance ratio test, automatic portmanteau test and generalized spectral test. The statistics from these tests allow us to run regression against indicators for market conditions and a set of control variables. Using a century-long Dow Jones Industrial Average index, we find evidence supporting the two testable implications of AMH, highlighting market conditions do matter for return predictability.

WoS Citation        Google Citation

Due to the dominance of absolute market efficiency, little is known about when, why and how stock market becomes efficient. This paper proposes a framework to measure the degree of market efficiency, and finds that emerging stock markets experience more frequent price deviations than developed markets, which can largely be attributed to low-income economies providing weak protection for private property rights. We argue that secure private property rights is a necessary and sufficient condition for ensuring stock prices move closer to a random walk. This is because informed arbitrageurs will not trade if they cannot keep their profits, leaving those markets dominated by sentiment-prone noise traders whose correlated trading causes stock prices to deviate from the random walk benchmarks for persistent periods of time.

This paper advocates the application of nonlinearity tests for weak-form efficiency studies given the possibility that returns series can be linearly uncorrelated but nonlinearly dependent. It is equally important to test the persistence of nonlinear predictability by means of sub-samples or time-varying parameter model because its presence could be localized in time.  Using data from 10 Asian stock markets, we find that all return series exhibit strong evidence of nonlinearity, but their appearance is episodic and transient in nature.

We further extend the empirical literature by conducting additional analyses to address new research questions: (1) what are the factors that  might account for the cross-country differences in relative efficiency? (2) Are major economic and political events responsible for the transient burst of nonlinear dependence? (3) Do market states play a role in generating the detected nonlinear dependence? (4) Is it possible to predict when the nonlinear dependence would occur? Our framework suggests moving beyond testing absolute market efficiency to exploring its determinants, preceding the cross-country work of Griffin et al. (2010), Lim et al. (2010) and Jain and Xue (2017).

Google Citation

This paper utilizes the framework proposed by Lim (2007, Physica A) to empirically assess the effects of the 1997 financial crisis on the efficiency of eight Asian stock markets. Applying the bicorrelation test in rolling windows, we are able to track the evolution of market efficiency over the sample period from 2 January 1992 to 31 December 2005, and observe graphically there is significant deviation from efficiency during crisis for all markets. By comparing the total number of rolling windows with significant nonlinearity over the three sub-periods of pre-crisis, crisis and post-crisis, we are able to gauge not only the relative efficiency of these Asian markets during each sub-period, but also to compare the efficiency of each stock market across the three sub-periods. Empirically, our finding of higher inefficiency during the crisis are not surprising as in that chaotic financial environment, investors would overreact not only to local news, but also to news originating in the other markets, especially when the news events were adverse. Thus, we argue that pursuing credible policy actions to calm the markets and restore investor confidence should be the priority for crisis management.

This paper has been in the Top 25 Hottest Articles for 35 consecutive quarters!

Google Citation

This paper extends the nonlinearity literature to show that the stylized fact of nonlinear serial dependence in stock returns is episodic and transient, implying the degree of return predictability varies continuously over time and thus market efficiency is not a static characteristic as widely assumed. We advocate applying nonlinearity tests in rolling sample approach because the latter enables researchers not only to track the evolving nature of market efficiency, but also to rank the relative efficiency of stock markets.

WoS Citation        Google Citation