Explore publications.

With an observational research design, using only publicly available stock price data, this paper aims to demonstrate that momentum extends beyond the medium term (3 to 12 months) when generating momentum profiles using a preset term-structure. The optimal holding period is a function of the formation period – the longer the formation period, the shorter the holding period and vice versa. Convention has it that once a stock begins to gain momentum it usually maintains it for 15 months before faltering, reversing some gains after 18 months. A full momentum cycle (formation plus hold) should complete within 24 months, peaking at 18 months. A dataset of 370 stocks, listed on the Johannesburg Stock Exchange (JSE) between 2007 and 2024, was used to generate momentum profiles from daily historical prices. Profiling shifts the focus onto the holding period to classify momentum cycles on the eventual outcome. Cycles are classified as mini, regular, ultra, super and hyper, depending on duration. Clusters of momentum appear when generating profiles for individual stocks, allowing one to identify optimal entry and exit points into active momentum cycles. Results show that major cycles comprise many minor cycles. Medium-term momentum or regular cycles, therefore, may consist of several mini cycles. The main finding of this paper is that there is potential to remain in active cycles for several years, thereby holding momentum. A custom momentum index, updated weekly with variable members, may give some indication of the state of momentum in a market at a particular point in time as an increase in index levels tend to correspond with an increase in active momentum cycles or momentum activity.

This study tests for the applicability of the Adaptive Market Hypothesis (AMH) and Bounded rationality theories on the Johannesburg Stock Exchange (JSE). This is in order to determine the influence of behavioural risk factors on the efficiency of the stock market. Behavioural theories show a gap in the theory of Efficient Market Hypothesis. Using quantile regression, our study established the applicability of the Adaptive Market Hypothesis on the JSE. Past market returns were shown to be significant in predicting future returns and thus did not follow a random walk. The lagged return increased at higher quantiles and differed with changes in market conditions (i.e. pre-financial crisis, financial crisis and post-financial crisis).  Thus, the predictability of returns varies with a change in market conditions. This paper is focused on only on the Johannesburg stock exchange in South Africa, more specifically the movement is the all share index. The findings should be able to be applied to emerging and developed economies. Business confidence is found to have a negative relationship with returns showing a lag in time for sentiment to be incorporated in prices. In contrast, consumer confidence is found to have a positive relationship with returns. In summary, investors are shown to be influenced by fundamental and behavioural factors.

The purpose of this article was to assess the impact of financial market liquidity on international capital flows in emerging markets. Specifically, the research investigates the effect of bond market liquidity and stock market liquidity on foreign portfolio investments using data for five emerging African countries, being Egypt, Kenya, Mauritius, Nigeria and South Africa, for the period 2000 to 2020. The data was sourced from the Bloomberg and World Bank (WDI) databases. Panel data analysis (fixed effects model) was undertaken using three different liquidity measures: the effective spread; Amihud’s (2002) illiquidity measure; and market impact as measured by trading volume. Our findings revealed mixed results. It was found that stock market liquidity attracted foreign portfolio investments. Although bond market liquidity, as measured by the volume of trade, promoted foreign portfolio investment, it was different for the effective spread, as the higher the effective spread, the higher the inward FPI flows, and vice versa. Results on the effects of the bond effective spread on FPI show that as long as the bonds are above the investable grade, investors are not discouraged by the cost of trading. Our findings thus confirm that FPI inflows are predisposed on liquid and efficient host country financial markets. Further, the entrance of foreign investors in the host country’s domestic financial markets, leads to the enhancing of liquidity in the local market, thus increasing risk sharing between local and foreign investors.