Three distinct but interrelated studies with their foundations in recent developments in the South African capital markets are presented in this thesis. The first study presents an empirical analysis of the systemic risk exposures and contributions of 125 financial institutions between 2003 and 2018. Using two popular measures of systemic risk, the marginal expected shortfall (MES) and conditional capital shortfall (SRISK), it is shown that banking institutions are collectively the largest contributors to systemic risk in the financial system. Surprisingly, further analysis reveals that despite the high levels of market concentration and interconnectedness, SRISK increases are not propagated across sectors. Notwithstanding the foregoing, the results provide support for previous empirical findings of the systemic importance of banking institutions. In addition, causality analysis of the relationship between SRISK in the banking sector and the prime lending rate provides new evidence that complements previous theories of systemic risk spillovers into the real economy, specifically through lending activity. Overall, the results illustrate the potential for the use of market based measures in supporting macroprudential oversight and informing policy decisions. The second essay addresses questions related to misconduct contagion and crowding. Crowding is a form of clustering in which the behaviour of market participants leads to congestion on one side of the market, otherwise known as crowded trades. We propose a measure of crowding, based on intraday trade data and use the measure to study changes in the trading environment following allegations of misconduct. Evidence of coincidental and significant changes in crowding and trade volumes is reported in the first set of empirical results, consistent with the notion of information contagion and how firm-specific developments may have significance for other firms. More importantly, the study demonstrates empirically, that crowding increases exposure to adverse spillover effects and deteriorates liquidity in the equity market. We further contributes to the literature, by documenting novel evidence of the asymmetric effects of intraday volatility and trade volume on MES and quoted spreads, respectively, that is dependent on the crowd direction. Relative to buy-crowds, sellcrowds amplify the effect volatility has on MES and reduce the effect trade volume has on quoted spreads. In the third study, the aim is to investigate the implications of domestic crosslistings for the market quality of twenty-six firms that cross-listed between April 2018 and April 2020, following a series of amendments to legislation. Evidence of significant improvements in market quality in the six months after a cross limited, even after adjusting for market quality changes of firms that do not cross-list. Additionally, our results offer no support for the hypothesis that there is a significant difference in market quality changes observed for high and low liquidity firms, in contrast to previous cross-listing studies. Lastly, by consolidating order books across exchanges, it is shown that the price dimension of execution quality can be improved across all venues, even after controlling for liquidity characteristics. We conclude that interoperability between venues can be effective in reducing the cost of trading, and is therefore necessary for a domestic cross-listing to be worthwhile. Collectively, the findings contribute to the ongoing debate around best execution standards and inter-market competition in South Africa's equity market.
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/35591 |
Date | 26 January 2022 |
Creators | Dube, Qobolwakhe Thomas |
Contributors | Taylor, David, Georg, Co-Pierre |
Publisher | Faculty of Commerce, African Institute of Financial Markets and Risk Management |
Source Sets | South African National ETD Portal |
Language | English |
Detected Language | English |
Type | Doctoral Thesis, Doctoral, PhD |
Format | application/pdf |
Page generated in 0.0016 seconds