This paper examines bubbles on the JSE All Share Index as well as the critical time of the stock market crash from 2/01/ 2004 – 27/03/2014. The underlying hypothesis define bubbles as extreme and begin as a group of small events which grow in a super exponential form explained by a log periodic power law model (LPPL model). The hypothesis is based on the assumption of investors’ herding behavior, where investors collude by making investment decision correlated with their counterparties. The paper implements a Savitzky Golary Algorithm to detect peaks and calculate the critical time of the crash from the peaks. An Ordinary Least Squares (OLS) method is used to determine both the value of stock market price index at the critical time and the increase in the stock market price index over the time before the crash. The remaining parameters of the LPPL model are estimated using a Maximum Likelihood Estimation method. On the empirical results; 68 peaks were detected, and the LPPL model at the critical crash time is estimated 34736.586. Five bubbles are detected; the 15/8/2005 bubble, 28/5/2013 bubble, 23/8/2013 bubble, 5/11/2013, and 1/20/2014.

Empirical evidence indicates that companies that reduce information asymmetry by increased voluntary disclosures achieve several benefits, such as lower cost of capital, improved pricing, and liquidity of their shares. Despite the possibility of such benefits, many studies report varying degrees of voluntary disclosure behaviour that is attributable to various factors. Recent studies indicate that investors’ investment horizon has a significant effect on actions taken by management. Companies with predominantly short-horizon investors spend less on research and development, invest in shorter-term projects that are less profitable than longer-term projects, and are more likely to manipulate earnings to meet short-term earnings expectations. This study investigates whether investors’ investment horizon has an effect on the quality of companies’ information environment. Long-horizon investors should be familiar with their investee company’s risks and rewards, using both their own internal information gathering processes and the cumulative information disclosed by management over time. Moreover, over the course of a long-term relationship, they can become familiar with management’s capability to deliver long-term sustainable returns. Long-horizon investors should therefore be less concerned with short-term fluctuations of earnings and management’s public explanations and disclosures thereof. I hypothesise that higher (lower) proportions of long-horizon investors are associated with lower (higher) quality voluntary disclosure. The shareholder familiarity hypothesis was tested in this study, using an ordinary least squares regression. Voluntary disclosures were observed via the channel of companies’ websites. A checklist was compiled of best practices for online investor relations, and content analyses were conducted on the websites of 205 companies listed on the Johannesburg Stock Exchange. Shareholder familiarity was proxied by shareholder stability, measured over nine years. The stability measure was lagged by one year to create a temporal difference between the shareholder profile and disclosure behaviour. I found that companies with a profile of unstable investors that are larger, younger, dual-listed and have a Big4 auditor have higher quality online investor relations practices. The hypothesis of a negative association between shareholder familiarity and voluntary disclosure quality is therefore accepted. This study extends the theory on information asymmetry and voluntary disclosure by providing evidence supporting the argument that investor horizon is a predictor of voluntary disclosure quality. The dictum of more is better does not hold in all scenarios. It is important for financial directors and investor relations officers to establish the investment horizon profile of their respective companies’ shareholders before they embark on extensive disclosure programmes. / Financial Intelligence

The disclosure of headline earnings is one of the JSE Limited (JSE)’s listing requirements. All listed entities are required to comply with this disclosure requirement. Guidance in the form of The Circular on headline earnings is issued by the South African Institute of Chartered Accountants (SAICA), and is updated every time when there are changes to the International Financial Reporting Standards (IFRS). The Circular adopts a rules-based approach and specifies what is included and excluded in the calculation of headline earnings. The rules consist of general rules, which apply to all entities other than those industry groups with special provisions such as the life insurance entities. This study questions whether a rules-based headline earnings approach or a principles-based approach is more appropriate for the calculation of headline earnings of listed property entities on the JSE, for economic decision-making purposes. The research method consisted of questionnaires that were sent out to stakeholders. The responses from the CFOs and investment analysts show that principles-based headline earnings guidance is preferred over rules-based headline earnings guidance. / Business Management / M. Phil (Accounting Sciences)

The disclosure of headline earnings is one of the JSE Limited (JSE)’s listing requirements. All listed entities are required to comply with this disclosure requirement. Guidance in the form of The Circular on headline earnings is issued by the South African Institute of Chartered Accountants (SAICA), and is updated every time when there are changes to the International Financial Reporting Standards (IFRS). The Circular adopts a rules-based approach and specifies what is included and excluded in the calculation of headline earnings. The rules consist of general rules, which apply to all entities other than those industry groups with special provisions such as the life insurance entities. This study questions whether a rules-based headline earnings approach or a principles-based approach is more appropriate for the calculation of headline earnings of listed property entities on the JSE, for economic decision-making purposes. The research method consisted of questionnaires that were sent out to stakeholders. The responses from the CFOs and investment analysts show that principles-based headline earnings guidance is preferred over rules-based headline earnings guidance. / Business Management / M. Phil. (Accounting Sciences)