Ajlouni, Moh'd Mahmoud
(2004)
Insider trading on the UK stock market: information contents and managerial incentives.
PhD thesis, University of Nottingham.
Abstract
This study investigates the managerial incentive of insider trading. A research subject that has not been empirically examined in the UK yet. In doing so, this study merges two parallels but related tracks: The information contents of insider trading which is a Finance subject backed by the Efficient Market Hypothesis (EMH) and the managerial incentives of insider trading which is a Management subject backed by the Agency Theory. In fact, the managerial incentives of insider trading only becomes testable once there is evidence that insider trading is profitable.
In detecting the information contents of insider trading, this research differs from previous literature in that (1) it employed three signal definitions, (2) it used dally data, (3) it used the security' return index, instead of share prices, (4) it used a most recent set of data, (5) it used the Capital Asset Pricing Model (CAPM) to estimate the expected returns, (6) it reports the results within a shorter event window and (7) it provides, for the first time, empirical evidence on the Executive Share Options (ESO) transactions, in addition to ordinary shares. In terms of the managerial incentive part of this research,a major contribution of this research is that it provides evidence for the first time on the managerial aspects of insider trading by directors in the UK.
The analysis has examined the short-term profitability of FTSEIOO directors trading in their own firm's ordinary shares and executive shares options, over recent years (1999 to 2000). The empirical results of this study shows that except for the executive share options portfolios, the empirical results significantly reject the null hypothesis that directors trading in their own company's securities are not profitable. Instead they suggest the alternative hypotheses that directors buying portfolios achieve positive abnormal return and those of selling ones avoid negative abnormal returns. The results of this study have been checked on by re-running the analysis taking into account thin trading, confounding events, year-by-year analysis, and firm size. The robustness check analysis shows that there is no thin trading problem in the sample securities, no firm key event announcement conoccured with the director dealing transactions, thus the director dealing abnormal returns are not a result of other events. In addition, smaller firms outperform larger ones, particularly in the longer event windows of buying transactions and directors buying in the year 2000 outperform those in 1999. However the year 1999 was a successful selling year for FTSE 100 directors.
Two important conclusions are suggested by employing different signal methodologies These are first, different signal definition produce different results, not only in terms of the level and sign of cumulative abnormal return (CAR), but also in terms of the significance of the statistical results. On one hand, multiple (MS) and quantitative (QSs) signals produced significant CARs at earlier days than single signal (SS). This leads to suggest that the market reacts significantly sooner to successive signals than to a single signal. On the other hand, none of the other signals produce significantly results that reject (accept) what has been accepted (rejected) by the single signal. The results of three tests, parametric (standard errors) and non-parametric (Wilcoxon Signed Ranks Test), about the equality of the means of CARs of SS and of MS, QSI, QS2, QS3 and QS4 might lead to suggest that while the magnitude of CARs across various signals is identical, the time when these CARs becomes significant is varied. CARs of compounded signals (MS and QSs) are significant at earlier days in the event window, while those of SS are significant at later days in the event window. This suggests that none of other signals can be considered as an alternative or a counterpart to the single signal with daily data. Second, each signal definition requires certain data frequency. Single signal produces robust results when daily data are used, while those of multiple and quantitative signals are mixed. Monthly data is recommended with multiple signals, whereas both monthly and daily data can be used with quantitative signal.
In the context of EMH, the empirical result of this study shows clearly that the stock exchange is significantly inefficient in terms of the strong level of market efficiency. On the other hand, the availability of abnormal returns to outsiders following the publicly known information, i.e. insiders' transactions, can be seen as a direct test to the semi-strong level of market efficiency. The empirical results indicate that abnormal returns can be earned by outsiders' imitating insiders' transactions. However, taking into account the transaction cost, such returns would end up with zero, if not negative returns.
In terms of the managerial incentive of insider trading, the model states that as the number of insiders in the firm increases, competition to insider trading increases and each insider's expected returns decreases. On the other hand, as number of insiders' increases, the explicit form of director's compensation should increase to offset his insider trading returns decreases. This concept leads to an empirically testable assumption that the director's expected compensation has two forms, (1) an explicit form (salary, bonuses, perks and other ex ante measurable incentives) which is predicted to be positively correlated with the number of insiders; and (2) an implicit form (his expected insider trading returns) which is predicted to be negatively correlated with the number of insiders in his firm.
The empirical prediction, presented in a multivariate model, was tested using, FTSE 100 chief executive officers (CEO) data. The data and the justification behind using each and every dependent and independent variable in the test is discussed. Also, the variables for the regression model, which is used to explain the relationship between the CEO explicit form of compensation (dependent variable) and the number of insiders in his firm, as well as his personal and job characteristics (independent variables), are analysed and assessed.
The results were very much in favour of the model. The positive relation between the explicit forms of CEO compensation and the number of insiders in the UK FTSE100 firms was found to be quite robust. That is, the significant relation does not depend on (i) whether the model accounts for CEO internal experience and industry sector, (ii) whether the model accounts for CEO's capacity to trade and his actual insider trading returns and (iii) whether an omitted variables problem is accounted for by using panel data. This leads to conclude that insider trading is an integral part of the director's total compensation package, and thus, can be considered as a managerial incentive.
A by-product finding from the analysis indicates that there is an indication that the labour market for top management in FTSEIOO might not be competitive. This conclusion is brought about by the positive association found between director's pay and his realized insider trading returns. However, this conclusion is subject to the definition of CAR used in the analysis.
| Item Type: |
Thesis (University of Nottingham only)
(PhD)
|
| Supervisors: |
Fenn, P.T.
Toms, J.S. |
| Keywords: |
London Stock Exchange, Insider trading in securities,Great Britain, Directors of corporations |
| Subjects: |
H Social sciences > HG Finance |
| Faculties/Schools: |
UK Campuses > Faculty of Social Sciences, Law and Education > Nottingham University Business School |
| Item ID: |
11093 |
| Depositing User: |
EP, Services
|
| Date Deposited: |
28 Jan 2010 10:51 |
| Last Modified: |
17 Oct 2017 06:16 |
| URI: |
https://eprints.nottingham.ac.uk/id/eprint/11093 |
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
