«Occasional Paper Series 23 ◆ Measuring market cleanliness Ben Dubow Nuno Monteiro Financial March 2006 Services Authority FSA OCCASIONAL PAPERS IN ...»
Occasional Paper Series 23
FSA OCCASIONAL PAPERS IN FINANCIAL REGULATION
We are committed to encouraging debate among academics,
practitioners and policy-makers in all aspects of financial regulation.
To facilitate this, we are publishing a series of occasional papers in
financial regulation, extending across economics and other disciplines.
These papers cover topics such as the rationale for regulation, the costs and benefits of various aspects of regulation, and the structure and development of the financial services industry. Since their main purpose is to stimulate interest and debate, we welcome the opportunity to publish controversial and challenging material and to publish papers that may have been presented or published elsewhere.
The main factor in accepting papers, which will be independently refereed, is that they should make substantial contributions to knowledge and understanding in the area of financial regulation. We encourage contributions from external authors as well as from within the FSA. In either case, the papers will express the views of the author and not necessarily those of the FSA.
If you want to contribute to this series, please contact Peter Andrews
or Maria-Jose Barbero at:
The Financial Services Authority, 25 The North Colonnade, Canary Wharf, London E14 5HS.
Telephone: (0)20 7066 3104 or 5808.
Email: firstname.lastname@example.org or email@example.com.
FSA Occasional Papers are available on our website: www.fsa.gov.uk.
We welcome comments on these papers; please address them to the contacts listed above.
NUNO MONTEIROFSA Occasional Paper © March 2006 Biographical note Ben Dubow and Nuno Monteiro are members of the Economics of Financial Regulation Department within our Finance, Strategy and Risk Division.
Acknowledgements We are grateful to our referee Kevin R. James of the London School of Economics for his detailed advice and comments on the methodology. We are also grateful to Ian Tonks, Isabel Argimon, Mohamed Serokh, Isaac Alfon, Lionel Stehlin, Hector Sants, Gay Huey Evans, Dilwyn Griffiths, Ruth Dent, Carlos Conceicao, Peter Andrews and Kari Hale. Any errors or omissions are, however, the sole responsibility of the authors.
1. Summary 5
2. Introduction 7
3. Key concepts 8
4. The impact of FSMA 12
5. Data and method 14
6. Results 21
1. Summary The FSA has a Statutory Objective to maintain confidence in the financial system.
The FSA’s business plan for 2006/07 notes that all that we do, whether in wholesale or retail markets, reflects our belief that efficient, orderly and fair markets are the most efficient way of delivering value to both users and providers of financial services. The FSA has primary responsibility for tackling market abuse in the UK and we consider the successful reduction of market abuse to be one of our highest priorities. This has been emphasised by FSA Senior Management in recent speeches.1 The FSA has committed to evaluate its overall performance and this piece of economic analysis looks to establish a robust methodology to measure the level of market cleanliness. In selecting this measure, we have sought an outcome-based approach. Our methodology aims to assess the overall level of market cleanliness and the deterrent effect of FSA regulation. This may be more informative than, for example, looking at the number of successful Enforcement cases, which provides no direct information about the impact the FSA is having on the level of market abuse.
This paper develops measures of market cleanliness based on the extent to which share prices move ahead of the regulatory announcements which issuers are
required to make to the market. We examine two broad kinds of announcement:
announcements relating to take-over bids using data for 2000 and 2004 and announcements about the trading performance of FTSE350 listed issuers using data from 1998 to 2003. Share price movements ahead of such announcements may reflect insider trading.2 The paper does not examine other forms of market abuse.3 We estimate statistical relationships between the returns to individual stocks and the return to the market as a whole. These relationships can be used to identify whether the returns to an individual stock at key times are “abnormal” in a purely 1 See e.g. Speech by Margaret Cole, Director of Enforcement, FSA to SII Compliance Forum, 18 January 2006 2 Throughout this paper the term “insider trading” is used to mean acting or causing others to act on material non-public information which could affect the value of an investment. This term is not a legal one but is intended to include the UK legal offences of insider dealing and misuse of information.
3 We believe that examining share price movements ahead of announcements is much less informative about the overall prevalence of other forms of market abuse, e.g. market manipulation.
Measuring market cleanliness
statistical sense. We interpret a large abnormal return around the time of a regulatory announcement as indicating that the announcement contained important news about the stock’s value; news which could be of interest to an insider trader.
We refer to these regulatory announcements as “significant announcements”.
Our measure of market cleanliness is based on the proportion of significant announcements where the announcement is preceded by an “informed price movement” (IPM). We define an IPM as an instance where there is an abnormal stock return before an announcement and that return is positive in the case of a good news announcement or negative in the case of a bad news announcement. We assert that IPMs can indicate insider trading, while asserting neither that most insider trading gives rise to IPMs nor that IPMs arise only as a result of insider trading.
We examine the measure for periods before and after 2001, when prosecuting the relevant forms of market abuse was made easier and greater penalties were introduced. Our analysis indicates that there was no change after 2001 in the level of market cleanliness in relation to the announcements made by FTSE350 issuers.
For another analysis performed using announcements of take-overs, the measures we calculate provide some evidence of a deterioration in market cleanliness.
A previous study by Bhattacharya and Daouk (2002) suggests that enforcement, rather than the existence of rules alone is required for regulation to improve market cleanliness. So a possible explanation for the measure failing to decline for the FTSE350 announcements could be that the first successful enforcement against insider trading under the new post-Financial Services and Markets Act (FSMA) regime did not take place until 2004. However, the analysis of take-overs did include data for 2004, after enforcement had taken place. This may suggest that the scale of the fines (£1,000 to £18,000) and nature of the cases in fact had a limited deterrent effect.
We believe that following detailed examination of alternative approaches, we have developed a useful measure of market cleanliness. Although changes in the measure from year to year may not show up as “statistically significant”4, we believe it may be useful to repeat the analysis presented in this paper in future.
4 The confidence intervals presented in section 9 below imply that with a 95% confidence interval an increase of at least 7.9% would have been required for us to find that the change in the measure associated with the FTSE350 announcements was significant; for the take-over announcements the figure is 4.1%.
2. Introduction The FSA has a Statutory Objective to promote market confidence. That objective requires the FSA to keep markets “clean” – i.e. free from market abuse and the FSA has primary responsibility for tackling market abuse in the UK.
But how clean are UK markets? And has the tightening of regulation in recent years had a visible effect – not just on the number of punishments meted out, but on the way markets perform? This paper reports work to develop such an “outcome-focused” measure of market cleanliness and to assess whether the Financial Services & Markets Act5 (FSMA), which came into force in 2001, had an impact on that measure.
The scope of the paper is limited in at least three ways. First, given the data available and existing techniques at our disposal, we have focused on one form of market abuse – insider trading – and have not looked at other forms of market abuse (i.e. market manipulation).6 Primarily, this reflects the fact there is a substantial economic literature investigating the impact of insider trading on markets which provides techniques directly applicable to our research aim. Secondly, we only look at “cash” equities, rather than their derivatives, including instruments such as contracts for difference. We explain below why we do not believe this to be a genuine limitation. Thirdly we are not seeking to measure the benefits of insider trading regulation. We note only that financial markets are built on trust and that insider trading erodes that trust and can increase costs for all market participants.7 The remainder of this paper is set out as follows. Section 3 introduces the key statistical and regulatory concepts required to understand, at an intuitive level, the measure of market cleanliness we have developed. Section 4 explains the regulatory changes which occurred in 2001 and discusses whether those should have changed the behaviour of market participants at that time so as to affect our measure of market cleanliness. Section 5 explains the data and method we have used. Section 6 presents the results and Section 7 the conclusions. Section 8 provides some descriptive statistics on our sample while Section 9 provides a more precise explanation of method and results for readers with some background in statistics.
5 and in particular the offence contained in section 118(2)(a) 6 Market manipulation can involve spreading false information about the value of a security or trading aimed at manipulating prices. Examples of the latter include “price positioning” and “abusive squeezes” and are given in FSA HANDBOOK MAR 1.6.12 and 1.6.18 respectively.
7 For a summary of the welfare effects of insider trading see Minenna (2003) Measuring market cleanliness
3. Key concepts All the analysis in this paper revolves around the identification of “abnormal” movements in the return to an individual stock around the time of the announcements which issuers are required to make to the market under the FSA’s Listing Regime.8 To identify whether stock returns are abnormal, we need a statistical model of normal or “expected” returns. The details of our model for expected returns are given in section 5 below but here we note only that the model can take into account movements in the market as a whole. The abnormal return on a given day is the difference between the expected return from our model and the actual return. By adding together abnormal returns over time we calculate cumulative abnormal returns (CARs)9. These concepts are illustrated in the diagram below.
Figure 1: Key concepts illustrated
8 Most relevant here is Disclosure Rule 2.2.1 which imposes a general obligation on issuers to publish important information as soon as possible.
9 If the actual return is equal to the expected return the CAR is equal to zero.
The flat dotted line along the bottom of the chart represents the expected stock price.
At a point in time represented by the vertical line, the firm makes a regulatory announcement to the market. Because this announcement contains new information (in this case good news) about the performance of the firm (and hence the firm’s value) the stock price increases. This gives rise to a positive post-announcement CAR.
These concepts can be used to identify possible instances of insider trading as illustrated in Figure 2 below. Here, a positive abnormal price movement occurs before the announcement of good news (a “positive pre-announcement CAR”). This could indicate that the positive news in the announcement had been traded on before it had been made public, in breach of FSMA. The analysis we will present neither requires us to assert that all such price movements must be the result of insider trading nor that all insider trading must give rise to such price movements.
Figure 2: Key concepts and insider training
We have now introduced all the concepts required to explain intuitively the measure we have developed. The first step is to identify a set of regulatory announcements which we know to contain a lot of new information. We can identify such announcements where the associated CARs are so large that the probability that these CARs are simply the result of random variation in stock returns is very small.
In fact, we require CARs to be so large that the probability of them arising purely by chance is just 1%. That is to say these CARs are “statistically significant at the Measuring market cleanliness 1% level”. To identify these relatively informative announcements we examine the sum of the pre-announcement and post-announcement CAR (the “total CAR”).10 We refer to the announcement identified in this way as “significant announcements”.