Do Financial Markets Discipline Corporate Maleficence by Driving Down Stock Prices ? Tackling Money Laundering Conference Utrecht, 2-3 November 2007 Peter-Jan.

Do Financial Markets Discipline Corporate Maleficence by

Driving Down Stock Prices ?

Tackling Money Laundering Conference

Utrecht, 2-3 November 2007

Peter-Jan EngelenUtrecht University, the Netherlands

p.engelen@econ.uu.nl

Slide nr.2

Background of the project

Relationship between discovery of illegal corporate behaviour and stock prices

Do shareholders punish companies by driving down stock prices?

Is there any price for corporate maleficence?

Disciplinary roleMagnitude of penalty

Slide nr.3

Background of the project – cont’d Event study methodology

Campbell, Lo and MacKinlay (1997), Chap.4

MacKinlay (JEL, 1997)McWilliams and Siegel (AMJ, 1997)Armitage (JES, 1995)

Two exploratory studiesLow countriesSample of 5 European countries

Slide nr.4

Methodology Event study

AARE aggregate of individual abnormal stock returns aligned in event time

Calculating individual ARs:

0:

0:

1

0

E

E

AARH

AARH

N

iEiE AR

NAAR

1,

1

tititi RERAR ,,,

Slide nr.5

Benchmark expected return models Market-adjusted model

Market model

Dimson model for thin trading correction

tmtiti RRAR ,,,

tmiititi RbaRAR ,,,ˆˆ

tmDi

Dititi RRAR ,,,

ˆˆ

iiiiiDi bbbbb ,2,1,0,1,2

ˆˆˆˆˆˆ

23

138,

23

138, 116

1ˆ116

1ˆ

t

ttm

Di

t

tti

Di RR

Slide nr.6

Test statistics

Traditional t-test of Brown and Warner (1985)

1~1,

Nt

N

SAR

testt

N

iEi

i

EiEi s

ARSARwith

ˆ,

,

Slide nr.7

Test statistics – cont’d

Potential problemsEvent-induced variance

• Variance during event window exceeds variance over estimation period

Thin trading• Non-normal return distribution

Traditional test statistics might be misspecified

Non-parametric tests do not depend on assumptions about probability distribution of returns

Exploratory study 1

Slide nr.9

Sample description

Preliminary study

The Low Countries (B, NL)

Listed on Euronext Brussels or Amsterdam

1994-2003

Public announcement of 57 cases of corporate malconduct

Leading financial newspapers (FD, FET)

Slide nr.10

Scopei. Impact of different types of illegal

behaviour Insider trading, corruption, tax fraud,

accounting fraud, miscellaneous

ii. Impact of company versus individual level

iii. Impact of phase Rumour Formal investigation (police, judicial)

iv. Impact of direct versus indirect effect Bottom line (direct) Reputation (indirect)

Slide nr.11

Hypotheses Hyp.1 – Stock prices of listed firm show a

negative abnormal return upon the announcement of the corporate malconduct

Hyp.2 – A value-impact corporate malconduct exhibits a larger negative abnormal return of stock returns than a maleficence with only an impact on the trust of shareholders

Hyp.3 – Corporate malconduct at the firm level has a larger negative abnormal return than at the individual level

Hyp.4 – The further corporate maleficence is along the formal investigation procedure, the larger the abnormal negative return

Slide nr.12

Empirical results – Total sample (Madj)

Day N AAR t-value-2 57 -0.73% -1.96-1 57 -0.41% -1.110 57 -0.94% -2.54*

+1 57 -1.03% -2.78**

+2 57 -0.13% -0.35+3 57 -0.09% -0.24+5 57 0.91% 2.45

Slide nr.13

Empirical results – subsamples Corruption

Only day 0 sign. at 5% level using MM (-1.77%) Other days no significant ARs

Tax fraud Sign. at 5% level at day 0 (-0.99%) Sign. at 1% level at day [+1] (-3.55%) (-

4.54%)

Insider trading Not sign. at day 0 (-0.66%) Sign. at 1% level at day [-1] (-2.13%)

Slide nr.14

Empirical results – subsamples

Accounting fraudSign. AR at day [-2] at 1% level (-10.40%)

Miscellaneous Sign. AR at day 0 at 1% level (-1.20%)

Slide nr.15

Empirical results – subsamples

StadiumNo price reaction for rumoursSign. neg. AR for court phase

Firm vs. individual levelNo difference

Bottom-line vs. trustHigher impact for bottom-line events

Exploratory study 2

Slide nr.17

Sample description Public announcement of 239 cases of

corporate malconduct Leading financial newspapers 1995-2005 Five countries

Belgium, France, Germany, the Netherlands, UK

Five types of corporate malconductInsider trading, tax and accounting fraud,

bribery, price fixing and market power abuse, miscellaneous

Slide nr.18

Sample description – cont’dCountryFrance 51Belgium 18UK 63The Netherlands 43Germany 64TypeInsider trading 62Fraud 51Bribery 23Price fixing 93

Slide nr.19

Empirical results (Madj)all

(239)

fraud

(51)

bribery

(23)

price

(93)

insider

(62)

-2 -0.09%

+0.17%

+0.01%

-0.20%

-0.28%

-1 -1.04%

+1.13%

-0.25%

-0.96%

-2.45%

0 -0.08%

-1.98%

+0.20%

+0.29%

-1.97%

+1 +0.06%

-0.06%

+0.15%

-0.13%

+0.60%

+2 -0.18%

-0.11%

-0.28%

-0.12%

-0.89%

[-5,+5]

-2.07%

-1.20%

-1.15%

-0.64%

-5.36%

Slide nr.20

Empirical results (Madj) – cont’d

All

(239)

France

(51)

Germ.

(64)

UK

(63)

Nether.

(43)

Belg.

(18)

-2 -0.09%

-0.23%

-1.14%

+0.23%

+0.57%

+1.67%

-1 -1.04%

-0.40%

-1.17%

-0.98%

-2.08%

-0.38%

0 -0.08%

+0.06%

-1.85%

-0.56%

-0.40%

-1.21%

+1 +0.06%

-0.19%

+0.82%

+0.09%

-0.43%

-0.89%

+2 -0.18%

-0.37%

-1.28%

+0.59%

+0.45%

+0.27%

[-5,+5]

-2.07%

-1.63%

-5.51%

+0.18%

-2.23%

+1.53%

Slide nr.21

Further research Larger sample with richer taxonomy of

corporate malconduct

Cross-country (five countries) – cultural differences

Differences in types across countries

Differences pre and post Enron (mental framing) or other time-effects

Interpretation and consequences of the results for business ethics