Table II. Descriptive statisticsa
Notes: aThe sample consists of 267 French listed companies selected from the Worldscope database over 1990-1999, i.e. 2,670 firm-year observations. Any firms with missing values are eliminated to facilitate comparability; Return ¼ annual investment return; NI ¼ annual earnings per share before extraordinary items/preferred dividends
index is less exposed to international market fluctuations than a company listed abroad but in a less important market. International investors own almost 40 per cent of the share capital of French CAC 40 companies (the 40 largest French listed companies making up the CAC 40 Index). We therefore chose to use stock index listing as a proxy for the degree of international financing.
A dummy variable is used dividing our sample into two groups: equal to 1 if a company belongs to at least one of the following three stock indexes: DJ Global, FTSP World or CAC 40; 0 otherwise. The only exception is Bull Soci´et´e Anonyme, which figures in both the New Zealand Stock Exchange 40 (NZSE) Index and in the Paris SBF 120 Index. Besides Paris, the company is also listed in Germany (Frankfurt, Berlin, Munich, Dusseldorf) and Switzerland (Zurich). We thus decided to include Bull in the International Financing group.
Auditing firms. The sample is divided into two sub-groups according to their auditors. If a company is audited by at least one Big Five accounting firm, the value of the dummy variable is 1, otherwise it is 0.
5. Results Using the simple or multiple regression and average t-test methods, we test our seven hypotheses one by one.
First, we will consider H1. Using the Ball and Brown (1968) model NIit ¼ 0 þ 1 R i t þ " i t f o r t i m e l i n e s s a n d t h e B a s u ( 1 9 9 7 ) m o d e l N I i t ¼ 0 + 1 R D i t þ 2 R i t þ 3 R i t R D i t þ " i t f o r c o n s e r v a t i s m , w e o b t a i n t h e f o l l o w i n g r e s u l t s . C o m p a r i n g R 2 i n t h e t i m e l i n e s s a n d c o n s e r v a t i s m m o d e l s , a c l e a r i m p r o v e m e n t c a n b e o b s e r v e d i n t h e s e c o n d ( F i g u r e 1 ) . T h e a v e r a g e s f o r R 2 a r e r e s p e c t i v e l (timeliness) and 0.139 (conservatism) (Table III). The t-test produces a t equal to 3.974 (0.003). The validity of Basu’s model is thus confirmed for the French context: The published earnings of French companies are timelier in reflecting publicly available bad news than good news. y 0 . 0 8 2
However, the above graph also indicates that there is no clear upward trend in the R2 using Ball and Brown’s model. The simple regression of timeliness R2 over the