published on line : 2011/01/08International audienceThis study examines the effects of customer satisfaction on analysts' earnings forecast errors. Based on a sample of analysts following companies measured by the American Customer Satisfaction Index (ACSI), we find that customer satisfaction reduces earnings forecast errors. However, analysts respond to changes in customer satisfaction but not to the ACSI metric per se. Furthermore, the effects of customer satisfaction are asymmetric; for example, analysts are more willing to use good news (i.e. an increase in customer satisfaction information) than bad news (i.e. a decrease in satisfaction). Similarly, customer satisfaction reduces negative deviation more than positive deviation of the analysts' forecasts from actual earnings. Furthermore, the effects of customer satisfaction depend upon the base level of satisfaction that the firm has achieved. Finally, the effects of customer satisfaction on analysts' forecast errors differ across firms with volatile satisfaction scores and those with stable satisfaction scores. We discuss the implications of our results for marketers and participants in financial markets

No abstract

Jean-François CASTA Cédric LESAGE 8.2.1. Financial Statements and the fundamentals of measurement in accounting Our discussion concerns the operating rules governing quantification used in accounting. These rules are based on a rigorous conception of "numericity" which relates back to a given state of mathematical technology-a truly invisible technology-linked to the concept of measurement used.

No abstract

We examine the consequences on impairment testing disclosures of auditor-pair choice made by French listed companies where two (joint) auditors are required by law. Managers are likely to manipulate impairment-testing disclosures since it relies on unverifiable fair value estimates (e.g., goodwill). From a simple game theory model, we demonstrate that a Big-4 auditor paired with a non-Big 4 auditor increase auditors' incentives to force firms to disclose more because Big 4 auditor fully bears reputation costs. Using a disclosure score for firms composing the French SBF 120 index from 2006 to 2009, we provide evidence that combination of Big 4 / non-Big 4 auditors generate higher impairment-related disclosures levels whereas the other combinations, i.e. two Big 4 or two non-Big 4, tend to decrease the level of impairment-related disclosures. These empirical results are consistent with our model predictions and robust to various controls variables (e.g., size, risk, year and firm fixed effects).