Proper Inferences or a Market for Excuses? The Capital-Market Effects of Mandatory IFRS
|Speaker:||Professor Christian Leuz, The University of Chicago Booth School of|
|Date:||Tuesday 8 October 2013|
|Time:||14:00 - 15:30|
|Location:||Bateman Lecture Theatre|
Barth and Israeli (2013) raise five serious concernsregarding the research design and interpretation ofChristensen, Hail, and Leuz (2013). They claim: (i) theevidence stands in stark contrast to Daske, Hail, Leuz,and Verdi (2008) and fails to replicate its priorfindings; (ii) the research design using fixed effectsleaves out main effects and two-way interactionswhich likely biases the estimated liquidity effectsaround IFRS adoption and changes in enforcement;(iii) the vast majority of sample observations do notcontribute to the identification which is misleading interms of the scope and the conclusions that can bedrawn from the study; (iv) the timing of IFRSadoption and enforcement changes is measuredimprecisely leading to low power tests; and (v) theevidence from Japan is irrelevant to the study. In thisnote, we show that all five claims are incorrect ormisleading. Our discussion also more broadlydescribes how to correctly interpret the fixed-effectspecifications in Christensen, Hail, and Leuz (2013).Since studies in accounting, finance, and economicsmake extensive use of fixed-effect models, a properunderstanding of this research design is important toavoid the interpretational mistakes of Barth andIsraeli. More generally, we discuss that properempirical identification and inferences are importantto international accounting research and IFRS studiesso that this area does not become a market forexcuses.