Markus Brendel, Christin Rudolph, Bernhard Schwetzler

• We promote the Oaxaca-Blinder decomposition as a new empirical approach to corporate finance-related research. Originated in labor economics, its primary field ofapplication is the examination of gender and race-related wage gaps. Allowing for an in-depth analysis of factors driving valuation differences between two distinctive groups, the method likewise provides an effective tool for corporate finance topics. We demonstrate its usefulness on the basis of an old yet still up-to-date problem, namelythe value discount associated with corporate diversification. In particular, we aim to disentangle how the different agency conflicts - the one between the corporate manager and shareholders and the other between majority and minority shareholders - work on the discount. Using a sample of CDAX firms from 2000 to 2009, we find the latter conflict to be the driving agency-related cost.