Premium
Is there Really No Conglomerate Discount?
Author(s) -
Ammann Manuel,
Hoechle Daniel,
Schmid Markus
Publication year - 2011
Publication title -
journal of business finance and accounting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.282
H-Index - 77
eISSN - 1468-5957
pISSN - 0306-686X
DOI - 10.1111/j.1468-5957.2011.02261.x
Subject(s) - economics , enterprise value , diversification (marketing strategy) , agency cost , leverage (statistics) , endogeneity , shareholder , debt , shareholder value , financial economics , equity value , conglomerate , cost of capital , monetary economics , microeconomics , business , econometrics , accounting , corporate governance , finance , incentive , external debt , marketing , machine learning , debt levels and flows , computer science , paleontology , biology , sedimentary rock
Recent research questions the existence of a conglomerate discount. This study addresses two of the most important explanations for the conglomerate discount and finds evidence in support of an economically and statistically significant discount. The first explanation is that the risk‐reducing effect of diversification increases debt value and consequently the use of the book value of debt leads to an underestimation of firm value in diversified firms. We show that the effect of using an imputed market value of debt reduces the conglomerate discount only by a small fraction. However, consistent with the value‐transfer hypothesis, we find the discount to increase in leverage and no discount for all‐equity firms. An agency cost‐based explanation, which reconciles these conflicting findings, is that managers in levered firms become aligned with creditors and reduce firm risk at the expense of shareholders. Hence, the diversification discount only occurs in levered firms and stems from conflicts of interest between managers and shareholders over corporate risk taking. Second, the conglomerate discount may emerge from a neglect of the endogenous nature of the diversification decision. We first show that the conglomerate discount in fact disappears when we account for endogeneity in a Heckman selection model. However, when we account for fixed effects, the conglomerate discount remains statistically and economically significant, also in a Heckman selection‐model or instrumental variables framework.