Survival and Conglomerate Value

This paper shows how a selection bias generates an apparent value reduction for firms that survive longer in the sample. We test our model on conglomerate firms and find that this bias reduces the value of surviving conglomerates with lower default probability relative to those with higher default probability. The data also indicate that only conglomerates closest to distress have the same value as their focused peers. We show that the selection bias disappears once we control for firm differential ability to survive in the sample. We thus shed new light on the effect of survival-enhancing coinsurance on firm value.