The determinants and effects of corporate diversification
Liebenberg, Andre Peter
MetadataShow full item record
This dissertation investigates the determinants and effects of corporate diversification using a sample of property-liability (P/L) insurers over the period 1995 to 2002. First, we canvas the extant literature across several disciplines in order to identify theoretical explanations for why managers diversify their fi rms. Prior research provides three prominent explanations for corporate diversification: the agency hypothesis, the efficiency hypothesis, and the coinsurance hypothesis. We test the ability of these explanations to explain observed variation in diversification status, extent, and strategy among P/L insurers. Our results suggest that existing theory is at least partially successful in explaining variation in diversification levels and strategies among insurance firms. Although the agency and efficiency view s are more successful than the coinsurance view in explaining both total and unrelated diversification, we are unable to find unambiguous support for either of these views. We do, however, find support for our reformulated managerial discretion hypothesis in the sub-sample of unaffiliated insurers. Next, we review theory and evidence regarding the performance effects of diversification. The strategic focus hypothesis predicts a negative relation between diversification and performance while the conglomeration hypothesis predicts a positive relation. We develop and test a model that explains performance as a function of line-of-business diversification and other correlates. We consistently find that undiversified insurers outperform diversified insurers in t erms of both ROA and ROE. Our results indicate that diversification is associated with a penalty of at least 1% of ROA or 3.5% of ROE. When we confine our analysis to diversified firms we find a negative relation between the extent of diversification and both ROA and ROE. While there is some evidence suggesting a nonlinear, U-shaped, relation between the extent of diversification and performance we find that highly diversified insurers do not outperform their more focused counterparts. Taken together, our findings provide strong support for the strategic focus hypothesis.