Are Conglomerates an Ancient Thinking: Should Companies Diversify or be Undiversified

Caspar Grauballe Nielsen

Student thesis: Master thesis


Conglomerates has historically been forged due to ownership type and antitrust legislations – combined with a historic limitation of benchmark and peer-group assessment data. Using the panel data models pooled OLS, random effect, and between effect on the most recent data, with regional diversification, this study addresses the conglomerate discount phenomenon. Based on data of 450 of the largest companies in the world, over the last eleven years, and 4950 variables in total, the thesis finds that there is a slight conglomerate discount in total shareholder return, as well as return on invested capital. The thesis finds four significant findings being 1) the rejection of the hypothesis that return on invested capital distribution for diversified and undiversified companies is the same, 2) that the conglomerate discount in the dataset is below recent literature findings on the topic, but indeed present, 3) that contrary to theory, conglomerates do not perform better in financial crisis’ and 4) that Asian conglomerates do not thrive better than European and North American conglomerates. These findings indicate that the pros of conglomerates, such as internal capital markets, economies of scale -and scope and diversification is outweighed by the cons such as agency costs, increasing complexity and lack of transparency. Companies should pursue divestment of non-core assets/businesses in order to obtain its full potential not only in the market, but also in terms of financial performance. Conglomerates are an ancient thinking belonging within history books.

EducationsMSc in Advanced Economics and Finance, (Graduate Programme) Final Thesis
Publication date2016
Number of pages100
SupervisorsCaspar Rose