Introduction Although initial public offerings (IPOs) recently have attracted public and academic attention, the special case of firms founded by families or single entrepreneurs has not been investigated in depth. Given the lack of empirical research on family-controlled businesses (McConaughy et al. 2001), this study is an attempt to remedy this deficiency. For family-owned firms, a change in ownership structure by going public is a significant change in the governance of the corporation, because for the first time outside shareholders come into play. This pattern typically is to be observed for IPOs of family-owned firms in Germany. (1) Our paper merges two strands of research: on the one hand, the entrepreneurial literature on family-controlled businesses, and on the other hand, the vastly increasing research on corporate governance and IPOs. After reviewing some of the related literature, we present empirical evidence about (1) the change in corporate governance (as proxied by a change in the ownership structure) following IPOs; and (2) its impact on the long-run stock performance of family-owned firms in Germany. We conclude with a summary of the major findings. Related Literature Franks and Mayer (2001) report that family business shareholdings account for one-third of total shareholdings in Germany. They argue that family investors, since their wealth is tied directly to that of the family firm, seem to exert strong disciplining of poorly performing management. For the U.S., Kang (2000) and McConaughy et al. (2001) find that family controlled firms have greater value and are operated more efficiently than other firms. In a recent article, Bhattacharya and Ravikumar (2001) show that the development of efficient capital markets accelerates the evolution and the sale of family businesses. Thus, the public sale of family-owned firms and their resulting performance is a topic of growing importance. In Germany there were at least three factors that prohibited founding families from selling their firms during our investigation period. First, tax regulation made the sale of privately held equity stakes less attractive. Although the profit on the sale of exchange-traded shares is tax-exempt after a holding period of one year, selling stakes in nonquoted companies is taxed at high rates. Second, low competition in the underwriting market and thereby high direct costs of an IPO led to capital-raising firms preferring Hausbank-financed debt or private placements. Three, German ideas on family values and the corporate mission of family-owned firms also resulted in many owners rather not taking public their firms (Schurmann and Korfgen 1997). Goergen and Renneboog (2001) provide a comparative study on the evolution of ownership and control in German and U.K. IPOs. They found that initial shareholders (including families) lose majority control on average six years after the IPO in Germany. However, in two-thirds of their sample of 52 German IPOs, initial shareholders retain absolute majority control, and the probability of retaining control concentration augments when nonvoting shares are issued. Goergen (1998) studied ownership retention in German and U.K. IPOs, the determinants of ownership retention, and its impact on IPO long-term performance. He found that the long-run performance of IPOs is not correlated with ownership retention. Overall, he suggests that the poor long-term performance of IPOs in his sample, compared to the stock market as a whole, cannot be explained by agency conflicts caused by the reduction in ownership by the original shareholders. We focus our attention on the strategic decisions of family-owners floating their firms. We restrict our analysis for the following reasons. First, family-owned firms have long been the most typical firms in Germany to undertake a change of ownership by means of an IPO. Second, a family-owned firm comes closest to one in which ownership and control are not separated until the IPO, making it ideal for investigating the performance effect of a change in the corporate governance structure. …
Read full abstract