Takeovers and Implicit Contracts during the First Great Merger Wave: Evidence from the Australian Banking Industry

It has been argued that hostile takeovers redistribute wealth from workers to shareholders by enabling the acquiring firm to revoke implicit contracts. This paper uses micro-data from personnel records to examine the consequences of the Union Bank of Australia’s 1892 takeover of the Bank of South Australia (BSA). The evidence confirms that older workers at the BSA were harmed. I show that they faced a high probability of losing their jobs following the merger, lost specific human capital due to the closure of branches, faced a flatter salary profile over the remainder of their career, and received a reduced pension. I gratefully acknowledge data collection and encoding by Brendon Avallone, David Choa, Jocelyn Cory, Jeremy Feiglin, Sarah Martin, Geoffrey Sinclair, Rita Yeoh, and Ming Yu. I also wish to thank Trevor Hart, Peggy Kennedy, and Tony Miller from the ANZ Group Archive for their considerable help with the primary documents. This research has been funded by a grant from the Australian Research Council. The industrial consolidations of the “first great merger wave” of the 1890s have generated considerable debate among economic historians on the consequences of monopolization on consumer welfare and the regulatory response of government. However, consumers were not the only stakeholders who were potentially adversely affected by the process of consolidation. If workers possessed specific human capital or were tied to their employer through long-term implicit contracts, mergers could have substantially reduced their lifetime incomes by allowing acquiring firms to renegotiate existing contracts. Recent research on the railroads and the banking industry suggest that there existed an environment of non-spot labor market arrangements, suggesting that it may have been possible for firms to forcefully transfer wealth from workers to employers through acquisitions. This paper examines the consequences of mergers for workers in the acquired firm during the first great merger wave using the case study of the takeover of the Bank of South Australia (BSA) by the Union Bank of Australia (UBA) in 1892. The banking industry is an excellent subject for the study of the labor market effects of mergers for several reasons. First, although caution should be taken when drawing inferences from a single case study, it is clear that the merger itself was unexceptional for the industry. Second, the banking industry developed sophisticated internal labor markets well before the 1890s. In other papers I have shown that contracts deviated substantially from what would be expected in a spot labor market. Third, there is extremely high quality personnel data from the banking industry that enables a more comprehensive analysis of the consequences of mergers for personnel practices than has heretofore been possible. The approach that I use to analyze the consequences of the merger for the BSA’s workers is similar in spirit to a series of studies beginning with a seminal article by Andei Schliefer and Lawrence Summers. They argue that the gains to shareholders from hostile takeovers may be due to redistribution from employees to shareholders, rather than to increased productivity. Managers in the acquired firm may have previously promised workers lifetime employment with deferred compensation, whereby young workers are paid less than their productivity and older workers are paid more than their productivity in manner described by Edward Lazear. Following the hostile takeover, the new management is able to renege on late-career salary overpayments that had been promised by the acquired firm. This reneging may take the form of dismissing older employees, flattening of tenure-salary profiles, or pension cuts. A popularized version of this view of hostile takeovers was the premise for the Hollywood film Wall Street, where the villain acquires an airplane manufacturer in order to replace its American plant with a Mexican plant staffed by low-wage labor For example see Lamoreaux, The Great Merger Movement. See Hamilton and MacKinnon, "Long-Term Employment Relationships”; Sammartino, Human Resorce Management; and Howlett, “Evidence” on internal labor markets at railroads. See Seltzer and Merrett, “Personnel Practices”; Seltzer and Simons, “Salaries and Career Opportunities” on internal labor markets at banks. Because the USA was dominated by unit banks, the first great merger wave largely bypassed the industry. However, in branch banking countries, such as Australia and the United Kingdom banking mergers, which had been fairly common at an even earlier time, began to increase following the depression of the 1890s. Seltzer and Merrett, “Personnel Practices”; Seltzer and Simons, “Salaries and Career Opportunities”. Schliefer and Summers, “Hostile Takeovers”. Lazear, “Why?”. and to raid the workers’ pension fund. Subsequent empirical studies have generally confirmed that following hostile takeovers there is an above-normal turnover rate among older workers at the acquired firm, a flattening of the wage profiles of retained workers from the acquired firm, and an above normal rate of reversion of pension fund surpluses. One drawback of these studies is that the takeovers being studied are relatively recent, and thus have on-going effects on many of the workers at the acquired firms. Another drawback is that these studies focus on firm-level averages, rather than the employment outcomes of individual workers. The historical nature of this study and the comprehensiveness of the data enable me to avoid both these shortfalls. The basic empirical approach to this study is to compare employment, wage, and pension outcomes of 1) BSA employees at the BSA prior to 1892 2) former BSA employees at the UBA from 1892, and 3) direct entrants to the UBA before and after 1892. In addition to the possibility that implicit contracts were not honored following the merger, I also consider whether some of the human capital possessed by the BSA’s workers was specific to the BSA and thus was not fully transferable after the merger. I assume that specific capital was increasing in seniority, and, critically, that some human capital was specific to the individual branch (for example, knowledge of the local economy or of customers). This implies that senior BSA employees in branches that were closed or merged into existing UBA branches following the takeover would have lost more specific human capital than other employees.