This thesis contributes to an understanding of factors important in the choice of auditors by public corporations. The study extends the costly contracting and governance implications of the theory of the firm to address the issue of factors likely to motivate selection (or retention) of a particular auditor. This approach, by endogenising the auditor within a contracting equilibrium in the firm, extends previous agency theory based research in this area by providing a potentially richer theory with which to address the auditor choice issue. The study provides analysis and evidence, that, following changes in the firm's production-investment opportunity set, changes of auditor (audit switches) reduce contracting costs by improving the efficiency of controls in claimholder contracts. Differences in firm production-investment attributes imply a demand for heterogeneity in the audit product market and, in response to which, auditors develop comparative advantages in the supply of particular "types" of audits. The thesis consists of two separate, but integrally related, empirical studies. Both address particular aspects of the auditor switch/retention decision. The first study investigates audit switches by public companies in Australia. To facilitate the explanation about the association between variations in auditee attributes and audit switches, three strategic hypotheses were developed which related the audit switch to auditee production-investment opportunity set shifts, auditee growth and the timing of these firm specific events. A number of proxy variables were derived and were tested on a sample of 132 switches which occurred in Australia over a ten year period between 1976 and 1984. The second study considers the more specific issue of the role auditor choice in takeovers. Because of the time series nature of the research design used in the first study, auditee firms which were targets in successful takeovers were excluded from that sample. However, because takeovers typically involve a significant reallocation of resources with a consequent realignment of firm contracting structures, the decision to retain or switch the auditor of an acquiree following a takeover also provides an interesting element in the auditor choice "puzzle". Hypotheses and proxy variables were developed which relate attributes of the acquiror, target and the takeover process itself to the specialist technology of the auditor to determine whether switching or retention of the auditor would be expected. A separate sample of 72 takeovers occurring between 1978 and 1985 was used to test the hypotheses. Both studies utilized quantitative and qualitative variables in an attempt to increase the power of the tests. From the studies the following major results emerged. Study 1:1. Auditees which switch auditors appear to do so about the time of production-investment shifts. This result is indicated by greater production-investment shifts in firms which switch auditors compared with those that do not. These production investment shifts appear to be stronger in switches where the incumbent and replacement auditor are of approximately equivalent size and more so, if both auditors in the switch are large. 2. No strong relationship appears to exist between asset growth (or decline) and the decision to switch to a larger (or smaller) auditor. However, auditees which switch from smaller to larger auditors are motivated by auditor reputation effects in the public debt and equity markets. 3. Management changes typically occur prior to an audit switch while other production investment shifts follow the switch. This result indicates that not only are shifts in management associated with production-investment shifts, but also management seeks to realign contract control structures in advance of production-investment shifts to minimise contracting costs. Study 2:1. Following takeovers, acquirors are more likely to switch the target's auditor to the acquiror's auditor when there are large differences in the production opportunity set between the acquiror and target. While this result is interesting, it is not consistent with the predictions of the theory developed in the thesis. The theory predicted that, because auditors supply competence with respect to contracts of the auditee (which are in turn derived from the auditees production-investment) configuration, contracting costs would be minimized if the auditor of the target were retained when the production-investment differences between acquiror and target are large. The anomalous result emerged despite use of an apparently robust qualitative proxy to measure production-investment differences between acquiror and aquiree. 2. Further to the result report above, two other hypotheses related to the demand for auditor expertise as a function of the auditee's production-investment configuration were also tested. The first addressed the auditor's expertise as a function of the existing portfolio of clients while the second addressed the "significance" of the switch decision as a function of the control obtained over the target. Neither produced significant results. 3. The second study also addressed the issue of whether the target management's recommendation about acceptance of the takeover offer influences the decision to switch or retain the target's auditor was addressed. Recommendations of acceptance are higher in switching firms but, again, not in the direction predicted. Because of the apparently anomalous results obtained in the second study, a number of alternative explanations related to the difficulty of modelling the takeover process, managerial and auditor inefficiencies, the role of auditors as acquisition specialists, motivations for vertical and horizontal integration and research design limitations, were canvassed.