This paper investigates the determinants of mutual fund portfolio manager ownership and its association with fund performance. Using hand-collected data of 1,420 U.S. equity funds from 32 fund families, we find that variations in fund manager holdings are broadly consistent with optimal contracting theory instead of the result of managers' personal investment consideration. Portfolio manager ownership is positively and significantly correlated with variables that proxy for intensity of agency conflicts. Specifically, portfolio managers hold more mutual fund shares when the size of concurrently managed hedge fund increases and when the advisor is affiliated to the bank. In addition, fund managers invest more in funds with primary investment in growth stock, non-index funds, and solo-managed funds. Regarding to the alternative governance mechanism, higher threat of dismissal for outsourced funds, stronger monitoring from institutional investors, and long-term performance based bonus work as substitutes of fund manager ownership while director ownership works as a compliment. Finally, we find little evidence supporting the notion that funds with higher portfolio manager ownership perform better.

The aim of this thesis is to investigate the effect of two specific external, to the principal-agent relationship, influences on executive pay practices in the UK, namely pay consultants and the introduction of the International Financial Reporting Standards (IFRS). The thesis consists of three essays. In the first essay, I examine the role of pay consultants in UK CEO pay practices. The results illustrate that their role is not consistent with the predictions of the managerial power theory. More specifically, pay consultants do not try to help managers towards the expropriation of shareholders' wealth; on the contrary I show strong indications that pay consultants urge firms towards the adoption of more incentive based CEO compensation. Moreover, I report that economic characteristics (e.g. firm size, complexity of the contract) rather than CEO power explain the firm's choice to hire a compensation consultant. These results are robust to selection bias controls. The results of this essay indicate that pay consultants play a less "sinister" role than what the managerial power theory suggests and that their advice and expertise can assist firms design an optimal executive pay contract. In the second essay, I examine the existence of managerial opportunism at the switch from UK GAAP to IFRS. I find strong indications that the restatements from UK GAAP to IFRS have not been manipulated by managers. I examine the existence of such behaviour under different specifications and for different types of CEOs that one would expect to engage in opportunistic behaviour to maximise the expected personal wealth. The research design that I adopt makes the results less prone to methodological issues common in studies in this area. Positive Accounting Theory literature has established that managerial opportunism seriously affects accounting choice. The results of this essay imply that with respect to IFRS restatements, where managers had strong incentives to manage future earnings, I find no signs of manipulation. This essay thus puts into question the Positive Accounting Theory Paradigm. In the third essay, I examine the effect of IFRS on the use of performance measures for evaluating and rewarding managers. This essay illustrates that firms make less use of accounting based performance measures due to the introduction of IFRS. I explain these results based on the predictions of optimal contacting theory. I claim that IFRS adds unnecessary "noise" to accounting numbers not relevant to the managers' actions. This is mainly due to the adoption of "fair value" accounting, which makes accounting earnings more value relevant and therefore useful for firm valuation purposes; however, "fair value" accounting also makes accounting numbers more volatile and sensitive to market movements. If this increase in volatility is related to events outside the managers' control, this makes the use of accounting based performance measures less useful for evaluating and rewarding managers. The results of this essay imply that IFRS might have made accounting earnings more useful for stock market purposes, e.g. firm valuation, but this has happened at the expense of other purposes that accounting serves, e.g. contracting.