Abstract
I examine the relationship between the strength of corporate governance and firm-level investment policies. While the link between governance and choice of investment policy has been examined by many, endogeneity in the choice of these variables has clouded inference. I overcome this drawback by examining changes to investment policies following an exogenous shock to governance with the passage of the Sarbanes-Oxley Act of 2002 and the associated changes to the listing requirements of major stock exchanges prompted by the high-profile corporate scandals of the early 2000s. Provisions of the Sarbanes-Oxley Act and stock exchanges examined in this study include the financial reporting, related party transaction, and the board and committee independence provisions, since these provisions have a high likelihood of impacting upon agency conflicts. I find that firms did in fact respond to the change in governance post-SOX by altering investment behaviour that is generally consistent with the agency theory. For example, I find that the investment-cash flow sensitivity for all firms reduces while the marginal value of capital expenditure increases in the new regulatory environment. Literature on SOX also suggests perverse outcomes, such as increased risk aversion and high compliance costs, which can potentially harm firm value. Consistent with this, I find a decrease in the marginal value of cash following the governance changes, suggesting either an increase in risk aversion or an adverse impact from compliance costs may be occurring. These results hold after controlling for measures of financing constraints. Since other contemporaneous events might be adding noise to the conclusions drawn, I adapt the difference-in-differences methodology laid out in Chhaochharia and Grinstein (2007) to isolate the effect of tightened governance following the new rules. This methodology permits me to examine the differential effects on firms that were compliant with the provisions versus those that were not over a seven-year time span centred on the event of interest. For the full sample, non-compliant firms in general were affected more, with the majority of the results consistent with a reduction in agency conflicts. Finer division of the samples based on ease of access to external markets proxied by size, payout, and creditworthiness also produced results generally consistent with this for firms with less-than-perfect access to external capital. While the overall findings imply a reduction in agency conflicts post-SOX, there is some evidence that certain firms found the costs of compliance to be excessive, while still others saw managers becoming excessively risk averse.