A review of Dey, A., 2008, “Corporate Governance and Agency Conflicts,” Journal of Accounting Research 46 (5), 1143-1181.
I. Research question and its importance
Drawing on the theory of corporate governance, the author empirically examines the notion that governance mechanism is a function of agency conflicts in a firm. This question is important because “there is little evidence on the factors that determine the endogenous presence or effectiveness of these (governance) mechanisms” (p. 1144). Thus, corporate governance literature had brought mixed results which this article aims to reconcile. The author proposes a set of (seven) factors that better measure governance and tests its association with agency conflicts documented in the literature. Moreover, the author conducts additional analyses that examines the association between governance and firm performance to increase relevance of her findings. Ultimately, she intends to document the cross-sectional difference of governance structures based on vary levels of agency conflicts.
II. Method and findings
To construct a set of factors pertaining to governance, the author employs exploratory principal components analysis to define (seven) factors that best capture governance, as “these factors can then be used to explain the underlying dimensions of governance” (p. 1145). She retrieves governance data from Schedule 14A for director information and merges with other sources. As a result, her sample contains 371 firms. OLS was used for regression analysis. The author shows agency conflicts in t-1 period is positively related to governance in t period, where “this finding is consistent with the theory that governance structures, particularly those related to the board of directors, the audit committee, and the auditor, are positively related to the level of agency conflicts in firms” (p. 1146). To sum up, her finding supports the conclusion that “governance structures arise endogenously to firms’ competitive and business environments” (p. 1146). To make this finding more relevant to practitioner, the author conducts additional analyses that examines the association between governance and firm performance via OLS regression. She finds this relation significantly positive primarily for firms in the “high agency conflicts group” that she has created using clusters analysis.
III. Limitation and future research
There are several critiques toward this paper. First, as was acknowledged by the author, her sample selection is biased toward relatively large firms, although she does not indicate why. Second, corporate governance has changed much since the infamous Enron scandal in 2001. Governance practice has very likely changed subsequent the event. Third, despite mentioning “agency problems vary across firms” (p. 1144) this variation warrants little managerial implication. Future research might examine this postulation in different corporate contexts. Forth, cross-sectional data has limited her ability in testing cause-effect. Alternatively, granger causality test might aides her in augmenting this argument. Finally, as in her conclusion, “evidence on the aspects of governance that need to be modified, and the types of firms for which such measures will be more effective” (p.1175) more works worth (and actually, have!) extending this topic for further examination.