Chipping Away At Financial Reporting Quality
Miami University of Ohio – Department of Finance
University of Alabama – Culverhouse College of Commerce & Business Administration
University of Pittsburgh – Accounting Group
November 6, 2015
Chief financial officers are responsible for managing the financial reporting process. We test whether the quality of a firm’s financial reports is a function of the effort expended by the CFO. Using golfing records to measure leisure consumption, we first show that CFOs consume more leisure when they have lower economic incentives to work. We show further that higher levels of CFO leisure are negatively associated with a number of indicators of financial reporting quality. The use of firm fixed effects and an instrumental variable analysis suggest that the observed relations are causal. Further tests indicate that higher leisure consumption is associated with shorter conference calls with a more uncertain tone. Finally, the effects of lower quality reporting are demonstrated by results linking CFO leisure with analysts’ forecast dispersion and weaker earnings response coefficients.
Chipping Away At Financial Reporting Quality – Introduction
Chief financial officers (CFOs) are responsible for financial reporting. Existing literature demonstrates the relevance of individual CFO characteristics such as gender, experience, and professional qualifications for the financial reporting quality of a firm (Aier et al. 2005; Chava and Purnanandam 2010; Jiang et al. 2010; Li et al. 2010). In this paper, we test a fundamental economic proposition that may be important in this context, but that has been largely overlooked in the literature. Motivated by seminal financial economics contributions to agency theory predicting that delegated managers may be tempted to shirk their responsibilities, we consider whether a firm’s financial reporting quality is a function of the amount of effort exerted by its CFO (Jensen and Meckling, 1976). Because of the difficulty in observing and measuring executive effort, prior research has been unable to research this question empirically. To overcome this issue, we use an observable measure of leisure consumption as an inverse proxy for the effort provided by a CFO, namely the amount of golf they play. To the best of our knowledge, this is the first endeavor of its kind in the literature.
Our findings can be summarized as follows. First, there is considerable heterogeneity in the amount of golf played by CFOs who maintain an official golfing handicap. Consistent with agency theory, the amount of golf CFOs play is a function of their economic incentives to create value in their jobs. We also find evidence that when CFOs spend more time on the golf course, the quality of their work suffers. In particular, more frequent golf play is associated with indicators of lower financial reporting quality, less informative communications with investors, and a weaker relationship between financial disclosures and stock prices.
To conduct our analyses, we make assumptions about the amount of effort a CFO is putting forth by evaluating the amount of leisure they consume. The particular measure we use is the number of rounds of golf played by golfing CFOs. We are able to collect this data from a detailed database of golf records maintained by the United States Golf Association (USGA) that is used to calculate and track the handicaps of participating golfers. The database contains round-by-round records and includes the calendar month and year of the round, the player’s score, the relative difficulty of the course, and if the round was played at a course where the golfer maintains a membership. For this study, we identify and hand collect golf records for 385 CFOs from 2008 to 2012. We submit that golfing frequency is a valid measure of leisure consumption because of the significant time commitment represented by a single round of golf. The average CFO in this sample records 20 rounds of golf per year, which represents a substantial amount of time away from the office. Assuming an average round takes five hours, this translates to 2.5 work weeks of the year allocated to playing golf. The data also reveals considerable variation in the time allocated to golf, as thirteen percent of CFOs in the sample do not record any rounds in a year and ten percent record 46 or more rounds in a year.
We begin by evaluating whether the amount of leisure a CFO consumes is related to their economic incentives. We find this to be the case. Using the delta of CFOs’ stock and option holdings in their firms as a measure of economic incentives to work hard, we find that that they play 22.7 percent less golf when their incentives are one standard deviation higher. This finding suggests that the number of rounds of golf a CFO plays annually is a reasonable proxy for the amount of leisure they consume, and that they consume less leisure when they have greater incentives to work.
See full PDF below.