The director and board performance from the perspectives of voting shareholders, the implementation of majority-approved shareholder proposals provide a more specific and tighter measure for the actions boards take to do their jobs. For this comprehensive exam paper, I will focus on votes withheld for director elections because implementation of majority-approved shareholder proposals requires non-trivial hand collection and might decrease the sample size significantly.
This measure has several important merits in multiple directorships setting compared with commonly used measures. First, it has been well documented that reputational concerns from the director labor market, the increasing prevalence of ‘ vote-no’ campaigns, and increased prevalence of corporates’ voluntary adoption of Majority Voting standards makes votes withheld from director elections a good proxy for shareholder general evaluation of director, committee, and board performance. (Georgeson 2000-2016). Second, while uncontested shareholder voting rarely remove specific directors, the voting outcomes of director elections are associated with subsequent board conduct, governance changes and firm performance (Del Guercio, Seery, and Woidtke , Cai, Garner, and Walkling , Fischer, Gramlich, Miller, and White , Ertimur, Ferri, and Muslu , Ertimur, et al. ). Third, academics and practitioners have consensus that the threshold of 20% of votes withheld represents substantial shareholder dissatisfaction. (Del Guercio et al. , Cai et al. , Fischer et al. , Ertimur, et al. , Ertimur et al. ). Fourth, data for this measure on individual director level is directly available, and can be easily aggregated to get measures on committee level and board level. For the aggregate level, I can use either the average percentage of votes withheld per director, or the percentage of directors with substantial shareholder dissatisfaction (at least 20 % votes withheld) of committees, or boards.
In sum, to better measure the effort exerted by directors and the value added by boards, I use a comprehensive list of complementary performance metrics: voting outcome of director election, serving on committees and attendance rate at board meetings on individual director level; voting outcome, CEO turnover sensitivity to performance, excessive CEO pay, and financial reporting misstatement on committee level, and voting outcome, market-to-book ratio (M/B) and return on sales (ROS) on firm level.
3. Sample, data, empirical results, and additional analyses
3. 1. Sample and data
To answer these questions, I study the multiple directorships policies for S&P 1500 firms for the period from 2000 to 2016. First, I construct a dataset of director-level data for S&P 1500 firms from the Institutional Shareholder Services ((formerly MSI, IRRC, andRiskMetrics) Directors Database from 2000-2016. I restrict my sample to S&P 1500 firms to make the data collection manageable. This dataset contains information on director attendance and a range of other director characteristics (e. g., name, age, tenure, gender, committee memberships, independence classification, primary employer and title, number of other public company boards serving on, shares owned, etc.) which are collected by ISS from company proxy statements, annual reports or company websites. My first multiple directorships measure, Multiple Directorships_traditional, are retrieved from this database directly. Based on the directors’ total number of meetings required to attend and whether they attended fewer than 75%, I also construct a measure, Current Distraction, to better capture the time demand and the distraction effect of multiple directorships.
To test H1 and H2, we need to know director attendance immediately before and immediately after firms adopt overboarding rules. However, only annual attendance data is publicly available. In addition, many overboarding policies set grace periods explicitly or implicitly. Therefore, I use the director attendance the year before adoption year to test H1, and the director attendance the year after adoption year to test H2.
Next, I require these firms to have accounting data from Compustat, stock return and pricing data from CRSP, corporate-governance provisions data from ISS Governance database, director elections voting data from ISS Voting Analytics (VA) database, and financial reporting restatement data from AuditAnalytics database. (access?) For firm performance, I measure themarket-to-book ratio (M/B) and return on sales (ROS), measured at the lagged fiscal year-end, to compare with previous literature (Fich and Shivdasani , Field, et al. ).
Then I use BoardEx database, which includes profiles of executives and directors (demographics, education and career history, compensation, board and committee memberships, etc.) and the connections among them, to construct my second sets of measures of multiple directorships, Multiple Directorships_public, Multiple Directorships_private, and Multiple Directorships_total. Using BoardEx database, I also construct two measures of director qualification, Accumulated Directorships Experience and General Qualification, which reflect director qualification but are not necessarily associated with their current busyness. BoardEx’s coverage of U. S. public companies is extremely limited prior to 2000, causing serious survivorship bias (see, Fracassi and Tate , Engelberg, Gao, and Parsons, ). Similar to previous literature which also uses the BoardEx database, I focus on the period after 2000 to mitigate these concerns. It needs to be noted that while BoardEx database suffers survivorship bias and some other issues, using BoardEx provides several important merits for my research. First, it provides information about directors’ seats on significant private firms and gives a better measure of directors’ total workload. Second, it allows me to test directors’ seats choices between public firms and private firms. Third, it provides information to construct a measure of directors’ resources/talent which is not necessarily connect with their current busyness.
Finally, I hand-collect information on firms’ overboarding policies for this sample. Normally, firms overboarding policies are disclosed in their governance principles/governance guidelines. I retrieve firms’ current overboarding policies from their websites, and their historical overboarding policies from the Internet Archive library, which archives over hundreds of billion historicalweb pageson the Internet. (https://archive. org/web/)
The paper "Research methodology for director elections analysis" was contributed to our database by a real student. You can use this work as a reference for your own writing or as a starting point for your research. You must properly cite any portion of this sample before using it.
If this work is your intellectual property and you no longer would like it to appear in our database, please request its deletion.Ask for Removal
Create a Citation on Research Paper
PaperPrompt. (2022) 'Research methodology for director elections analysis'. 4 August.
PaperPrompt. (2022, August 4). Research methodology for director elections analysis. Retrieved from https://paperprompt.com/research-methodology-for-director-elections-analysis/
PaperPrompt. 2022. "Research methodology for director elections analysis." August 4, 2022. https://paperprompt.com/research-methodology-for-director-elections-analysis/.
1. PaperPrompt. "Research methodology for director elections analysis." August 4, 2022. https://paperprompt.com/research-methodology-for-director-elections-analysis/.
PaperPrompt. "Research methodology for director elections analysis." August 4, 2022. https://paperprompt.com/research-methodology-for-director-elections-analysis/.
"Research methodology for director elections analysis." PaperPrompt, 4 Aug. 2022, paperprompt.com/research-methodology-for-director-elections-analysis/.
Get in Touch with Us
Do you have more ideas on how to improve Research methodology for director elections analysis? Please share them with us by writing at the [email protected]