 |
|
To isolate the effect of UTB_Settle% on the dependent variable, we include a number of control variables in Eq. 1). As noted earlier, Hoopes et al. (2012) demonstrate that the likelihood that the firm will be subject to IRS audit is negatively to its tax avoidance activities. As
such, we use publicly available data from Transactional Records Access Clearinghouse (TRAC) to construct a proxy for ex ante IRS audit
probability, IRS_TRACct, and we include this variable in Eq.(1). Among other data, TRAC collects and reports the number of corporate audits performed by the IRS during the year for 12 groups of companies, where a company is categorized based on its total assets (as denoted by the subscript c in the variable acronym). Following Hoopes et al.(2012), we measure IRS_TRAC as the number of federal corporate audits performed for each size group in a given year divided by the number of federal corporate returns filed for the size group during the year. Further, due to the documented association between IRS_TRAC and firm size in Hoopes et al. (2012), we include the natural log of total assets(Size) in the regression model.......[Citing TRAC research].
|
|
|
|
