Through lobbying for reduced taxes, some firms argue they are engaging in good corporate citizenship. Davis et al. test this theory by estimating Equation using a logistic regression model, with dependent variable Lobby (0/1), equal to 1 if the firm incurs expenditure on tax lobbying, and 0 otherwise. The results indicate a significant positive relation between tax lobbying and measures of CSR. The findings were deemed economically significant, as firms in the highest quantile of CSR_index have approximately 158% higher probability of lobbying for taxes than other firms. The same test was repeated using Lobby_exp, the amount spent by firms on tax lobbying activities scaled by lagged assets during the year. Davis et al. found that firms in the highest quartile of CSR_Index spend approximately 1.3% more (of...

Through lobbying for reduced taxes, some firms argue they are engaging in good corporate citizenship. Davis et al. test this theory by estimating Equation using a logistic regression model, with dependent variable Lobby (0/1), equal to 1 if the firm incurs expenditure on tax lobbying, and 0 otherwise. The results indicate a significant positive relation between tax lobbying and measures of CSR. The findings were deemed economically significant, as firms in the highest quantile of CSR_index have approximately 158% higher probability of lobbying for taxes than other firms.

The same test was repeated using Lobby_exp, the amount spent by firms on tax lobbying activities scaled by lagged assets during the year. Davis et al. found that firms in the highest quartile of CSR_Index spend approximately 1.3% more (of lagged assets) on lobbying for taxes. The overall results are consistent with the view that CSR and taxes act as substitutes. The tests suggest that not only do firms that are more socially responsible pay less corporate taxes but also engage in more tax lobbying.

Sensitivity Tests – Alternative measures

Davis et al. test alternative measures of CSR to gauge what effect this will have on results found previously. The initial tests follow Kim et al. by summing the strengths and concerns of five of the seven KLD categories. Consistent results were found summing all seven. It could also be argued that the strengths and concerns are not equivalent, or that the association between negative CSR activities and taxes is not the same as that of positive CSR activities and taxes. To test this, Davis et al. follow Hoi et al. and create three new measures of CSR. The new variables were created using strengths and concerns from the MSCI ESG five social rating categories: community, diversity, employee relations, environment, and product. The Sum of Negative CSR is the sum of concerns using the MSCI.

Conversely, the Sum of Positive CSR is the sum of strengths. An indicator variable is also created called High Negative CSR, equal to 1 for firm’s displaying three or more concerns. Re-estimations of Equation show that in each estimation of Sum of Positive CSR there is a significant negative coefficient and insignificant coefficients for the other new variables. Therefore, supporting the previous findings, suggesting that firms that engage in more socially responsible activities avoid more taxes.