Page 212 - 34-1

The Effect of SEC’s Comment Letters on the Comparability of Non-GAAP Earnings

How do SEC comment letters enhance the comparability of non-GAAP earnings?
When a company receives a SEC comment letter, it is required to respond to the issues
raised in the comment letter, including correcting disclosure content. In support of the
idea that comment letters can improve disclosure quality, Johnston and Petacchi (2017)

find that companies receiving comment letters can reduce information asymmetry, and the
earnings response coefficient also increases. Bozanic et al. (2017) discover that companies
receiving comment letters experience a decrease in information asymmetry and litigation

risk. Therefore, when a company receives a regulatory letter regarding non-recognized
earnings, its information content is more likely to align with the SEC’s requirements for
non-GAAP earnings. It will make non-GAAP earnings more comparable.
However, some studies also find comment letters cannot improve the quality for
disclosure if the requirements of comment letters are related with manager’s private

information (Robinson, Xue, and Yu, 2011; Kim, Kim, and Musa, 2018; Cassell, Dreher,
and Myers, 2013; Cunningham, Johnson, Johnson, and Lisic, 2020). This is because
disclosing additional information may increase disclosure costs, such as by raising

potential competition costs and processing costs (Beyer, Cohen, Lys, and Walther, 2010;
Cassell et al., 2013; Robinson et al., 2011; Blankespoor, deHann, and Marinovic, 2020).
We believe that company managers will consider the costs and benefits of complying
with the requirements stated in comment letters. Therefore, we propose the following
hypothesis:

H1: SEC comment letters related to non-GAAP earnings do not affect their comparability.

3. Research design

Our study explores whether SEC comment letters can enhance the comparability
of non-GAAP information. We obtain analyst forecast data from the I/B/E/S database,
financial condition from the Compustat, and comment letters from Audit Analytics. The
full sample contains 2,000 observations. We follow prior research by focusing on firms

receiving SEC comment letters related to their use of non-GAAP measures. NG_COMP
represents the comparability of non-GAAP earnings. We use the model provided by De
Franco et al. (2011) to establish the dependent variable (NG_COMP). The empirical model
is described as follows:

204