<p class="MsoNormal" style="text-align: justify; margin: 0in 0.5in 0pt; mso-pagination: none;"><span style="font-size: 10pt;"><span style="font-family: Times New Roman;">Prior research (Thomas, 1989; Das and Zhang, 2003; Jordan et al., 2008a) presents evidence that managers manipulate income to round up earnings per share (EPS) to key user reference points when unrounded EPS fall only slightly below these breakpoints.<span style="mso-spacerun: yes;">&nbsp; </span>In the U.S., studies on audit quality suggest that, relative to small audit firms, large CPA firms provide higher quality audits and, as such, more aggressively constrain their clients&rsquo; attempts to manage earnings in general (e.g., see Francis and Krishnan, 1999; Krishnan, 2003).<span style="mso-spacerun: yes;">&nbsp; </span>However, research outside the U.S. finds no consistent audit quality differential based on auditor size (e.g., see Piot and Janin, 2007; Maijoor and Vanstraelen, 2006).<span style="mso-spacerun: yes;">&nbsp; </span>The current study examines whether audit quality, as proxied by auditor size, in the U.S. constrains earnings management to effect user reference points in EPS.<span style="mso-spacerun: yes;">&nbsp; </span>Using the Big 4 versus non-Big 4 dichotomy as the measure of auditor size and audit quality, results suggest that audit quality significantly restricts management&rsquo;s attempts at rounding up EPS as clients of Big 4 firms show no major signs of this manipulative behavior while non-Big 4 auditees appear to round up the first digital position right of the decimal point in EPS across zero to increase the digit immediately left of the decimal point by one.</span></span></p>

<p class="MsoNormal" style="text-align: justify; margin: 0in 0.5in 0pt;"><span style="font-family: Times New Roman; font-size: x-small;">Significant research (e.g., Carslaw, 1988; Thomas, 1989) provides evidence that managers manipulate earnings to reach cognitive reference points in income.<span style="mso-spacerun: yes;">&nbsp; </span>More specifically, when the second-from-the-left earnings digit falls just below zero, management finds ways to round earnings up to just above this breakpoint so that the first earnings digit increases by one.<span style="mso-spacerun: yes;">&nbsp; </span>The current study demonstrates that for a sample of large publicly-traded U.S. companies this same type of manipulative behavior appears to be occurring with respect to reported sales revenue.</span></p>