【正文】 第一篇：Audit Office Size, Audit Quality, and AuditPricing JongHag Choi, Chansog(Francis)Kim, JeongBon Kim, and YoonseokZang SUMMARY: Using a large sample of client firms over the period 2000–2005,this paper investigates whether and how the size of a local practice office within anaudit firmhereafter, office size is a significant, engagementspecific factor determiningaudit quality and audit fees over and beyond audit firm size at the national level andauditor industry leadership at the city or office our empirical tests, audit qualityis measured by unsigned abnormal accruals, and the office size is measured in twodifferent ways: one based on the number of audit clients in each office and the otherbased on a total of audit fees earned by each results show that the officesize has significantly positive relations with both audit quality and audit fees, even aftercontrolling for nationallevel audit firm size and officelevel industry relations support the view that large local offices provide higherquality auditspared with small local offices, and that such quality differences are priced in themarket for audit :audit office。office size。audit quality。audit Availability:Data are publicly available from sources identified in the way we think about an accounting firm changes dramatically when we shift the unit of analysisaway from the firm as a whole, to the analysis of specific citybased offices within a termsof DeAngelo’s1981b argument, a Big 4 accounting firm is not so big when we shift to theofficelevel of example, while Enron represented less than 2 percent of Arthur Andersen’s national revenues from publicly listed clients, it was more than 35 percent of such revenues in the Houston As alluded to in the above quote, the size of a citybased audit engagement office could bea more crucial determinant of audit quality and thus audit feesthan the size of anationallevel audit firm because the citybased office is a semiautonomous unit withinan audit firm with its own client is an officebased engagement partner or audit team, notnational headquarters, who actually administers and implements individual audit engagement contracts, including the delivery of audit services and the issuance of an audit this regard,Wallman(1996)and Francis(2004)argue that the assessment of auditor independence needs tofocus more on the individual office level rather than the entire firm level because most of the auditdecisions with respect to a particular client are made within each individual anecdotalevidence on the collapse of Enron, which was audited by the Houston office of Arthur Andersen,is a good example that demonstrates the importance of officelevel audit , much of extant audit research has focused its attention on two nationallevel audit firm characteristics asfundamental determinants of audit quality, namely: audit firm size(., Simunic and Stein 1987。Becker et 。Francis and Krishnan 1999。Kim et 。Choi and Doogar 2005), andauditor industry leadership(., DeFond et 。Balsam et 。Krishnan 2005).These studies find, in general, that large audit firms with international brand names(., Big4 auditors)or industry expertise provide higherquality audit services than small audit firms which lack such brand names or industry in this line of research is the assumptionthat audit quality is homogeneous across offices of various sizes located in different cities withinthe same audit a result, we have little evidence on crossoffice differences in audit quality,and in particular, whether and how the size of a local engagement office has an impact on auditquality and/or audit natural questiontoaskis:Istheofficesizean additionalengagementspecific factor determining audit quality and thus audit pricing over and beyond auditfirm size and industry leadership? We aim to provide direct evidence on this unexplored recent studies provide indirect evidence suggesting that audit quality may differacross different engagement offices within an audit example, in the first thatuses each engagement office as the unit of analysis, Reynolds and Francis(2000, 375)find thatwhen client size is measured at the office level using officespecific clienteles, “Big 5 auditorsreportmoreconservativelyforlarger clients.”Further,Ferguson et al.(2003)and Francis et al.(2005)find that cityspecific, officelevel industry leadership, when bined with the nationallevel leadership, generates the highest audit fee premiums and thus, by inference, higher auditquality in the Australian and markets, respectively, while nationallevel industry leadership alone has no , Francis et al.(2006)document that client earningsquality proxied by abnormal accruals is higher when auditors are citylevel industry leaders alone,or they are both citylevel and nationallevel industry differently, their results indicatethat nationallevel industry leadership alone has no significant impact on audit , Choi et al.(2007)show that the geographical proximity of the citybased engagementoffice to clients’ headquarters is positively associated with the accrual quality of clients, suggesting that the geographical location of the auditor’s office is an important engagementspecificdeterminant of audit above findings, taken together, suggest that citybased, officelevel characteristics may play an important role in determining audit quality and thus audit should be pointed out, however, that none of these studies has paid attention to the question ofwhether the size of a local engagement office is systematically associated with audit quality andfees paid to bridge this gap in our knowledge, we investigate a hitherto underresearched question ofwhether, and how, the size of a local engagement office hereafter, office sizeis associated withaudit quality and audit first hypothesize that office siz