초록 close

본 논문은 피감사기업의 소송위험과 감사보수 사이의 관계를 살펴본 연구이다. 감사인이 감사업무 수행시 부담하는 법적 책임인 감사위험이 감사보수에 영향을 미친다는 연구는 다수 수행된 바 있다. 선행연구들은 감사위험이 높을수록 감사시간이 증가하며, 그 결과로 감사보수가 증가된다고 보고하였다. 이런 선행연구들에서 감사위험을 나타내기 위해 사용된 자료는 대부분 재무제표에서 측정된 재무비율 변수들로서, 수익성(손실여부나 총자산이익률)이나 안전성(부채비율이나 유동비율), 감사의견(비적정 의견이나 계속기업존속여부의 불확실성 의견)이었다. 이런 재무비율 변수와는 별개로, 재무제표상에 직접 포함되지는 않지만 임원배상책임보험료 수치를 이용해 감사위험을 측정한 후 분석한 연구와 감사인들에게 직접 조사한 서베이(survey) 자료를 감사위험의 측정치로 이용한 연구도 있다. 이와 달리 본 연구는 과거 선행연구들에서 분석되지 않았던 변수인 피감사기업의 소송위험을 이용하여 피감사기업의 위험수준을 보다 직접적으로 측정하였다. 피감사기업의 소송위험은 현재 피감사기업을 피고로 한 소송이 진행중인지 여부, 피감사기업이 피고인 경우에 진행중인 소송의 건수, 소송이 감사보고서에 특기사항으로 기재된 여부, 소송가액의 합계 등으로 나누어 측정한 후 분석하였다. 현재 소송이 진행중인 기업, 그중에서도 소송가액이 큰 기업들은 소송의 진행결과에 따라 기업의 재무적 안전성이나 순이익에 큰 영향을 미칠 수 있다. 따라서 소송 관련사항은 감사인이 부담해야 할 감사위험을 증가시킬 수 있으므로, 감사인이 피감사기업에게 요구하는 감사보수가 달라질 수 있다. 2002년부터 2007년까지 한국거래소에 상장된 유가증권상장기업 2,821개 기업/연도 표본을 이용하여 분석한 결과는 이런 가설을 지지하고 있다. 즉 소송진행여부나 해당 소송사항이 감사보고서상에 특기사항으로 보고된 경우 및 소송의 건수, 소송가액 등은 모두 감사보수와 유의적인 양(+)의 관계를 가지는 것으로 나타났다. 이런 분석결과는 감사보수에 영향을 미치는 감사의 범위, 감사업무의 복잡성과 재무비율로 측정된 피감사기업의 감사위험 변수들, 그리고 감사인의 특성변수들을 모두 통제한 후에 얻어진 결과이다. 또한 선행연구에서 사용된 피감사기업의 임원배상책임보험료 금액을 추가로 통제한 후에도 유의한 결과로 나타났다. 즉 본 연구의 발견은 상당히 강건하다고 할 수 있다. 따라서 소송대상기업은 감사인의 감사위험 중 고유위험을 증가시켜 감사인의 감사보수를 할증시키는 것으로 나타났다.


This study examines the association between litigation risk of client firms and audit fees. Prior studies report that audit risks borne by auditor are reflected in determining audit fees. They report that as audit risk increases, audit fee increases. The studies measure audit risk mostly by financial variables collected from financial statements, such as profitability (for instance, loss indicator variable, return on assets, or return on equity), solvency (for instance, leverage ratio or current ratio), and audit opinion (for instance, non-clean opinion or going-concern uncertainty opinion). There are many examples of such studies that use the variables mentioned above. Choi et al. (2008) examines the association between country-level auditors' legal liability and audit fees, in addition to firm-level risk variables, such as loss indicator variable, return on assets, and leverage ratio. They show that audit fee monotonically increases as country-level as well as firm-level legal liability measures increase. Seetharaman et al. (2002) also show that U.K. firms cross-listed in the U.S. pay higher audit fees than non-cross-listed firms after controlling for firm-level financial risks because cross-listing in the U. S. causes higher legal liability for auditors. Similarly, Venkataraman et al. (2008) report that initial public offering firms pay higher audit fees due to greater litigation risk, compared with existing listed-firms. This is because initial public offering firms are likely to be related to higher risks. Other than financial variables measured from financial statements, Lee et al. (2007) use directors and officers liability insurance premiums for the client firms to measure the audit risk. Similar in spirit, Pratt and Stice (1994) survey auditors for perceived audit risk of client firms. Both Lee et al. (2007) and Pratt and Stice (1994) use the measured audit risk to examine if the risk is reflected in setting audit fees. Unlike prior studies, this study uses a different and more direct measure of audit risk, the litigation risk of client firms. The litigation risk is measured by four different variables: (1) the existence of litigation for the client firms, (2) the number of litigation cases for the client firms, (3) the existence of record on the litigation in the audit report for the client firms, and (4) the amount of damage claims for the auditor for the litigation. As a result, we examine the association between audit fees and four risk measures. These four measures are not examined in prior studies. Although a prior study of Palmrose(1988) examines the association between number of litigation against auditors and audit fees, it is the study on the number of litigation against auditors, not against client firms. This study focuses on the litigation against client firms, which is different from the focus of prior studies. This is a direct measure of audit risk, because the results of the litigation greatly influence on the financial status of the firms. However, the financial variables measured from financial statements do not exactly reflect such risks. If a firm is under litigation as a defendant (measure 1), the audit risk is expected to be greate than the risk for the firms not under litigation. Second, among firms under litigation, if the number of litigation cases increases (measure 2), it is expected that the risk is greater. Third, an auditor would record the existence of the litigation only if he judges that the litigation can cause serious damage to the client firms. Thus, the existence of litigation case for which auditor records it in the audit report (measure 3) can be used to measure the audit risk. Finally, if the amount of damage claims against the auditors is greater, the result of one court case can greatly influence the financial status of a client firm. Thus, the amount (measure 4) can be used as a proxy for the audit risk. Because a firm is under litigation by suing another party (firm, individual, or governmental authorities), we use an alternative measure by calculating the net amount of damage claims (damage claims for the defendant minus damage claims as a plaintiff). For empirical analyses, we collect 2,821 firm-year observations from the companies listed in Korean Stock Exchange over the period of year 2002 to 2007. Among them, about 36.5 percent of firms (average 173 firms per year) are under litigation and the frequency of litigation increases as time passes. It may reflect the trend of increasing audit risk related to the passage of Korean version of Sarbanes-Oxley Act (so called K-SOX) in year 2002 and Class-Action Litigation Act in year 2003 and 2004. In addition, average of 146 firms have number of litigation data per year and each firm under litigation have average 9 ongoing litigation. The amount of damage claims against a client firm is 49,256,000 Korean won. Both univariate and multivariate results support the predictions. For the firms under litigation pay significantly higher audit fees than other firms not under litigation, even after controlling for other 16 variables that are reported to be related to audit fees in prior studies. These other variables represents the extent, complexity, and financial risks of audit service as well as industry and year dummies. The audit fee premiums are greater when the existence of litigation is written on audit report as a special items. Both the number of litigation and amount of damage claims against the auditors are reflected in audit fee when the regression analyses are performed with only firms that are under litigation. Various sensitivity analyses results also support the findings. Because Lee et al. (2007) uses directors and officers liability insurance premiums to measure for audit risk, this study include both client firms' litigation-related variables and the insurance premiums in the same regression model and examines if the results hold. This study consistently finds that the litigation-related variables are significantly associated with audit fees, even though the insurance premiums are insignificant or only marginally significant. The results imply that the litigation-related variables reflect the audit risk better than directors and officers liability insurance premiums. In addition, the information on premiums are available only for limited number of companies, which restricts the generalizability of the findings, compared with that of this study. In summary, the empirical findings support that auditor reflects audit risk related to the litigation of client firms in setting audit fees. Based on this finding, future studies, regulators and other information users should use the information on the litigation in evaluating client risk.