STATEMENT ON AUDITING STANDARDS NO. 59:

A CODIFICATION OF EXISTING PRACTICE?

ABSTRACT

The Auditing Standards Board (ASB) issued Statement on Auditing Standards (SAS) No. 59, “The Auditor’s Considerations of an Entity’s Ability to Continue as a Going Concern” to satisfy public concern about whether companies being audited can continue as going concerns. The standard’s objective was to reconcile the different beliefs between financial statement users and auditors regarding auditors’ responsibilities related to the going concern question. SAS No. 59 increased auditors’ responsibility for evaluating a company’s ability to continue as a going concern by requiring auditors to take an active role in seeking and evaluating evidence pertinent to the going concern question. The objective of this study is to evaluate the impact of SAS No. 59 on auditors' going concern decisions.

The findings of this study suggest that SAS No. 59 had a modest impact on auditors’ going concern decisions. Though auditors were more likely to issue going concern opinions (GCOs) to bankrupt companies after the issuance of SAS No. 59 than before, the post-SAS No. 59 GCO companies were not financially stronger or larger than pre-SAS No. 59 GCO companies. Additionally, except for the change in debt to total assets measure, auditors’ reliance on the financial characteristics listed in SAS No. 59 (and 34) did not change under the provisions of SAS No. 59. It seems that the ASB simply codified existing practice when they issued the more stringent standard.

INTRODUCTION

In April 1988, the Auditing Standards Board (ASB) issued Statement on Auditing Standards (SAS) No. 59, “The Auditor’s Considerations of an Entity’s Ability to Continue as a Going Concern.” The ASB designed the standard to satisfy public concern about whether companies being audited can continue as going concerns (Ellingsen, Pany, and Fagan 1989). The standard’s objective was to reconcile the different beliefs between financial statement users and auditors regarding auditors’ responsibilities related to the going concern question. SAS No. 59 increased auditors’ responsibility for evaluating a company’s ability to continue as a going concern (Carcello, Hermanson, and Huss 1995; Raghunandan and Rama 1995; and Bell and Tabor 1991). Specifically, the standard requires auditors to take an active role in seeking and evaluating evidence pertinent to the going concern question (Asare 1990).[1]

An objective of this study is to evaluate the impact of SAS No. 59 on auditors’ going concern decisions. SAS No. 34, “The Auditor’s Considerations When a Question Arises About an Entity’s Continued Existence,” provided the authoritative guidance to auditors for evaluating the going concern issue prior to SAS No. 59. Though SAS No. 59 superseded SAS No. 34 and increased the auditor’s responsibility in assessing the going concern issue, certain parts of the standards are similar. Specifically, the financial characteristics suggested for auditors to identify conditions relevant to the going concern problem are the same in both SAS Nos. 34 and 59.[2] Though the financial characteristics listed in the standards are the same, auditors may use a different process under SAS No. 59 than they used under SAS No. 34 in deciding whether to issue going concern opinions (GCOs) (Chen and Church 1992). This study evaluates whether SAS No. 59 had an impact on the degree of reliance auditors place on the financial characteristics in evaluating the going concern question.

Since the ASB sought to create the perception among users and others that SAS No. 59 increased auditors’ responsibilities, it is plausible that it increased the costs associated with not issuing GCOs when companies subsequently fail (Raghunandan and Rama 1995). Consequently, auditors may be more conservative in their going-concern evaluations, which would increase the likelihood of auditors issuing GCOs. Additional tests of SAS No. 59’s impact evaluate whether the propensity of auditors to issue GCOs to companies that later declared bankruptcy increased after the issuance of SAS No. 59. Also, this study evaluates whether the propensity of auditors to issue GCOs to financially distressed, but not bankrupt, firms increased subsequent to the standard.

The impact of SAS No. 59 on the financial condition and size of firms that receive GCOs also is evaluated in this study. Pre-SAS No. 59 studies suggest that financial condition and size are the most important determinants of whether auditors issue GCOs (Mutchler 1986; McKeown, Mutchler, and Hopwood 1991; and Chen and Church 1992). In general, smaller companies and companies in poorer financial conditions are more likely to receive GCOs. As previously mentioned, the increased costs associated with not issuing GCOs to companies that subsequently fail likely increased the level of conservatism exhibited by auditors when assessing the going concern issue. Not only is the auditor’s propensity to issue GCOs likely to increase, but also the size and degree of financial health of companies that receive GCOs are likely to change subsequent to SAS No. 59. More specifically, firms that receive GCOs subsequent to the standard may be larger and less financially distressed than those firms that received GCOs in periods prior to the standard. This study evaluates whether the financial condition and size of firms that received GCOs subsequent to SAS No. 59 are different from those of firms that received GCOs prior to the standard.

MOTIVATION

This study is motivated by the continued interests of the ASB, government, and financial statement users in the effects of SAS No. 59. SAS No. 59 was one of nine expectation gap standards issued by the ASB to reconcile the different beliefs between financial statement users and auditors regarding auditors’ responsibilities. The standard imposed on the auditor a responsibility to evaluate the going concern assumption as part of every audit. By issuing SAS No. 59, the ASB implied that GCOs are important signals of impending failure to financial statement users and that auditors can and should take more responsibility for assessing the ability of their clients to continue as going concerns (Ellingsen, Pany, and Fagan 1989). The ASB and the AICPA’s Auditing Standards Division recognized that SAS No. 59 ushered in significant changes in some fundamental and long-standing auditors’ responsibilities; consequently, the ASB and the AICPA have exhibited continued interest in the standard’s impact on the profession.

The ASB and the AICPA’s Auditing Standards Division held a joint conference in 1992 to evaluate the status of the implementation of SAS No. 59. An objective of the conference was to stimulate research directly related to the more stringent standard. The results of the conference, reported in the Proceedings of the Expectations Gap Roundtable, highlighted the relationship between the going concern status and bankruptcy as one of the most significant issues. Furthermore, the proceedings indicated that the extant research is limited by the fact that most of it was conducted prior to SAS No. 59 (Carmichael and Pany 1993).

The ASB’s continued interest in SAS No. 59 and GCOs also is evidenced by the issuance of additional standards. SAS No. 64, “Omnibus Statement on Auditing Standards -- 1990,” further tightened the professional standards related to GCOs by prohibiting the use of conditional terminology in GCOs (Carcello, et al. 1995 and AICPA 1990).[3] Subsequent to SAS No. 64, the ASB recognized that auditors often circumvented the purpose of GCOs by using conditional terminology such as, “If the company is unable to obtain refinancing, there may be substantial doubt about the company’s ability to continue as a going concern.” As a result, in 1995 the ASB set forth additional guidelines with the issuance of SAS No. 77, “Amendments to Statements on Auditing Standards No. 22, Planning and Supervision, No. 59, The Auditor’s Consideration of an Entity’s Ability to Continue as a Going Concern, and No. 62, Special Reports.” SAS No. 77 prohibited the use of conditional language in the GCO and indicated that auditors must clearly state whether substantial doubt exists.

The impact of SAS No. 59 also should interest the public and government. The ASB’s actions related to the going concern issue, evidenced by the issuance of SAS No. 59, were in response to public and legislative expectations. From 1985 to 1993, the U.S. House of Representatives held a series of hearings about the public accounting profession. Congressmen Dingell and Wyden criticized the accounting profession for not using GCOs to provide early warning signals of the subsequent failure of companies. Support for their criticism was reported in prior research that indicated that GCOs precede bankruptcy in only about half of the cases (e.g., Koh 1991; Altman 1982; and Altman and McGough 1974). When companies fail shortly after the issuance of non-going concern opinions (NGCO), there often is a public perception that the financial statement users should have received an early warning signal from the auditors (Berton 1985; Ellingsen, Pany, and Fagan 1989; AICPA 1989; and Carmichael and Pany 1993). The public clearly views these business failures as audit failures (Berton 1985).

CONTRIBUTION AND HYPOTHESES DEVELOPMENT

This section discusses prior studies from the audit opinion literature that are relevant to this study. These studies can be categorized into two distinct categories: (1) studies that evaluated the ability of financial ratios to foreshadow auditors’ GCOs and (2) studies that evaluated the propensity of auditors to issue GCOs to both bankrupt and other financially distressed companies

Usefulness of Financial Ratios to Predict GCOs

SAS No. 34 provided the auditors’ authoritative guidance related to the going concern issue before the issuance of SAS No. 59. Though SAS No. 59 increased auditors’ responsibilities in assessing the going concern issue, certain sections of the standards were unchanged. Specifically, both standards identified (1) recurring operating losses, (2) working capital deficiencies, (3) negative cash flows from operations, and (4) adverse key financial ratios as financial characteristics auditors should investigate for negative trends to identify firms with going concern problems. The remainder of this section summarizes prior studies that modeled the auditor’s GCO decision using financial ratios.

Levitan and Knoblett (1985) compiled a list of twenty-six variables and classified each into one of the four categories identified in SAS No. 59 (and No. 34). They developed a model that discriminated between 32 companies that received GCOs from 32 companies that received NGCOs in the 1980-1981 period. Using stepwise discriminant analysis, the set of variables that best discriminated between the GCO and NGCO firms was: (1) net worth to total debt, (2) a dummy integer representing how many of the previous three years’ cash flows were negative, (3) slope of the trend line of the three years’ current ratios, (4) dummy integer representing how many of the previous three years reported negative net income, and (5) total debt to total assets. Their model correctly classified 93.6% of the companies as those that received GCOs and those that received NGCOs. Since the sample used by Levitan and Knoblett (1985) terminated at the point that SAS No. 34 was issued, their study was directed toward those financial factors which foreshadowed GCOs prior to 1982 (Levitan and Knoblett 1985).[4]

Mutchler’s (1985) study was designed to test the extent to which auditors’ GCO decisions could be predicted using publicly available information. Her sample consisted of 119 companies that received GCOs and 119 companies that received NGCOs during the 1981-1982 period. She developed a multiple discriminant analysis model that discriminated between GCO and NGCO companies using financial ratios identified by auditors as useful cues in evaluating whether to issue GCOs.[5] Specifically, the variables were: (1) cash flow to total liabilities, (2) current assets to current liabilities, (3) net worth to total liabilities, (4) total long-term liabilities to total assets, (5) earnings before interest and taxes to net sales.

Mutchler tested her model using two samples: (1) the entire sample of 119 GCO and 119 NGCO firms and, (2) a subset of the entire sample that included 42 firms that received GCOs for the first time and 42 NGCO companies. Since she was interested solely in the predictive power of the variable set, she reported only the model’s classification accuracies for both samples (83%).[6] Consequently, her results have limited use in evaluating the ratios’ abilities to foreshadow GCOs since she did not disclose the coefficients and significance levels for the variables in the model. Other limitations of Mutchler’s study relative to the research proposed in this paper are (1) the sample included firms from periods in which SAS Nos. 34 and 59 were not effective and (2) the ratios used in the model did not include any of the trend variables specifically identified in the auditing standards.

Menon and Schwartz (1987) continued the inquiry into whether financial ratios have the ability to predict GCOs. Their sample included 89 bankrupt firms, 37 of which received GCOs, from the 1974-1980 period. By restricting their sample to only bankrupt companies, they provided insights into the financial characteristics of failing companies that received GCOs and those that received NGCOs.[7] They developed a logit model using seven predictor variables: (1) current ratio, (2) change in current ratio, (3) retained earnings to total assets, (4) debt to total assets, (5) income to total assets, (6) recurring operating losses, (7) cash flow from operations to total liabilities.

Menon and Schwartz validated their model using samples of bankrupt and nonbankrupt companies. The bankrupt sample included 39 firms, 14 of which received GCOs, that filed bankruptcy during the 1981-1983 period. The nonbankrupt sample included 46 nonbankrupt firms, 11 of which received GCOs, that reported net losses and negative retained earnings in 1981. For both samples, the model exhibited a 78% accuracy for classifying companies as those that received GCOs and those that received NGCOs.

Menon and Schwartz reported that the change in current ratio and recurring operating losses were statistically significant in the model; however, these findings cannot be interpreted within the context of auditors’ opinion decisions pursuant to the provisions of SAS Nos. 34 and 59. As was the case in the Mutchler (1985) and Levitan and Knoblett (1985) studies, the model was estimated using a sample selected from periods before SAS Nos. 34 and 59 were issued. Additionally, except for the two significant variables identified above, Menon and Schwartz did not evaluate the financial characteristics identified in the standards.

Chen and Church (1992) developed a logit model to predict GCOs using 127 (127) firms that received (did not receive) GCOs during the 1983 to 1986 period. The primary objective of their study was to evaluate the usefulness of companies’ default status in predicting GCOs. The variables used to develop their model were (1) cash flows from operations to total liabilities, (2) current assets to current liabilities, (3) long-term debt to total assets, (4) earnings before interest and taxes to sales, (5) one year change in current ratio, (6) log of total assets, (7) a dummy variable of 1 if the company experienced two consecutive years of operating losses, and (8) default status. The results of their logit model indicated that current assets to current liabilities, long-term debt to total assets, log of total assets, and default status were useful to auditors in the going concern decision.

Though Chen and Church used a sample selected from a period in which SAS No. 34 was effective, the results are still dated (Chen and Church 1992). Under SAS No. 59, auditors are required to consider the prospects that a company will be unable to continue as a going concern as part of every engagement; as such, it is possible that under the new standard auditors may use a different process than was used previously in deciding whether to issue GCOs. Also, as was the case with the other studies previously discussed, Chen and Church failed to consider ratio trends.

For the most part, the financial characteristics listed in SAS No. 59 (and 34) have been ignored in prior research.[8] If auditors are to use them in assessing the going concern question, it is necessary to evaluate the abilities of the characteristics to foreshadow GCOs. Also, prior studies used models for auditors’ opinion decisions developed with data from pre-SAS No. 34 periods; consequently, it is not clear how to interpret their findings under the current provisions of the auditing standards. Tests in the current study evaluate the financial characteristics’ usefulness in auditors’ going concern evaluations performed under SAS No. 59. The testable hypothesis can be set forth as follows:

Hypothesis 1: The financial characteristics listed in SAS No. 59 (and 34) are useful to auditors when evaluating the going concern assumption for post-SAS No. 59 companies.

As previously indicated, the ASB did not amend the list of financial characteristics that may signal companies with going concern problems when the board issued SAS No. 59. However, since SAS No. 59 significantly increased auditors’ responsibilities related to the going concern question, it is plausible that auditors’ reliance on the characteristics increased in post-SAS No. 59 periods. Tests in the current study evaluate auditors’ reliance on the financial characteristics listed in the standards using pre and post-SAS No. 59 companies. Specifically, the tests evaluate whether the auditors’ reliance on the characteristics increased with their responsibilities in post-SAS No. 59 periods. The testable hypothesis is stated as follows:

Hypothesis 2: When evaluating the going concern assumption, auditors’ decisions are more consistent with financial characteristics identified by SAS No. 59 (and 34) after the ASB issued SAS No. 59 than before.

Auditors’ Propensities to Issue GCOs

Carcello et al. (1995), Johnson and Khurana (1995), and Raghunandan and Rama (1995) investigated whether the proportion of firms that received GCOs prior to bankruptcy increased after SAS No. 59 became effective.[9] The latter two studies reported that auditors were more likely to issue GCOs prior to bankruptcy after SAS No. 59 was implemented. Raghunandan and Rama (Johnson and Khurana) used samples of 82 (78) and 93 (107) bankrupt companies from periods before and after the standard’s effective date.[10] The proportion of bankrupt companies that received GCOs prior to bankruptcy for the pre (post) SAS No. 59 samples ranged from 39% to 46% (57% to 62%). Both studies reported that their logistic regression results that used audit opinions as the dependent variables indicated that time (pre or post SAS No. 59) was significant.[11] The findings of the studies were interpreted as evidence that auditors were more likely to provide early warning disclosures for bankrupt companies after SAS No. 59 became effective.

The results of the Carcello et al. (1995) study were inconsistent with those of the Johnson and Khurana (1995) and Raghunandan and Rama (1995) studies discussed above. Their sample included 211 (119) bankrupt companies from the pre (post) SAS No. 59 period.[12] They reported that the proportions of firms that received GCOs prior to bankruptcy were not significantly different before (51.7%) and after (54.6%) the issuance of SAS No. 59. Additionally, their logistic regression results that used audit opinions as the dependent variables indicated that time (pre or post SAS No. 59) was not significant.[13] Carcello et al. (1995) concluded that even though the ASB issued the new standard in response to a going concern expectations gap, it does not appear that the gap between audit firm reporting and users’ expectations has been reduced. That is, the audit opinions were not more likely to signal early warnings of potential failure subsequent to SAS No. 59.

Raghunandan and Rama (1995) is the only study that examined effects of SAS No. 59 on auditors’ propensities to issue GCOs to financially distressed companies other than bankruptcies.[14] They hypothesized that auditors were more likely to issue GCOs after SAS No. 59 became effective. Their sample included 174 and 188 nonbankrupt, but financially stressed, companies from periods before and after SAS No. 59’s effective date. They reported that 22% (35%) of the pre (post) SAS No. 59 companies received GCOs. Their logistic regression results using audit opinions as the dependent variables indicated that time (pre or post SAS No. 59) was significant. They interpreted their findings as evidence that auditors are more likely to issue GCOs to financially distressed companies subsequent to the issuance of SAS No. 59; furthermore, they contended that their results were consistent with the position that the efforts of the ASB, in issuing SAS No. 59 as an expectation gap standard, were successful.

Given the level of interest expressed by the ASB, financial statement users, and government in the going concern standards, the extant research subsequent to SAS No. 59 is sparse. Furthermore, the studies do not provide consistent evidence about the standard’s effect on auditor GCOs. Only two of three studies reported that audit opinions were more likely to provide early warning signals of impending bankruptcy after SAS No. 59 became effective. Raghunandan and Rama (1995) is the only study that considered the standard’s impact on audit opinions for financially distressed companies other than bankruptcies. They reported that companies in financial distress were more likely to receive GCOs subsequent to SAS No. 59.

In general, the studies discussed above indicated that the expectations gap narrowed based on the higher proportions of GCO firms after the standard became effective; however, additional research is warranted to add credence to these findings. A potential problem with two of the studies discussed above relates to their use of bankruptcy prediction models. Carcello et al. (1995) and Johnson and Khurana (1995) used model predictions to control for the financial condition of firms.[15] To the extent these models do not accurately measure the financial stress of firms in time periods and industries different from those used to develop the model, the results of these studies may be open to question. This study provides additional evidence related to the propensity of auditors to issue GCOs before and after the effective date of SAS No. 59 using bankrupt and other financially distressed companies. The testable hypotheses related to the propensity of auditors to issue GCOs are set forth as follows:

Hypothesis 3: Auditors are more likely to issue GCOs to firms that file bankruptcy in post-SAS No. 59 periods than in pre-SAS No. 59 periods.

Hypothesis 4: Auditors are more likely to issue GCOs to financially distressed companies other than bankruptcies in post-SAS No. 59 periods than in pre-SAS No. 59 periods.

Financial Condition and Size

Prior research suggests that financial condition and size are the most important determinants in whether auditors issue GCOs (Carcello et al. 1995, Raghunandan and Rama 1995, Johnson and Khurana 1995, Chen and Church 1992, Mckeown et al. 1991, and Mutchler 1986). In general, studies have reported that relatively smaller companies and companies in poorer financial conditions are more likely to receive GCOs. Post-SAS No. 59 studies that evaluated auditors’ GCOs included control variables in their logistic models to ensure that the observed differences in the proportion of GCOs before and after SAS No. 59 were attributable to the standard, not the financial condition and size of firms (e.g., Carcello et al. 1995, Raghunandan and Rama 1995, and Johnson and Khurana 1995). However, these studies did not assess whether the financial condition and size of firms receiving GCOs changed subsequent to SAS No. 59.

As previously discussed, the ASB sought to create the perception among users and others that SAS No. 59 increased auditors’ responsibilities for evaluating the going concern question. Raghunandan and Rama (1995) and Hopwood, McKeown, and Mutchler (1994) suggest that, subsequent to the standard, higher costs are associated with not issuing GCOs to companies that subsequently fail; consequently, the level of conservatism exhibited by auditors in going concern evaluations likely increased. As indicated in the Carcello et al. (1995), Raghunandan and Rama (1995), and Johnson and Khurana (1995) studies, it is plausible to posit that the propensity of auditors to issue GCOs increased subsequent to SAS No. 59; however, it is just as plausible to suggest that auditors may issue GCOs to companies from broader ranges of financial conditions and sizes subsequent to the standard.

This study evaluates the financial condition and size of firms that received GCOs before and after the issuance of SAS No. 59. Since no evidence exists to suggest that the ASB intended to broaden the range of firms that receive GCOs, it is necessary to evaluate whether the standard had this effect on the going concern decision. The above discussion leads to the following testable hypotheses:

Hypothesis 5: Companies that received GCOs in post-SAS No. 59 periods were financially stronger than those that received GCOs in pre-SAS No. 59 periods.

Hypothesis 6: Companies that received GCOs in post-SAS No. 59 periods were larger than those that received GCOs in pre-SAS No. 59 periods.

RESEARCH DESIGN

As previously indicated, the overall objective of this study is to evaluate the impact of SAS No. 59 on auditors’ GCO decisions. This section describes the samples and tests used to evaluate hypotheses related to the study's objective. Specifically, this section describes the selection criteria used to identify the distressed companies. Also, it explains the methodology employed to evaluate the effects of SAS No. 59 on auditors’ opinion decisions.

Sample

The analyses in this study used a 1985-1987 (pre-SAS No. 59) sample and a 1988-1991 (post-SAS No. 59) sample, with each sample including distressed firms.[16] Distressed companies were defined as those reported by Compustat as meeting one or more of the following conditions: [17]

· Chapter 11 Bankruptcy: Compustat’s Industrial Annual Research File (CIAR) contains companies that were deleted from the Industrial Annual File (CIA) because of bankruptcy. CIAR identifies bankrupt firms with an 02 code for footnote 35.[18]

· Chapter 7 Liquidation: CIAR also contains companies that were deleted from CIA because of liquidation. CIAR identifies liquidated firms with an 03 code for footnote 35.

· Bonds vulnerable to default: Both CIAR and CIA report bond ratings for companies evaluated by S&P. Companies with bonds rated CCC or below were included in the distress sample. These companies were identified by codes 19-24 for data item 280.

· Low stock rating: CIAR and CIA also report stock ratings for companies evaluated by S&P. Companies whose stock was rated as “lower B” and below were included in the distressed sample. These companies were identified by codes 18-22 for data item 282.

Since analyses in this study distinguishes between bankruptcy and other financially distressed companies, the samples were partitioned into two categories: (1) those identified as distressed because of bankruptcy; (2) those identified as distressed for reasons other than bankruptcy.[19] As reported in Table 1, the pre-SAS No. 59 (post-SAS No. 59) sample included 103 (108) bankrupt companies and 50 (53) companies that were identified as financially distressed because of reasons other than bankruptcy.

The companies’ audit opinions were needed for this study’s analyses. Codes for auditors’ opinions reported on CIAR and CIA were used to identify whether the companies received GCOs or NGCOs. Companies with GCOs were defined as those reported by Compustat as meeting one of the following conditions:[20]

· Unqualified opinion with explanatory language: Both CIAR and CIA code data item 149 as 4 when auditors expressed an unqualified opinion regarding the financial statements by adding explanatory language to the standard report.[21]

· Disclaimer: Both CIAR and CIA code data item 149 as 3 when auditors refused to express opinions regarding companies’ abilities to sustain operations as going concerns.[22]

Pursuant to SAS No. 59, when auditors have substantial doubt about companies’ abilities to continue as going concerns, they are required to issue either unqualified with an explanatory paragraph or disclaimer opinions. However, virtually no authoritative guidance or published research exists that auditors could use for determining which type of opinion to issue (LaSalle and Anandarajan 1996). Furthermore, LaSalle and Anandarajan (1996) indicated that no evidence exists to suggest that the differences in auditors’ reporting decisions related to going concern decisions are systematic; consequently, this study included both unqualified with an explanatory paragraph and disclaimer opinions as GCOs. However, Table 2 shows that substantially all of the GCOs used in this study were unqualified opinions with explanatory paragraphs.

The remainder of this section discusses the tests used to evaluate the impact of SAS No. 59 on auditors’ going concern decisions. Table 3 provides a brief summary of the method and objective for each test.

Predictive-Ability of Financial Characteristics

The financial characteristics listed in SAS No. 59 (and 34) to assist auditors in their going concern evaluations are the following trends: (1) recurring operating losses, (2) working capital deficiencies, (3) negative cash flows from operations, and (4) adverse key financial ratios. Test 1 evaluated the whether auditors’ use of these financial characteristics changed after the issuance of SAS No. 59. The pre- and post-SAS No. 59 samples were partitioned into two groups: (1) those companies that received GCOs and, (2) those companies that received NGCOs. A logistic regression model was developed using audit opinions (GCO and NGCO) as dependent variables and measures of the financial characteristics set forth in the auditing standards as the independent variables.

As previously indicated, SAS No. 59 increased auditors’ responsibilities for evaluating the going concern question; however, the financial characteristics set forth by the ASB as conditions that may alert auditors to companies with going concern problems are the same for SAS Nos. 34 and 59. This test evaluated whether auditors’ use of these characteristics in going concern evaluations increased after the ASB issued SAS No. 59.

For test 1, the following logistic regression model for auditors’ opinion decisions was developed:

GC = B₀ + B₁TIME + B₂NOI + B₃NOI*TIME + B₄ CR + B₅ CR*TIME (1)

+ B₆ NOCF + B₇NOCF*TIME + B₈DTA + B₉DTA*TIME

+ B₁₀ SIZE + B₁₁ SIZE*TIME + e

where

GC = 1 for GCO, 0 otherwise;

TIME = 1 for post-SAS No. 59, 0 otherwise;

NOI = number of the previous three years with negative operating income;

CR = change in the current ratio measured as (CR_t - CR_t-2) / CR_t-2;

NOCF = number of the previous three years with negative operating cash flows;

DTA = change in the debt to total assets ratio measured as (DTA_t - DTA_t-2) / DTA_t-2;[23]

SIZE = natural log of total assets_t;[24]

e = error term.

For test 1, the variables of interest in equation (1) are those that measure the interactions between financial ratios and TIME (pre and post-SAS No. 59). If the logistic regression results indicate that the interaction variables are significant, then the evidence suggests that auditors’ reliance on the financial characteristics listed in the standards changed after the issuance of SAS No. 59.[25] Also, the significance of individual coefficients from equation (1) was used to evaluate the usefulness of the financial characteristics in auditors’ going concern judgments under the provisions of SAS No. 59.[26]

Propensity of Auditors to Issue GCOs

Tests 2 and 3 evaluate the propensity of auditors to issue GCOs before and after the issuance of SAS No. 59. The proportions of companies that received GCOs were measured using the following samples:

· Pre-bankruptcy sample (pre-B): Subset of the pre-SAS No. 59 sample containing only bankrupt companies.

· Post-bankruptcy sample (post-B): Subset of the post-SAS No. 59 sample containing only bankrupt companies.

· Pre-financial distress sample (pre-FD): Subset of the pre-SAS No. 59 sample containing only financially distressed companies other than bankruptcies.

· Post-financial distress sample (post-FD): Subset of the post-SAS No. 59 sample containing only financially distressed companies other than bankruptcies.

For each sample, the proportion of firms that received GCOs was calculated by dividing the number of firms that received GCOs by the total number of firms in the sample.

Test 2 used the proportions of bankrupt firms that received GCOs from the pre-bankruptcy and post-bankruptcy samples to evaluate whether the likelihood of auditors to issue GCOs to bankrupt companies changed under the provisions of SAS No. 59. As previously discussed, SAS No. 59 requires auditors to actively investigate the going concern question for all financial statement audits. Arguably, the auditors’ increased responsibilities also increased the cost of issuing NGCOs to companies that subsequently fail; consequently, auditors are more likely to issue GCOs to bankrupt companies after the issuance of SAS No. 59. For test 2, binomial tests were used to compare the proportion of GCO companies in the pre-B sample to that of GCO companies in the post-B sample. A significant increase in the proportion of GCO companies in the post-B sample would suggest that auditors are more likely to provide early warning signals for bankrupt companies after the issuance of SAS No. 59.

Test 3 used the proportions of GCO companies in the pre-financial distress and post-financial distress samples to evaluate whether the likelihood of auditors to issue GCOs to financially distressed companies, other than bankruptcies, changed under the provisions of SAS No. 59. Since SAS No. 59 increased auditors’ responsibilities related to the going concern issue, it is plausible that they are more conservative in their evaluations of the going concern question. As a result, auditors are more likely to issue GCOs to financially distressed companies, other than bankruptcies, under SAS No. 59 than they were before the issuance of the standard. For test 3, binomial tests were used to compare the proportion of GCO firms in the pre-FD sample to that of GCO firms in the post-FD sample. A significant increase in the proportion of GCO companies in the post-FD sample would suggest that auditors are more likely to issue GCOs to financially distressed companies, other than bankruptcies, after the issuance of SAS No. 59.

Financial Condition and Size of GCO Companies

Tests 4 and 5 evaluated whether the financial condition and size of companies that received GCOs prior to the issuance of SAS No. 59 were different from those that received GCOs subsequent to the standard. As previously discussed, prior studies indicated that the most important determinants in whether auditors issued GCOs were companies’ financial health and size. Though the post-SAS No. 59 studies controlled for these factors, they did not evaluate whether GCO companies’ financial condition and size changed after the issuance of SAS No. 59. Tests 4 and 5 evaluated GCO companies’ financial health and size using the pre- and post-SAS No. 59 samples.[27]

The analyses for Tests 4 and 5 required the following financial measures: (1) leverage--debt to total assets, (2) liquidity--cash flow from operations to total assets, (3) solvency--current assets to current liabilities, (4) performance--net income to total assets and, (5) size--natural log of total assets.[28] Testing procedures for comparing two population means were used to evaluate the financial condition (e.g., ratios) and size of GCO companies before and after the issuance of SAS No. 59.[29] Evidence that GCO firms in the post-full sample are financially stronger, or larger, than those in the pre-full sample would suggest that auditors’ GCO decisions are more conservative after the issuance of SAS No. 59. That is, they issue GCOs to both larger and financially stronger companies under the provisions of SAS No. 59.

RESULTS AND DISCUSSION

This section reports the findings of tests used to evaluate the impact of SAS No. 59 on auditors’ opinion decisions. The predictive ability of the financial characteristics listed in SAS No. 59 (and 34) are discussed. Also, the propensity of auditors to issue GCOs, as well as the financial health and size of GCO companies before and after SAS No. 59, are presented.

Financial Characteristics

Test 1 evaluated an auditors’ opinion decision model developed using the financial characteristics listed in SAS Nos. 34 and 59. Table 4 reports the univariate results for the financial characteristics used to develop the auditors’ opinion decision model shown in equation (1). The univariate results indicated that the GCO firms were significantly different from the NGCO firms for two characteristics: (1) the number of previous three years with negative operating income and (2) the number of the previous three years with negative operating cash flows. The number of previous three years with negative operating income and cash flows was greater for the GCO companies than for the NGCO companies.[30]

Test 1 evaluated whether the financial characteristics listed in SAS No. 59 (34) were useful to auditors when evaluating the going concern assumption for post-SAS No. 59 companies. Table 5 reports the logit results for the auditors’ opinion decisions model shown in equation (1). The p-values for the F-statistics reported in Table 5 indicated that the change in debt to total assets variable was useful to auditors in their going concern evaluations under the provisions of SAS No. 59. However, the number of the previous three years with negative operating income, change in the current ratio, and number of the previous three years with negative operating cash flows variables were not significant in the post-SAS No. 59 period.[31]

Test 1 also evaluated whether the auditors’ reliance on financial characteristics listed in SAS Nos. 59 (and 34) changed after the issuance of SAS No. 59. The variables of interest were those that measure the interactions between financial ratios and TIME (pre and post-SAS No. 59). Table 5 reports that the change in debt to total assets measure was the only variable that exhibited a significant interaction with TIME.[32] This suggests that auditors’ reliance on the debt to total assets variable in going concern evaluations changed after the issuance of SAS No. 59. The variables for the number of previous three years with negative operating income, the number of previous three years with negative operating cash flows, and the change in the current ratio did not exhibit significant interactions with TIME. That is, the auditors’ reliance on these financial characteristics was not affected by the issuance of SAS No. 59.

The results reported in Table 5 generally failed to support hypotheses 1 and 2.[33] Though the ASB identified negative trends for operating income, working capital, operating cash flows, and key financial ratios (i.e. leverage ratios) as signals of companies with going concern problems, only the number of previous three years with negative operating cash flows (change in debt to total assets) was significant to auditors’ opinion decisions in the pre (post) SAS No. 59 period. The financial characteristics evaluated in equation (1) are those listed in both SAS Nos. 34 and 59. As previously indicated, SAS No. 59 increased auditors’ responsibilities for evaluating the going concern question; however, auditors’ reliance on the financial characteristics in going concern evaluations was unchanged except for the change in debt to total assets measure. These findings suggest that SAS No. 59 had a modest effect on the auditors’ reliance on the financial characteristics listed in the standards. A possible explanation is that the ASB simply codified existing practice with the issuance of SAS No. 59.

Propensity of Going Concern Opinions

Test 2 evaluated the propensity of auditors to issue GCOs to bankrupt companies in the pre and post-SAS No. 59 periods. The results reported in Table 6 indicated that the proportion of bankrupt companies with GCOs in the post-SAS No. 59 period (53.7%) was significantly higher than that in the pre-SAS No. 59 period (39.8%). The significant increase in the proportion of bankruptcies with GCOs in the post-SAS No. 59 period provides support for hypothesis 3. That is, auditors were more likely to issue GCOs to companies that subsequently declared bankruptcy under the provisions of SAS No. 59.[34] This result supports the notion that the increased responsibilities imposed on auditors by SAS No. 59 also increased the costs associated with not issuing GCOs when companies that subsequently fail; consequently, auditors were forced to be more conservative in their going-concern evaluations.

Test 3 evaluated the propensity of auditors to issue GCOs to financially distressed companies other than bankruptcies, before and after the issuance of SAS No. 59. The results in Table 6 indicated that the proportion of other distressed companies with GCOs in the post-SAS No. 59 period (49%) was higher than that in the pre-SAS No. 59 period (40%). The higher proportion of other distressed companies with GCOs in the post-SAS No. 59 period is consistent with hypothesis 4; however, the binomial test indicated that he proportions of other distressed companies with GCOs were not significantly different between the pre and post-SAS No. 59 periods.[35] Though auditors were more conservative in their going concern evaluations for bankruptcies under the provisions of SAS No. 59, the standard did not significantly alter their GCO decisions for financially distressed firms other than bankruptcies.

Financial Condition and Size of GCO Firms Under SAS No. 59

Tests 4 and 5 evaluated the financial condition and size of companies that received GCOs before and after the issuance of SAS No. 59. Financial health was measured using financial ratios that proxy for firms’ leverage, liquidity, solvency, and performance. The results reported in Table 7 indicated that the total liabilities to total assets (leverage) variable for companies that received GCOs in the post-SAS No. 59 period was significantly higher than that for companies that received GCOs in the pre-SAS No. 59 period. However, the operating cash flows to total assets (liquidity), current assets to current liabilities (solvency), and net income to total assets (performance) variables were not significantly different for the GCO companies between the pre and post-SAS No. 59 periods. Also, the results reported in Table 7 indicated that the size of companies that received GCOs before and after the issuance of SAS No. 59 was not significantly different.

The findings of Tests 4 and 5 did not provide support for hypotheses 5 and 6.[36] Overall, the financial condition and size of GCO firms in the post-SAS No. 59 period were similar to those of GCO firms in the pre-SAS No. 59 periods. As previously discussed, prior research suggests that financial condition and size are the most important determinants in whether auditors issued GCOs to problem companies. Though SAS No. 59 likely increased the level of conservatism exhibited by auditors in their going concern evaluations, it did not result in auditors issuing GCOs to firms that were financially stronger and larger than those prior to the issuance of the standard.[37]

SUMMARY

This study evaluated an auditors’ opinion decision model developed using the financial characteristics listed in SAS No. 59 (and 34). The results indicated that the change in debt to total assets variable was useful to auditors in going concern evaluations under the provisions of SAS No. 59. However, the number of the previous three years with negative operating income, change in the current ratio, and number of the previous three years with negative operating cash flows variables were not significant in post-SAS No. 59 going concern evaluations. Variables that measured the interaction between the financial characteristics and time (pre and post-SAS No. 59) also were included in the model. The change in debt to total assets measure was the only financial characteristic that exhibited a significant interaction with time. That is, the issuance of SAS No. 59 generally did not affect auditors’ reliance on the financial characteristics even though the standard increased auditors’ responsibilities related to going concern evaluations.

The proportion of post-SAS No. 59 bankruptcies that received GCOs was significantly higher than that of pre-SAS No. 59 bankruptcies that received GCOs. Thus, auditors were more likely to issue GCOs to bankrupt firms for audits performed after the issuance of SAS No. 59. The proportion of post-SAS No. 59 financially distressed firms other than bankruptcies that received GCOs was higher than that of pre-SAS No. 59 financially distressed other than bankruptcies that received GCOs; however, these proportions were not significantly different between the pre and post-SAS No. 59 periods. Additionally, post-SAS No. 59 GCO companies were not significantly different than pre-SAS No. 59 GCO companies in terms of financial health and size.

In sum, the findings of this study suggest that SAS No. 59 had a modest impact on auditors’ going concern decisions. Though auditors were more likely to issue GCOs to bankrupt companies after the issuance of SAS No. 59 than before, the post-SAS No. 59 GCO companies were not financially stronger or larger than pre-SAS No. 59 GCO companies. Additionally, except for the change in debt to total assets measure, auditors’ reliance on the financial characteristics listed in SAS No. 59 (and 34) did not change under the provisions of SAS No. 59. It seems that the ASB simply codified existing practice when they issued the more stringent standard.