An empirical investigation of IPO underpricing and the change to the LLP organization of audit firms.
This paper provides evidence on the relationship between the change in audit firm legal structure to limited liability partnerships (LLPs) and the pricing of initial public offerings (IPOs). During the early 1990s, the auditing profession asserted that it faced a liability crisis." (1) In response to the adverse litigation climate, the profession successfully lobbied for changes to reduce audit firms' legal liabilities. One particularly significant change was the AICPA's by-law changes in 1992 that allowed audit firms to practice under any organizational form allowed by state laws. This change enabled audit firms to practice under limited liability (LL) organizational forms. As state laws were changed to allow LL forms of organization, many audit firms converted to LL forms of practice. By late 1994, all Big 6 audit firms converted from general partnerships to LLPs.
Audit firms provide implicit insurance in the case of corporate failures (Wallace 1980). Since audit firm liability arises from opining upon misleading financial statements, a firm's association with the financial statements provides insurance to investors and financial intermediaries who rely on the financial statements. Under a LLP, personal assets of individual partners, who were not directly involved in a client's litigation, are no longer available to pay a partnership's liabilities. Thus, the LLP form of practice reduces the implicit insurance provided by audit firms. (2) In addition, the reduction in auditor legal liability associated with the LLP form of practice can also reduce audit quality. In a general partnership, every partner is personally liable for all services provided by the audit firm, which provides an incentive for partners to monitor the quality of the services provided by their fellow partners. Under LLP provisions, nonnegligent partners are absolved of personal liability. To the extent that intrafirm monitoring incentives are reduced, audit quality could be reduced subsequent to the adoption of a LLP organizational form. (3)
Several features of the IPO market increase the importance of the services provided by audit firms that are associated with the IPOs. First, there is relatively little public information available about IPO organizations, so the proxy statement, financial reports, and the audit firm's opinion are vital sources of information for underwriters and investors. Second, companies entering the IPO market often have a high level of financial risk. For example, IPO companies more often exhibit financial distress than established companies (Ritter 1991; Beatty 1993). Finally, IPOs are covered by the 1933 Securities Act, a law with features favorable to plaintiffs for recovering damages.
We use IPO underpricing to measure the security pricing implications of the change in auditor liability status. Underpricing occurs when the initial offer price of a security is less than its trading price shortly after the security begins trading. Tinic (1988) proposes that underpricing is an implicit insurance premium charged by underwriters. Underwriters offer the securities at a discount to protect themselves and stock issuers against legal liabilities and associated damages. Generally, the IPO offer price sets an upper limit on the maximum damages that investors can sue for under the 1933 Securities Act. Lowering the initial offer price serves as a form of insurance for underwriters against post-offering litigation losses. In essence, IPOs contain a "put option"--that is, an option to sue to recover losses, given to the IPO purchasers, and underpricing reduces the cost of the "put option" to the underwriter.
The change to LLP status reduced the funds available to settle litigation related to failed IPOs. Underwriters may respond rationally to the reduction in funds available from audit firms as potential joint and several co-defendants in litigation by increasing IPO underpricing for IPOs with a significant litigation risk. The increase in underpricing would enable underwriters to protect themselves from the reduction in funds available from audit firms as co-defendants. Further, audit partners are likely to have less incentive to monitor the decisions of their fellow audit partners under LLPs since they no longer bear the risk of personal liability for other partners' actions. Beatty (1989) argues that lower audit quality increases the information asymmetry between entrepreneurs and potential investors, which may be associated with an increase in the amount of underpricing of IPOs to compensate investors for the increased risk they face when purchasing shares in IPOs. If the reduction in incentives to monitor reduces audit quality, then underwriters may further underprice IPOs in the post-LLP period.
We use high-technology industry membership, the size of the offering, and the 1POs subsequent standard deviation of returns as proxies for litigation risk, and find that IPOs with greater litigation risk experience more underpricing after audit firms changed to LLP status. Our findings suggest that underwriters perceived a decline in the value of services provided by audit firms associated with the audit firm's change in legal structure, although we are not able to distinguish whether the decline in value results from a reduction in implicit insurance or a reduction in audit quality. Even so, the analysis contributes to the understanding of the economic role of auditors by specifically examining whether the auditors' liability status is associated with valuation decisions made by underwriters and companies going public.
The remainder of this paper is organized as follows. The next section presents a review of prior literature and develops the research hypothesis. We then provide a description of the methods and variables that we use, followed by the empirical analysis and results. Finally, we provide a summary and discussion of our findings.
PRIOR LITERATURE AND HYPOTHESIS
Limited Liability Partnerships
An audit firm's liability depends partially on its organizational form of practice. As a general partnership, the assets of an audit firm and the personal wealth of all partners are subject to loss from litigation. Similar to general partnerships, LLPs subject the firm's assets to attachment in a lawsuit. In addition, a partner's personal wealth is at risk for his or her malfeasance or negligence. However, a partner's personal wealth is not subject to attachment for the malfeasance or negligence of other partners of the audit firm. Theoretically, general partnerships and LLPs differ in the provision of implicit insurance by the aggregate wealth of all partners not associated with a particular client. (4)
Prior to 1992, audit firms were limited in their choice of organizational form, typically operating as general partnerships. During 1992, the AICPA amended its by-laws to allow accounting firms to operate in any organizational form allowed by state laws (Simonetti and Andrews 1994). Since the amendment to the AICPA's by-laws, most firms have adopted a LLP form of organization. Under LLP provisions, partners are afforded LLP protection only for acts that occur after the partnership is registered as an LLP. For acts committed prior to registration as an LLP, the partnership is treated under the firm's legal form at the time of the act, principally general partnerships (McGaughey 1996). (5)
The Insurance Hypothesis
The insurance hypothesis posits that audit firms provide implicit insurance in the case of corporate failures via the potential for investors to sue the auditors for attesting to misleading information (Wallace 1980). While intuitively appealing, evidence regarding the validity of the insurance hypothesis is inconsistent. For example, Schwartz and Menon (1985) study audit firm switches by failing companies and hypothesize that those companies might switch to larger audit firms because such firms provide insurance against potential claimants in the event of financial failure. They find that failing companies are more likely to switch audit firms, but they find no evidence that failing companies switch to larger audit firms with more potential resources in the event of litigation. The results appear inconsistent with the insurance hypothesis, but may reflect the unwillingness of the large audit firms to accept the potential liability presented by these riskier clients. Menon and Williams (1994) find that the bankruptcy announcement of Laventhol and Horwath (L&H) adversely affected the stock prices of L&H clients. However, it is not clear whether the adverse affect was caused by a reduction of implicit insurance provided by the audit firm, or whether it was simply attributable to investors having lost confidence in the assurance provided by L&H. This conclusion is supported by a similar study of the L&H bankruptcy conducted by Baber et al. (1995), who were also unable to discriminate between the two possibilities.
In a related study examining LLP status and the insurance hypothesis, Chung et al. (1998) find statistically significant negative abnormal returns for Big 6 clients around the date that the Big 6 audit firms announced the change to LLP status. They also provide evidence that weaker and riskier companies exhibit a stronger negative market reaction to the LLP formation by their audit firms. Both results are consistent with the insurance hypothesis. The interpretation offered in Chung et al. (1998) is that the shift to LLP status reduced the amount of insurance provided by audit firms.
The Audit Quality Hypothesis
Principal agent models demonstrate how auditor liability mitigates the moral hazard problem associated with the information asymmetry between auditors and their clients by motivating auditors to exert the appropriate level of effort (Antle 1982). Dye (1993) presents an analytical model in which the audit plays both an information role and an insurance role. Under certain conditions, Dye (1993) demonstrates that the auditor's wealth at risk from litigation becomes a "bond" for audit quality. Other studies have also examined the link between auditor liability and audit quality (e.g., Dye 1995; Schwartz 1997; Nelson et al. 1988). Because auditors can reduce expected litigation and damages by performing higher-quality audits, the threat of litigation provides an economic incentive for auditors to perform high-quality audits. The change to the LLP form of practice reduces the litigation risk faced by partners not directly involved in the litigating client. As a result, the change to LLP may result in less peer monitoring within the firm. Partners could potentially become less concerned with the risks taken by their fellow partners, and may even begin to admit more risk-taking partners to the partnership. On average, the reduction in auditor liability associated with the shift to LLP status could reduce the economic incentive for auditors to perform high-quality audits.
IPO Underpricing As a Response by Underwriters
Prior empirical evidence demonstrates that IPOs typically have large positive abnormal returns during the period immediately following the initial offering, a phenomenon called "underpricing" (i.e., underwriters and issuing companies set the initial offer price of a security lower than its expected post-offer trading price). (6) Early research by Ibbotson (1975) notes that underwriters and the issuing companies may use underpricing as a form of insurance against litigation. Underpricing acts as insurance because the upper bound on litigation exposure for both the issuing company and the underwriter is determined by the offer price. (7) Consistent with this idea, Tinic (1988) finds that IPO underpricing is higher during the post- 1933 Securities Act period, concluding that underwriters reduced offer prices to minimize possible litigation losses after the law passed.
Prior research suggests an association between the riskiness of the underlying IPO and underpricing. For example, Beatty and Ritter (1986) demonstrate that underpricing varies directly with measures of ex ante uncertainty. Examining ex ante uncertainty in a different context, Willenborg (1999) studies the small deal segment of the IPO market, by partitioning the sample of IPOs based upon whether the company is or is not a Development Stage Enterprise (DSE). He finds that underpricing is greatest for the DSE sample, which is expected if such companies have a higher ex ante risk of failure. This evidence is consistent with underwriters and issuers setting lower offer prices to (1) provide implicit insurance against litigation and (2) compensate initial investors for increased risk.
Prior to the 1995 Securities Litigation Reform Act, liability under securities laws was joint and several among the defendants. Auditors and underwriters were typically considered "deep pockets" by plaintiffs and in many cases provided the only potential recovery when an IPO failed. Under joint and several liability, auditors and underwriters shared liability costs. Removal of a portion of the auditors' resources available to pay those costs as the result of the change to LLP status potentially transferred greater liability to the underwriters. The logical response by underwriters is to increase underpricing as a means to reduce potential legal damages. In addition, several studies of the IPO market show an inverse relationship between auditor quality and IPO underpricing (Simunic and Stein 1987; Balvers et al. 1988; Beatty 1989; Michaely and Shaw 1995). These studies attribute their findings to the information-signaling role of audits. That is, audits are quality differentiated based upon audit firm size. (8) Larger audit firms provide higher-quality audits, and higher-quality audits reduce ex ante uncertainty for investors. The reduction in ex ante uncertainty, in turn, reduces underpricing. Therefore, if the conversion to LLPs reduced audit quality through reduced intrafirm monitoring, it should result in greater ex ante uncertainty and an increase in IPO underpricing, particularly for riskier securities.
For reasons described under "The Insurance Hypothesis" and "The Audit Quality Hypothesis" subsections, we expect greater underpricing after audit firms shift to LLP status. Further, the impact of the LLP change is strongest for IPOs with the greatest risk of post-IPO litigation, so we expect that the greatest underpricing occurs for riskier securities in the post-LLP period. (9) This discussion suggests the following hypothesis:
H1: IPO underpricing increases for high litigation risk companies in the post-LLP period compared to high litigation risk companies in the pre-LLP period.
The sample includes 486 companies listed on the Securities Data Company database for initial public offerings between November 1, 1993 and May 31, 1995. (10) Of these, 56 have missing data, so our final sample includes 430 IPO companies. (11) This specific time period provides an equal time period (nine months) before and after audit firms changed to LLP status, which occurred during August and September 1994. (12) Table 1, Panel A shows that the manufacturing (46.8 percent) and service (26.7 percent) industries make up the largest segments in our sample. Our sample also includes a significant number of high-technology companies (158 of 430 companies, or 36.7 percent).
Our primary empirical tests consider how audit firm LLP status is associated with the underpricing of riskier IPOs. The dependent variable in our models is In (1 + UNDERPRICING), which is the natural log of the difference between the security's offer price and the first-day closing market price, divided by the offer price. (13) We specify pre-LLP and post-LLP periods according to the audit firm's LLP status on the date each security was issued. For example, LLP = 1 (LLP = 0) for a security issued after (before) September 1, 1994 with Arthur Andersen as the audit firm, since Arthur Andersen became a LLP on September 1, 1994.
Proxies for High Litigation Risk Securities
We use three alternative proxies for high litigation risk securities, in order to provide assurance that results of our hypothesis tests are not driven by the choice of a particular risk proxy: (1) high-technology companies, (2) larger size offerings, and (3) stocks that subsequently have a higher standard deviation of returns. High-technology companies are a proxy for litigation risk because such companies are the most likely to be involved in securities litigation (Jones and Weingram 1996). We measure a company's high-technology industry membership (HIGHTECH) using the categories developed by Francis and Schipper (1999) (i.e., the computer, electronics, pharmaceuticals, and telecommunications industries). (14) The second proxy for litigation risk is the offer size (OFFERSIZE). While larger offerings reduce ex ante uncertainty (i.e., because institutional investors have greater incentive to search for private information prior to the IPO), Bohn and Choi (1996) show that the majority of IPO litigation is concentrated in larger IPOs because plaintiff attorneys have greater incentives to pursue larger recoveries. In the test of the interaction between OFFERSIZE and LLP, OFFERSIZE is a dichotomous variable equal to 1 for IPOs larger than the median offer size in our sample, and equal to 0 otherwise. The third proxy for litigation risk is the volatility of the security price following the IPO (STDRETURNS). To the extent underwriters anticipate this volatility, the proxy measures the litigation risk associated with significant declines in stock price.
The focus is on the interactions between the LLP variable and each of the litigation-risk variables. However, the three interaction variables are highly correlated with each other and LLP. If all three variables are included in the same model, then the collinearity inflates the standard error of the coefficients, making inferences difficult. (15) To alleviate the collinearity issue, we present three separate regressions, one that includes the HIGHTECH x LLP interaction, a second that includes the OFFERSIZE x LLP interaction, and a third that includes the STDRETURNS x LLP interaction. (16)
To assess the interactive effects of LLP and our three proxies for risky securities, we control for other factors known to affect UNDERPRICING. First, litigation involving IPO companies is most likely when these companies suffer financial hardship. St. Pierre and Anderson (1984) provide evidence of a direct relationship between bankruptcies and litigation brought against audit firms. They also provide evidence that audit firm lawsuits are frequently associated with significant losses by the client company even without the client company declaring bankruptcy. Similarly, Carcello and Palmrose (1994) show that audit firms are defendants in 74 percent of litigation cases that arise from a company's bankruptcy, and that the litigation rate against audit firms is higher for bankrupt companies than an estimated rate of litigation for all public companies. Lys and Watts (1994) also provide evidence that audit firms are more likely to be sued if client firms experience financial difficulty or have poor stock price performance. These studies suggest that the risk of securities litigation increases with financially distressed clients.
However, recent research suggests that up to 25 percent of companies issuing small IPOs receive going-concern opinions (Willenborg and McKeown 2000), demonstrating the severity of these companies' financial weakness. Willenborg and McKeown (2000) show that underpricing is lower for companies with going-concern opinions. They interpret this result as consistent with the going-concern opinion clearly signaling the risks of the IPO and thereby reducing future liability to the underwriter and audit firm. Overall, prior evidence with respect to the financial risk of IPOs is mixed, so we include financial condition as a control variable but make no prediction on its sign. We measure a company's financial condition (FINCONDITION) using the index developed by Zmijewski (1984), (17) where higher values of the index indicate weaker financial condition (i.e., higher likelihood of bankruptcy). (18)
We control for the reputation of the IPO's underwriter (UNDERWRITER) using the Carter-Manaster underwriter ranking index (Carter and Manaster 1990). High-quality underwriters are expected to reduce ex ante uncertainty, thereby reducing underpricing (e.g., Carter and Manaster 1990). We also control for the percent of ownership retained by existing owners (PCTINSIDER). Grinblatt and Hwang (1989) develop an analytical model from which they derive the proposition that underpricing is positively related to the retained ownership. However, it has also been proposed that a larger percentage of retained ownership can be a signal of quality that reduces ex ante uncertainty, which in turn reduces underpricing (Beatty 1989). The empirical evidence is mixed, where both the percentage of ownership retained (Beatty and Welch 1996; Willenborg 1999) and the percentage of insider ownership offered (Beatty 1989) are shown to be positively related to underpricing. Therefore, we make no directional prediction for the PCTINSIDER variable.
We control for the average market return in the 15 days prior to the issuance of the IPO (MRKTRETURN), because prior research shows that underpricing is positively associated with those returns (Loughran and Ritter 2002; Lowry and Schwert 2002). We control for the absolute percentage change in the offer price between the final offer price and the average expected offer price ([DELTA]OFFERPRICE) because Hanley (1993) shows that this price revision reflects important information revealed in the pre-issue market that is positively associated with underpricing. (19) Finally, we control for whether the IPO is listed on a major stock exchange (EXCHANGE = 1 if issued on NYSE or AMEX; = 0 otherwise), because such listings are associated with reduced underpricing (Lowry and Schwert 2002).
Model of IPO Underpricing
We model IPO underpricing as a function of LLP status, the three proxies for litigation risk, LLP x litigation-risk proxy interactions, and control variables. Three separate regressions are presented, and each includes HIGHTECH, OFFERSIZE, and STDRETURNS as indicators for litigation risk, respectively:
ln(1 + UNDERPRICING) = [b.sub.0] + [b.sub.1] LLP + [b.sub.2] HIGHTECH + [b.sub.3] lnOFFERSIZE + [b.sub.4] STDRETURNS + [b.sub.5] (LLP x LITIGATION RISK PROXY) + [b.sub.6] FINCONDITION + [b.sub.7] UNDERWRITER + [b.sub.8] PCTINSIDER + [b.sub.9] MRKTRETURN + [b.sub.10] [DELTA] OFFERPRICE + [b.sub.11] EXCHANGE + u where:
UNDERPRICING = (first day closing market price--offer price)/offer price;
LLP = 1 if first trade date is after August 1, 1994 for companies audited by Coopers & Lybrand, Ernst & Young, and Price Waterhouse, after August 15, 1994 for companies audited by KPMG Peat Marwick and Deloitte & Touche, and after September 1, 1994 for companies audited by Arthur Andersen; 0 otherwise;
OFFERSIZE = number of shares offered x offer price per share (in millions);
HIGHTECH = 1 if a member of one of the high technology company categories identified as in Francis and Schipper (1999) (i.e., SIC codes 283, 357, 360, 361, 362, 363, 364, 365, 366, 367, 368, 481, 737, and 873); 0 otherwise;
STDRETURNS = the standard deviation of monthly market adjusted returns for the IPOs first year of trading;
LLP x OFFERSIZE = LLP x an indicator variable equal to 1 if the IPOs offer size is greater than the median offer size in our sample; 0 otherwise;
FINCONDITION = the financial condition index derived in Zmijewski (1984), where higher values indicate weaker financial condition;
UNDERWRITER = the Carter-Manaster underwriter ranking index (Carter and Manaster 1990), where higher values indicate more prestigious underwriters;
PCTINSIDER = percentage of ownership in company retained by owners who held an interest prior to the IPO;
MRKTRETURN = average market return in 15 days prior to the offer date; [DELTA]OFFERPRICE = |(offer price per share - average expected offer price per share)/average expected offer price per share|; and
EXCHANGE = 1 if the stock is traded on the NYSE or AMEX; 0 otherwise.
Table 1, Panel B presents descriptive statistics for the pre- and post-LLP periods. Two variables have statistically significant differences between the pre and post LLP periods. UNDERPRICING is significantly higher in the post-LLP period. Mean underpricing increases from 8 percent to nearly 16 percent. Given the mean offer size for our full sample of $45 million, an 8 percent increase in underpricing represents about $3.6 million in additional underpricing in the post-LLP period. The only other variable that differs between periods is MRKTRETURN, which increases in the post-LLP period. Our multivariate tests take on additional importance given that the univariate results with respect to UNDERPRICING could reflect the positive correlation between MRKTRETURN and UNDERPRICING.
Table 1, Panel C presents univariate tests of our hypothesis that IPO underpricing increases for high-litigation-risk companies in the post-LLP period compared to high-litigation-risk companies in the pre-LLP period. The results are consistent with this expectation. Using STDRETURNS as a litigation-risk proxy, UNDERPRICING increases from 7.4 percent in the pre-LLP period to 18.1 percent in the post-LLP period (t = 5.133, p = 0.000) for the higher-litigation-risk securities. Using HIGHTECH as a litigation-risk proxy, UNDERPRICING increases from 9.7 percent in the pre-LLP period to 20.9 percent in the post-LLP period (t = 4.199, p = 0.000) for the higher-litigation-risk securities. Using lnOFFERSIZE as a litigation-risk proxy, UNDERPRICING increases from 7.5 percent in the pre-LLP period to 15.6 percent in the post-LLP period (t = 3.387, p = 0.001) for the higher-litigation-risk securities. A similar increase is evident in the univariate tests for the lower-litigation-risk securities. Although the results of t-tests are not significant for the STDRETURNS comparison (t = 0.738, p = 0.461), they are marginally significant for the HIGHTECH comparison (t = 1.743, p = 0.082) and the t-test is significant for the lnOFFERSIZE comparison (t = 2.369, p = 0.019). Overall, the univariate results suggest the strongest effect of the shift to LLP status is for the higher-litigation-risk securities. Of course, these results need to be confirmed in more-powerful multivariate tests that control for other potentially confounding factors, and we report results of those tests below.
Table 2 presents a correlation matrix. Underpricing is positively associated with LLP, HIGHTECH, STDRETURNS, UNDERWRITER, PCTINSIDER, MRKTRETURN, and [DELTA]OFFERPRICE, and is negatively associated with FINCONDITION and EXCHANGE. Underpricing is also positively correlated with the three interaction variables, consistent with the prediction of HI. The interaction variables are also highly correlated with each other and with LLP, which supports the tactic of not including all three interaction terms in the same model.
Multivariate Analysis of IPO Underpricing
We regress IPO underpricing on LLP status, the three measures of litigation risk (HIGHTECH, lnOFFERSIZE, and STDRETURNS), the interactions of LLP x HIGHTECH, LLP x lnOFFERSIZE, and LLP x STDRETURNS, and control variables in three separate linear regression models. (20) Table 3 shows that the models are significant, explaining about 15 percent of the variation in IPO underpricing, which is consistent with prior research (e.g., Willenborg 1999).
The results show greater underpricing in the post-LLP period for HIGHTECH securities (t = 2.539, p = 0.005), for larger offers (t = 1.830, p = 0.034), and for securities with higher STDRETURNS (t = 2.357, p = 0.009). These results strongly support our hypothesis that the pricing of IPOs with the greatest potential liability is strongly affected by the change in audit firm liability status that reduced the amount of recovery available from audit firms.
Results for control variables are consistent with expectations as well. Underpricing is lower for financially weaker clients, and for those traded on major exchanges (e.g., in Model 1a, t = -2.170, p = 0.031 and t = -2.055, p = 0.020, respectively). In addition, underpricing is greater when the MRKTRETURN is higher immediately prior to the offer date (e.g., in Model 1a, t = 1.985, p = 0.024), and the price revision ([DELTA]OFFERPRICE) is greater (e.g., in Model 1a, t = 1.991, p = 0.024). UNDERWRITER and PCTINSIDER are insignificant across all three models.
In this section, we explain sensitivity tests conducted to consider potential alternative explanations for our results. Two events occurred during our sample period that potentially affect IPO underpricing. The Central Bank of Denver U.S. Supreme Court case (April 19, 1994) reduced the ability to impose liability on auditors via the Racketeer Influenced and Corrupt Organizations Act (RICO) (Hanson and Rockness 1994). The practical effect of the ruling is that auditors can no longer be held liable as "aiders and abettors" of a primary wrongdoer under Section 10b of the 1934 Securities Exchange Act. This arguably makes it more difficult to hold auditors liable, because plaintiffs must argue that the auditor is a primary wrongdoer. Dampening the potential effect of this event on our results, the Reves et al. U.S. Supreme Court case in early 1993 (i.e., preceding our sample period) had already limited auditors' liability under RICO (Bunim and Jacobson 1994).
The second event was the Republican Congressional victory in the November 1994 election. During their congressional campaigns that year, Republicans proposed significant tort reform, which could thereby limit auditors' liability. It is important to note that simply electing the Republican Congress did not change auditor or underwriters' liability, and even when laws are enacted, they are generally not retroactive. Thus, the fact that auditors ultimately got additional liability protection in late 1995 has little effect on auditor or underwriters' liability for IPOs conducted in 1994.
Given the potential effects of these events, we conducted univariate and multivariate tests to evaluate their potential impact. In our additional tests, we substituted the two alternative event dates instead of the LLP dates we use in our study. Univariate and multivariate comparisons suggest the Denver case had no effect on underpricing. Our tests based on the election of the Republican Congress are more difficult to interpret. In our univariate tests, the Republican election date is statistically significant, but is less significant than when the LLP date is used. In multivariate tests, the Republican election date and risk interactions are also significant, but again less so than when we use the LLP-based interactions. We ran the model with both the LLP interactions and the Republican Congress interactions and only the LLP interactions were significant, which suggests that the change in LLP status explains the increase in underpricing rather than the election of the Republican Congress in 1994.
SUMMARY AND DISCUSSION
The purpose of this paper is to investigate the relationship between the change in audit firm legal structure to limited liability partnerships (LLPs) and the pricing of initial public offerings (IPOs). In conducting this investigation, we note that the analysis is subject to the common limitation associated with event studies in that we attribute the results to the change in LLP status, whereas an alternative explanation may be influencing our results. We strengthen the reliability of our results by incorporating control variables known to explain underpricing, and by defining the event window narrowly enough to reduce the risk of other events influencing our findings. We also examine and discuss alternative explanations for the findings: the two important events that occurred during our sample period, including the Central Bank of Denver case, and the Republican Congressional victory in the November 1994 election. We contend on both theoretical and empirical grounds that these potential alternative explanations are unlikely explanations for our results.
Ultimately, our results provide evidence on an important audit and societal issue, documenting that audit firms' change to a LLP form of practice is associated with underwriters' underpricing and further demonstrating the important economic role of audits. This finding is consistent with both the theory that auditors provide implicit insurance as potential co-defendants in IPO-based litigation and the theory that underwriters perceive audit quality as a factor in determining the prices of securities.
TABLE 1 Summary Statistics Panel A: Industry Distribution SIC Industry Description n % # HIGHTECH 10-14 Mining 12 2.8 0 15-17 Construction 3 0.2 0 20-39 Manufacturing 201 46.8 78 40-49 Transportation, communications 39 9.2 10 50-51 Wholesale trade 23 5.4 0 52-59 Retail trade 38 8.9 0 70-89 Service 114 26.7 70 Total 430 100.0 158 Panel B: Descriptive Statistics (n = 430) Pre-LLP Post-LLP (n = 251) (n = 179) Variable Mean Median Mean Median UNDERPRICING 0.080 0.039 0.158 0.080 OFFERSIZE 42.840 25.990 50.043 31.050 FINCONDITION 0.499 5.602 0.199 4.909 STDRETURNS 0.156 0.069 0.161 0.061 UNDERWRITER 6.710 3.129 6.898 3.109 PCTINSIDER 0.384 0.248 0.410 0.233 MRKTRETURN 0.008 0.022 0.015 0.019 [DELTA]OFFERPRICE 0.129 0.122 0.142 0.124 HIGHTECH 0.347 0.477 0.397 0.491 EXCHANGE 0.139 0.347 0.156 0.364 t-test Wilcoxon Test t p-value Z p-value UNDERPRICING -4.28 0.000 3.90 0.000 OFFERSIZE -0.94 0.348 1.84 0.066 FINCONDITION 0.58 0.565 0.34 0.735 STDRETURNS -0.67 0.505 1.26 0.200 UNDERWRITER -0.62 0.538 1.28 0.207 PCTINSIDER -1.11 0.263 0.99 0.323 MRKTRETURN -3.61 0.001 3.60 0.000 [DELTA]OFFERPRICE -1.08 0.279 1.09 0.274 HIGHTECH -1.06 0.289 1.06 0.285 EXCHANGE -0.49 0.625 0.49 0.624 Panel C: ComparisonS of UNDERPRICING in Pre-LLP and Post-LLP periods for High- versus Low-Risk IPOs (n = 430) t-test of Differences between Pre-LLP Post-LLP Pre- and Post-LLP Periods n = 251 n = 179 t-test (two-tailed p-value) Low STDRETURNS 0.063 0.073 0.738 (0.461) High STDRETURNS 0.074 0.181 5.133 (0.000) HIGHTECH = 0 0.054 0.080 1.743 (0.082) HIGHTECH = 1 0.097 0.209 4.199 (0.000) Low lnOFFERSIZE 0.063 0.102 2.369 (0.019) High lnOFFERSIZE 0.075 0.156 3.387 (0.001) Variable Definitions: UNDERPRICING = (first day closing market price--offer price)/offer price; OFFERSIZE = number of shares offered X offer price per share (in millions); FINCONDITION = the financial condition index derived in Zmijewski (1984), where higher values indicate weaker financial condition; STDRETURNS = the standard deviation of monthly market adjusted returns for the IPO's first year of trading; UNDERWRITER = the Carter-Manaster underwriter ranking index (Carter and Manaster 1990), where higher values indicate more prestigious underwriters; PCTINSIDER = percentage of ownership in company retained by owners who held an interest prior to the IPO; MRKTRETURN = average market return in 15 days prior to the offer date; [DELTA]OFFERPRICE = |(Offer price per share--Average expected offer price per share)/ Average expected offer price per share|; LLP = 1 if first trade date is after August 1, 1994 for companies audited by Coopers & Lybrand, Ernst & Young, and Price Waterhouse, after August 15, 1994 for companies audited by KPMG Peat Marwick and Deloitte & Touche, and after September 1, 1994 for companies audited by Arthur Andersen; 0 otherwise; HIGHTECH = 1 if a member of one of the high technology company categories identified as in Francis and Schipper (1999) (i.e., SIC codes 283, 357, 360, 361, 362, 363, 364, 365, 366, 367, 368, 481, 737, and 873); 0 otherwise; and EXCHANGE = 1 if the stock is traded on the NYSE or the AMEX, 0 otherwise. TABLE 2 Correlation Matrix (n = 430) 1 2 3 4 1. lnUNDERPRICING 2. LLP 0.21 *** 3. HIGHTECH 0.27 *** 0.05 4. lnOFFERSIZE 0.02 0.07 -0.17 *** 5. STDRETURNS 0.13 *** 0.03 0.32 *** -0.28 *** 6. FINCONDITION -0.09 * -0.03 -0.01 -0.07 7. UNDERWRITER 0.10 ** 0.03 0.05 0.58 *** 8. PCTINSIDER 0.10 ** 0.05 0.07 -0.27 *** 9. MRKTRETURN 0.13 *** 0.17 *** -0.04 -0.01 10. [DELTA]OFFERPRICE 0.14 *** 0.05 0.12 ** -0.01 11. EXCHANGE -0.14 *** 0.02 -0.25 *** 0.44 *** 12. LLP x OFFERSIZE 0.34 *** 0.53 *** 0.58 *** -0.03 13. LLP x HIGHTECH 0.27 *** 0.90 *** 0.18 *** 0.01 14. LLP x STDRETURNS 0.22 *** 0.62 *** 0.01 0.41 5 6 7 1. lnUNDERPRICING 2. LLP 3. HIGHTECH 4. lnOFFERSIZE 5. STDRETURNS 6. FINCONDITION 0.28 *** 7. UNDERWRITER -0.15 *** -0.14 *** 8. PCTINSIDER 0.11 ** -0.07 -0.12 ** 9. MRKTRETURN 0.04 0.02 -0.05 10. [DELTA]OFFERPRICE 0.16 *** 0.06 0.19 *** 11. EXCHANGE -0.26 *** -0.01 0.10 ** 12. LLP x OFFERSIZE 0.25 *** 0.02 0.10 ** 13. LLP x HIGHTECH 0.29 *** 0.03 0.04 14. LLP x STDRETURNS -0.01 -0.07 0.26 *** 8 9 10 1. lnUNDERPRICING 2. LLP 3. HIGHTECH 4. lnOFFERSIZE 5. STDRETURNS 6. FINCONDITION 7. UNDERWRITER 8. PCTINSIDER 9. MRKTRETURN 0.05 10. [DELTA]OFFERPRICE -0.03 0.08 * 11. EXCHANGE -0.19 *** 0.01 -0.03 12. LLP x OFFERSIZE 0.09 * 0.10 ** 0.11 ** 13. LLP x HIGHTECH 0.07 0.17 *** 0.09 * 14. LLP x STDRETURNS 0.03 0.14 *** 0.07 11 12 13 1. lnUNDERPRICING 2. LLP 3. HIGHTECH 4. lnOFFERSIZE 5. STDRETURNS 6. FINCONDITION 7. UNDERWRITER 8. PCTINSIDER 9. MRKTRETURN 10. [DELTA]OFFERPRICE 11. EXCHANGE 12. LLP x OFFERSIZE -0.17 ** 13. LLP x HIGHTECH -0.09 * 0.64 *** 14. LLP x STDRETURNS 0.18 *** 0.30 *** 0.53 *** This table reports Pearson correlations among independent variables, with *, **, and *** indicating that the estimated coefficients are significantly different from zero at the 10 percent, 5 percent, and 1 percent levels, respectively, in two-tailed tests. TABLE 3 IPO Underpricing Regressions ln(1 + UNDERPRICING) = [b.sub.0] + [b.sub.1] LLP + [b.sub.2] HIGHTECH + [b.sub.3] lnOFFERSIZE + [b.sub.4] STDRETURNS + [b.sub.5] (LLP x LITIGATION RISK PROXY) + [b.sub.6] FINCONDITION + [b.sub.7] UNDERWRITER + [b.sub.8] PCTINSIDER + [b.sub.9] MRKTRETURN + [b.sub.10] [DELTA]OFFERPRICE + [b.sub.11] EXCHANGE + u Model 1a HIGHTECH Predicted Interaction Variable Sign B (t-test) Intercept -0.052 (-1.289) LLP + 0.022 (1.263) HIGHTECH + 0.037 (1.921) ** lnOFFERSIZE + 0.018 (1.688) ** STDRETURNS + 0.128 (1.071) LLP x HIGHTECH (H1) + 0.072 (2.539) *** LLP x lnOFFERSIZE (H1) LLP x STDRETURNS (H1) Control Variables FINCONDITION +/- -0.003 (-2.170) ** UNDERWRITER - 0.001 (0.114) PCTINSIDER +/- 0.001 (1.322) MRKTRETURN + 0.645 (1.985) ** [DELTA]OFFERPRICE + 0.115 (1.991) ** EXCHANGE - -0.046(-2.055) ** Adj. [R.sup.2] = 0.151 F = 7.914 *** Model 1b Model 1c lnOFFERSIZE STDRETURNS Interaction Interaction Variable B (t-test) B (t-test) Intercept -0.040(-0.910) -0.040(-0.938) LLP 0.276 (1.521) * -0.034(-0.896) HIGHTECH 0.066 (4.304) *** 0.063 (4.104) *** lnOFFERSIZE 0.011 0.018 (1.699) ** STDRETURNS 0.125 -0.032(-0.227) LLP x HIGHTECH (H1) LLP x lnOFFERSIZE (H1) 0.044 (1.830) ** LLP x STDRETURNS (H1) 0.522 (2.357) *** Control Variables FINCONDITION -0.003(-1.906) * -0.003(-1.929) * UNDERWRITER 0.001 (0.222) 0.001 (0.088) PCTINSIDER 0.001 (1.158) 0.001 (1.529) MRKTRETURN 0.664 (2.036) ** 0.653 (2.007) ** [DELTA]OFFERPRICE 0.110 (1.907) ** 0.119 (2.056) ** EXCHANGE -0.052(-2.334) ** -0.043(-1.916) ** Adj. [R.sup.2] = 0.144 0.149 F = 7.579 *** 7.817 *** *, **, *** Indicate significant effects < 0.10, < 0.05, and < 0.01, respectively. Tests of directional expectations are one-tailed.
We thank the participants at the research workshop at the University of Wisconsin-Green Bay. We also acknowledge the helpful comments of Tim Pollock, John Eichenseher, and James Loebl. We appreciate the comments of Bill Messier (editor) and two anonymous reviewers.
(1) For example, the cover of the Journal of Accountancy had these headlines associated with articles covering liability: "Special Report: The Liability Crisis in the United States" (November 1992), "Attacking the Profession's Biggest Problem" (December 1992), and "Liability Protections, Now!" (April 1994).
(2) The difference in the amount of implicit insurance provided by the two organizational forms (general partnerships and LL firms) could be substantial. The collapse of the accounting firm of Laventhol & Horwath (L&H) provides an example of the amount of the difference. L&H filed for bankruptcy in 1990, citing litigation claims as a major factor in seeking bankruptcy protection (Stice 1991). The total amount assessed against the partners' personal assets averaged $76,000 per partner--an amount that is over 40 percent of their equity in the firm (Stone 1994). Under a LLP organizational form, the partners' personal assets, totaling $47.3 million, would not have been subject to attachment in an investor lawsuit. This example demonstrates that the difference in the liability form is economically significant for audit firms.
(3) Audit firms have other controls in place to ensure a standard of audit quality despite audit partners' potential incentives associated with a reduction in their liability (e.g., professional standards of conduct, concurring partner review, and regulatory sanctions).
(4) The protection provided to auditors by the LLP has yet to be determined in the courts. The Enron case might determine the effectiveness of the LLP in protecting non-negligent partners' personal assets (e.g., Neil 2002).
(5) While the laws governing LLPs are established by each individual state, most states have adopted laws that are very similar. For example, in an article contrasting Nebraska's law with the laws of other states, McGaughey (1996) describes the different state regulations as consistent regarding the protection afforded each partner against the wrongful acts of those not under his or her control.
(6) For a review of the literature related to IPO underpricing, see Ritter and Welch (2002).
(7) Essentially, an investor can only claim a loss equal to the offer price for a worthless stock. So by setting a lower offer price, the underwriter and issuer can reduce their liability. Of course, this approach is costly in that it directly reduces the proceeds of the offering received by the issuer and commission received by the underwriter.
(8) DeAngelo (1981) asserts that larger audit firms will supply higher levels of quality than smaller audit firms. She proposes that large audit firms have larger investment in reputational capital and greater incentive to provide high-quality audits. Larger audit firms are motivated to provide higher quality audits because they will lose more in the way of "quasi-rents" if their reputation is harmed.
(9) We use the term "post-LLP period" to describe the time after each audit firm changed to a LLP.
(10) We omit certain companies from our sample. We omit foreign-listed companies because of potential differences in litigation-related rules for these securities. We omit IPOs that include unit offerings because it is difficult to determine the value of the individual components included with packages of securities, such as stock bundled with warrants. We omit companies issuing limited partnership interests because the features of such securities differ from those of common stock. We omit non-Big 6 auditing firm audited companies to avoid issues involving market segmentation by auditor type. Finally, we omit financial services companies such as commercial banks, mutual funds, real estate investment trusts, and insurance companies because of their unique operational and risk features.
(11) There is no difference in the percentage of companies with missing data in the pre- and post-LLP periods (Chi-square = 0.054, p = 0.817).
(12) The audit firms of Coopers & Lybrand, Ernst & Young, and Price Waterhouse became LLPs on August 1, 1994. KPMG Peat Marwick and Deloitte & Touche became LLPs on August 15, 1994. Arthur Andersen became an LLP on September 1, 1994 (Chung et al. 1998). If we use an even event window of three months before and after the shift in LLP status, our subsequent results remain qualitatively the same, but due to the reduced sample size, statistical significance declines. Our results also are robust to an 18-month window. We believe the nine-month window reflects a reasonable trade-off between giving us enough power to detect an effect while minimizing the risk that other factors that arise over time may be driving our results.
(13) Prior literature in this area (e.g., Beatty 1989; Beaay and Ritter 1986; Willenborg 1999; Ritter and Welch 2002) uses the dependent variable underpricing as we have defined it. We use the log of underpricing in order to mitigate the econometric problems arising from the distribution of this variable described by Willenborg (1999). The offer price is the price that investors must pay for allocated shares in an IPO; it is not the opening price. Reported results do not differ if we market-adjust the underpricing variable (e.g., see Willenborg 1999).
(14) Companies with the following three-digit SIC codes are classified as HIGHTECH: 283, 357, 360, 361,362, 363, 364, 365, 366, 367, 368, 481, 737, 873.
(15) If we include all three interaction terms in our model, the variance inflation factor is over 10 (the suggested cut-off for problematic multicollinearity). Nonetheless, all three interactions are still significant (one tailed p-values less than 0.05.)
(16) Separate models are often estimated using sets of interactions for situations involving multiple indicators of a similar underlying construct in order to avoid collinearity. For examples of such models in the accounting and finance literatures, see Gul and Tsui (1998), Kang et al. (2000), Busaba et al. (2001), and Huson et al. (2001).
(17) The financial condition index is calculated using amounts from the balance sheet included in the registration statement for the most recent period prior to the IPO. Generally, this is the IPO company's last quarter-and preceding the filing of the registration statement. Net income is the annual net income for the period end corresponding to the balance sheet date.
(18) We band-collected audit opinions for all the IPOs in our sample. We were able to locate opinions in the prospectus for all but 62 observations. For those 62 observations, we went to Compustat and collected the first opinion code listed. In total, our sample includes 35 going-concern opinions out of 486 observations (7.2 percent), Note that this is a much smaller percentage of going-concern opinions than that found in Willenborg and McKeown (2000), whose sample includes about 25 percent of IPOs with going-concern opinions. (The difference in frequency between our study and theirs is that their sample includes only small development-stage IPOs, whereas we include non-development-stage IPOs.) Still, we included a dummy variable in our regressions, coded equal to 1 if the IPO had a going-concern opinion, and 0 otherwise. Including this dummy variable has no effect on our reported results and is itself not significant. Therefore, we do not include it in the reported model.
(19) Adding a dichotomous variable to capture whether the price revision is positive or negative (as in Lowry and Schwert 2002) does not lead to any significant changes in the reported results or inferences in hypothesis testing, so we omit it for simplicity.
(20) The models do not violate OLS regression assumptions. Variance inflation factors and condition indices indicate no collinearity problems (Belsley et al. 1980). For example, the highest variance inflation factor is 8.456 and the highest condition index is 21.844.
Antle, R. 1982. The auditor as an economic agent. Journal of Accounting Research 20 (Autumn): 503-527.
Baber, W., K. Kumar, and T. Verghese. 1995. Client security price reactions to the Laventhol and Horwath bankruptcy. Journal of Accounting Research 33 (Autumn): 385-395.
Balvers, R., B. McDonald, and R. Miller. 1988. Underpricing of new issues and the choice of auditor as a signal of investment banker reputation. The Accounting Review 63 (October): 605-622.
Beatty, R., and J. Ritter. 1986. Investment banking, reputation, and the underpricing of initial public offerings. Journal of Financial Economics 15 (January/February): 213-233.
--. 1989. Audit firm reputation and the pricing of initial public offerings. The Accounting Review 64 (October): 693-709.
--. 1993. The economic determinants of audit firm compensation in the initial public offerings market. Journal of Accounting Research 31 (Autumn): 294-303.
--, and I. Welch. 1996. Issuer expenses and legal liability in initial public offerings. Journal of Law and Economics 39 (October): 545-602.
Belsley, D., E. Kuh, and R. Welsch. 1980. Regression Diagnostics Identifying Influential Data and Sources of Collinearity. New York, NY: John Wiley & Sons.
Bohn, J., and S. Choi. 1996. Fraud in the new issue market: Empirical evidence on securities class actions. University of Pennsylvania Law Review 144 (January): 903-982.
Bunim, M. J., and A. B. Jacobson. 1994. Liability issues. Journal of Accountancy 178 (October): 119-124.
Busaba, W., L. Benveniste, and R. J. Guo. 2001. The option to withdraw IPOs during the premarket: Empirical analysis. Journal of Financial Economics 60 (April): 73-102.
Carcello, J., and Z-V. Palmrose. 1994. Audit firm litigation and modified reporting on bankrupt clients. Journal of Accounting Research 32 (Supplement): 1-30.
Carter, R., and S. Manaster. 1990. Initial public offerings and underwriter reputation. The Journal of Finance 45 (September): 1045-1067.
Chung, S., R. Narasimhan, and P. Wei. 1998. Market reaction to LLP formations: A test of the insurance hypothesis. Working paper, Rowan University.
DeAngelo, L. 1981. Auditor size and audit quality. Journal of Accounting and Economics 3 (December): 183-199.
Dye, R. 1993. Auditing standards, legal liability, and auditor wealth. The Journal of Political Economy 101 (October): 887-914.
--. 1995. Incorporation and the audit market. Journal of Accounting and Economics 19 (February): 75-114.
Francis, J., and K. Schipper. 1999. Have financial statements lost their relevance? Journal of Accounting Research 37 (Autumn): 319-352.
Grinblatt, M., and C. Hwang. 1989. Signalling and the pricing of new issues. The Journal of Finance 44 (June): 393-420.
Gul, F. A., and J. S. L. Tsui. 1998. A test of the free cash flow and debt monitoring hypotheses: Evidence from audit pricing. Journal of Accounting & Economics 24 (December): 219-237.
Hanley, K. W. 1993. The underpricing of initial public offerings and the partial adjustment phenomenon. Journal of Financial Economics (34): 231-250.
Hanson, R. K., and J. W. Rockness. 1994. Gaining a new balance in the courts. Journal of Accountancy 178 (August): 40-45.
Huson, M. R., R. Parrino, and L. T. Starks. 2001. Internal monitoring mechanisms and CEO turnover: A long-term perspective. The Journal of Finance 56 (December): 2265-2297.
Ibbotson, R. 1975. Price performance of common stock new issues. Journal of Financial Economics 2 (September): 235-272.
Jones, C., and S. Weingram. 1996. The determinants of 10b-5 litigation risk. Working paper, Stanford University.
Kang, J. K., A. Shivdasani, and T. Yamada. 2000. The effect of bank relations on investment decisions: A investigation of Japanese takeover bids. The Journal of Finance 55 (October): 2197-2218.
Loughran, T., and J. R. Ritter. 2002. Why don't issuers get upset about leaving money on the table in IPOs? Review of Financial Studies 15 (Special): 413-443.
Lowry, M., and G. W. Schwert. 2002. IPO market cycles: Bubbles or sequential learning? The Journal of Finance 57 (June): 1171-1200.
Lys, T., and R. Watts. 1994. Lawsuits against audit firms. Journal of Accounting Research 32 (Supplement): 65-93.
McGaughey, M. S. 1996. Limited liability partnerships: Need only professionals apply? Creighton Law Review 30 (December).
Menon, K., and D. Williams. 1994. The insurance hypothesis and market prices. The Accounting Review 69 (April): 327-342.
Michaely, R., and W. Shaw. 1995. Does choice of auditor convey quality in an initial public offerings? Financial Management 24 (Winter): 15.
Neil, M. 2002. Partners at risk: Euron could test the effectiveness of limited liability partnerships. ABA Journal 88 (August): 44-47, 57.
Nelson, J., J. Ronen, and L. White. 1988. Legal liabilities and the market for auditing services. Journal of Accounting, Auditing & Finance 3 (Summer): 255-285.
Ritter, J. 1991. The long-run performance of initial public offerings. The Journal of Finance 46 (March): 3-27.
--, and I. Welch. 2002. A review of IPO activity, pricing, and allocations. The Journal of Finance 57 (August): 1795-1828.
St. Pierre, K., and J. Anderson. 1984. An analysis of the factors associated with lawsuits against public accountants. The Accounting Review 59 (April): 242-263.
Schwartz, K., and K. Menon. 1985. Audit firm switches by falling firms. The Accounting Review 60 (April): 248-261.
Schwartz, R. 1997. Legal regimes, audit quality, and investment. The Accounting Review 72 (July): 385-406.
Simonetti, G., and A. Andrews. 1994. Limiting accountants' personal liability won't solve the country's liability crisis! Journal of Accountancy 177 (April): 46.
Simunic, D., and M. Stein. 1987. Product Differentiation in Auditing: Auditor Choice in the Market for Unseasoned New Issues. Vancouver, British Columbia: The Canadian Certified General Accountants' Research Foundation.
Stice, J. D. 1991. Using financial and market information to identify pre-engagement factors associated with lawsuits against audit firms. The Accounting Review 66 (July): 516-533.
Stone, M. 1994. Incorporated CPA firms. Journal of Accountancy 177 (March): 33.
Tinic, S. 1988. Anatomy of initial public offerings of common stock. The Journal of Finance 43 (September): 789-822.
Wallace, W. 1980. The Economic Role of the Audit in Free and Regulated Markets. New York, NY: Touche Ross Foundation. New York, NY.
Willenborg, M. 1999. Empirical analysis of the economic demand for auditing in the initial public offerings market. Journal of Accounting Research 37 (Spring): 225-238.
--, and J. McKeown. 2000. Going-concern initial public offerings. Journal of Accounting & Economics 30 (December): 279-313.
Zmijewski, M. 1984. Methodology issues related to the estimation of financial distress prediction models. Journal of Accounting Research 22 (Supplement): 59-82.
Steven R. Muzatko is an Assistant Professor at the University of Wisconsin-Green Bay. Karla M. Johnstone is an Associate Professor, Brian W. Mayhew is an Assistant Professor, and Larry E. Rittenberg is a Professor, all at the University of Wisconsin-Madison.
Submitted: May 2002
Accepted: August 2003
|Printer friendly Cite/link Email Feedback|
|Author:||Muzatko, Steven R.; Johnstone, Karla M.; Mayhew, Brian W.; Rittenberg, Larry E.|
|Publication:||Auditing: A Journal of Practice & Theory|
|Date:||Mar 1, 2004|
|Previous Article:||Professional service fees and auditor objectivity.|
|Next Article:||Audit committee characteristics and restatements.|