Auditors’ Response to Organized Labor in Client Firms
Simon Yu Kit Fung
The Hong Kong Polytechnic University
Email:
Woo-Jong Lee
The Hong Kong Polytechnic University
Email:
Bin Srinidhi
University of Texas at Arlington
Email:
Lixin (Nancy) Su
The Hong Kong Polytechnic University
Email:
Preliminary version
November 28, 2014
______
Acknowledgements: We are grateful to Mark DeFond, Ferdinand Gul, Sophia Hamm, Chul W. Park, Haina Shi, Dan Simunic, Jing Zhou, and workshop participants at Shanghai University of Finance and Economics, Fudan University and University of Hong Kong for their helpful comments.Nancy Su acknowledges the financial support from The Hong Kong Polytechnic University (Grant #: G-UA1A).
Auditors’ Response to Organized Labor in Client Firms
Abstract
Using a sample of US firms for the period 2000-2011, we examine whether organized laborin audit client firms affects their financial statement quality and auditor decisions such as audit fees and going concern qualifications. We do not find evidence supporting the association between unionization and financial statement quality metrics such as future accounting restatements, discretionary accruals and the likelihood of reporting small profits. However, we find that labor unionization is associated with higher audit fees and higher likelihood of going-concern qualifications but shorter audit report lags. Because the auditor’s response seems to be unrelated to inherent earnings quality differences, these findings support the contention that the auditors mitigate the higher litigation risk in unionized firms by giving more going concern qualifications (thereby deflecting the legal threat away) and by charging higher audit fees to compensate for the additional risk.
Keywords: Labor union; audit fees; going-concern audit opinions; audit report lag
JEL Classification codes: M4; M42; J51
Auditors’ Response to Organized Labor in Client Firms
1.Introduction
In this paper, we examine the effect of organized labor in client firms on the earnings quality and auditor choices. We are motivated to study this issue because unionization could alter auditor’s incentives in different and often opposing ways and the resolution of how the auditors respond to them is an empirical issue. On the one hand, the labor union could act as a monitor of managers’ reporting and mitigate the opportunities and incentives for earnings management and misstatement while on the other hand, unionization could constrain value-adding activities by managers and increase the exposure of the client firm and its auditor to greater litigation risk. Few prior studies have examined this issue.
Several studies in micro-economics and financeliterature have examined the effect of unionization on a firm’s costs and value.Klasa et al. (2009) argue that in order to gain negotiating strength over the unions, unionized firms strategically hold less cash than they would otherwise have held. Chen et al. (2011) argue that labor unions increase firms’ costs ofequity by decreasing their operating flexibility. Faleye et al. (2006) show that although unions have a voice in corporate governance, they use it in a way that slows down growth, impedes investment, reduces risk-taking and together, adversely affect shareholder value. A common thread in these studies is that unionized firms compromise on shareholder value and economically optimal decisions in negotiating with the unions. In effect, these compromises in unionized firms are sub-optimal and are likely to attract more litigation from disaffected shareholders and other stakeholders. This increased business risk and the resultant increase in litigation against the client increases the exposure of the associated auditors to litigation.As a result, the auditors could potentially bear both the direct litigation costs such as attorney fees and indirect litigation costs such as lost auditor time, reputationdamage and the potential loss of future business. At the same time, unions are interested in ensuring that the earnings are correctly stated so that the labor gets a fair share. The managers’ incentive to overstate the earnings is mitigated by the need to share a greater part of the (pre-wage) earnings with labor. The unions are also wary of managers taking excessive risks that could result in losses and consequent retrenchment of labor. These restraints on managers fuelled by the union could blunt earnings management incentives and opportunities and thereby reduce misstatement risk. In effect, reduced misstatement risk could thereby help the auditors in planning less substantive tests in their audits.
Auditors choose their effort intensity and make pricing decisions, going concern qualification decisions and audit planning decisions based both on the Audit risk of there being a material misstatement in the client’s accounting process as well as the Business riskcomprised of litigation, regulatory sanctions and reputation loss resulting from being associated with the client(Bell et al. 2001). The firm’s financial statements consist of assertions that involve the inherent risk of misstatement which are mitigated by the internal control system (respectively known as inherent and control risks). Auditing Standard 8 requires the auditor to plan the audit to be able to detect material misstatements to reduce the audit risk to an acceptable level. Firms that take more operational risks, have high growth, have incentives for earnings management, and have weaker governance and internal control systems, are more likely to have errors and irregularities in their assertions and the auditor will have to expend greater effort and cost[1] in detecting and correcting the errors. Accounting research documents that in the presence of a strong labor union, managers have incentive to hide good news while revealing bad news to curtail the rent-seeking behavior of the union and the employees (DeAngelo and DeAngelo 1991; D’Souza et al. 2001; Bova 2013; Chung et al. 2014). Therefore, the pre-audit financial reportsare less likely to contain material misstatements in unionized firms compared to similar non-unionized firms.In turn, auditors could potentiallyreduce the audit effort and correspondingly,the audit fee.[2]
In contrast, the auditor faceshigher business risk by associating with the client, i.e., the auditor facesrisk or injury to the professional practice from litigation, adverse publicity, or other events arising in connection with a client irrespective of whether or not an audit failure is asserted(AS No. 8; AU Section 312; Arens et al. 2012, p. 420).[3]Unlike the audit risk,business risk cannot be eliminated (or reduced below a threshold level)even if the auditor devotes greater effort to the audit(Bell et al. 2001).Prior studies document that labor unions are rent-seekers, which may harm the firm’s overall performance and thereby the firm value (Hirsch, 1991; Faleye et al. 2006). Rent-seekingby the unions could result in strikes and lock-downs, cause negative publicity and increase the likelihood of shareholder litigation. Therefore, unionization could be a threat to the client’s (and its shareholders’) well-being and in extreme cases, could threaten itssurvival.Auditors of unionized clients in turn face higher potential litigation or reputation loss (Arens et al. 2012, p.237).[4] Since auditors cannot influence the business risk through their effort, they couldrespond through (i) charging an audit fee premium; and/or (ii) issuing the going-concern qualification. In this case, the audit fee premium will not reflect additional effort but instead, reflects the compensation for the additional risk. The auditor’s other choice is to issue going concern qualification. The going concern qualification will reduce the likelihood of litigation and protect the auditor from legal penalties even when sued.[5]This reasoning suggests that auditors of more unionized clients are more likely to issue going-concern opinions. However, it remains anopen empirical question as to whether the audit fees charged to such audit clientsare lower (consistent with the lower risk of financial misstatement) or higher (consistent with the higher business risks) than for less unionized clients.
Using a firm-level constructof labor union influence inaudit clients in the US for the period 2000-2011, we show that signed discretionary accruals, the likelihood of future restatement and the likelihood of reporting small profits do not vary with union strength. Absolute discretionary accruals are in fact significantly lower for more unionized clients. Consistent with the findings of prior studies, these results show that the curtailing effect of unionization on risk taking and aggressive reporting reduces the earnings misstatement risk and thereby improves earnings quality. Therefore, unionized clients do not demand higher effort by auditors to control the audit risk. On the other hand, we find a positive association between the strength of labor unions in client firms and the fee charged by the auditors.[6]Specifically, moving from a weak union group to a strong union group increases audit fees by 4.16% (Column (2), Table 4), which corresponds to an increase of $22,500 (=0.0416*exp(13.20)) based on our sample average.We document a negative relationship between the strength of labor union and client firm’s financial health (proxied by Altman Z-score or Ohlson O-score), which supports our argument that labor unions impose business risks on the client firms.Consistent with our argument on how auditors respond to the increased business risk, we show that the auditor’s propensity to issue going concern qualification increases with the strength of labor union in the client firm.Further, we use the reporting lag between the end of the fiscal year and the date of release of financial statements to measure audit effort, and find that reporting lagsare shorter for unionized clients. Given that the increased audit fee does not reflect additional effort, we conclude that the auditors of unionized clients charge higher audit fees as a compensation for the higherbusiness risk.
Our study contributes to the auditing literature by showing that unionization affects auditors’ choicessuch as issuing going-concern opinion and decisions such as charging a higheraudit fee. We do not find evidence of differences in the risk of financial misstatements between unionized clients and their less unionized counterparts. This finding shows that the additional misstatement risk that might arise from the below-par performance of unionized firms is compensated by the monitoring and incentive effects of unions on reporting. Our finding of higher audit fee in unionized firms in the absence of differences in earnings quality relative to non-unionized firms supports the interpretation that the audit fee differential is in fact the premium to compensate for the higher business risk in unionized firms. Auditing literature has documented the association between financial reporting quality and audit fees, but is relatively silent on the effect of client business risk (e.g. Hill et al. 1994; Bell et al. 2001; Lyon and Maher 2005). Our study provides the evidence on auditor response to business risk by examining the unique context of unionized firmsthat have high business risk accompanied with low misstatement risk. We complementthe literature that examines the effect of ownership characteristics on audit outcomes (e.g. Velury et al. 2003; Kane and Velury 2004; Mitra et al. 2007; Han et al. 2013)by showing that auditors respond differently to other stakeholdersassociated with the audit engagement such as organized labor.
Our study contributes to the literature on labor unionsby providing evidence that unions increase the business risk of a firm, as assessed by the external auditors (measured by the higher audit fees and the higher likelihood of receiving going-concern opinions). This is consistent with and complementary to the literature that documents lower firm values and higher cost of capital for unionized firms (Chen et al. 2011).Our results also provide inputs to social and policy decisions concerning the desirability of union strength in business operations.[7]
The remainder of this paper is organized as follows. Section 2 reviews relevant researchesand develops our hypotheses. Section 3 presents the research methodology and data description, which is followed by the discussion of empirical results in Section 4. Section 5 provides concluding remarks.
2. Literature Review and Hypotheses Development
2.1 Effect of Labor Unions on Managers
Unionization of labor impactsa firm’s operations andmanagerial decisions in several ways. First, the presence of a strong labor union constrains the managers’ actions and renders the firm’s operations more inflexible than in a less unionized firm. Unions impose higher employment termination costs making it less attractive for firms to hire employees to cater to short term needs. In anticipation that their pay structure cannot be easily adjusted, unionized firms areless flexible in their hiring and termination decisions than non-unionized firms.
Second, unions are akin to fixed claimants on the firm’s resources. Faleye et al. (2006) claim that similar to risky debt, unionized labor’s wage contract(current and retired labor’s stream of promised wages and benefits) is a fixed claim on the firm’s resources less a put option, the exercise price of which is the expected value of labor’s claims in bankruptcy. Therefore, the downside risk in payoff is more meaningful than its upside potential to the union workers. The employees perceive deteriorating firm performance as an increase in unemployment risk and the union helps them to effectively demand monetary and non-monetary compensation such as higher wages, additional benefits, and improved working conditions (Agrawal and Matsa, 2013; Chemmanur et al., 2013). Stronger labor unions can more effectively force their demand on managers (Chemmanur et al., 2013). In effect, firms with higher unemployment risk pay higher wage to their employees and this relation is more pronounced for firms with stronger unions. Anecdotal evidence also indicates that unions could be vocal in their support for managerial termination in bad times when they perceive higher unemployment risk.[8]Prior research supports the contention that unions’ assessments of managers’ investment decisions are similar to and consistent with thecreditors’ perspective, suggesting that unions prefer lowerinvestment risk than shareholders or managers (Faleye et al. 2006; Chen et al. 2012). For example, Chen et al. (2012) show that firms in more unionized industries undertake less risky investments. Similarly, Connolly et al. (1986) find that unionization produces a limiting influence on research and development investment. Faleye et al. (2006) study a sample of firms with union block (equity) ownership and find that, despite unions’ significant equity stakes that could potentially increase in value with firm risk, these firms avoid risks and exhibit lower total factor productivity. This finding suggests that unions seek short-term stability at the expense of long-term firm value.
In order to seek their fixed claims and other demands on the firms, labor unions could credibly threaten work stoppages or strikes. These threats derive their credibility from the costs they can impose on the firm with work stoppages, strikes, deliberately inefficient production, loss ofreputation in the labor market and negative publicity among investors and other capital market participants. Their equity block ownership further enables them to pursue their interest throughthe firm’s corporate governance structure. Prior studies show that unions with block equity ownershiphave succeeded in lowering executive compensation (Banning and Chiles, 2007; Gomez and Tzioumis, 2006) and limiting the risk-taking behaviour by the managers (Faleye et al. 2006).
In response to the threats of strikes and other methods employed by the unions to extract quasi-rents from firms (Baldwin 1983; Grout 1984), managers have incentives to shelter firm resources from strong labor unions. Consistent with this arguments, prior studies have shown that managersin firms with stronger labor unions hold smaller cash reserves (Klasa et al., 2009), issue more debt (Bronars and Deere, 1991;Matsa, 2010), disclosegoodnews lessfrequently (Chung et al., 2014) andwithhold news to bolster their bargaining power against organized labor (Hilary, 2006). These managerial actions are likely to have an impact on a firm's financial reporting quality (e.g. Leung et al. 2012; Mora and Sabater, 2008; Farber et al. 2012; Bova 2013) and accounting policy choices (Cullinan and Knoblett, 1994; D’Souza et al. 2001).
Prior studies provide mixed evidence on income-decreasing earnings management in unionized firms. Liberty and Zimmerman (1986) find no evidence of income-decreasing earnings management prior to the labor renegotiation periods in their examination of 105 unionized companies from 1968 to 1981. In contrast, DeAngelo and DeAngelo (1991) document income-decreasing earnings management of seven U.S. steel companies during the 1980s.
Findings of studies examining the effect of unionization on accounting policy choice are also inconclusive. Cullinan and Knoblett (1994) do not find an association between the presence of an organized workforce and the inventory and depreciation method choice. On the other hand, D’Souza et al. (2001) show that firms with a unionized workforce tend to use immediate recognition when adopting Financial Accounting Standards 106 (Accounting for Postretirement Benefits).