Competitive Rivalry in Audit Markets
Simon Dekeyser a*
Ann Gaeremynck a
W. Robert Knechel a,b
Marleen Willekens a,c
a KU Leuven
Faculty of Economics and Business
Leuven, BELGIUM
b University of Florida
|Fisher School of Accounting,
FL, United States
c BI Norwegian Business School,
Oslo, Norway
October 2016
Competitive Rivalry in Audit Markets
Abstract
In this paper we argue that audit firms compete rationally and consider the potential actions of other firms when deciding how fiercely to compete with market rivals. Based on prior literature in the field of industrial organization, we hypothesize that competing with the same audit firms across different industries within a geographical region (which we label “multi-industry contact”) leads to less competition overall, which suggests mutual forbearance among rivals. However, client concentration within an industry increases the immediate benefits of vigorous competition inducing audit firms to compete more aggressively. Further, a drop in quality for an audit firm can adversely affect the firm’s reputation, making the firm more vulnerable to aggressive competition from other audit firms. We measure rivalry using two dynamic measures of competition (i.e., market-share mobility and leader dethronement) and find that multi-market contact, market concentration and reputation damage all affect competitive rivalry as predicted.
Keywords: competition, leader reputation, market instability, multimarket contact
JEL- classification : M42
* The authors are indebted to Liesbeth Bruynseels, Joseph Gerakos and John-Christian Langli for useful comments as well as to participants at the 2016 EIASM Audit Quality workshop in Florence (Italy), the 2015 Auditing Section Midyear Meeting, the 2015 EARNET conference and workshops at BI Norwegian Business School, University of Exeter, Tulane University and KU Leuven. Simon Dekeyser gratefully acknowledges financial support from the Research Foundation – Flanders (FWO).
Competitive Rivalry in Audit Markets
Introduction
The level of competition in audit markets has been a major concern of regulators over the past decade (European Commission, 2011; U.S. Government Accountability Office [GAO], 2003, 2008). The US Government Accountability Office (GAO) clearly articulated these concerns in 2008: “Dominant sellers, in this case accounting firms, may be more likely or more able to engage in coordinated interaction in ways that can affect auditing practices or prices” (GAO, 2008). Such “coordinated interaction” can be explicit (collusion) or implicit (strategic or mutual forbearance). In this paper, we examine factors that affect how aggressively audit firms compete with each other based on an analysis of the US market by industry and location (Metropolitan Statistical Area, or MSA). More specifically, we assume that audit firms consider the potential actions of other firms when deciding how to compete in a specific market (industry, MSA). We do not assume or require active collusion among audit firms, although such collusion is not ruled out by our analysis. We then analyze how this strategic forbearance influences the behavior of participants across audit market segments.
The audit market can be viewed as oligopolistic because a small number of individual audit firms (e.g., the Big Four) are large enough to alter market conditions through their own actions. Consequently, their decisions concerning how fiercely to compete in a market are dependent on the potential and expected reactions of other large firms in the same market (Melvin and Boyes, 2002). A competitive action by one firm can significantly alter market conditions, leading rivals to alter their own competitive strategy which, in turn, will further impact market conditions. We argue that audit firms compete rationally and will consider both the immediate benefits and future costs when deciding how vigorously to compete in a specific market. Potential benefits include an increase in the number of clients and revenues obtained by taking clients away from competing audit firms. Future costs arise because rivals may retaliate through increased price competition (even on retained clients) and targeting of the aggressor’s own clients, possibly resulting in a loss of industry market share and profits. In general, audit firms will compete more fiercely the higher the benefits and the lower the costs of their competitive actions (Motta, 2004). As a result, audit firms can choose either to compete aggressively, and risk retaliation, or act passively to decrease the effect of potential competition either explicitly or implicitly.
We investigate three factors that can impact the cost and benefits of competition: multimarket contact between audit suppliers, buyer concentration, and reputation damage incurred due to a drop in audit quality. First, empirical evidence shows that multimarket contact leads to higher prices, profits and lower sales growth rates in markets for aviation, banking, and mobile phone (Barros, 1999; Evans and Kessides, 1994; Gimeno, 2002; Greve, 2008; Parker and Röller, 1997). Audit rivals are likely to compete in multiple industries within an MSA (i.e., firms have direct contact across multiple industries in a given location). In such situations, they may choose not to compete heavily in each other’s focal industries because that could result in retaliatory and vigorous competition in all industries. The net gain from competing aggressively in one industry may therefore be reduced by losses across other industries in which audit firms compete. This reduces the incentives for audit firms to compete fiercely in all industries in which other firms also compete (Bernheim and Whinston, 1990). Consistent with these arguments, we hypothesize that audit firms that compete in multiple industries within a locale (MSA) have lower incentives to compete aggressively against other firms in the same markets (Hypothesis 1).
Second, large clients can exert their bargaining power by negotiating lower fees from their current auditor (Casterella et al., 2004; Huang et al., 2007; Mayhew and Wilkins, 2003) or increase competition among auditors by threatening to switch suppliers (Motta, 2004). The benefit of attracting a large client is substantial and could exceed future losses caused by any retaliatory reactions by rivals.[1] Consistent with this argument, the industrial organizational literature finds that markets in which buyer concentration is high exhibit greater variation in suppliers’ market shares over time (Caves and Porter, 1978; Kato and Honjo, 2006). We therefore predict that audit client concentration is positively associated with the aggressiveness of competition among audit firms in a market segment (Hypothesis 2).
Third, we test the effect of damage to an audit firm’s reputation as measured by accounting restatements experienced by the clients of a firm. Our perspective is based on evidence from prior studies show that firms with a large market share provide higher audit quality and/or have a reputation for high quality (Craswell et al., 1995; Ferguson et al., 2003; Francis et al., 2005; Reichelt and Wang, 2010). As a result, they may become vulnerable if the market believes that their audit quality has declined. We presume that restatements of financial statements by a firm’s clients negatively affect the incumbent audit firm’s reputation since restatements result in negative capital market consequences for the client (Palmrose et al., 2004) and have adverse implications for the auditor-client relationship (Huang and Scholz, 2012). Damage to an audit firm’s reputation is likely to increase the willingness of its clients to switch audit firms, make it easier for competitive rivals to attract the firm’s clients, and make it harder for the incumbent to retaliate due to this loss of reputation. We therefore predict that restatements by clients will be associated with an increase in competitive aggressiveness in the industry market segment where the restatement occurred (Hypothesis 3).
The academic literature typically measures audit market competition in a single-period, static setting by focusing on supplier concentration (Bandyopadhay and Kao, 2004; Feldman, 2006; Pearson and Trompeter, 1994), industry specialization (Craswell et al., 1995; Ferguson et al., 2003; Francis et al., 2005), or market-share distance from the closest competitor (Numan and Willekens, 2012). These static measures, however, conceal much of the dynamic competitive processes in markets (Davies and Geroski, 1997). Substantial variation over time in leading firms’ market shares may exist, even in markets where competition is labeled as low using static competition measures (Bujink et al., 1998; GOA, 2008; Scherer and Ross, 1990). Prior industrial organizational literature argues that market instability is a sign of high competition and inter-firm rivalry (Kato and Honjo, 2006; Schmalensee, 1989; Staigler and Wolak, 1992). We therefore use measures of market-share instability as our proxies for competition in an audit market (Caves and Porter, 1978; Ferrier et al., 1999; Schmalensee, 1989).
Our analysis uses a U.S. sample of 3,279 market-segment-years at the MSA level over the period 2003–2012.[2] In previous literature, dynamic market competition and market instability is proxied by changes in market share and relative rankings of incumbents and entrants over time (Caves and Porter, 1978; Ferrier et al., 1999; Schmalensee, 1989). In line with this literature, we capture competitive rivalry using two measures: (1) market-share mobility and (2) leadership dethronement.[3] Market-share mobility is the sum of the year-on-year market share changes of all competitors within a market segment (Caves and Porter, 1978; Kato and Honjo, 2006; Sakakibara and Porter 2001). Leader dethronement is the year-on-year change in the identity of the market leader in a market segment. We include this variable since the leadership position is particularly valuable as leaders in many industries have strong reputations, can exploit economies of scale, and can charge higher prices (Armstrong and Collopy, 1996; Ferrier et al., 1999), which makes the leadership position highly contested in competitive markets.
In general, our results support our hypotheses. When audit firms compete in multiple industries within an MSA, competition is less fierce, as evidenced by a negative association between our measure for multi-industry contact and both market-share mobility and leadership dethronement. This result suggests that audit firms follow a strategy of “mutual forbearance” when they are in potential competition in many market segments (industries), that is, they refrain from competing aggressively for each other’s existing clients. Our results also show that client concentration is associated with more aggressive competition in an audit market segment (i.e., the market is made up of larger clients). Finally, we also find a significant change in overall market-share mobility in industry market segments where reputation damage occurred due to client accounting restatements. Moreover, we also find that an accounting restatement by a client of the industry leader increases the likelihood that the leader will lose its leadership position. This indicates that the leader’s reputation may be seriously damaged by the restatement, opening the door for more aggressive competition from rivals. As changes in auditor market shares can be caused by client switches or changes in fees, we perform supplemental analyses to investigate if these are responsible for the observed market instability. In general, the supplemental analysis reveals that multi-industry contact between audit suppliers is negatively associated with the amount of client switching, but is not associated with fee changes of non-switching clients. These results are also consistent with audit firms engaging in mutual forbearance.
Our paper offers a number of contributions to the literature on audit competition. First, we empirically study factors that affect the aggressiveness of the competitive rivalry among audit firms, presenting evidence suggesting that both the costs and benefits can influence how vigorously an audit firm may compete in a market segment. Second, unlike previous studies that have used static measures of audit competition (Bandyopadhay and Kao, 2004; Feldman, 2006; Pearson and Trompeter, 1994), we introduce and test two dynamic measures. Because static measures conceal much of the dynamic competitive processes, especially in highly concentrated markets (Bandyopadhay and Kao, 2004; Davies and Geroski, 1997), studying dynamic competition measures adds to the literature because they are a good indicator of rivalry in concentrated markets (Caves and Porter, 1978). Third, we also link competition to a proxy for reputation damage, i.e., we demonstrate that market instability is larger in industry market segments in which clients have accounting restatements, suggesting that such restatements result in damage to a firm’s reputation which makes them more vulnerable to aggressive competition.. Finally, we contribute to the regulatory debate about whether there is sufficient competition in the audit market (European Commission, 2011; U.S. Government Accountability Office [GOA], 2003, 2008). Since our evidence suggests that the fierceness of competition depends on audit market characteristics, we illustrate that one-size-fits-all regulation to encourage audit market competition may not be optimal.
The remainder of the paper is organized as follows. In section 2, we develop our hypotheses. Section 3 presents the research design, while section 4 describes the sample selection procedure. Section 5 presents the results and section 6 concludes.
1. Hypotheses
We characterize the audit market as a quality-differentiated oligopoly, dominated by a few suppliers (Numan and Willekens, 2012). A key feature of oligopoly is that each supplier’s competitive moves affect market conditions, including the market clearing price, and that suppliers’ actions are interdependent. Competitors will respond to the actions of one firm by adjusting their own competitive strategies. The adjustments they make in their strategies will, in turn, alter market conditions again. Competitive rivals will take into account the direct effect of their market decisions as well as secondary effects that follow from the reactions of other firms. Rational firms competing in such markets will weigh the immediate benefits of competing aggressively to obtain new clients and increased revenues against potential future costs arising from the reactions of rivals. In the extreme, the market dynamics may result in a “price war” among rivals. This paper focuses on how the benefits and costs of aggressive competition can influence a firm’s actions across the various markets in which it competes with its rivals (i.e., compete for clients in different industries).
1.1. Multi-industry contact
Prior research divides the audit market into segments based on industries within MSAs (Francis et al., 2005; Numan and Willekens, 2012; Reichelt and Wang, 2010). Thus, the same competing audit firms/offices try to attract clients in multiple industries/market segments. As a result, audit firms may find it more profitable to focus on some key industries rather than competing aggressively in all industries, especially in a single geographical area. From an economic perspective, the gain from an aggressive approach in one industry segment may be outweighed by rivals’ reactions in other industry segments. Competing firms might therefore practice mutual forbearance, refraining from competing aggressively in their rivals’ focal industries to avoid aggressive competition in their own focal industries. Edwards (1955) first argued that multimarket links could affect competition: “Firms that compete against each other in many markets may hesitate to fight vigorously because the prospects of local gain are not worth the risk of general warfare.” A formal analysis by Bernheim and Whinston (1990) shows that competing in multiple market segments decreases suppliers’ incentives to compete vigorously. Empirical studies consistently show that multimarket contact negatively affects competition because of mutual forbearance. As a result, multimarket contact has been shown to result in higher prices, greater profits, higher survival rates, and decreases in the rate of sales growth (Barros, 1999; Evans and Kessides, 1994; Gimeno, 2002; Greve, 2008; Li and Greenwood, 2004) in industries such as air travel (Evans and Kessides, 1994), banking (De Bonis and Ferrando, 2000) and insurance (Greve, 2008). Note further that it has also been shown that the retaliatory response of firms is reinforced when differences in market shares across market segments exist (Bernheim and Whilston, 1990).