Auditor Independence as an Economic Decision

  • Benito Arruñada


This Chapter examines the economic factors which weigh on auditor independence. When issuing his report, an auditor’s decision is influenced by a wide range of considerations, amongst which his professional and personal ethics usually play a fundamental role. A positive analysis of these personal and professional ethical reasons is particularly difficult, however. Moreover, it is in any event desirable that there should be incentives which lead in the right direction even in the absence of such ethical principles. For these reasons, attention will focus here on analyzing the economic incentives weighing on this decision to ascertain the circumstances in which such incentives will ensure correct action.


Bargaining Power Economic Decision Audit Firm Audit Quality Specific Asset 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    A large part of the debate on auditing has centered around independence, perhaps because it is “difficult to prove and easy to challenge” (Mednick, 1990, p. 86). It is thus understandable that defining independence has also attracted considerable attention in both the academic literature and works of professional associations and legislators. See, for example, Mautz and Sharaf (1961), Nichols and Price (1976), Rittenberg (1977) and, for a more formal treatment, Antle (1984).Google Scholar
  2. 2.
    To the Federation of European Accountancy Experts (known as FEE, the Fédération des Experts Comptables Européens),independence includes “independence of mind”, understood as “the state of mind which has regard to all considerations relevant to the task in hand but no other” and “independence in appearance”, consisting of “the avoidance of facts and circumstances which are so significant that an informed third party would question the statutory auditor’s objectivity. When independence is dealt with in laws and professional rules, it is only independence in appearance which is addressed” (FEE, 1996, p. 24).Google Scholar
  3. 3.
    They may occasionally be so when losing a client results in excess capacity, as mentioned in Section 3.3.2. This problem will be all the more serious the more that possibilities for moving resources between audit activities and providing non-audit services are limited.Google Scholar
  4. 4.
    It should be recalled that this formulation is merely for explanatory purposes, and thus does not rule out the possible existence of non-linear effects despite using linear expressions; similarly, it is assumed that there are no interdependencies between the asset and liability blocks.Google Scholar
  5. 5.
    The studies by Palmrose (1988a, 1991) support consideration of this situation as reference since they show that low audit quality is normally associated with client insolvency.Google Scholar
  6. 6.
    The diagrammatic representation in Figures 3.1 and 3.2 should not be interpreted as indicating that the auditor change and client evolution nodes are simultaneous. On the contrary, it could in general be expected that client continuity will be decided in a shorter period than the latter’s financial evolution.Google Scholar
  7. 7.
    See, on this aspect, Arrubada and Paz Ares (1997, pp. 32–5).Google Scholar
  8. 8.
    As in Grout et al. when dealing with this same problem: “For simplicity, assume that the client firm and auditor have equal bargaining power and split the surplus” (1994, p. 329).Google Scholar
  9. 9.
    On the other hand, many studies did not find systematic discounts in initial audits. This result is of little interest, however, because they were unable to reject the hypothesis of no fee cutting.Google Scholar
  10. 10.
    See, for example, Roth (1995) for a description and analysis of the available evidence on bargaining experimentation.Google Scholar
  11. 11.
    Although in a different context, since a prior decision to increase capacity is not necessarily present, these assets would have a similar nature to those Williamson calls “dedicated assets” (1985, p. 96) or those which other authors have attributed with “temporal specificity” ( Masten, Meehan and Snyder, 1991 ).Google Scholar
  12. 12.
    See Buijink et al. (1996, pp. 69–70).Google Scholar
  13. 13.
    See DeAngelo (1981a, pp. 119–20), Beck et al. (1988a, pp. 52–4) and Grout et al. (1994, pp. 325–6).Google Scholar
  14. 14.
    Let us briefly examine the meaning and reality of these assumptions: (a) Assuming that the audit market is competitive is justified by the nature of the market in terms of its entry conditions and functioning, as well as the empirical evidence. (The pioneering work in this respect is that of Simunic [1984]). This evidence can be translated to other geographical spheres thanks to the uniformity of the service and the presence of operators of different sizes organized in networks of international scope. However, for a model which initially assumes imperfect competition, see Gigler and Penno (1995). (b) Auditing is also treated as a uniform product amongst potential providers. This is equivalent to keeping the analysis restricted to a set of potential providers of comparable quality. (c) Likewise, a particular intertemporal cost structure is assumed, characterized by the presence of start-up or learning costs. (d) It is also assumed that all costs, both start-up and recurrent, are the same for all provider firms, as well as the extemal effects or cost savings generated by the joint provision of services. (This cost equality eliminates any possibility of contractual friction and, in particular, guarantees that the present value of the future quasi-rents are fully offset by the discount in the initial period). Finally, (e), all income and expenses related to the services are assumed to arise at the beginning of each period, a simplification which does not mean a loss of generality, at least in the qualitative plane.Google Scholar
  15. 15.
    The effect of poor audit quality on relationships with other clients was formally studied by Balachandran and Ramakrishnan (1987, pp. 117–9) in an agency context, treating the reputational repercussions as contingent contracts, entered into by a firm consisting of several auditors. They show that under certain conditions the price necessary to motivate auditors is less than when the audit is contracted separately, without the audit quality affecting the compensation which the firm receives through other audits carried out on other clients.Google Scholar
  16. 16.
    See Menon and Williams ( 1994, p. 341 ). In this case the clients lost all the auditor-specific quasi-rents, which may not have occurred if the firm had continued to exist. The authors consider that this effect is minor, but they might underestimate it by concentrating only on the start-up costs.Google Scholar
  17. 17.
    Liability rules can also have a purely compensatory function, instead of generating incentives (see, for an introduction, Shapiro [1991] and other works in the monographic issue of Journal of Economic Perspectives of summer 1991). Auditors, however, are in a poor position to fulfil this function, for at least two reasons. Firstly, the shareholders of client companies can diversify risks in the capital market at less cost than auditors are able to with their client portfolio. Secondly, many countries require, and the traditional organizational pattern of this activity seems to make it advisable, that firms and their partners be subject to unlimited liability. Nevertheless, the American courts have based certain judgements in cases against audit firms on considering them be financial insurers, which makes little economic sense. The phenomenon has thus at times been explained as the consequence of a failure in the political process (for example, Lys and Watts, 1994, p. 66, n. 1). For an analysis of the role of the auditor as insurer, as opposed to that of providing contractual assurance or safeguard used here, see for example, Hill, Metzger and Schatzberg (1993). These authors also provide clear indications that prices tend to rise as the client risk rises, but cannot identify whether this is due to greater audit intensity or to the fact that the price incorporates the higher cost because of the increase in expected litigation.Google Scholar
  18. 18.
    See, for example, Balachandran and Ramakrishnan (1987, p. 118).Google Scholar
  19. 19.
    See, in particular, DeJong (1985), Palmrose (1988b), Melumad and Thoman (1990) and Balachandran and Nagarajan (1991). There are also some less solid indications to the contrary, however, (e.g. Shibano, 1994 ).Google Scholar
  20. 20.
    See Pratt and Stice (1994, pp. 641–2) for a summary and commentary on earlier works in this field.Google Scholar
  21. 21.
    In fact, liability has ceased to be seen as a complementary but has become a fundamental aspect of the audit service itself. For example, Dye (1993), when formally analyzing the consequences of different liability rules on auditor conduct, considers that the audit price depends on two factors: its informational value and the value of the option of those using financial statements in terms of the wealth of the auditor in the event that it is shown that the audit has been defective.Google Scholar
  22. 22.
    The meaning of joint and several liability can be clarified with a couple of examples, the first anecdotal but real. In 1987, Walt Disney World was sued by Ms. Wood. The plaintiff had suffered an accident on a Disney World bumper car track and sued Disney and her own boyfriend. The jury attributed 14% of the blame for the accident to her, 85% to the boyfriend and the remaining 1% to Disney. However, as the boyfriend had no resources, Disney had to pay 86% of the damages (Narayanan, 1994, p. 40). Translated into the auditing field, in a typical example taken from Hanson and Baker (1996) a client overstated his inventory and his auditor did not detect the problem. Based on the financial statements, a bank gave credit to the client and the latter became insolvent and could not repay it. The judge or the jury attributed 90% of the fault to the client and 10% to the auditor and evaluated the total loss at 400,000 dollars. Under joint and several liability, the bank could make up its total losses from the auditor. Under proportionate liability, however, the auditor would only be liable to pay 40,000 dollars.Google Scholar
  23. 23.
    Liability insurance premiums have reached 8% and 14.3% of turnover of the main audit firms in Great Britain (1994) and the United States (1992) despite which in some cases they have not been able to obtain it even at these prices. (Data taken on Great Britain from “British Accounting Liability: Big Six PLC” [The Economist,7 October 1995, pp. 109–12] and for the US from Mednick and Peck [1994, p. 891]). Between 1985 and 1992, in the US, the insurance premiums of the then Big Six firms multiplied by three and the deductibles by six (Narayanan, 1994, p. 40, n. 2).Google Scholar
  24. 24.
    During the 1990s different legal and legislative decisions have been reducing the liability of auditors in the United States. In this respect, see Hanson and Baker (1996).Google Scholar
  25. 25.
    See, for example, “La ley debe delimitar mejor la responsabilidad del auditor” (Expansión, 21–2 September 1996 ).Google Scholar

Copyright information

© Springer Science+Business Media New York 1999

Authors and Affiliations

  • Benito Arruñada
    • 1
  1. 1.Pompeu Fabra UniversityBarcelonaSpain

Personalised recommendations