1 For reviews, compilations, and references, see Karyl A. Kinsey, “Theories and Models of Tax Cheating,” 18 Crim. Just. Abstracts 403 (1986); Jeffrey A. Roth, John T. Scholz, & Ann Dryden Witte, eds., Tax Compliance, vol. 1: An Agenda for Research (Philadelphia: University of Pennsylvania Press, 1989; Jeffrey A. Roth & John T. Scholz, eds., Tax Compliance, vol. 2: Sod Science Perspectives (Philadelphia: University of Pennsylvania Press, 1989); Long, Susan B. & Swingen, Judyth A., “Taxpayer Compliance: Setting New Agendas for Research,” 25 Law B Soc'y Rev. 637 (1991); Joel Slemrod, ed., Why People Pay Taxes: Tax Compliance and Enforcement (Ann Arbor: University of Michigan Press, 1992) (“Slemrod, Why People Pay Taxes”).CrossRefGoogle Scholar

2 Long & Swingen, 25 Law B Soc'y Rev. at 669.Google Scholar

3 For some exceptions see Kent W. Smith, “Will the Real Noncompliance Please Stand Up? Complexity and the Measurement of Noncompliance” (presented at IRS Research Conference, Washington, D.C., 17 Nov. 1988) (“Smith, ‘Will the Real Noncompliance Please Stand Up?’”) Long & Swingen, 25 Law B Soc'y Rev. Google Scholar

4 Jonathan S. Feinstein, “Reassessing the Tax Gap: New Estimates of Filer Income Tax Evasion and Its Detection,” 45 Tax Notes 1003 (1989).Google Scholar

5 Henk Elfers, Henry S. J. Robben, & Dick J. Hessing, “Reliability and Validity of Tax Audits: Can We Have Confidence in the Opinion of the Dutch IRS on Noncompliance?” Working Paper (Erasmus, The Netherlands: Erasmus Centre for Sociolegal Tax Research, 1989); Henk Elfers, Henry S. J. Robben, & Dick J. Hessing, “Under-reporting Income: Who Is the Best Judge—Tax-payer or Tax-Inspector?” 1541. Royal Statistical Soc'y, ser. A, 125 (1991); Henk Elffers, Income Tox Evasion: Theory and Measurement (Deventer, The Netherlands: Kluwer, 1991). We should note that tax audits in the Netherlands are based on the returns and third-party information only. Unlike Oregon state and 1RS audits, they do not involve a review of the taxpayers' own books and records or discussions with the taxpayers or their representatives.CrossRefGoogle Scholar

6 Long & Swingen, 25 Law B Soc'y Rev. at 659 n.38.Google Scholar

7 Internal Revenue Service (IRS), Internal Revenue Service Annual Report 1989, Pub. 55, table 34 (Washington, D.C.: Internal Revenue Service, 1989).Google Scholar

8 Kent W. Smith & Loretta J. Stalans, “Encouraging Tax Compliance with Positive Incentives: A Conceptual Framework and Research Directions,” 13 Law B Pol'y 35 (1991); Stalans, Loretta J., “Citizens' Procedural Expectations for an Upcoming Tax Audit: Their Na. ture and Formation,” 5 Soc. Just. Res. 93 (1992); Loretta J. Stalans, “Lay Evaluations of Encounters with Government Officials: Do Expectations Serve as Filters and Standards?”in Linda Heath et al., eds., 3 Applications of Heusistics and Biases to Social Issues (New York: Plenum Press, forthcoming) (“Stalans, ‘Lay Evaluations’”); Loretta J. Stalans & Kent W. Smith, “Procedural Criteria in Taxpayers’ Evaluations of Their Audit Process: Differences across Persons and Situations,” ABF Working Paper #9205 (Chicago: American Bar Foundation, 1992) (“Stalans & Smith, ‘Procedural Criteria’”); Loretta J. Stalans & Kent W. Smith, “How Taxpayers’ Anticipations Influence Their Audit Experience: Cognitive and Interpersonal Processes” (presented at Law & Society Association annual meeting, Chicago, 27 May 1993) (“Stalans & Smith, ‘How Taxpayers’ Anticipations Influence Audit Experience”); Kent W. Smith, “The Cultural Embeddedness of Tax Issues: The Nature of Disputes in Audits” (presented at Law & Society Association annual meeting, Chicago, 27 May 1993) (“Smith, ‘Cultural Embeddedness’”).CrossRefGoogle Scholar

9 The scale is a mean of the auditors' ratings of the complexity, unusualness, and ambiguity of the tax issues in the returns (Cronbach's α=.80).Google Scholar

10 We are using “law” to refer to the cumulative body of related statutes, regulations, administrative rulings, and court cases.Google Scholar

11 E.g., Mather, Lynn & Yngvesson, Barbara, “Language, Audience, and the Transformation of Disputes,” 15 Law B Soc'y Reu. 775 (1981); Merry, Sally Engle & Silbey, Susan S., “What Do Plaintiffs Want? Reexamining the Concept of Dispute,” 9 Jusr. Sys. J. 151 (1984); Miller, Richard E. & Sarat, Austin, “Grievances, Claims, and Disputes: Assessing the Adversary Culture,” 15 Law B Soc'y Reu. 525 (1981); Neil Vidmar, “The Origins and Consequences of Procedural Fairness,” 15 Law B Soc. Inquiry 877 (1990); Vidmar, Neil & Schul-ler, Regina A., “Individual Differences and the Pursuit of Legal Rights: A Preliminary Inquiry,” 11 Law & Hum. Behau. 299 (1987).CrossRefGoogle Scholar

12 E.g., Eugene Bardach & Robert A. Kagan, Going by the Book: The Problemems of Regulatory Unrensonabkness (Philadelphia: Temple University Press, 1982); John Braithwaite, To Punish or Persuade: Enforcement of Coal Mine Safety (Albany: State University of New York Press, 1985); John Braithwaite, John Walker, & Peter Grabosky, “An Enforcement Taxonomy of Regulatory Agencies,” 9 Law B Pol'y 323 (1987); Keith Hawkins, Environment and Enforcement: Regulation and the Social Definition of Pollution (Oxford: Clarendon Press, 1984) (“Hawkins, Environment and Enforcement”); Bridget Hutter, “Variations in Regulatory Enforcement Styles,” 11 Law B Pol'y 153 (1989); Scholz, John T., “Cooperation, Deterrence, and the Ecology of Regulatory Enforcement,” 18 Law d Soc'y Rev. 179 (1984); Scholz, John T. & Wei, Feng Heng, “Regulatory Enforcement in a Federalist System,” 80 Am. Pol. Sci. Reu. 1249 (1986); Neal Shover et al., “Regional Variation in Regulatory Law Enforcement: The Surface Mining Control and Reclamation Act of 1977,” in Keith Hawkins & John M. Thomas, eds., Enforcing Regulation (Boston: Kluwer-Nijhoff Publishing, 1984); Robert A. Kagan, “On Regulatory Inspectorates and Police,” in id. at 37-64 (“Kagan, ‘On Regulatory Inspectorates’”)'CrossRefGoogle Scholar

13 We are following the convention of using “citizen” to refer to the disputants who are not government officials conducting the inquiry or enforcing the law, even though the parties are often not legal citizens of the political jurisdiction within which the dispute occurs. Alternatives such as “civilian” seem to have even more unintended and distracting connotations. In most instances, the term refers to individuals acting on their own behalf or as representatives and decision makers for a collective entity, but we mean the term to be construed broadly to encompass those entities under inspection or enforcement by an official. Like “taxpayer,” it can refer to individuals, partnerships, corporations, trusts, and even other government agencies subject to review by the official.Google Scholar

14 E.g., William L. F. Felstiner, Richard L. Abel, 8 Austin Sarat, “The Emergence and Transformation of Disputes: Naming, Blaming, Claiming.” 15 Law B Soc'y Rev. 631 (1981); Merry & Silbey, 9 Just. Sys. J.; Miller & Sarat, 15 Law B Soc'y REV.; Vidmar & Schuller, 11 Law B Hum. Behau. CrossRefGoogle Scholar

15 Felstiner et al., 15 Law B Soc'y Reu. Google Scholar

16 Smith, “Will the Real Noncompliance Please Stand Up?” (cited in note 3).Google Scholar

17 Felstiner et d., 15 Law & Soc'y Reu. Google Scholar

18Occasionally taxpayers who do not reject the claim during the audit begin the dispute after the completion of the audit by filing an appeal.Google Scholar

19 felstiner et al., 15 Law d Soc'y Rev. Google Scholar

20 Thibaut, John & Walker, Laurens, “A Theory of Procedure,” 66 Cal. L. Rev. 541 (1978).CrossRefGoogle Scholar

21 The fact that government officials do the initial naming and claiming in tax and most regulatory disputes has a measurement implication. Some of the research on disputes between government officials and citizens has used grievances and complaints by citizens as the sampling unit; e.g., Daniel Katz et d., Bureaumatic Encounters: A Pilot Study in the Evaluation of Gommt Services(Ann Arbor, Mich.: Institute for Social Research, 1975); Vidmar & Schuller, 11 Luw B Hum. Behau.(cited in note 11). Consequently, they have tended to focus on only the PEAL-based disputes that, within the enforcement arena, were either unresolved or unsatisfactorily resolved from the perspective of the citizen and that became overlaid by the citizen with PIES. Research on regulation, particularly that looking at the behavior of front-line inspectors, provides some insights into the negotiation of regulatory disputes; see, e.g., Bardach & Kagan, Going by the Book; Braithwaite, To Punish or Persuade; Gilboy, Janet A., “Deciding Who Gets in: Decisionmaking by Immigration Inspectors,” 25 Law B Soc'y Reu. 571 (1991); Hawkins, Environment and Enforcement; Hutter, 11 Law & Pol'y. However, much of this research takes infractions of the law as givens and focuses on alternative enforcement styles, their causes, and their consequences for future compliance. Because of this emphasis on the regulatory process after infractions have been named and claimed, the research tends to look only tangentially at our primary focus.CrossRefGoogle Scholar

22 Mather & Yngvesson, 15 Low & Soc'y Reu. (cited at note 11).Google Scholar

23 See, e.g., E. Allan Lind & Tom R. Tyler, The Social Psychology of Procedural Justice, especially ch. 10 (New York: Plenum, 1988) (“Lind & Tyler, Social Psychology of Procedural Justice”); Vidmar, 15 Law B Soc. Inquiry (cited in note 11).Google Scholar

24 Some auditors' styles of investigating and making claims about errors, however, may give citizens the impression that the auditors distrust citizens and assume from the beginning that errors have been made. Stalans, 5 SOC. Just. Res.; Stalans & Smith, “Procedural Criteria” (both cited in note 8).Google Scholar

25 In other arenas, another type of dispute often ensues after a dispute about either a PEAL or a PIE has been settled with at least a partial victory for the party making the claim: a dispute about what the sanctions (penalties, compensation, corrective actions, and so forth) should be for having misapplied the law or contributed to the injury. This type of dispute is analytically distinct from those about both PEALS and PIES in that it does not concern claims about what was done in the past, although such claims may enter into the negotiation. These disputes often take on the character of a bargaining situation, and in the regulatory literature they have been modeled as ongoing games between citizens and officials (see especially Ian Ayres & John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (New York: Oxford University Press, 1992); Scholz, 18 Law & B Soc'y Rev. (cited in note 12)). These disputes that may follow disputes about PEAL and PIEs are beyond the purview of this article, but it is relevant in this context to note that disputes about sanctions in enforcement arenas are probably constrained by the focus on legal correctness to a greater degree than are such disputes between citizens.Google Scholar

26 Some taxpayers who engage professional representatives, particularly after the audit has begun, may view them as serving to some extent as mediators who can view the issues more objectively than they can and who can help the two sides reach an agreement. The possibility of third-party intervention may structure the bargaining between large corporations and tax officials more than it does the bargaining between tax officials and individual or small-business taxpayers. Appeals are much more common in large cases, and both sides are probably more likely to be concentrating on building a case for appeal and possible litigation.Google Scholar

27 Merry & Silbey, 9 Just. Sys. 1. (cited in note 11); Lind & Tyler, Social Psychology of Procedural Justice. Google Scholar

28 William, L. F. Felstiner, Avoidance as Dispute Processing: An Elaboration,” 9 Law & Soc'y Rev. 695 (1974); and Merry & Silbey, 9 Just. Sys.J., and Vidmar & Schuller, 11 Law & Hum. Behav.(both cited in note 11).Google Scholar

29 Felstiner, 9 Law & Soc'y Rev. at 695.Google Scholar

30 For a discussion of a similar range of goals among frontline regulatory inspectors, see especially Bardach & Kagan, Going by the Book (cited In note 12).Google Scholar

31 Thibaut & Walker, 66 Cal L. Rev. at 541 (cited in note 20).CrossRefGoogle Scholar

32 Id. at 545.Google Scholar

33 See especially Erika Apfelbaum, “On Conflicts and Bargaining,” 7 Advances Experimental Soc. Psychology 103 (1974); Robert M. Axelrod, The Evolution of Cooperation (New York: Basic Books, 1984); Morton Deutsch, The Resolution of Conflict: Consmcctive and Destructive Processes (New Haven, Conn.: Yale University Press, 1973); Harold H. Kelley & Anthony J. Stahelski, “Social Interaction Basis of Cooperators' and Competitors' Beliefs about Others,” 16 J. Personality B Social Psychology 66 (1970); Dean G. Pruitt & Jeffrey 2. Rubin, Social Conflict: Escalation, Stalemate and Settlement (New York: Random House, 1986).Google Scholar

34 Much of the dispute processing and procedural justice literature distinguishes between the process stage and decision stage of dealing with a dispute. See, e.g., Lind & Tyler, Social Psychology of Procedural Justice (cited in note 23); Thibaut & Walker, 66 Cal. L. Rev. (cited in note 20); Vidmar, 15 Law & Soc. Inquiry (cited in note 11). In this framework, our discussion of negotiating strategies focuses on the decision stage. Strategies for negotiating disputes are conceptually and empirically distinguishable from the procedural styles officials employ during the process stage to establish the interpersonal tone, elicit cooperation, and manage the pace of the audit. In our study, there is a moderate relationship between the auditors' reported procedural styles and their negotiating preferences (Kendall's τ=.18, p=.09). Those who report they are flexible in running the audit tend to be cooperative in resolving disputes, and those who are controlling tend to hold firm in negotiations. The auditors who take a tit-for-tat approach to dealing with taxpayers are also the ones who are most likely to emphasize convincing or persuading the taxpayers to accept their positions on disputes.Google Scholar

35 We are using the awkward phrase “likely correctness” to emphasize that correctness is often a matter of probability rather than certainty in areas such as taxation that have both complex fact situations and complex laws. Even with agreement on the legal interpretation, the two sides may place different probabilities on correctness, as potentially determined by appeals and litigation. Smith, “Will the Real Noncompliance Please Stand Up?” (cited in note 3). When we use the term “correct” here, we are implicitly meaning “most likely correct.”Google Scholar

36 Figure 2 is an idealized depiction of the strategies a party may adopt for dealing with a dispute, not a portrayal of possible resolutions. Issues may often be resolved, for instance, with one party believing the resolution is a new correct solution and the other party feeling that it is making a compromise that is not the most likely correct interpretation and application of the law.Google Scholar

37 For supporting evidence, see Stalans, 5 Soc. Just. Res. (cited in note 8).Google Scholar

38 Kent W. Smith, “Reciprocity and Fairness: Positive Incentives for Tax Compliance,” in Slemrod, Why People Pay Taxes 223 (cited in note 1) (“Smith, ‘Reciprocity and Fairness’”); Smith & Stalans, 13 Law B Pol'y (cited in note 8). Some research has examined how the actual resolution of disputes during audits are shaped by taxpayers' expectations about how auditors will treat them personally. Stalans, “Lay Evaluations” (cited in note 8), finds that taxpayers who anticipated a conflictual interaction (i.e., expected the auditor to be distrustful and rude) are more likely after the audit to disagree with the auditor about the correct legal solution for a specific issue than are taxpayers who anticipated an impartial and respectful interaction. Stalans & Smith, “How Taxpayers' Anticipations Influence Audit Experience” (cited in note 8), have also found that taxpayers who anticipated a conflictual interaction are more likely to assert their positions during audits. Stalans, “Lay Evaluations,” suggests that these patterns support the first part of the self-fulfilling prophecy: taxpayers' expectations about the nature of the interaction affect their own behavior toward auditors. These expectations, moreover, focus on the interpersonal nature of the interaction rather than on the instrumental aspects such as decision control.Google Scholar

39 In calculating the response rate, we excluded those who had moved and whose new telephone number or address could not be located through the several sources we checked, those who had moved out of the sampling areas, and those whose audits were dropped by the DOR after their initial contact.Google Scholar

40 The response rate in the rural area was 82%; in the smaller urban area, 68%; and in the large urban area, 50% (p <.00l).Google Scholar

41 The auditors and their managers at no point knew which taxpayers were in our sample or which ones had been interviewed by us, except in a few cases where the taxpayers volunteered the information in the course of the audit.Google Scholar

42 In one of the offices where we conducted our research, the manager assigns cases to the auditors, often with an eye toward the continuing professional development of the auditors.Google Scholar

43 By happenstance, this range is the other side of the one-in-three likelihood of being sustained on its merits that the IRS uses as the criterion for “realistic possibility” in its proposed amendments to Circular 230, the federal government's guide to conduct by lawyers, CPAs, and enrolled agents practicing before the IRS (8 Oct. 1992, Notice of Proposed Rulemaking).Google Scholar

44 During the pretesting we discussed the differences among the options with taxpayers and auditors, and they generally seemed to be making the distinctions we intended. Clearly, these options are an imperfect empirical capturing of the ideal-type strategies in fig. 2, and more work could be done on refining the measurement of strategies through responses to a hypothetical dispute. The measurement error is probably greatest along the dimension of reaching agreement on legal correctness.Google Scholar

45 For our analysis we did some minor editing of the strategies reported by the respondents. We ignored any strategies made after the respondents said they would capitulate or indicated that they did not know what they would do. Ah, if a hold firm choice was followed immediately by one of the more specific assertive strategies, the order of the two was switched. This adjustment is a post-interview approximation of forcing the respondents to chose among the more specific assertive strategies first. The change, which does not affect our substantive conclusions, was made for 1 auditor and 17 (8%) of the taxpayers. Finally, one auditor first chose the two assertive strategies and then repeated one of them as the third choice. We changed the third choice to holding firm.Google Scholar

46 The bracketed reference to a representative was included if the response to the previous question indicated that the respondent would let a tax expert handle some of the audit.Google Scholar

47 We assigned 3 to those with missing values whose response to the open-ended question was that issues would be decided by the “law” or figures, 4 to those who said by negotiation or “in our favor,” and 2 to those said the decisions would be made arbitrarily or in favor of the state. We did not substitute the extreme values of 1 or 5. See Stalans, 5 Soc. Just. Res. (cited in note 8), for a detailed description of how the coding of decision control was distinguished from other concepts.Google Scholar

48 The distributions of first, second, and third choices for taxpayers and auditors are all statistically significantly different with pc.05 for both x2 and τb (with hold firm placed before persuade).Google Scholar

49 One taxpayer selected holding firm as the second strategy followed by going along. We coded that respondent as holding firm.Google Scholar

50 Einhom, Hillel J. & Hogarth, Robin M., “Ambiguity and Uncertainty in Probabilistic Inference,” 92 Psychological Rev. 433 (1985).Google Scholar

51 Bardach & Kagan, Going by the Book (cited in note 12). See also Gilboy, 25 Law & Soc'y Rev (cited in note 21).Google Scholar

52 Bond, Rod A. & Lemon, Nigel F., “Training, Experience, and Magistrates' Sentencing Philosophies,” 5 Law B Hum. Behav. 123 (1981); Nigel Lemon, “Training, Personality and Attitude as Determinants of Magistrates' Sentencing,” 14 Brit. J. Criminology 34 (1974).CrossRefGoogle Scholar

53 Using the t-test for the difference in proportions, all these comparisons have one-tailed probabilities of.05 or less except the difference in the proportions including both agreement strategies among their three choices, which has a probability of.06.Google Scholar

54 We initially included several other variables in our analysis, either as controls or as checks on alternative hypothesis. None of them, however, had a significant net effect on the odds ratio, and their inclusion did not appreciably change the results given in table 7. The other variables explored were past audit experience; the scale for the legitimacy of government (see table 2); an attitude item about whether the amount of taxes paid by the respondent is too high; three measures of additional aspects of the perceived fairness of the DOR and its auditors (Stalans, 5 Soc. Just. Res.); and self-reported ratings of dominance, competence, and flexibility.Google Scholar

55 Smith, “Reciprocity and Fairness” (cited in note 39); Smith & Stalans, 13 Law & Pol'y (cited in note 8).Google Scholar

56 16 J. Personality & Soc. Psychology (cited in note 34).Google Scholar

57 Lind & Tyler, Social Psychology of Procedural Justice (cited in note 23); Stalans, & Smith, , “Procedural Criteria” (cited in note 8); Tom R. Tyler, “The Role of Perceived Injustice in Defendants' Evaluations of Their Courtroom Experience,” 18 Law & Soc'y Rev. 51 (1984); id., “What Is Procedural Justice? Criteria Used by Citizens to Assess the Fairness of Legal Procedures,” 22 Law & Soc'y Rev. 103 (1988).Google Scholar

58 See also Smith & Stalans, 13 Law & Pol'y at 35; Stalans, 5 SOC. Just. Res. at 93 (both cited in note 8).Google Scholar

59 See, e.g., Apfelbaum, 7 Advances of Experimental Soc. Psychology at 103 (cited in note 34); Larry B. Heuer & Steven Penrod, “Procedural Preferences as a Function of Conflict Intensity,” 51J. Personality B Soc. Psychology 700 (1986); Kelley & Stahelski, 16 J. Personality & Soc. Psychology (cited in note 34); Robert J. Stemberg & Lawrence J. Soriano, “Styles of Conflict Resolution,” 47 J. Personality B Soc. Psychology 115 (1984); Robert J. Stemberg & D. M. Dobson, “Resolving Interpersonal Conflicts: An Analysis of Stylistic Consistency,” 52 J. Personality B SOC. Psychology 794 (1987); Vidmar & Schuller, 11 Luw & Hum. Behav. (cited in note 11).Google Scholar

60 For related analyses and findings, see Smith & Stalans, 13 Law & Pol'y; Stalans & Smith, “How Taxpayers' Anticipations Influence Audit Experience”; Smith, “Cultural Embeddedness” (all cited in note 8).Google Scholar

61 E.g., Merry & silbey, 9 Just Sys J;Vidmar, 15 Law & Soc Inquiry; vidmar & Schuller, 11 Low & Hum. Behav. (all cited in note 11); and Lind & Tyler, Social Psychology of Procedural justice (cited in note 23).Google Scholar

62 The term used by Kagan, “On Regulatory Inspectorates” (cited in note 12).Google Scholar

Strategic Preferences 367Google Scholar

63 This grounding of regulatory inspection in legal correctness is at times overlooked in discussions and disputes about alternative enforcement strategies. In some respects, the move in many regulatory realms toward cooperative and participatory procedures can be interpreted as an attempt to redefine regulation in nonlegal terms in order to avoid protracted disputes about PEALS. For instance, the main theme in a recent newspaper article about an agreement between a utility and a state environmental agency was that the agreement is an example of the “evolution” toward cooperative rather than confrontational regulation (like Peterson, “Utility and Trenton Agree on Plan to Grow Fish Eggs,” N.Y. ‘Times, 13 July 1993, Late Ed.– Final, p. 85, col. 1). Well into the article, the reporter noted, “The state likes the idea of avoiding what promised to be a protracted and potentially losing legal battle with the utility, which maintains that the fish screens that it now uses meet the requirements of the law.” The reporter also quoted environmental groups as complaining that the regulators are “losing sight of the law” and their obligation to enforce it.Google Scholar

64 Kagan, “On Regulatory Inspectorates” at 54.Google Scholar

65 Id. Google Scholar

66 As one reviewer noted, tax laws place greater emphasis on the classification of activities and economic transactions than do many other types of regulation. However, the classifications in tax law are often seen as overly inclusive, just as are many types of regulations: the law's classification of what is taxable is overly inclusive, too many types expenses must be capitalized, too many expenses are considered personal, and so forth. All these examples, of course, can be turned around and stated in terms of some other category being too narrow, but the general thrust of the complaints are that tax laws are overreaching and too burdensome.Google Scholar

67 Stalans, 5 Soc. Just. Research; id., “Lay Evaluations” (both cited in note 8).Google Scholar