BobekHagemanHatfieldJATA2010 PDF

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/228131978
The Role of Client Advocacy in the Development of Tax Professionals’ Advice
Article in Journal of the American Taxation Association · August 2009

DOI: 10.2308/jata.2010.32.1.25
CITATIONS READS
23 557
3 authors, including:
Donna D. Bobek Amy M. Hageman

University of South Carolina Kansas State University
29 PUBLICATIONS 475 CITATIONS 35 PUBLICATIONS 469 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Amy M. Hageman on 06 February 2015.
The user has requested enhancement of the downloaded file.

JATA American Accounting Association
Vol. 32, No. 1 DOI: 10.2308/jata.2010.32.1.25
Spring 2010
pp. 25–51
The Role of Client Advocacy in the

Development of Tax Professionals’Advice
Donna D. Bobek, Amy M. Hageman, and Richard C. Hatfield
ABSTRACT: A primary responsibility of tax professionals is to be an advocate for their

clients 共AICPA 2000兲. Prior studies have shown mixed results on how the advocate role
influences tax professionals’ decision processes and outcomes 共e.g., Cloyd and Spilker
1999; Davis and Mason 2003; Barrick et al. 2004; Kahle and White 2004兲. In this study,
we consider how advocacy may be at least partially context-specific, introduce the
construct of client-specific advocacy, and thoroughly examine the influence of advo-
cacy attitudes on a number of steps in the judgment and decision-making process.
Consistent with attitude theory, we report experimental results that suggest client char-
acteristics influence tax professionals’ advocacy attitudes. We also find that client-
specific advocacy influences process variables such as the weighting of evidence and
decision outcomes such as the recommendation of tax advice. The results of the study
indicate that tax professionals may be unintentionally influenced by client attributes
when making judgments and may have difficulty separating their advocacy and evi-
dence evaluation roles.
Keywords: client advocacy; tax advice; client risk; client importance.
Data Availability: Data are available from the first author.
INTRODUCTION
ne of the primary roles of a tax professional is to function as a client advocate 共AICPA
O 2000兲. Unlike auditors, who are required to be independent of their clients, tax profes-
sionals are bound by professional standards to advocate for favorable tax positions that
fall within statutory boundaries. In executing their role as client advocates, however, tax profes-
sionals must still “objectively evaluate all relevant facts and tax authorities when preparing ad-
vice” 共Kadous and Magro 2001, 453兲. Failing to accurately and objectively assess the underlying
facts and the relevant law can result in taxpayer and tax professional penalties as well as negative
Donna D. Bobek is an Associate Professor at the University of Central Florida, Amy M. Hageman is a
Research Fellow at the University of Central Florida, and Richard C. Hatfield is an Associate Professor at
The University of Alabama.
We thank participants at the 2008 ABO Conference and the 2009 JATA conference for their helpful comments. We also
gratefully acknowledge the helpful comments of our discussants, Tracy Noga and Jennifer Kahle Schafer, as well as Dale
Bandy, Carlin Dowling, Richard Sansing 共JATA editor兲, and an anonymous reviewer; and the technical assistance of Joe
Schmitt.
Submitted: September 2008

Accepted: July 2009
Published Online: February 2010
25
26 Bobek, Hageman, and Hatfield
outcomes for tax professionals’ firms 共Hatfield 2000兲. Thus, tax professionals must serve as client
advocates while still evaluating evidence in an objective, unbiased manner.
The construct of advocacy has been an important part of the literature on tax professionals’
judgment and decision-making for over two decades 共e.g., Ayres et al. 1989; Roberts 1998兲, with
more recent studies tending toward direct construct measurement of advocacy attitudes 共e.g.,
Mason and Levy 2001; Stephenson 2007兲. Several prior studies have recognized the importance of
advocacy and its effect on judgment 共e.g., Johnson 1993; Davis and Mason 2003兲, although some
have failed to establish a direct connection between measured general advocacy levels and tax
professionals’ recommendations 共e.g., Kadous and Magro 2001; Barrick et al. 2004兲. These prior
studies have not explicitly considered how advocacy may be context-specific, or how, as an
attitude, it may change based on the external circumstances 共e.g., see Ajzen and Fishbein 1980兲.
Given that advocacy can influence tax professionals’ cognitive processes 共e.g., evidence search
and evaluation; Johnson 1993; Andre 2008兲, the variables that affect advocacy have the potential
to unintentionally affect their judgments and decisions. These unintentional effects may contribute
to biased judgments that could lead to flawed or deficient outcomes.
The purpose of this study is twofold. First, we investigate whether client characteristics
influence the degree to which tax professionals will function as advocates for their clients. Second,
we examine how this client-specific advocacy affects the manner in which tax professionals make
judgments and decisions. Specifically, we conduct an experiment with 101 tax professionals to
examine the degree to which advocacy, client characteristics, and other factors affect tax profes-
sionals’ decision processes 共weighting of evidence and IRS likelihood assessments兲 and decision
outcomes 共recommendation and allowance of favorable tax positions兲. Based on the experiment,
we assess whether tax professionals exhibit differing levels of advocacy for different clients, and
whether the influence of client-specific advocacy results in biased judgments.
The results of this study provide evidence that some client characteristics affect the advocacy
attitudes of tax professionals, and that client-specific advocacy influences tax professionals’ deci-
sion processes and outcomes. We used path analysis to simultaneously control for the direct and
indirect relationships among the study’s constructs. ANCOVA and regression analysis supple-
mented and confirmed the path analysis findings. Specifically, tax professionals in our study
displayed lower levels of client-specific advocacy for riskier clients. Furthermore, client-specific
advocacy influenced elements of tax professionals’ decision processes as well as their recommen-
dations. While client importance did not directly affect client-specific advocacy, it did have a
separate direct effect on the weighting of evidence and the allowance of favorable taxpayer
treatment. This suggests that tax professionals may be unintentionally influenced by both their
advocacy attitudes and attributes of the client when making judgments. Overall, the results indi-
cate that client-specific advocacy influences multiple steps in tax professionals’ judgment and
decision-making processes.
These results contribute to the taxation literature in several ways. First, this study extends
research on advocacy by examining its influence on a broader range of process and outcome
variables than have been considered in past research 共e.g., Davis and Mason 2003兲. Second, the
study introduces the construct of client-specific advocacy. As the results indicate that client char-
acteristics influence tax professionals’ advocacy attitudes toward a particular client, future research
should consider that advocacy itself is a function of client and context-specific characteristics.
Third, despite the fact that tax professionals should objectively evaluate evidence and the law, the
study’s findings demonstrate that tax professionals’ level of client-specific advocacy influences
their judgment processes. This adds to the growing body of literature suggesting that tax profes-
sionals may have difficulty separating their advocacy and objective evaluator roles 共e.g., Kadous
and Magro 2001; Kadous et al. 2008兲.
Journal of the American Taxation Association Spring 2010

American Accounting Association
Client Advocacy in the Development of Tax Professionals’Advice 27
The remainder of this paper proceeds as follows. The next section presents prior research and
develops hypotheses. The third section describes the research method. The fourth section presents
results, and the fifth section presents conclusions and limitations.
PRIOR RESEARCH AND HYPOTHESES

judgment and decision making for nearly two decades. In Table 1, we provide a summary of the
prior studies that have included advocacy as a construct of interest. Panel A identifies studies that
have inferred an advocacy effect, while Panel B includes the studies that have directly measured
the construct. Most early studies 共Johnson 1993 is a notable exception兲 did not directly measure
the advocacy attitudes of their participants. Instead, they inferred an advocacy effect based on
participants’ preferences for the tax-minimizing position 共e.g., LaRue and Reckers 1989; Carnes et
al. 1996; Cloyd and Spilker 1999; Hatfield 2000兲. More recent studies examining advocacy have
tended toward direct construct measurement 共e.g., Kadous and Magro 2001; Davis and Mason
2003; Barrick et al. 2004; Kahle and White 2004; Stephenson 2007; Pinsker et al. 2009兲. As
shown in Table 1, Panel B, we have identified eleven prior studies 共including one that is a working
paper and one that is an unpublished dissertation兲 that directly measured the advocacy attitudes of
their participants. The majority of these studies have relied on a scale developed by Mason and
Levy 共2001兲.
Based on the AICPA’s Statement of Responsibilities in Tax Practice and prior academic tax
literature, Mason and Levy 共2001, 127兲 define client advocacy as:
Advocacy is a state of mind in which one feels one’s primary loyalty belongs to the taxpayer. It is
exhibited by a desire to represent the taxpayer zealously within the bounds of the law, and by a
desire to be a fighter on behalf of the taxpayer. 共emphasis added兲
Thus, client advocacy refers to the degree of taxpayer loyalty exhibited by the tax professional
共Mason and Levy 2001; Davis and Mason 2003兲. However, a tax professional’s role as an advo-
cate should not bias the decision-making process. Tax professionals should evaluate evidence in an
objective, unbiased manner. Further, both taxpayers and tax professionals are subject to a variety
of monetary penalties should they take positions that do not meet appropriate standards.1 Thus, it
is incumbent on tax professionals in their role as client advocates to accurately and objectively
evaluate evidence and the law in order to provide their clients with accurate advice regarding the
merits of the preferred position.
However, most prior research that has studied client advocacy has found that it affects not
only the tax professionals’ recommendations to clients, but also many judgment processes. For
example, advocacy levels have been found to affect confirmation bias 共e.g., Johnson 1993兲, as tax
professionals generally place heavier weight on evidence that supports the client-preferred posi-
tion. A tax professional’s degree of advocacy also affects the review process 共Hatfield 2001兲.
Furthermore, advocacy may affect search behavior 共e.g., Cuccia 1994; Cloyd and Spilker 1999;
Andre 2008兲 and the weighting of evidence used to determine a final recommendation 共e.g., Levy
1996; Davis and Mason 2003兲. Prior studies have shown mixed results regarding its effect on final
recommendations to the client. Johnson 共1993兲 and Levy 共1996兲 found a significant effect, while
Kadous and Magro 共2001兲 and Barrick et al. 共2004兲 did not.
1
For example, taxpayers are subject to a “substantial understatement” penalty if they substantially understate their taxes
by taking an undisclosed position that does not have substantial authority, or a disclosed position that does not have a
reasonable basis 共IRC §6662兲. The tax preparer penalty has changed several times in the last two years. However, at the
time the data were collected for this study 共2007兲, a preparer was subject to a penalty for taking an undisclosed position
that did not have a realistic possibility of success 共IRC §6694兲.

Journal of the American Taxation Association
28
TABLE 1
Prior Advocacy Studies
Panel A: Inferred Advocacy Effect
Advocacy Advocacy Dependent
Study Participants Conceptualization Measure Variable(s) Advocacy Finding
LaRue and Reckers 110 Big 8 tax Preference for the Indirect Reporting Aggressive 共pro-taxpayer兲
1989 managers client-favored recommendation reporting recommendation
position influenced by interaction
between experience, payment
status, and tax savings.
Ayres et al. 119 tax Preference for the Indirect Reporting Due to their professional
1989 CPAs; 49 client-favored recommendation advocacy role, CPAs were
non-CPAs position more likely to recommend
client-favored positions in
ambiguous situations than
were non-CPAs.
Duncan et al. 133 Big 8 tax Preference for the Indirect Reporting Tax professionals sought
1989 managers client-favored recommendation to mitigate client biases by
position recommending more
aggressive 共pro-taxpayer兲
positions for less aggressive
clients.
Helleloid 181 tax Preference for the Indirect Reporting CPAs gave more pro-
1989 CPAs; 96 client-favored recommendation taxpayer recommendations
masters’ position when greater ambiguity was
students present, but were not
influenced by client attitude.
Cuccia 45 non-CPA Preference for the Indirect Information CPAs were less pro-client in
Bobek, Hageman, and Hatfield

1994 tax preparers; client-favored search behavior their recommendations than
36 tax CPAs position 共effort兲, reporting were non-CPAs. For CPAs,
recommendation increased penalties resulted
in increased effort in
information search.
(continued on next page)

Spring 2010
Client Advocacy in the Development of Tax Professionals’Advice

Cuccia et al. 67 Big 6 tax Preference for the Indirect Reporting Due to their inherent
1995 managers client-favored recommendation, advocacy role, tax
position probability of professionals recommended
success, pro-taxpayer positions under
interpretation of verbal standards 共exploitation
authority of standard兲 and numerical
standards 共evidence
interpretation兲.
Carnes et al. 77 tax Preference for the Indirect Reporting Greater risk propensity was
1996 professionals client-favored recommendation associated with pro-taxpayer
position recommendations.
Spilker et al. 63 Big 6 tax Preference for the Indirect Reporting Tax professionals were more
1999 professionals client-favored recommendations; aggressive in recommending
position explanation of client-favored positions in
recommendations ambiguous tax compliance
共as opposed to planning兲
situations.
Cloyd and 72 Big 5 tax Preference for the Indirect Information Due to their advocacy role,
Spilker 1999 professionals client-favored search behavior, tax professionals overweight
position likelihood client-favorable cases in
assessments, information searches
reporting 共confirmation bias兲, which
recommendations can lead to overly aggressive
recommendations.
Hatfield 56 tax staff Preference for the Indirect Evaluation bias Confirmation bias is
2000 accountants client-favored 共confirmation dependent upon a preference

position bias兲, reporting for client-favorable opinions.
recommendation
Hatfield 65 tax Preference for the Indirect Weight given to Due to their inherent
2001 managers client-favored a staff advocacy role, supervisors
position accountant’s may still be biased toward a
research report preference for client-
favorable positions.
Spring 2010
29
30
Kadous et al. 63 tax Preference for the Indirect Information Tax professionals’
2008 professionals client-favored search behavior, information search and
position likelihood judgments are less consistent
assessments, with client preferences for
reporting clients with high practice
recommendations risk.
Panel B: Direct Measurement of Advocacy Attitudes

Johnson 1993 109 Big 6 tax Preference for the Developed Weighting of Advocacy contributes to tax
professionals client-favored 17-item evidence; professionals’ use of
position advocacy scale reporting confirmatory processes
recommendation 共confirmation bias兲 and
directly affects the strength
of their recommendations.
Levy 1996 138 tax Degree of taxpayer Mason and Evidence Advocacy influenced
professionals loyalty exhibited Levy judgments, evidence judgments and
by the tax 共2001兲 reporting reporting judgments;
professional scale judgments, evidence that environmental
共attitude兲 reporting conditions affect advocacy.
recommendations
Cuccia and McGill 94 Big 5 tax Beliefs regarding Johnson Probability Advocacy scores influenced
2000 professionals appropriate 共1993兲 estimates, judgments 共but were not
behavior when tax scale reporting related to evidence order兲.
law is ambiguous recommendations

Mason and Levy 34 tax CPAs Degree of taxpayer Developed NA Developed a valid, reliable
2001 and 30 IRS loyalty exhibited nine-item scale to measure client
agents by the tax advocacy advocacy.
professional scale
共attitude兲
Spring 2010


Kadous and Magro 86 practicing Degree of taxpayer Mason and Reporting Advocacy scores did not
2001 tax loyalty exhibited Levy 共2001兲 recommendations explain a significant amount
professionals by the tax scale and the of variance in judgments of
professional weighting of recommendations and
共attitude兲 evidence weighting of evidence.
Davis and Mason 91 tax CPAs Degree of taxpayer Mason and Weighting of Advocacy scores influenced
2003 and 59 IRS loyalty exhibited Levy 共2001兲 evidence 共tax the weighting of evidence
agents by the tax scale authority 共stronger advocates were
professional judgments兲 more influenced by
共attitude兲 favorable
judicial precedents兲.
Barrick et al. 55 practicing Degree of taxpayer Mason and Reporting Measured advocacy scores
2004 tax loyalty exhibited Levy recommendations, did not influence
professionals by the tax 共2001兲 probability recommendations or
professional scale estimates probability estimates.
共attitude兲共However, due to their
inherent advocacy role, tax
professionals were more
persuaded by biased
memoranda that incorrectly
supported the client-favored
position.兲
Kahle and White 53 practicing Degree of taxpayer Mason and Belief revision Due to tax professionals’
2004 tax loyalty exhibited Levy inherent advocacy role,
professionals by the tax 共2001兲 belief revision is greater
professional scale when it supports client
共attitude兲
preferences.
Stephenson 252 non-CPA Degree of taxpayer Mason and Client advocacy CPAs and non-CPAs
2007 tax preparers; loyalty exhibited Levy scores demonstrated similar client
286 tax CPAs by the tax 共2001兲 advocacy scores; tax
from regional professional scale 共dropped preparers exhibited higher
or local firms 共attitude兲 one item兲 advocacy levels than
perceived by their clients.
Spring 2010
31
32
Andre 2008 64 Adoption of Measured Confirmatory Advocacy scores influenced
accounting client- with seven information the degree of confirmatory
masters’ preferred position items 共e.g., searches and information searches 共e.g.,
students; 27 as a reference importance evaluations supporting cases examined兲.
law students; frame of client
11 Big 4 tax preferences兲
professionals
Pinsker et al. 44 auditors, Degree of taxpayer Five items Belief revision The type of decision
2009 38 tax loyalty exhibited from the environment 共audit or tax兲
professionals, by the tax Mason and moderates the influence of
and 80 professional or Levy advocacy on judgments.
masters’ auditor 共attitude兲共2001兲
students scale

Spring 2010
In addition to the previous studies that directly measure tax professionals’ advocacy attitudes,
two prior studies that examine the influence of client preferences on tax professionals’ judgment
are particularly relevant to our study: Cloyd and Spilker 共1999兲 and Kadous et al. 共2008兲. Cloyd
and Spilker 共1999兲 examine how client preferences may bias tax professionals’ decision processes
and outcomes, as tax professionals’ information search and likelihood assessments are more likely
to support the client-preferred position, leading to overly aggressive recommendations. Building
upon this finding, Kadous et al. 共2008兲 find that tax professionals’ bias in information search and
recommendations is less prevalent when advising riskier clients. Taken together, these two studies
suggest that tax professionals’ inherent advocacy role may lead to biased judgments and decisions,
but that characteristics of the clients, themselves, may help mitigate some of this bias.
While many prior studies have demonstrated the importance of advocacy, the prior studies
that have examined this construct either have not directly measured advocacy attitudes or have
measured the general level of advocacy that respondents exhibit toward a “generic” taxpayer.
These studies have not explicitly investigated how client characteristics may lead tax professionals
to exhibit more or less advocacy toward a specific client. Advocacy is commonly conceptualized
as an attitude held by the tax professional 共Roberts 1998; Mason and Levy 2001兲. According to
Ajzen and Fishbein 共1980兲, an attitude that relates to a particular behavior 共for example, advocacy
as it relates to a tax professional’s recommendation兲 is actually a function of an individual’s
underlying beliefs about what will occur if he engages in that particular behavior 共in this case,
being an advocate兲. Thus, advocacy as an attitude can be expected to change based on the possible
economic or non-economic outcomes associated with acting as an advocate in a particular instance
or context. It is then likely that the link between a tax professional’s advocacy attitude toward a
specific client and his judgments and decision-making processes will be stronger than that found
by prior research, which has measured subjects’ advocacy attitudes in general.
Client Risk
A key client attribute that may influence a tax professional’s advocacy attitude is client risk.
Client risk refers to the risk borne by the tax professional in making inappropriate client recom-
mendations. Kadous and Magro 共2001兲 use the term practice risk rather than client risk and
identify a number of factors associated with this construct. Client risk includes potential exposure
to monetary costs imposed by the IRS; liability to the client for additional taxes, penalties, and
interests on those taxes; other compensatory and punitive damages and the cost of legal represen-
tation; and non-monetary costs including the loss of reputation, damage to relationship, and
AICPA sanction. Client risk is higher for certain client characteristics identified in the tax practice
literature, including uncooperativeness, frequent involvement in litigation, participation in high-
risk industries, fee pressure, financial difficulties, and involvement in suspicious transactions 共see
Bandy 1996; Kahan 1996; Pascarella 1996; Fiore 1998; Bobek et al. 2004兲.
Kadous and Magro 共2001兲 used an experiment to show that tax professionals weighted infor-
mation differently when client 共practice兲 risk was higher. In a follow-up study, Kadous et al.
共2008兲 established that biased information searches and aggressive recommendations were more
common with low-risk clients. These studies suggest that because client risk has the potential to
alter a tax professional’s beliefs concerning ultimate outcomes, tax professionals may exhibit
differing levels of advocacy for clients with differing perceived risk. This leads to the first hy-
pothesis:
H1: Tax professionals will exhibit greater 共weaker兲 client advocacy for low 共high兲 risk
clients.

Client Importance
Another salient client characteristic that may influence tax professionals’ beliefs about likely
outcomes is client importance. As conceptualized by prior research 共see Roberts 关1998兴 for a
review兲, client importance relates to the current and future income streams associated with a
particular client, either directly or through their influence on other current or potential clients.
Kahle and White 共2004兲 identify tax professionals’ desire to retain important clients and how this
influences their decision making as an area for future research that explores the consequences of
client characteristics. Evidence from prior experimental research 共e.g., McGill 1988; Reckers et al.
1991兲 suggests that the more important a client is, the more aggressive a tax professional is willing
to be; self-reports from interviews show a similar relationship 共Milliron 1988兲. Thus, tax profes-
sionals may exhibit higher levels of advocacy for clients that are more important to their firms,
because acting as an advocate for such clients has the potential to improve outcomes for the tax
professionals. This leads to our second hypothesis:
H2: Tax professionals will exhibit greater 共weaker兲 client advocacy for more 共less兲 important
clients.
A primary reason for studying the effect that client characteristics have on client advocacy is
because prior studies have found a relationship between general advocacy levels and tax profes-
sionals’ decisions 共e.g., Roberts 1998; Davis and Mason 2003兲 as well as the processes they
engage in to make these decisions 共e.g., Johnson 1993; Davis and Mason 2003; Andre 2008兲. If
tax professionals are following a strategy of utility maximization, client characteristics such as risk
and importance may be expected to have an effect on recommendations directly and/or indirectly
through their effect on client-specific advocacy 共i.e., tax professionals may strategically provide
marginally different advice to clients with differing characteristics兲. Specifically, the riskiness/
importance of the client can, and perhaps should, directly affect how aggressive a tax professional
is willing to be when providing a recommendation. Given the same set of facts and law, objec-
tively analyzed, it is reasonable to conclude that a tax professional might be willing to provide
more 共less兲 aggressive advice for an important 共risky兲 client, due to the potential benefits 共costs兲
associated with the specific client. However, advocacy 共general in prior literature兲 has been shown
to bias decision making 共i.e., to affect tax professionals’ cognitive decision processes兲. Thus, if
client characteristics have an effect on the evaluation of facts and evidence indirectly through their
effect on client-specific advocacy 共as opposed to just affecting the final recommendation兲, the
potential for biased decisions becomes more pronounced due to the multiple channels through
which biases could manifest.
Therefore, we consider the influence of client characteristics and client-specific advocacy on
several aspects of the decision-making process. We first consider their effect on decision-
processing variables, and second, we consider their effect on decision outcomes. It should be
noted that client characteristics may only operate through client-specific advocacy 共i.e., client-
specific advocacy may fully mediate the relationship of client characteristics to other constructs兲.
However, since prior research 共e.g., Kadous and Magro 2001; Pinsker et al. 2009兲 has found that
characteristics of the client or task can have direct effects on judgment variables and that client
advocacy also affects judgment variables 共e.g., Johnson 1993; Davis and Mason 2003; Barrick et
al. 2004; and Andre 2008兲, we expect that client characteristics and client-specific advocacy will
influence tax professionals’ weighting of evidence and judgments about the likelihood of favorable
IRS treatment 共both important inputs into the decision-making process兲. This leads to the follow-
ing two hypotheses:
H3: Client characteristics 共specifically, risk and importance兲 will influence tax professionals’
decision processes 共specifically, weighting of evidence and likelihood judgments兲.

H4: Client-specific advocacy will influence tax professionals’ decision process 共specifically,
weighting of evidence and likelihood judgments兲.
Additionally, we consider the effect of client characteristics and client-specific advocacy on
the decision outcome. A direct effect of these variables on recommendations does not necessarily
represent biased decision making, but may merely be utility-maximizing behavior as described
previously. Therefore, it is important to consider both the decision processes 共which should not be
affected by client characteristics or client-specific advocacy兲 and the decision outcome 共which
could normatively be expected to be affected by these variables兲. Most prior literature that has
directly measured tax professionals’ advocacy 共e.g., Johnson 1993; Levy 1996; Kadous and Magro
2001; Barrick et al. 2004兲 has evaluated the tax professional’s recommendation to the client as the
decision outcome variable, with mixed results.
We also consider a second outcome measure: whether the tax professional would “allow” the
taxpayer to take the tax-minimizing position. Prior literature has suggested that the distinction
between the “recommendation” and the “allowance” of a particular tax position may be important,
given the differing levels of responsibility in the decision 共e.g., Reckers et al. 1991兲. Even though
tax professionals may recommend a more conservative choice, if the client presses the issue, they
may still be willing to sign a return that has taken a more aggressive position.
Prior research has suggested that client characteristics such as client risk and client impor-
tance may influence decision outcomes such as the recommendation and allowance of favorable
tax advice to the client 共e.g., Reckers et al. 1991; Kadous and Magro 2001兲. This leads to our fifth
hypothesis:
H5: Client characteristics 共specifically risk and importance兲 will influence tax professionals’
decision outcomes 共specifically, recommendations and allowance of favorable tax
treatment兲.
Finally, we consider the separate influence of client-specific advocacy on decision outcomes.
The prior mixed results on how tax professionals’ advocacy affects their recommendations 共e.g.,
Johnson 1993; Levy 1996; Kadous and Magro 2001; Barrick et al. 2004兲 may be due to a tendency
to examine tax professionals’ general levels of advocacy rather than the advocacy that they exhibit
for a specific taxpayer. Since client-specific advocacy may positively influence tax professionals’
recommendations, it follows that the same relationship may be seen between client-specific advo-
cacy and the allowance of a favorable tax position. This leads to our final hypothesis:
H6: Client-specific advocacy will be positively related to tax professionals’ pro-client deci-
sion outcomes 共specifically, recommendations and allowance of favorable tax treatment兲.
Figure 1 depicts an overall model of the predicted relationships.
RESEARCH METHOD
Participants
We directly contacted firms to obtain the majority of the 101 experienced tax professionals
who participated in the study. We contacted 211 tax professionals and received 93 responses
共response rate of 44 percent兲. We encouraged participation by offering a chance to win a $25 gift
certificate or a donation to one of five possible charitable organizations 共chosen by participant兲

FIGURE 1
Model of Hypothesized Relationships
upon completion.2 Analysis indicated that participants did not significantly differ between the four
experimental conditions on any demographic variable, with the exception of gender. Male tax
professionals were more likely to be assigned to the “low importance” condition than the “high
importance” condition 共p ⫽ .065兲. We consider the effect of demographic variables in the addi-
tional analyses section of the paper.
Table 2 reports demographic information regarding the study’s participants. The majority of
participants were female 共67 percent兲, CPAs 共62 percent兲, and at the staff or senior level 共60
percent兲. They spent, on average, between 20–30 percent of their chargeable time providing tax
advice that required some amount of research, and the average participant had 8.28 years of tax
experience.
Experimental Task and Procedures

All participants completed the task in an online environment. We manipulated two indepen-
dent variables, client risk and client importance, at two levels 共high and low兲, creating four
experimental conditions. Participants provided demographic information and answered questions
measuring their general level of client advocacy. Next, participants received facts about the client.
These were manipulated across experimental conditions, with participants receiving different in-
formation based on whether they were in the high or low importance condition and the high or low
risk condition. Appendix A contains the differing client facts for each of the four experimental
conditions. As a manipulation check, we then instructed participants to answer “how important”
the hypothetical client was relative to other clients at their firm, and “how much risk” was created
by maintaining the hypothetical tax clients.
2
In addition to the direct firm contact participants, we also solicited participants through a mailing to 100 AICPA
members. The prospective participants received a cover letter with an electronic link to the online experiment. To
encourage participation, a $25 gift certificate was randomly awarded to participants based on the number of responses.
We received eight responses from this mailing 共response rate of 8 percent兲. Participants from the AICPA and participants
obtained from direct firm contacts did not vary significantly in their responses to the study’s independent or dependent
variables.

TABLE 2
Demographic Information
Samplea
Years of Tax Experience
Mean 8.28 years
共S.D.兲共8.06兲
Firm Type
International 32%
National 11%
Regional 17%
Local 40%
Position in Firm
Partner/Owner 11%
Senior Manager 12%
Manager 17%
Senior 24%
Staff 36%
Gender
Female 67%
Male 33%
Professional Status
CPA 62%
a Percentages are calculated based on the total participants providing information for that category.
Participants next responded to a simulated engagement to provide tax advice to the hypotheti-
cal client. They were given a short vignette about the client’s “hobby loss” scenario.3 We selected
a hobby loss decision context because it involves a high degree of ambiguity 共uncertainty兲. Many
of the facts suggested that the taxpayer’s activities should be considered a business activity 共e.g.,
the taxpayer carried out the activity in a businesslike manner and there was a likelihood of future
profits兲, whereas others suggested it was a hobby 共e.g., the presence of personal pleasure in a
recreation activity兲. Appendix B reproduces the text of the tax issue; all participants received these
same facts.
Next, we provided participants with information from the applicable tax authority for this
issue, Treasury Regulation Sec. 1.183-2. This regulation provides nine factors that should be
considered when evaluating whether an activity is engaged in for profit; participants indicated how
relevant they found each factor in determining the appropriate tax treatment of the hypothetical
client’s activities. Subsequently, participants answered questions regarding their tax advice for the
decision context, including how they would advise the client, what tax treatment they would
allow, and the likelihood that the IRS would consider the activity to be a business. Finally,
3
“Hobby loss” rules pertain to IRC Section 183, which prohibits an individual from deducting expenses from an activity
that are in excess of that activity’s income if the activity is “not engaged in for profit.” Thus, if an activity is considered
a “business,” then a net loss from the activity can be deducted; however, if an activity is considered a “hobby,” while
any income earned must be included in gross income, expenses can only be deducted up to the amount of such income,
and are considered miscellaneous itemized deductions.

participants again responded to the client advocacy scale 共Mason and Levy 2001兲, but in terms of
their level of client-specific advocacy, as held toward the hypothetical client.4
Manipulated Variables
We manipulated “client risk” and “client importance” as between-subject factors across ex-
perimental conditions. Client risk was either high or low based on an adaptation of the risk
manipulations in Kadous and Magro 共2001兲.5 In the high client risk manipulation, the client had
unreliable and incomplete financial records, often proposed various questionable tax savings strat-
egies, and previously had sizable adjustments and penalties from IRS audits. In the low client risk
manipulation, the client had reliable and complete financial records, always avoided questionable
tax savings strategies, and had never had adjustments or penalties from IRS audits. To test the
effectiveness of this manipulation, participants rated “how much risk” was created in maintaining
the hypothetical tax clients relative to other clients at their firm, ranging from 1 共no risk兲 to 6 共very
high risk兲. The means and standard deviations of the high and low client risk ratings were 4.02
共1.21兲 and 2.09 共0.88兲, respectively, and were significantly different at p ⬍ 0.001, suggesting an
effective manipulation.6
Client importance was high or low, as modified from past literature 共e.g., Reckers et al. 1991兲.
In the high client importance manipulation, the client was a local bank president who was one of
the office’s largest clients with substantial audit fees, the firm had prepared the client’s personal
tax return for many years, and the client had been a source of past referrals. In the low client
importance manipulation, the bank was not a firm client, the firm’s only relationship was prepa-
ration of the client’s personal tax return for the past two years, and the client had not been a source
of referrals. To test the effectiveness of this manipulation, participants rated “how important” the
hypothetical tax clients would be relative to other clients at their firms, ranging from 1 共very
unimportant兲 to 7 共very important兲. The means and standard deviations of the high and low client
importance ratings were 5.50 共1.47兲 and 4.26 共1.87兲, respectively, and were significantly different
at p ⫽ 0.001, suggesting an effective manipulation.
Measured Variables
We measured both general and client-specific advocacy attitudes using a nine-item scale
developed and validated by Mason and Levy 共2001兲. Mason and Levy’s 共2001, 127兲 scale mea-
sures the degree to which a tax professional exhibits “primary loyalty to the taxpayer” 共emphasis
added兲; it has been validated through its use by other tax researchers, including Levy 共1996兲,
Kadous and Magro 共2001兲, Davis and Mason 共2003兲, Barrick et al. 共2004兲, and Stephenson 共2007兲.
We measured general advocacy prior to the hypothetical tax scenario, capturing general attitudes
pertaining to “the taxpayer.” We measured client-specific advocacy after the hypothetical tax
4
Twenty-one practicing tax professionals pilot-tested the instrument and provided extensive feedback in terms of the
length of the experiment, realism of the scenario, and clarity of the questions. Based on these pre-testers, we revised the
instrument; in particular, we made changes to strengthen the manipulation of “client importance.”
5
Kadous and Magro 共2001兲 used the term practice risk to describe this construct. To use consistent language describing
client characteristics, we refer to “practice risk” as “client risk.”
6
The six-point scale included the options “no risk,” “very little risk,” “some risk,” “moderate risk,” “high risk,” and “very
high risk,” including a neutral option 共which would make the measurement out of a seven-point scale兲 was not logical
in this context. Furthermore, the mean of perceived risk for the high-risk condition 共4.02兲 corresponds to slightly above
“moderate risk” on the measured scale. This is slightly lower than the average for the “high practice risk” condition in
the Kadous and Magro 共2001兲 study; however, the client in the high practice risk condition for Kadous and Magro
共2001兲 had also been involved in a lawsuit against its former accountant. The condition was absent in our study since
the hypothetical client was an individual taxpayer. Furthermore, Kadous and Magro 共2001, 462兲 asked participants to
evaluate “the riskiness of the client relative to the population of all possible clients,” whereas our study asked partici-
pants to rate how much risk was created by maintaining the clients relative to other clients at their firm.

scenario, capturing specific attitudes toward the experiment’s hypothetical client 共by substituting
the names of the hypothetical client in place of “the taxpayer” on the scale’s items兲.
Table 3 contains the items included on the Mason and Levy 共2001兲 advocacy scale, along with
the mean responses to each question for both general and client-specific advocacy. Paired sample
t-tests demonstrate that participants differed in their responses for the general and client-specific
advocacy scale measures for five of the nine scale items 共p ⬍ 0.10兲. Three of these items dem-
onstrated strongly statistically significant differences 共p ⬍ 0.01兲 between the two scales: Question
1 共“In an instance where no judicial authority exists with respect to an issue and where the Code
and Regulations are ambiguous, I feel that the taxpayer is entitled to take the most favorable tax
treatment”兲; Question 3 共“I feel I should apply ambiguous tax law to the taxpayer’s benefit”兲; and
Question 5 共“I believe it is important that I encourage the taxpayer to pay the least amount of taxes
possible”兲. On average, participants demonstrated higher levels of advocacy for taxpayers in
general 共general advocacy兲 than for the hypothetical client in the scenario 共client-specific advo-
cacy兲 for nearly all of these measures.7
Both the general and client-specific advocacy scales demonstrate acceptable levels of reliabil-
ity, with Cronbach’s alpha of 0.81 and 0.84, respectively; scale reliability is comparable with the
Cronbach’s alpha of 0.83 reported by Mason and Levy 共2001兲.8 Therefore, we summed the items
in each of these scales to form new measures.
We used several measures to capture tax professionals’ judgment and decision-making pro-
cesses. The decision process variables we collected 共used to test H3 and H4兲 were tax profession-
als’ perception of the likelihood that the IRS would consider the activity to be a business 共mea-
sured on an 11-point Likert-type scale, where 1 ⫽ 0 percent and 11 ⫽ 100 percent兲 and their
weightings of the relevance of the pertinent treasury regulation factors governing this issue 共mea-
sured on a 7-point Likert-type scale, where 1 ⫽ very relevant and 7 ⫽ very irrelevant兲. The mean
共standard deviation兲 of participants’ responses was 5.74 共2.58兲 for likelihood and 5.52 共1.47兲 for
weightings. We also measured the participants’ decision outcomes 共used to test H5 and H6兲. Of
primary interest was how tax professionals would advise the client 共measured on seven-point
Likert-type scale, where 1 ⫽ definitely a hobby activity and 7 ⫽ definitely a business activity兲 and
what tax treatment they would allow 共measured on a seven-point Likert-type scale, where 1 ⫽
definitely not allow business treatment and 7 ⫽ definitely allow business treatment兲. Tax profes-
sionals’ advice was more aggressive 共and more beneficial to taxpayers兲 to the extent that the
activity was considered a business.9 The mean 共standard deviation兲 of participants’ responses was
3.78 共1.86兲 for advise and 4.18 共1.95兲 for allow. Thus, it appears we were successful at creating an
ambiguous scenario.
RESULTS
As illustrated in Figure 1, the study’s hypothesized relationships are interconnected. Thus, we
employed path analysis 共using AMOS兲 to simultaneously test all of the study’s hypotheses and to
7
Several demographic characteristics were related to differences in the level of general advocacy. Consistent with prior
research, tax professionals that were male 共p ⬍ .10兲 or CPAs 共p ⬍ .02兲 had higher levels of advocacy, as did those with
higher positions in the firm 共p ⬍ .001兲, more experience 共p ⬍ .001兲, and who worked at larger firms 共p ⫽ .06兲.
8
The mean responses to the advocacy scale of our participants indicates that they are similar to participants used in prior
studies. Davis and Mason 共2003兲 provided information on the mean responses to the advocacy scale for CPAs and IRS
participants. Out of a range from 9 to 63, CPA participants in their study averaged a total score of 46.03. CPA
participants in our study averaged 44.55; this is in line with the average of 43.88 reported by Kahle and White 共2004兲.
9
We also considered whether tax professionals would advise the hypothetical clients to file amended returns for prior
years treating the activity as a business. While the likelihood of favorable IRS treatment 共a processing variable兲 was
related tax professionals’ decision to file amended returns 共p ⬍ .05兲, none of the study’s hypothesized independent
variables 共client-specific advocacy attitudes and client characteristics兲 significantly affected this decision.

TABLE 3
Mason and Levy (2001) Advocacy Scale
Client-
General Specific
Advocacy Advocacy
Question 1: In an instance where no judicial authority exists 5.33*** 4.97***
with respect to an issue and where the Code and Regulations
are ambiguous, I feel that the taxpayer is entitled to take the 共1.45兲共1.54兲
most favorable tax treatment.
Question 2: Generally speaking, my loyalties are first to the tax 3.02** 3.35**
system, then to the taxpayer. 共Item is reverse-coded兲共1.69兲共1.75兲
Question 3: I feel I should apply ambiguous tax law to the 4.87*** 4.52***
taxpayer’s benefit. 共1.68兲共1.63兲
Question 4: When examining a tax return, I tend to point out to 5.77 5.93
taxpayers reasonable positions they could have taken which
would have contributed to minimizing their tax liability. 共1.30兲共1.09兲
Question 5: I believe it is important that I encourage the 4.98*** 4.48***

taxpayer to pay the least amount of taxes possible. 共1.67兲共1.82兲
Question 6: I always interpret unclear/ambiguous laws in favor 4.26 4.24

of the taxpayers. 共1.74兲共1.67兲
Question 7: It is important to use trends in the law by trying to 4.96 5.04

establish a pattern of more favorable treatment for the
taxpayer and then extending this pattern to the taxpayer’s 共1.22兲共1.34兲
position.
Question 8: Where no judicial authority exists with respect to 5.10* 4.93*

an issue, I feel that the taxpayer is entitled to take the most
favorable tax treatment. 共1.37兲共1.42兲
Question 9: The taxpayer has the right to structure transactions 5.14 5.02
in ways that yield the best tax result, even if the law is
unclear in an area. 共1.45兲共1.41兲
Overall 43.43** 42.43**

共8.61兲共9.17兲
Cronbach’s Alpha 共Scale Reliability兲 0.812 0.843
*, **, *** Statistically significantly different at p ⬍ .10, p ⬍ .06, and p ⬍ .01, respectively, 共two-tailed兲, based on paired
sample t-tests.
Scores reflect the means 共standard deviation兲 for each item for all participants that answered the applicable item on both the
general advocacy and client-specific advocacy scales. Overall scales compare participants that measured all items on both
scales.
All items are on a seven-point Likert-type scale, ranging from strongly disagree to strongly agree. Adding up all nine items
results in a 共nine-point minimum兲 63-point maximum scale.

consider both the direct and indirect relationships among the constructs of interest. The path
analysis results are complemented with traditional hypothesis testing using ANCOVA and regres-
sion.
We estimate a model of the relationships proposed in Figure 1 in order to test all direct and
indirect relationships among client characteristics, client-specific advocacy, and judgment and
decision-making processes and outcomes. It is important to note that although we manipulated
client risk and client importance, in order to include these constructs in the path analysis, we use
respondents’ perceptions of relative client risk and client importance as opposed to condition
assignment. Use of tax professionals’ perceptions of risk and importance also affords a finer-
grained test of how these client characteristics influence tax professionals’ advice. As participants
were from a wide range of firm sizes, they also may have had different perceptions of relative
client characteristics.
Our initial model that tests all of the hypothesized relationships exhibits acceptable model fit
共e.g., CFI ⫽ 0.977, in excess of the recommended threshold of 0.90 per Bentler and Bonnett
关1980兴兲.10 However, many of the paths from client characteristics 共client risk and client impor-
tance兲 to decision processes 共IRS likelihood and weighting of evidence兲 and to decision outcomes
共recommendation and allowance of favorable tax advice兲 are statistically insignificant. Thus, we
delete statistically insignificant paths in order to develop a model that better describes the rela-
tionships among the constructs 共Schumacker and Lomax 2004兲.
Figure 2 depicts the model of the relationships among the study’s constructs after deleting the
statistically insignificant paths between client characteristics and the process and outcomes vari-
ables. The overall model fit is acceptable, and thus, is a good fit for the data.11
Hypotheses 1 and 2
As shown in Figure 2, H1 is supported in that tax professionals exhibited a lower level of
client-specific advocacy for clients they perceived to be more risky. The path from client risk
perceptions is negative and statistically significant 共p ⬍. 01兲. However, contrary to our prediction
in H2, respondents’ level of client-specific advocacy was not affected by client importance.
While we used respondents’ perceptions of client risk and importance in the path analysis, we
confirmed these results using the condition assignment as well. We provide descriptive statistics in
Table 4, Panel A and an analysis of covariance 共ANCOVA兲 in Table 4, Panel B. The ANCOVA
results show that when controlling for participants’ baseline levels of general advocacy as a
covariate, the manipulation of client risk has a statistically significant and negative influence on
client-specific advocacy 共p ⬍ 0.05兲. However, client importance did not influence client-specific
advocacy 共p ⬎ 0.4兲.
Hypotheses 3 and 4
Hypothesis 3 investigates the influence of client characteristics on the decision-making pro-
cess 共i.e., likelihood judgments and weighting of evidence兲, whereas H4 examines whether client-
10
The path analyses are conducted only for the participants with no omitted data, resulting in 88 participants.
11
Hair et al. 共1998兲 recommend the evaluation of several fit indices in assessing whether appropriate model fit has been
achieved. The Chi-square statistic is insignificant 共17.165; degrees of freedom ⫽ 12; p ⫽. 144兲, and the ratio of the
Chi-square statistic to degrees of freedom of 1.43 is well less than 2, indicating adequate model fit 共Hair et al. 1998兲.
The Comparative Fit Index 共CFI兲 of 0.985 exceeds the recommended threshold of 0.90 共Bentler and Bonnett 1980兲,
while the Root Mean Square Error of Approximation 共RMSEA兲 of .07 is under the recommended ceiling of .08 共Hair et
al. 1998兲. In addition to the global measures of model fit, incremental fit measures 共which compare the path model to the
null model兲 are also useful. The Tucker-Lewis Index 共TLI兲 of 0.966 and the Normed Fit Index 共NFI兲 of 0.955 both
exceed the recommended threshold of 0.90 共Hair et al. 1998兲.

FIGURE 2
Standardized Path Coefficients for Path Analysis
*, **, *** p < .10, p < .05, and p < .01, respectively, all tests two-tailed.
Model Chi-Square = 17.165 (df = 12), ratio = 1.43, p = .144.
Comparative Fit Index (CFI) = 0.985.
Root Mean Square Error of Approximation (RMSEA) = .07.
specific advocacy influences these judgment processes. In our experimental task, we present the
hobby loss scenario so that the factor “elements of personal pleasure or recreation” is the primary
factor that suggests hobby loss treatment 共i.e., unfavorable treatment for the taxpayer兲. This factor
is also one of the most difficult elements to surmount in successfully arguing that an activity is
engaged in with the objective of making a profit 共see Schwartz 1993兲. Advocacy has previously
been linked to the weighting of evidence 共e.g., Levy 1996; Cuccia and McGill 2000; Davis and
Mason 2003兲, and participants’ weighting of the relevance of this factor may be a reflection of
their advocacy attitudes.
As shown in Figure 2, client importance perceptions had a marginally significant 共p ⬍ 0.10兲
influence on the weighting of evidence, and client-specific advocacy was significantly related 共p ⬍
0.01兲 to both weighting of evidence and the tax professionals’ likelihood assessments of favorable
IRS allowance of pro-taxpayer treatment. Thus, to the extent that tax professionals advocate for a
specific client, they are more likely to view a pro-taxpayer position for that client as more sus-


TABLE 4
Analysis of Client-Specific Advocacya
Panel A: Marginal Means (Standard Error) of Client-Specific Advocacy, Evaluated at General Advocacy Covariate
Low Importance High Importance Row Means (Risk)
Low Risk 44.38 42.65 43.50
共1.06兲共1.06兲共0.75兲
n ⫽ 21 n ⫽ 21
High Risk 41.41 41.49 41.49
共1.09兲共0.95兲共0.72兲
n ⫽ 20 n ⫽ 26
Column Means 42.88 42.07 42.43
共Importance兲共0.75兲共0.70兲共0.52兲
Panel B: Analysis of Covariance (Using Client Risk/Importance Manipulations)

Significance Level
F-statistic (Two-Tailed)
Client Risk 3.946 .050
Client Importance 0.619 .434
Client Risk ⴱ Importance 0.720 .399
Interaction
General Advocacy 223.975 ⬍.001
Model Statistics
F-statistic ⫽ 57.859
Significance Level ⫽⬍ .001
Adjusted R2 = .723
a Includes the responses for participants who answered all items on both the general advocacy and client-specific advocacy scales. The client-specific advocacy and general advocacy
measures are explained in Table 3.
Spring 2010
43
TABLE 5
Tax Professionals’ Decision Processes
Factor
Likelihooda Weightinga
Constant 1.884 6.161***
共1.683兲共.914兲
Client Risk Perceptions .003 .108

共.193兲共.105兲
Client Importance .138 .179**

Perceptions 共.156兲共.085兲
Client-Specific Advocacy .073** ⫺.042**

共.030兲共.016兲
Adjusted R2 .050 .080

F-statistic 2.522* 3.517**
*, **, *** p ⬍ .10, p ⬍ .05, and p ⬍ .01, respectively, all tests two-tailed.
a Coefficients 共Standard Errors兲 for OLS Regression Models, by Dependent Variable.
“Likelihood” is an 11-point scale that pertains to the likelihood the IRS would consider the activity to be a business, and
“Factor Weighting” is a seven-point scale that refers to tax professionals’ weighting of the relevance of personal pleasure
in determining the appropriate tax treatment. Client risk perceptions are measured on a six-point scale; client importance
perceptions are measured on a seven-point scale, and client-specific advocacy is measured on a 共minimum nine-point兲
maximum 63-point scale.
tainable. There is no normative reason that advocacy should affect this judgment process. The fact
that tax professionals with higher levels of client-specific advocacy view favorable IRS treatment
as more likely indicates a bias in judgment not explained by pure utility-maximizing behavior.12
The path analysis results are also confirmed by regression analysis, reported in Table 5. The
first column addresses whether client characteristics and client-specific advocacy explain the like-
lihood assessments of favorable IRS allowance of pro-taxpayer tax treatment. Client characteris-
tics are not related to tax professionals’ likelihood assessments; however, client-specific advocacy
is statistically significant in explaining these judgments 共p ⬍ 0.05兲.
Results for weighting of evidence are shown in the second column of Table 5. Consistent with
H4, client-specific advocacy influences tax professionals’ weighting of evidence, as tax profes-
sionals with higher degrees of client-specific advocacy rated this factor as less relevant 共p ⬍ 0.05兲.
12
Alternative analysis used participants’ “general advocacy” scores, rather than their measures of client-specific advocacy;
results demonstrated that general advocacy is not related to tax professionals’ likelihood assessments or weighting of
evidence 共both p ⬎ .40兲. This finding is consistent with Barrick et al. 共2004兲, which did not find that general advocacy
attitudes explained tax professionals’ probability assessments of favorable judicial outcomes for client scenarios.

This further suggests that advocacy can bias tax professionals’ judgment processes.13 Furthermore,
although perceptions of client risk were not related to participants’ weighting of evidence, the
results show that client importance was significantly related to this decision process variable.14
Overall, H3 is partially supported, whereas H4 is fully supported. Regarding H3, client char-
acteristics are not directly related to IRS likelihood assessments, but tax professionals do weight
applicable evidence as more relevant for clients that are perceived to be more important. The
prediction of H4 that client-specific advocacy influences tax professionals’ decision processes is
supported for both the likelihood judgments and weighting of evidence. Taken together, these
findings suggest that tax professionals may not be objective in their evaluation of facts and the law
and may suffer from biased judgments.
Hypotheses 5 and 6
Hypotheses 5 and 6 address whether client characteristics and/or client-specific advocacy
affect the outcomes of tax professionals’ decision processes 共specifically, the recommendation and
allowance of favorable tax advice兲. In addition to the decision process variables 共IRS likelihood
assessments and weighting of evidence兲, client-specific advocacy was related to both tax profes-
sionals’ recommendations 共p ⬍ 0.10兲 and their allowance of favorable tax treatment 共p ⬍.05兲. In
addition, client importance had a significant direct effect 共p ⬍ 0.01兲 on respondents’ allowance of
favorable tax treatment, but was not directly related to their recommendations. Since perceived
importance does not affect tax professionals’ recommendations, the fact that it is related to allow-
ance of beneficial tax treatment suggests that important clients may have additional leverage over
tax professionals that their less important peers lack.
Table 6 confirms these results with regression analysis. Regression analysis shows that the
elements of the judgment process we included in the experiment—weighting of evidence and IRS
likelihood assessments—were important factors in respondents’ decision-making, and, overall the
variables in the regression model explain a large percentage of the variance 共adjusted R2 of 0.637
for recommendation and 0.721 for allowance兲. Consistent with the path analysis results, client-
specific advocacy is also related to both the recommendation and allowance of favorable treat-
ment, and client importance perceptions are related to allowance of favorable tax treatment.
To summarize, H5 and H6 predict that client characteristics and client-specific advocacy will
both influence the final outcomes of tax professionals’ decision-making. The predicted relationship
in H5 between client characteristics and decision outcomes is partially supported, as tax profes-
sionals are more likely to allow favorable tax treatment for clients perceived as more important
共corroborating Reckers et al. 关1991兴兲. Furthermore, the prediction of H6 that client-specific advo-
cacy will influence these outcomes is supported for both the recommendation and allowance of
favorable tax advice.15
13
Additional analyses demonstrated that client-specific advocacy is related to four additional factors 共all p ⬍ .05兲: the
manner in which the taxpayer carries out the activity, the expertise of the taxpayer or his advisors, the time and effort
expended by the taxpayer in carrying on the activity, and the expectation that assets used in the activity may appreciate
in value. However, none of these other factor weightings were significantly related to the participants’ decisions.
Additional analysis substituted participants’ general advocacy scores in place of the measure of client-specific advocacy;
while general advocacy was not related to the factor weighting used in the primary analysis 共element of personal
pleasure兲, it was related to two of the other factors: the time and effort expended by the taxpayer in carrying on the
activity and the expectation that assets used in the activity may appreciate in value.
14
Since these two dependent variable measures of tax professionals’ decision-making processes may be related, additional
analysis uses multivariate analysis of variance to examine the hypothesized relationships. Results are inferentially
identical.
15
The dependent variables for judgment processes and outcome were typically measured on a seven-point Likert-type
scale 共however, the IRS likelihood assessments were measured on an 11-point Likert-type scale兲. Results using ordinal
共rather than OLS兲 regression were inferentially identical.

TABLE 6
Tax Professionals’ Decision Outcomes
Recommendationa Allowancea
Constant 1.062 .670
共0.929兲共.870兲
IRS Likelihood Assessment .496*** .529***

共.048兲共.045兲
Factor Weighting ⫺.234*** ⫺.252***

共.089兲共.083兲
Client Risk Perceptions .039 ⫺.052

共.084兲共.079兲
Client Importance Perceptions .010 .151**

共.070兲共.066兲
Client-Specific Advocacy .023* .031**

共.014兲共.013兲
Adjusted R2 .637 .721

F-statistic 31.547*** 45.899***
*, **, *** p ⬍ .10, p ⬍ .05, and p ⬍ .01, respectively, all tests two-tailed.
a Coefficients 共Standard Errors兲 for OLS Regression Models, by Dependent Variable.
“Recommendation” is a seven-point scale that refers to which tax treatment tax professionals would advise the client
共hobby or business activity兲, and “Allowance” is a seven-point scale that refers to which tax treatment tax professionals
would allow their clients to take. Client risk perceptions are measured on a six-point scale; client importance perceptions
are measured on a seven-point scale, and client-specific advocacy is measured on a 共minimum nine-point兲 maximum
63-point scale.
Additional Analyses
Several additional analyses tested the robustness of the study’s results. First, we investigated
whether all of the advocacy scale items formed a unidimensional construct, as reported by Mason
and Levy 共2001兲. Principal components analysis using varimax rotation of the nine items on the
client-specific advocacy scale indicated that three separate factors emerged. The “primary factor”
explaining 48 percent of the variance in the responses, consisted of seven of the nine scale items;
Question 2 共Generally speaking, my loyalties are first to the tax system, then to the taxpayer兲 and
Question 4 共When examining a tax position, I tend to point out to taxpayers reasonable positions
they could have taken which would have contributed to minimizing their tax liability兲 each loaded
on a separate factor and explained approximately 13 percent and 12 percent of the variance,
respectively. Questions loading on the primary factor generally pertain to tax professionals’ ex-
ploitation of ambiguity.
Supplemental analysis investigates the relationships hypothesized in H1 and H2 using scales
comprised of the item共s兲 loading on the three factors.16 Results indicate that the relationship
16
Cronbach’s alpha for the scale formed with the seven items loading on the “primary” factor was 0.884.

between risk perceptions and client-specific advocacy was driven primarily by tax professionals’
tendencies to exploit ambiguity. Interestingly, while client importance did not influence client-
specific advocacy when examining the scale in total, perceptions of client importance were linked
to the tendency to encourage tax minimization 共Question 4兲, with tax professionals acting more
proactively for more important clients 共p ⬍ 0.06兲. This suggests that differing client characteristics
affect different elements of client-specific advocacy.
The relationships between client-specific advocacy and elements of the decision process 共the
weighting of evidence and the IRS likelihood assessments兲 are explained by tax professionals’
propensities to exploit ambiguity 共p ⬍ 0.05兲, but not by the other two items. Interestingly, the
relationship is flipped when examining the effect of client-specific advocacy items on decision
outcomes. Both items that loaded on separate factors 共loyalty to the taxpayer and encouragement
of tax minimization兲 are related to tax professionals’ allowance of favorable tax treatment, where
tax professionals that have greater loyalty to the taxpayer than to the tax system are more willing
to allow the taxpayer preferred position. These findings help demonstrate the root causes of the
effects of client-specific advocacy on different elements of judgment and decision-making.17
Second, we examined whether there were any differences in the results based on the demo-
graphics of the participants. None of the demographic variables had a statistically significant
influence on the study’s findings, with the exception of gender. When controlling for client char-
acteristics, client-specific advocacy, and elements of the decision process, female tax professionals
were much less likely to recommend a favorable tax position than their male colleagues were 共p ⬍
0.01兲, and were somewhat less likely to allow a client to take advantage of favorable tax treatment
共p ⬍ 0.10兲.18 The difference between male and female tax professionals in their decision outcomes
may be due to gender differences in risk aversion. Prior findings on gender differences in tax
professionals’ judgment and decision-making have pointed to a discrepancy in risk-taking behav-
ior 共e.g., McGill 1988; Sanders and Wyndelts 1989; Roberts 1998; Kelliher et al. 2001兲, as females
tend to be more risk-averse than males 共see Byrnes et al. 1999兲.
Limitations
The results of the study should be interpreted in light of its limitations. First, as with any
hypothetical case scenario, participants’ responses about their intended behavior may not match
with their behavior in reality. However, behavioral intentions are theoretically linked to behavior
when an individual perceives some degree of control in a situation 共Ajzen 1991兲. Another limita-
tion is the study’s use of only one case, raising concern that the findings may have been case-
specific. To alleviate this concern, we relied on a common ambiguous tax dilemma 共Krawczyk
1994; Carnes et al. 1996兲. Finally, there may have been potential demand effects, since the study’s
instrument measured two sets of similar questions 共general advocacy and client-specific advo-
cacy兲. However, we ensured that sufficient time elapsed between the completion of these two
scales by placing the questions measuring general and client-specific advocacy at the beginning
and end, respectively, of the study’s experiment. We did not randomize the order of these measures
since responses to the client-specific advocacy scale had to be collected after the participants were
exposed to the client description.
17
We also tested whether client-specific advocacy mediated the relationship between client risk/client importance and the
decision outcome variables. When separately examining client risk and controlling for the processing variables, there
was a significant relationship between client risk and the allowance of favorable tax treatment that was mediated by
client-specific advocacy. However, client-specific advocacy did not mediate the relationship between client importance
and the outcome variables.
18
Interestingly, when controlling for gender and the processing variables, client-specific advocacy no longer affects
recommendations, but still affects allowance 共p ⬍ .05兲.

CONCLUSION
Overall, the study’s results provide evidence that client characteristics influence tax profes-
sionals’ advocacy attitudes, and that client-specific advocacy influences a number of steps in the
judgment and decision-making process. The study’s results were tested using path analysis to
simultaneously consider the direct and indirect relationships among the constructs. Hypotheses 1
and 2 examine whether client characteristics influence the level of advocacy a tax professional
exhibits for a particular client. Results indicate that tax professionals exhibit lower levels of
client-specific advocacy for riskier clients 共consistent with Kadous and Magro 关2001兴兲. Supple-
mental analysis reveals that tax professionals are also more likely to act as advocates by high-
lighting tax-minimizing positions to clients perceived as more important.
Hypotheses 3 and 4 investigate how client characteristics and client-specific advocacy affect
tax professionals’ judgment processes. From a normative perspective, these factors should not be
related to the objective analysis of the underlying facts and law. However, our results show that
client importance is related to the weighting of evidence and client-specific advocacy is related to
both the weighting of evidence and the likelihood judgments of favorable IRS treatment. This
indicates that tax professionals may be unintentionally influenced by client attributes when making
judgments and may make biased judgments when they weight evidence and analyze the applicable
legal authority. These results are in line with prior research, which has consistently shown that tax
professionals are subject to a variety of cognitive biases related to their role as a client advocate
共e.g., Johnson 1993; Davis and Mason 2003; Barrick et al. 2004兲. These biased judgments are
particularly problematic in light of tax professionals’ increased responsibility under FIN No. 48 to
objectively evaluate uncertain tax positions.
Finally, H5 and H6 address the influence of client characteristics and client-specific advocacy
on tax professionals’ decision outcomes; results show that client-specific advocacy affects both the
recommendation and allowance of favorable tax advice, although its influence on the recommen-
dation is somewhat weaker. Client importance is related only to the allowance of favorable tax
advice. In all, the findings point to the importance of client-specific advocacy in multiple steps of
the judgment and decision-making process and echo concerns identified by prior research regard-
ing tax professionals’ objectivity. Our study comprehensively examines the influence of client-
specific advocacy on a number of decision process and outcome variables. The fact that prior
studies have found somewhat mixed results on the role of advocacy attitudes may be due to their
failure to consider how client characteristics could affect these attitudes 共e.g., Davis and Mason
2003; Barrick et al. 2004兲.
In conclusion, this study contributes to the behavioral taxation literature by providing pre-
liminary support that client characteristics influence advocacy attitudes, and that client-specific
advocacy influences multiple steps in tax professionals’ judgment and decision-making processes.
These results provide further evidence to the growing body of literature suggesting that tax pro-
fessionals may have difficulty separating their advocacy and objective evidence evaluator roles
共e.g., Kahle and White 2004; Kadous et al. 2008兲. Implications for staff training programs and tax
education include a need to focus on objectivity when evaluating facts and applying the law in
order to promote effective decision-making.
APPENDIX A
MANIPULATION OF CLIENT FACTS
Low Risk/Low Importance Condition
Jim and Jane Roe are high-income taxpayers. Jim is a local bank president and Jane is a
dietician at an area hospital. The bank is not a client of your firm. The Roes are not considered
important clients because your firm’s only relationship with them has been preparing their per-

sonal tax return for the last two years, and they have not been a source of referrals. Prior to
engaging your firm, IRS audits of the Roes’ returns have not led to any adjustments or penalties,
and their financial records are reliable and complete. They have always avoided questionable
tax-saving strategies.
Low Risk/High Importance

dietician at an area hospital. Jim’s bank is one of your office’s largest clients. The Roes are
considered important clients due to the bank audit fees, your firm’s preparation of the Roes’
personal tax returns for a number of years, and because they have been a source of past referrals.
Previous IRS audits of the Roes’ returns have not led to any adjustments or penalties, and their
financial records are reliable and complete. They have always avoided questionable tax-saving
strategies.
High Risk/Low Importance

dietician at an area hospital. The bank is not a client of your firm. The Roes are not considered
important clients because your firm’s only relationship with them has been preparing their per-
sonal tax return for the last two years, and they have not been a source of referrals. Prior to
engaging your firm, IRS audits of the Roes’ returns have led to sizable adjustments and penalties.
You have also found that their financial records are sometimes unreliable and incomplete. In your
dealings with them, they have proposed various questionable tax-saving strategies.
High Risk/High Importance

dietician at an area hospital. Jim’s bank is one of your office’s largest clients. The Roes are
considered important clients due to the bank audit fees, your firm’s preparation of the Roes’
personal tax returns for a number of years, and because they have been a source of past referrals.
Prior to engaging your firm, IRS audits of the Roes’ returns had led to sizable adjustments and
penalties. You have also found that their financial records are sometimes unreliable and incom-
plete. In your dealings with them, they have proposed various questionable tax-saving strategies.
APPENDIX B
HYPOTHETICAL CLIENT VIGNETTE
On June 15, 2007, Jim and Jane Roe came to you requesting your advice on the following tax
issue:
Since the beginning of 2005, Jane has become increasingly active in off-road bicycle racing.
Since the beginning of 2005 she has participated in approximately 60 races. The National Off-
Road Bicycle Association has recently ranked Jane in both the dual slalom and downhill races.
This ranking has led to some sponsorships that have provided her with bicycles, clothing and,
occasionally, entry fees. Although she has won a few of the races she has entered, her annual
expenses related to racing 共e.g., bicycle repairs, transportation, lodging, and entry fees兲 far exceed
her racing income.
Jane wants to make racing a business. Jim and Jane tell you that she trains daily. During the
racing season she works lighter hours to allow her to make the most of the racing events. Jane has
kept records of all of her earnings and expenses for the last three years 共including 2007兲. For 2005
and 2006 her winnings were $990 and $2,180, respectively. Her expenses for those years were
$11,200 and $16,700, respectively. With the 2007 season halfway over, she estimates that her

earnings will be around $4,000 while her expenses will be around $17,000. Cash prizes run as
high as $8,000 or $9,000 and she believes that her earnings will eventually surpass her expenses.
The issue on which you must decide is whether Jane can treat her bicycle racing activity as a
business 共making the loss resulting from the activity in 2007 deductible on their tax return兲 or
whether the activity should be treated as a hobby 共limiting their ability to deduct expenses asso-
ciated with this activity兲. They would also like to file amended returns for 2005 and 2006 so they
can use the losses to offset income from those periods.
Treasury Regulation §1.183-2 provides guidance regarding when an activity is considered to
be carried on for profit 共deductions allowed under sections 162/212兲. On the following screen are
excerpts from this regulation to aid you with your decision. You should assume that there are no
reported tax cases that deal specifically with bicycle racing and hobby loss rules, or any other
relevant authority other than this treasury regulation.
REFERENCES
Ajzen, I. 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes 50
共2兲: 179–211.
——–, and M. Fishbein. 1980. Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs,
NJ: Prentice Hall, Inc.
American Institute of Certified Public Accountants. 共AICPA兲. 2000. Statements on Standards for Tax Services
Nos. 1–8. New York, NY: AICPA.
Andre, S. 2008. Contributing factors of confirmatory information searches and evaluations. Working paper,
University of Wisconsin. Available at: http://ssrn.com/abstract⫽1093370.
Ayres, F. L., B. R. Jackson, and P. S. Hite. 1989. The economic benefits of regulation: Evidence from
professional tax preparers. The Accounting Review 44 共2兲: 300–312.
Bandy, D. 1996. Limiting tax practice liability. The CPA Journal 共May兲: 46–49.
Barrick, J. A., C. B. Cloyd, and B. C. Spilker. 2004. The influence of biased tax memoranda on supervisors’
initial judgments in the review process. The Journal of the American Taxation Association 26 共1兲:
1–21.
Bentler, P. M., and D. G. Bonnett. 1980. Significance tests and goodness of fit in the analysis of covariance
structures. Psychological Bulletin 88: 588–606.
Bobek, D. D., R. C. Hatfield, and S. S. Kramer. 2004. Professional liability suits against tax accountants:
Some empirical evidence regarding case merit. Advances in Taxation 16: 3–23.
Byrnes, J., D. Miller, and W. Schafer. 1999. Gender differences in risk taking: A meta-analysis. Psychological
Bulletin 125: 367–383.
Carnes, G., G. Harwood, and R. Sawyers. 1996. The determinants of tax professionals’ aggressiveness in
ambiguous situations. Advances in Taxation 8: 1–26.
Cloyd, C. B., and B. C. Spilker. 1999. The influence of client preferences on tax professionals’ search for
judicial precedents, subsequent judgments and recommendations. The Accounting Review 74 共3兲:
299–321.
Cuccia, A. D. 1994. The effects of increased sanctions on paid tax preparers: Integrating economic and
psychological factors. The Journal of the American Taxation Association 16 共1兲: 41–57.
——–, K. Hackenbrack, and M. W. Nelson. 1995. The ability of professional standards to mitigate aggressive
tax reporting. The Accounting Review 70 共2兲: 227–249.
——–, and G. McGill. 2000. The role of decision strategies in understanding professionals’ susceptibility to
judgment biases. Journal of Accounting Research 38 共2兲: 419–435.
Davis, J. S., and J. D. Mason. 2003. Similarity and precedent in tax authority judgment. The Journal of the
American Taxation Association 25 共1兲: 53–71.
Duncan, W. A., D. W. LaRue, and P. M. J. Reckers. 1989. An empirical examination of the influence of
selected economic and noneconomic variables in decision making by tax professionals. Advances in
Taxation 2: 91–106.

Fiore, N. 1998. From the tax adviser: Minimizing risks in taking on new clients. Journal of Accountancy 185
共2兲: 36.
Hair, J., B. Black, B. Babin, R. E. Anderson, and R. L. Tatham. 1998. Multivariate Data Analysis. Upper
Saddle River, NJ: Prentice Hall, Inc.
Hatfield, R. 2000. The effect of accountability on the evaluation of evidence: A tax setting. Advances in
Taxation 12: 105–125.
——–. 2001. The effects of staff accountant objectivity in the review and decision process: A tax setting. The
Journal of the American Taxation Association 23 共1兲: 61–74.
Helleloid, R. 1989. Ambiguity and the evaluation of client documentation by tax professionals. The Journal
of the American Taxation Association 11 共Fall兲: 22–36.
Johnson, L. 1993. An empirical investigation of the effects of advocacy on preparers’ evaluation of judicial
evidence. The Journal of the American Taxation Association 15 共1兲: 1–22.
Kadous, K., and A. Magro. 2001. The effects of exposure of practice risk on tax professionals’ judgments and
recommendation. Contemporary Accounting Research 18 共3兲: 451–475.
——–, ——–, and B. Spilker. 2008. Do effects of client preference on accounting professionals’ information
search persist with high practice risk? The Accounting Review 83 共1兲: 133–156.
Kahan, S. 1996. Fire that client! The Practical Accountant 共October兲: 46–53.
Kahle, J., and R. White. 2004. Tax professional decision biases: The effects of initial beliefs and client
preference. The Journal of the American Taxation Association 26 共1兲: 1–29.
Kelliher, C., D. Bandy, and A. Judd. 2001. An examination of the “realistic possibility” standard’s influence
on tax practitioner aggressiveness. Advances in Taxation 13: 95–121.
Krawczyk, K. 1994. The influence of tax law requirements and organization of client facts on professional
judgment. Behavioral Research in Accounting 6 共Supplement兲: 97–120.
LaRue, D., and P. Reckers. 1989. An empirical examination of the influences of selected factors on profes-
sional tax preparers’ decision process. Advances in Accounting 7: 37–50.
Levy, L. 1996. Evidence evaluation and aggressive reporting in ambiguous tax situations. Ph.D. dissertation,
University of Colorado.
Mason, J., and L. Levy. 2001. The use of the latent constructs method in behavioral accounting research: The
measurement of client advocacy. Advances in Taxation 13: 123–139.
McGill, G. 1988. The CPA’s role in income tax compliance: An empirical study of variability in recommend-
ing aggressive tax positions. Ph.D. dissertation, Texas Tech University.
Milliron, V. 1988. A conceptual model of factors influencing tax preparers’ aggressiveness. In Contemporary
Tax Research, edited by S. Moriarity, and J. Collins. Norman, OK: The Center for Economic and
Management Research, the University of Oklahoma.
Pascarella, S. 1996. Client screening: How to reduce malpractice exposure. The Tax Advisor 共March兲: 179–
180.
Pinsker, R., R. Pennington, and J. K. Schafer. 2009. The influence of roles, advocacy, and adaption to the
accounting decision environment. Behavioral Research in Accounting 21 共2兲: 91–111.
Reckers, P., D. Sanders, and R. Wyndelts. 1991. An empirical investigation of factors influencing tax prac-
titioner compliance. The Journal of the American Taxation Association 13 共2兲: 30–46.
Roberts, M. 1998. Tax accountants’ judgment/decision-making research: A review and synthesis. The Journal
of the American Taxation Association 20 共1兲: 78–121.
Sanders, D. L., and R. W. Wyndelts. 1989. An examination of tax practitioners’ decisions under uncertainty.
Advances in Taxation 2: 41–72.
Schumacker, R. E., and R. G. Lomax. 2004. A Beginner’s Guide to Structural Equation Modeling. Mahwah,
NJ: Lawrence Erlbaum Associates.
Schwartz, J. 1993. Avoiding the treasury stigma of a hobby. The CPA Journal 63 共5兲: 72–74.
Spilker, B. C., R. G. Worsham, and D. F. Prawitt. 1999. Tax professionals’ interpretations of ambiguity in
compliance and planning decision contexts. The Journal of the American Taxation Association 21 共2兲:
75–89.
Stephenson, T. 2007. Do clients share preparers’ self-assessment of the extent to which they advocate for
their clients? Accounting Horizons 21 共4兲: 411–422.

Summaries of Papers in This Issue 81
The Role of Client Advocacy in the Development

of Tax Professionals’Advice
Donna D. Bobek, Amy M. Hageman, and Richard C. Hatfield
One of the primary roles of a tax professional is to function as a client advocate. In executing
their role as client advocates, however, tax professionals must still “objectively evaluate all rel-
evant facts and tax authorities when preparing advice” 共Kadous and Magro 2001, 453兲. Failing to
accurately and objectively assess the underlying facts and the relevant law can result in taxpayer
and tax professional penalties as well as negative outcomes for tax professionals’ firms.
judgment and decision-making for over two decades. Prior studies have recognized the importance
of advocacy and its effect on tax professional judgment, although some of this research has failed
to establish a direct connection between measured general advocacy levels and tax professionals’
recommendations. These prior studies have not, however, explicitly considered how advocacy may
be context-specific, or how, as an attitude, it may change based on the external circumstances.
Given that advocacy can influence tax professionals’ cognitive processes 共e.g., evidence search
and evaluation兲 the variables that affect advocacy have the potential to unintentionally affect their
judgments and decisions.
The purpose of this study is twofold. First, we investigate whether client characteristics 共e.g.,
client risk and client importance兲 influence the degree to which tax professionals will function as
advocates for their clients. Second, we consider how this client-specific advocacy affects the
manner in which tax professionals make judgments and decisions. To examine these issues, we
conduct an experiment with 101 tax professionals, where we manipulate client characteristics and
measure general and client-specific advocacy. We then identify the extent to which advocacy and
client characteristics affect tax professionals’ decision processes 共weighting of evidence and IRS
likelihood assessments兲 and decision outcomes 共recommendation and allowance of favorable tax
positions兲.
The results of this study provide evidence that some client characteristics affect the advocacy
attitudes of tax professionals, and that client-specific advocacy influences tax professionals’ deci-
sion processes and outcomes. We used path analysis to simultaneously control for the direct and
indirect relationships among the study’s constructs. ANCOVA and regression analysis supple-
mented and confirmed the path analysis findings. Specifically, tax professionals in our study
displayed lower levels of client-specific advocacy for riskier clients. Furthermore, client-specific
advocacy influenced elements of tax professionals’ decision processes as well as their recommen-
dations. This suggests that tax professionals’ judgments may be unintentionally influenced by their
advocacy attitudes, as well as client attributes, which act as antecedents to these attitudes. Overall,
the results indicate that client-specific advocacy influences multiple steps in tax professionals’
judgment and decision-making processes.
These results contribute to the taxation literature in several ways. First, this study extends
research on advocacy by examining its influence on a broader range of process and outcome
variables than have been considered in past research. Second, the study introduces the construct of
client-specific advocacy. As the results indicate that client characteristics influence tax profession-
als’ advocacy attitudes toward a particular client, future research should consider that advocacy
itself is a function of client- and context-specific characteristics. Third, despite the fact that tax
professionals should objectively evaluate evidence and the law, the study’s findings add to the
growing body of literature suggesting that tax professionals may have difficulty separating their
advocacy and objective evaluator roles.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
View publication stats

BobekHagemanHatfieldJATA2010 PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

BobekHagemanHatfieldJATA2010 PDF

Hochgeladen von

Copyright:

Verfügbare Formate

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

The Role of Client Advocacy in the Development of Tax Professionals’ Advice

Article in Journal of the American Taxation Association · August 2009

Donna D. Bobek Amy M. Hageman

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

The Role of Client Advocacy in the

ABSTRACT: A primary responsibility of tax professionals is to be an advocate for their

Keywords: client advocacy; tax advice; client risk; client importance.

Data Availability: Data are available from the first author.

Submitted: September 2008

Journal of the American Taxation Association Spring 2010

PRIOR RESEARCH AND HYPOTHESES

Journal of the American Taxation Association Spring 2010

Bobek, Hageman, and Hatfield

(continued on next page)

Client Advocacy in the Development of Tax Professionals’Advice

2000 accountants client-favored 共confirmation dependent upon a preference

(continued on next page)

Panel B: Direct Measurement of Advocacy Attitudes

Bobek, Hageman, and Hatfield

(continued on next page)

Client Advocacy in the Development of Tax Professionals’Advice

(continued on next page)

Bobek, Hageman, and Hatfield

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Experimental Task and Procedures

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Question 5: I believe it is important that I encourage the 4.98*** 4.48***

Question 6: I always interpret unclear/ambiguous laws in favor 4.26 4.24

Question 7: It is important to use trends in the law by trying to 4.96 5.04

Question 8: Where no judicial authority exists with respect to 5.10* 4.93*

Overall 43.43** 42.43**

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Client Advocacy in the Development of Tax Professionals’Advice

Panel B: Analysis of Covariance (Using Client Risk/Importance Manipulations)

Client Risk Perceptions .003 .108

Client Importance .138 .179**

Client-Specific Advocacy .073** ⫺.042**

Adjusted R2 .050 .080

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

IRS Likelihood Assessment .496*** .529***

Factor Weighting ⫺.234*** ⫺.252***

Client Risk Perceptions .039 ⫺.052

Client Importance Perceptions .010 .151**

Client-Specific Advocacy .023* .031**

Adjusted R2 .637 .721

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Low Risk/High Importance

High Risk/Low Importance

High Risk/High Importance

Journal of the American Taxation Association Spring 2010

Journal of the American Taxation Association Spring 2010

Question 5: I believe it is important that I encourage the 4.98* 4.48*

Overall 43.43 42.43

Client-Specific Advocacy .073 ⫺.042

IRS Likelihood Assessment .496* .529*

Factor Weighting ⫺.234* ⫺.252*