What Are the Behavioral Options That Would Create Change

Document Sample
What Are the Behavioral Options That Would Create Change Powered By Docstoc
					                            Narrowing the Tax Gap: Behavioral Options

                                         Joshua D. Rosenberg 1    TF   FT

1. Introduction: The Cycle of Human Interactions
All human behavior, including paying taxes and evading taxes, is determined not simply
by rational decision-making, but also by unconscious learned behaviors, emotions,
perceptual and cognitive frameworks and schemas, and by self-fulfilling prophecies.
Human behavior is systemic in nature: our thoughts, feelings, perceptions, cognitive
frameworks, and behavior influence, and are influenced by, each other.
               To elaborate just a bit on the workings of human behavior, most people would
acknowledge that not all human behavior is entirely rational all the time. Few would
doubt that emotions can at least at times significantly impact our thoughts--both what we
think about and the way we think about it. Perhaps the most obvious example of this
phenomenon is the behavior of anyone in the throes of love (or of the hatred that, for
some, tends to follow the disillusionment of lost love). Such a person may find it
difficult to think about things other than the object of her intense feelings; and the content
of those thoughts will depend entirely on which 'throes' the person is in (that is, "in love,"
or in "hate") at the time. A person who thinks certain behaviors are cute while she is in
love may well start to think of those same behaviors as offensive once that love turns to
               Emotions not only affect not only the content of thoughts, but also the quality of
thinking. Strong feelings can prevent us from having cognitive access to otherwise
available information and memories. During, and immediately following, the grip of
emotions, "thinking cannot incorporate information that does not fit, maintain or justify
the emotion” being experienced. 2 Rather than functioning as it otherwise would, the
                                         TPF   FPT

T1 Professor of Law, University of San Francisco School of Law.

TPPaul Ekman, Emotions Revealed 39, 63 (2003). See also Memory for Everyday and Emotional Events

(Nancy L. Stein et al. eds., 1997).

emotionally triggered brain retrieves and focuses only on memories and thoughts related
to and reinforcing of the emotion being experienced. 3             PF   FP

               Of course, most taxpayers are not dealing with either current or former lovers
while preparing tax returns. Nonetheless, while many people may be unaware of
experiencing any emotions at all until the emotions reach a level that makes them
impossible to ignore, even low-level emotions that escape conscious awareness can
impact thinking.
               In addition to impacting both the way we think and what we think about,
emotions can essentially bypass the conscious thought process and act directly on our
behavior. A person who feels guilty about something he did (or failed to do) may well
act quite differently from how he might otherwise respond to an identical situation; and
the difference in response is likely not simply the result of a conscious cost-benefit
analysis that takes into account the otherwise nonexistent benefit of assuaging guilt.
Similarly, the person who feels particularly empathetic to a particular person or situation
will likely react based at least in part in response to that emotion, and not simply because
he factors that emotion into a rational cost-benefit analysis, but because the emotional
reaction itself directly affects behavior. 4  TPF   FPTP   P

               Indeed, it is not even just how we think and act that is affected by emotions. Our
very perceptions themselves are determined, in part, by current emotional experiences.
Emotions precipitate changes in the autonomic nervous system. 5 At a micro level, these
                                                                             PF   F   P

changes in the autonomic nervous system change not only the ability to think and act, but
also the ability to perceive. Both the focus of our attention and our ability to take in data

TP "[W]e evaluate what is happening in a way that is consistent with the emotion we are feeling, thus

justifying and maintaining the emotion. Expectations are formed, judgments made, that typically serve to
maintain rather than diminish the felt emotion." Ekman, supra note 30, at 63; R.W. Levenson et al.,
Emotion and the Autonomic Nervous System Activity in the Minangkabau of West Sumatra, 62 J.
Personality & Soc. Psych. 972-78 (1992). 'That awareness--of how our emotions affect what we are doing--
is the fundamental emotional competence. Lacking that ability, we are vulnerable ... to being sidetracked
by emotions run amok..' Daniel Goleman, Working with Emotional Intelligence 55 (1998).
4 See generally Joshua D. Rosenberg, Interpersonal Dynamics, 58 Miami L. Rev. 1225 (2004).
T          T

TP These changes include increasing the heart rate, changing breathing patterns, skin changes such as

perspiration or blushing, and redirecting blood flow (anger has been found to direct blood to the hands,
presumably for combat; fear has been shown to redirect blood to the legs, presumably for running). Ekman,
supra, at 68.

are altered by emotional states. Perceptions become differently focused as a result of
emotions, and they may also become less accurate. 6           TPF   FPT

              Of course, emotions only supplement, and do not replace, the role of thinking.
Obviously our strategic thinking impacts our behavior, but “thinking” takes into account
more than just strategic decision-making, and as such, it does much more than simply
provide a logical basis for action. Thinking, as much as emotional states, impacts not
only how we process what we perceive, but also what it is that we perceive in the first
place. We all have not only specific rational thoughts at any given time, but also more
general "frames of reference"--ways we tend to understand the world. These frames of
reference impact on us constantly, determining, in part, both how we see things and what
things we see (or pay attention to). 7    PF   FP

              In addition to the fact that both our perceptions and behavior are impacted by
emotions and by frames of reference, the extent to which the exact opposite is also true is
worth noting. What we do significantly impacts how we think and what we think about, 8                      PF   FP

6 Ibid.
T         T

TPE.g. Jackson, S. and R. Hatfield. 2005. A note on the relation between frames, perceptions and taxpayer

behavior. Contemporary Accounting Research 22: 145-164.
TPIn order to make this point to students, I have conducted in some classes a simple experiment in which

some students are chosen for each of two groups, and the remaining students (who have previously, and
secretly, been instructed on how to act) are designated as 'observers' of each group. Each group, with its
designated 'observers,' is sent to a different room and asked to toss pennies to see how close they can get
them to the wall. One group's tossing is met with complete silence by its 'observers,' who pretend to busy
themselves with note taking. The other group receives constant praise from its 'observers' (for example, for
their ability to get pennies close to the wall, for their good form, etc.). Not surprisingly, the second group
invariably continues long after the first group stops.
          When asked why they stopped, the first group typically replies that they had other things to do
(such as 'reviewing' the reading they were supposed to have done for that day's class). When asked why
they continued for as long as they did, the second group typically responds with statements such as 'it
reminds me of when I was a kid, so it brings back fond memories,' or 'it was fun,' or simply "you told us
to." None in the second group responds that she continued because she was getting cheered on by others,
and none in the first group suggests or believes that she stopped because her 'observers' were silent. Each
group thought differently about the tossing they had done, and each individual had very reasonable and
logical thoughts about why she did what she did. In each case, however, these thoughts were the result of
their behavior, rather than its cause.
          Equally as important as the fact that their actions drove their reasoning is the effect of those
actions on the general topics about which they thought during the experiment. The group that continued
tossing pennies was thinking about how to get the pennies close to the wall--how to improve their outcomes
and their technique. The other group was thinking about the content they were reading (after they quickly
stopped either throwing the dumb pennies against the wall or thinking about the pennies). While each
group might have been thinking very rationally, what they were thinking about was determined not by
rational choice, but by their behavior (which, in turn, was determined by the actions of their 'observers').
           Students have pointed out that penny-throwing was more enjoyable for the group that had the
active and engaged observers than it was for the group that had silent note-takers, so that the decision of

how we feel, and what and how we perceive. Research suggests that, contrary to rational
expectations, people act out of habit or other conditioning, and then rationalize their
actions afterwards, just as frequently as they think rationally and then act as a result of
that forethought. 9    PF   FP

               In addition, because the way we act necessarily has significant impact on those
with whom we interact, our behavior in large part determines the way that others act and
react towards us. Our perceptions of their reactions, in turn, both influence our immediate
emotions and behavior and also have a longer term impact on our cognitive frames and
the way we see and understand the world.
               Self-Fulfilling Prophecies: In sum, our thoughts, feelings, behaviors and
perceptions influence each other. We react to our perceptions of the world around us
while our own behavior impacts on the world. Basically, because of our particular frame
of reference (and thoughts, feelings, etc.), we expect people to act in certain ways, and we
act toward them in ways that tend to precipitate the behaviors we expect 10 . When peopleTPF   FPT

do act in the ways we expected, we interpret that behavior in line with our expectations,
and we react in certain predictable ways (which tend to confirm to us the validity of our
earlier expectations).
               Consider, for example, person P at a party who looks at person AC and thinks AC
is arrogant and cold, and then looks at person FW and thinks FW is friendly and warm. It
is almost inevitable that by the end of the event, P's initial perspective will prove (to P, at
least) correct (regardless of the actual personality of either AC or FW). In all likelihood,
P will approach and be receptive to FW, who in response will likely act friendly. On the
other hand, in what P believes is simply self-protection, she will likely retreat from AC,

one group to continue while the other group quickly quit is entirely "reasonable." My point, however, is
not about whether the behavior of both groups appears reasonable to an objective observer. It is instead
that those who participated were not thinking about their own behavior in that way. They may have been
acting according to known principles of behavioral psychology, but they thought they were acting for other
reasons entirely. In addition to affecting our thoughts, how we act also affects our emotions. We all know
that there are certain activities that make us feel better (sports, relaxation, being with close friends and
family, etc.) and others that make us feel worse (some kinds of legal work, being with certain people, etc.).
More recently, researchers have shown that merely adopting certain postures or facial expressions has
immediate impact on emotions, regardless of the reason the postures are adopted and regardless of whether
the positions are physically comfortable, uncomfortable, stressful or relaxing. Similarly, how we act while
experiencing emotions significantly impacts the course of those emotions, regardless of the way that our
behavior impacts on others who might be the cause or target of those emotions. Ekman
TP Joshua D. Rosenberg, The Psychology of Taxes, 15 Va. Tax Rev. 155 (1997)

TP  Id.   PT

who in turn will be less likely to act warmly toward P. P will then leave the party
unaware of how her own feelings and beliefs impacted her behaviors, or of how her own
behaviors impacted AC and FW, but acutely (albeit inaccurately) aware of her own
insight and ability to predict human behavior. These self-fulfilling prophecies and other
generally unconscious learned responses significantly impact the outcome of most
negotiations and most other interpersonal interactions.

2. How This Relates to Taxes
Many people tend to believe that the impact of emotions and cognitive and perceptual
frames on tax compliance is non-existent, and that reactions to tax are essentially a
product of a basically rational cost-benefit analysis. In fact, most people do have quite
definite emotional reactions to taxes, quite specific cognitive frames around taxes, and
learned but unconscious behavioral reactions that can significantly impact taxpaying (or
tax-evading) behavior.
       For too many Americans, the cycle that plays out around taxes is a basically
negative one. They may begin by simply attempting to avoid taxes when an opportunity
presents itself. The more they attempt to avoid their legal responsibility, the more they
tend to retrospectively justify their actions. They may begin to believe that taxes are bad
and wrong and unfair, in large part because such beliefs can serve to legitimize their tax-
evasive behaviors. Their anti-tax beliefs in turn may lead them to resent taxes and resent
and distrust the IRS. They may tend to think about taxes and the IRS as doing nothing
other than taking away our money. They talk negatively about taxes, and positively about
tax avoidance. Based on these conversations, many of those they talk to tend to believe
that others engage in significant tax avoidance and evasion. As a result, they, in turn, may
try to avoid paying their own share, believing that if they do pay what they owe, they are
being made the fool, because others get away with evasion, and the more this negative
cycle grows.
       This negative, self-fulfilling prophecy about taxes is, unfortunately, reinforced by
economic incentives. Taxpayers obviously have much to gain by avoiding their own

taxes. 11 Employees of taxpayers who work at tax planning or tax preparation may tend to
        TPF   FPT

see their Employer’s self-interest as tantamount to their own (if for no reason other than
that the Employer pays their salary and controls their employment future); and tax
advisors and preparers in general may not only tend to see their proper job as minimizing
their client’s tax liability, but they may also believe that the more they save for their
clients, the more their business will grow. 12 Low audit rates provide little economic
                                                      TPF   FPT

incentive for honest self-reporting, and aside from the new and basically unknown
whistleblower laws, financial incentives for accurately reporting about others are
              The negative cycle surrounding tax can be deep seated, and can affect not only
ordinary citizens, but lawyers, legislators and administrators alike. 13 Indeed, I imagine
                                                                                   TF   FT

that some will react to almost any efficient proposals for narrowing the tax gap with
notions that they are unworkable, ridiculous or even treacherous. It is worth pointing out
that the almost instinctive negative reaction to changes that would increase tax
compliance may itself be a powerful indicator of the cognitive frames and the feelings we
have around taxes. When at-risk communities join to cut down on other crimes by
becoming more observant and by working with local police (as opposed to engaging in
vigilantism), few, if any, people complain. If groups propose getting people involved
(within the law) to help provide information about perpetrators of crimes such as rape,
burglary, battery or robbery, or drug-dealing, very few Americans would have anything
like an instinctive negative reaction. Nonetheless, it is often such negative reactions that
greet proposals to efficiently cut down on tax evasion.
              Of course, tax evasion is different from some other crimes in that the “victim” is
not an individual, but is the huge entity we know as “the government.” When the

11 . Indeed, as Joel Slemrod has explained, if taxpayers were entirely rational and fully informed actors, the
T   T

economic incentives in place would cause them to engage in more tax evasion than they currently do.
Whether the (economically) unexpected degree of honesty indicates that many taxpayers have a much more
positive reaction to taxes, or whether it is the result of the fact that many taxpayers are unaware of the
actual likelihood of success of tax evasive behavior and/or of the limited consequences if they are caught,
or whether risk aversion combines with uncertainty to produce less risky behavior is unclear. I suspect that
these all contribute in some way to the “phenomenon” of economically unexpected tax compliance.
12 . I do not mean to suggest that such is the reaction of all tax preparers. As Prof. Hasseldine has
T   T

explained, tax preparers, as well as taxpayers, have numerous motivations and different frames and
behaviors. I mean to suggest only that these motivations likely impact more than a few preparers.
13 Again, as I explain above, e.g. footnote 10, I do not mean to suggest that this cycle exists universally.
T   T

Nonetheless, I believe that its existence is sufficiently widespread to merit attention.

“victim” is a large organization such as government or even a large corporation,
observers may tend to feel a little less personal empathy for the victim and, as a result, a
little less outrage at the crime. Indeed, many people may tend to see tax evasion as even
more “victimless” than even other crimes against the government. One might imagine a
terrorist group that decided to proceed by damaging only (federal) government-owned
property, and which caused damage ranging in amount from $10,000 to hundreds of
millions of dollars per incident, resulting in annual damage to the federal government of
hundreds of billions of dollars. Few if any people would react negatively to legislation
that helped government put the perpetrators in jail. Although only the government may be
the victim, the damage is at least visible to all.
        The costs of tax evasion are in the hundreds of billions of dollars annually, but
until recently, it has drawn very little real national attention. When it comes to taxes, this
failure to view the victims of tax fraud as anything other than the government, and the
apparently “hidden” nature of the harm, is itself in part a result of a distorted perspective
towards taxes. While tax revenues ultimately help individuals, we tend to dissociate the
benefits (to individuals) of taxes from the payment of taxes (see discussion at , infra.). 14
                                                                                           TPF   FPT

In addition, we tend to see tax evasion as even less “bad” than even other “invisible”
crimes against government. People seem to believe that stealing from the government by
evading taxes is somehow either not stealing at all, or is different from stealing from the
government by other means, such as embezzlement or robbery.
        This overly tolerant attitude towards tax evasion is due to a wide range of factors.
Likely, the “passive” nature of tax evasion (merely holding onto money, rather than
paying taxes and then stealing them back) contributes to the perception that many have
that tax evasion should be acceptable. In addition, as suggested earlier, many people
engage in tax evasion themselves (even if in rather small ways). Because of the cycle of
reactions explained above, if a person engages in tax evasion herself, she is more likely to
see that tax evasion as acceptable, and thus more likely to view the same tax evasion by
others as also acceptable (to avoid the cognitive dissonance that would arise if she
believed tax evasion to be “bad,” believed herself to be “good,” but knew that she

14 See generally Rosenberg, The Psychology of Taxes, supra.
T   T

willfully evaded taxes). 15 In addition, even those who do not themselves evade taxes
                             TPF   FPT

hear too often from others, including friends and relatives, about engaging in tax evasion.
Others who do not brag about their own tax evasion often at least compliment clever tax
“minimization,” if not actual tax fraud; and even governments and government officials
may send out mixed messages about taxes.
              One might get a real sense of the way even lawmakers think about taxes
differently from how they think about other kinds of laws by looking briefly at the history
of attitudes towards non-tax government regulation. About 70 years ago, nontax
government regulation was justified primarily by a common understanding that it was
government’s right, if not its responsibility, to protect citizens. Regulations of behavior
and of property drew assent and acceptance because of the presumed correlation of the
regulations with public morality and public welfare. More recently (about 20 years ago),
people started analyzing regulations using “regulatory scorecards,” which subjected
regulations to a form of economic cost-benefit analysis. 16 These scorecards quickly led
                                                                    TPF   FPT

to conclusions that most government regulation was at best irrational, wasteful, and
wrong (in other words, people began to view these regulations in the same light they have
always viewed tax laws). The use of these scorecards resulted in a sea-change in opinion
about government regulation: suddenly, such regulations became wildly disliked and
ridiculed. 17  TPF   FPT

              More recently, the use of these scorecards itself has come under some harsh
criticism by defenders of regulation. The major criticisms include statements that the
scorecards have unfairly resulted in the trashing of regulation because (1) scorecards are
overly effective in turning lawmakers against regulation in large part because they draw
attention to numbers by generating a false “appearance of precision;” 18 (2) the scorecards
                                                                                  TPF   FPT

15 See generally Psychology of Taxes, In fact, the tendency of people to believe that behaviors are
T         T

acceptable if they engage in those behaviors (as opposed to first deciding whether behaviors are acceptable
and only then deciding whether or not to engage in those behaviors) may well account for the fact that
different types of tax evasion are met with more or less acceptance in different communities. Communities
that tend to engage in certain kinds of tax evasion (for example, the use of “tax shelters” in some
communities, or deducting meals and other personal expenses in other communities, etc) are more likely to
accept as appropriate those same kinds of evasion in others.
16 Richard W. Parker, GRADING THE GOVERNMENT, 70 U. Chi. L. Rev. 1345, at 1407 (2003).
T         T

17 Ibid.
T         T

TP “One of the most striking features of the Hahn, Morrall, and Tengs/Graham scorecards--and, one

suspects, a key to their great influence--is the precision of their numbers, which they typically report to

significantly undervalue benefits of regulations; and (3) the scorecards simply ignore the
significance of moral values and ethical principals. 19      TPF   FPTP   P

          I have no comment on the use of these regulatory scorecards, other than to point
out the implications for taxes of both the powerful affects and the expressed deep
concerns of these scorecards. The primary way we have always thought about taxes is
precisely the way that started the strong anti-regulatory movement when people started
thinking about regulations that way.
          The attractive appearance of precision: As Professor Graetz said a decade ago, in
tax “Congressional decisionmakers routinely suffer from illusions of precision. Congress
today seems to want tax policymaking to turn on simple numerical answers.... Armed
with mathematical answers to both revenue and distributional questions, tax
policymakers routinely eschew the difficulties of exercising judgment to strike an
appropriate balance among ambiguous and often conflicting normative goals; in the
process, they put aside the massive empirical uncertainties they inevitably face. Instead,
they constrain themselves to write laws that conform to misleading or wrongheaded
mathematical                                                                          straightjackets.” 20
                                                                                                         TPF   FPT

Undervaluing Benefits: To suggest that most analysis of tax laws “undervalues” the
benefits of taxation is, to say the least, an understatement. It would be more accurate to
state that tax analysis at times completely disregards even the possibility of any value at
all coming from taxation. Often people assume that tax laws are only about the
government taking money, and when they focus on taxes, they tend to focus on the
extent, and the victims, 21 of that taking.
                            TPF   FPT

          True, revenue estimates note the potential dollars that might be raised by taxes,
but even then there seems to be no conscious connection between that revenue and actual
benefits to citizens. We do not think "without taxes, we would have no roads and could
never drive anywhere;" we do not think “without taxes, we would have no ability to enact
or enforce product safety requirements;” when we sit down to eat, we do not think “some

three or four significant digits.Yet the appearance of precision is highly misleading.” Richard W. Parker,
GRADING THE GOVERNMENT, 70 U. Chi. L. Rev. 1345, at 1407 (2003).
TP Id at 1350-55.

TP Michael J. Graetz, PAINT-BY-NUMBERS TAX LAWMAKING, 95 Colum. L. Rev. 609, 613 (1995).

TP See generally Psychology of Tax
     PT                                    For example, the tax distributional tables focus on whom we take
more from and whom we take less from, but always on whom we take from.

of the money went to make sure that the food we are about to eat is not dangerous."
When we think about tax laws, we do not think of any benefits at we receive from
government. We may know, on an intellectual level, that these connections exist, but any
such logical connection is at best only purely intellectual, is rarely in our consciousness
and, as a result, has little if any impact on our feelings towards taxes in general.
            Ignoring Moral values and ethical principals. Even in the face of regulatory
scorecards, most nontax laws both reflect current morality (usually by penalizing
undesirable conduct as tortious, criminal, etc.) and help define morality (by specifying
what behaviors are appropriate or inappropriate). When people evaluate those laws, they
pay at least some attention to whether the behavioral norms embodied in those laws are
ones with which they agree (e.g., is it bad to assault people, is it wrong to allow people to
die from preventable workplace accidents, etc.), and the laws tend to draw some respect
and legitimacy from this correspondence between the behavioral mandates of the law and
society's moral norms.
            On the other hand, the effect of taxes on behavior is generally either ignored or
viewed as an unfortunate but necessary inefficiency. As a result, not only do tax laws not
draw any positive attention from any correlation with moral or ethical values, but any
such correlation may well be viewed as problematic. In fact, tax laws can be significantly
more effective than other laws in bringing about desired behaviors, 22 and, as scorecard
                                                                                    TPF   FPT

analysis points out, other laws can do as much redistribution and as much to impact
economic relations as do tax laws. 23     TPF   FPT

            Despite the fact that taxing and spending, on the one hand, and government
regulation, on the other, can often be interchangeable, we have always thought that taxes
ought not to interfere with private economic decision-making, while regulations should

TP Unlike criminal and other civil laws, tax laws allow not just for punishments (in the form of increased

tax liability) for those who do not do as we want, but also for positive reinforcement, by means of tax
deductions or credits, for those who do engage in the desired behaviors. In addition, the huge sanctions of
the criminal law (and the less severe sanctions of contract law and tort law) are applied only infrequently,
and often after the lapse of significant time. As a result, they are relatively ineffective at bringing about
behavior change. Tax laws alone are enforced quarterly (or, in the case of withholding, either weekly or
biweekly, depending on how frequently paychecks are distributed). The consistency and temporal
contiguity of tax law's impact to taxpayer behavior is strong, and, consequently, so is the law's effect on
TP Virtually any regulatory goal can be accomplished by some version of taxing and spending. Rather than

requiring A to take some action to benefit B, government could tax A and give the funds to B or spend
them to provide B with the intended benefit.

be addressed specifically to that end. 24 Whether or not we ought to continue to operate
                                       TPF   FPT

under this assumption generally when enacting and evaluating laws may be open to
question. 25 What is not questionable at this point is that these assumptions result in a
         TPF   FPT

very distinct anti-tax bias among too many of the general public and too many lawmakers
and administrators. As a result of an unconscious association of taxes with their burdens
rather than with their benefits, too often when people think about taxes, the good
disappears and only the bad remains.

3. Making Changes

a. Difficult but possible
The inertia created by the self-reinforcing cycle of human interactions can be sufficiently
powerful so as to make change difficult. Negative attitudes towards tax make it difficult
to adopt legislative and administrative solutions to the tax gap. The absence of strong
administrative and legislative action enables taxpayers to continue to avoid and disparage
taxes, and little may change.
        Nonetheless, understanding the cycle of human interactions provides some hope
for concerted change. Because cognitive frames, behavior, thoughts and feelings affect
each other, anyone seeking to make changes in the system can effectively intervene at all
of these levels to maximize change. Just as making people feel better about taxes will
increase their compliance with the tax laws, increasing their compliance with tax laws
will likely make them feel better about taxes. Similarly, just as changing the reactions of
third parties to an individual’s tax avoidance or tax compliance will affect that individual,
so will that individual’s tax-related behavior change the reactions of third parties. By
making self-reinforcing changes at several different levels simultaneously, much might
be accomplished.
        Just as an individual seeking self-improvement needs to put effort into changing
her behaviors, thoughts, and feelings, any system seeking to change must do the same.
Effective change requires, at the least, effective communication between the various

24 See Psychology of Taxes
T   T

25 See Psychology of Taxes
T   T

system components. An individual seeking to change needs to become aware of what she
is doing that she wants to change. She must become conscious of self-defeating
behaviors, the feelings that trigger them and that result from them, the cognitive frames
that contribute to these feelings and behaviors, the actions of others to which she is
reacting, and the way that her own cognitive and emotional frames impact the
motivations she attributes to others. Similarly, efficiently changing the tax culture
requires at the least communication between the parts of the system involved.
            Obviously the tax administration needs to know about the taxpayers. It needs not
only information to enforce compliance but also information to be able to assist taxpayers
who need help and to be able to encourage, facilitate and reward compliance.
            Perhaps less obvious, but also important, tax compliance can be helped if
taxpayers are more aware of the tax laws and policies. Many (admittedly not all)
complaints about tax unfairness are likely based on at best incomplete information, about
tax objectives and the use of tax dollars. Many (again not all) of those who may not
report accurately may either not know what do or at least cannot find out what to do and
how to do it without some difficulty. Many others who may avoid taxes simply may not
fully understand the importance of reporting honestly. They may have been told, on
occasion, that they should do so, but they have never been made to really understand its
importance, as we all understand the importance of so many products we see and hear
advertised on a daily basis.
            In suggesting approaches to closing the tax gap, then, I believe it important
whenever possible to intervene on several levels. I begin with some interventions directed
specifically at behavior, and end with possible interventions focused on changing
cognitive, emotional and perceptual frames. With respect to all of these, the underlying
goal is to increase the amount and the accuracy of communication from taxpayers to the
IRS and from the IRS to taxpayers.

b. Starting at the End: Imagining the Ideal
Developments in the field of Positive Psychology 26 and public health strongly suggest
                                                      TPF   FPT

that while it may be useful to note problems with any current system of behavior in order

TP     PT

to fix those problems, it can be significantly more productive to keep in mind not simply
defects in the current system, but an idea of what a well-working positive system would
look like. So far, when it comes to tax-paying, we have spent remarkably little time or
effort even imagining what such a well-functioning system might look like. While I do
not profess to have the exact answer, I feel obliged to at least offer a few notions.
       In contrast to the current cycle of actions and reactions surrounding taxes, one
might imagine an “ideal” (at least from the perspective of government) cycle of actions
and reactions around taxes. In such a system, people would feel about tax laws the same
way they feel about criminal laws, contract laws, and property laws—that they are an
important part of government and are enacted for our benefit. We would no more brag
about cheating on taxes than we would brag about robbing our neighbors or committing
vandalism. We would pay our taxes and feel good about it. We would believe that others
pay their fair share, we would expect them to do so, we would be disturbed when they did
not, and we would do what we could to ensure that the tax laws were properly enforced
and that the IRS had all the information it needed. We would see tax compliance, rather
than tax avoidance, as the norm, and we would do what we could to ensure our own
compliance and that of others.
       In this ideal tax world, cognitive frames and emotional reactions that supported
taxpaying would be accompanied by economic incentives that strongly supported tax
compliance. Rather than enabling a system that rewarded complicity in tax avoidance (by
or on behalf of employers, employees, tax advisors or their clients), there would be in
place a system that rewarded assistance in tax compliance. Accurate self-reports and
accurate reports about other taxpayers would be made easy, be encouraged, and be
rewarded. Tax avoidance and behaviors that enabled others to avoid taxes would be met
with results other than promotions and more business.

4. Changing Behavior: Some Specifics
While dramatic changes in behavior are not likely to occur without changes in cognitive,
emotional and perceptual frames, and while behavior change itself will often help bring
about accompanying changes in cognitive, emotional and perceptual frames, behavior
change is often directly accomplished most often by a combination of (1) modeling, (2)

stimulus control, and (3) response control. Essentially, modeling is simply showing
people how to do something, stimulus control refers to establishing an environment that
is conducive to the desired behavior, and response control refers to a system that rewards
desired behaviors and punishes undesired behaviors so as to shape peoples’ reactions in a
favorable way. The basic way to change behavior is thus to show people what they are
supposed to do; make it easy to do it, and make it difficult to avoid doing it; and to
reward those who do it and punish those who do not. 27        TPF   FPT

                Although this model is simple, it is not often fully adopted. Too often
governments attempt to change individual behavior by something more akin to a basic
“command” model. They tend to issue commands ("do this," or "don't do that"). When
individuals do not comply, government then attempts to punish the disobedient actor.
Without the other elements of basic behavior change and without accompanying changes
to cognitive, emotional and perceptual frames, commands followed by punishments are
relatively ineffective at bringing about significant behavior change, and they and tend to
create unnecessary anxiety in the individual who is punished. 28          TPF   FPT

i. What Behaviors to Change: More Complete and Accurate Reporting
The vast majority of the tax gap is the result not of taxpayers failing to pay what they
admittedly owe based on accurate reporting of income and deductions, but of taxpayers
failing to accurately report in the first place. Taxpayers, and often their employees,
employers and advisors, have intimate knowledge of all the information necessary to
accurately assess their tax liability. The IRS, on the other hand, may have only very
limited access to that same information. Instead, it relies primarily on self-reports, third-
party informational reporting where it is required, and whatever access it can obtain to

TP Although this model is simple, it is not often fully adopted. Most often governments attempt to change

individual behavior by something more akin to a basic “command” model. They tend to issue commands
("do this," or "don't do that"). When individuals do not comply, government then attempts to punish the
disobedient actor. Without the other elements of basic behavior change and without accompanying changes
to cognitive, emotional and perceptual frames, commands followed by punishments are relatively
ineffective at bringing about significant behavior change, and they and tend to create unnecessary anxiety
in the individual who is punished.
28 Of course another possible result of mere commands to change behavior is that government will issue
T           T

the command, people will not follow it, and government will fail to enforce it because to do so would mean
to punish so many people that government would simply not have either the resources or the credibility to
do it.

other financial records that might suggest items of unreported income or over-reported
            Complicating the IRS’s general lack of access to information is the fact that our
tax system is based on specific transactions rather than on accrual of wealth, and those
transactions may be structured by taxpayers, their advisors, employees and the others
involved in order to mislead and misdirect the IRS. The result is at times something akin
to the huddle of a football team. In on the huddle and strategic tax planning are the
taxpayer, his advisors, employers, employees and others with the shared interest of
minimizing tax payments; and the plan and strategy is developed to be played out against
the IRS, whose “team” is not only excluded from the planning, but whose intended
deception or misdirection often lies at the very heart of the strategy designed in the
huddle. 29  TPF   FPT

            Information Reporting by third parties: It is no surprise that tax compliance is
highest where third party (typically employer) reporting is present. When government
knows about a person’s taxable receipts, enforcement (at least at the income side, if not
the deduction end of the equation) is fairly straightforward.                                In recent years,
developments in technology have made third-part reporting of possibly tax-relevant
information significantly less cumbersome in many areas (for example, in capital gains
reporting, and for corporate payees). It may go without saying that wherever additional
tax-related reporting is made, tax compliance will be more likely. 30            TPF   FPT

TP Of course, often the taxpayer alone plans and executes the entire tax-avoidance strategy by means as

straightforward as cashing checks and not reporting income or accepting cash payments and not reporting
them as income. In these cases, the planning aspect may exist only in the taxpayer’s mind. Obviously, until
taxpayers’ cognitive and emotional frames lead them to value tax honesty over tax avoidance (either
because of the benefits associated with tax honesty or the burdens, including possible detection and penalty,
associated with tax avoidance), gaining the taxpayer’s cooperation in planning may be difficult. Even in
cases where only a single individual is involved in the planning, though, it is likely that at least two other
persons are involved in the implementation—the person who pays the taxpayer and the person paid by the
taxpayer). See discussion at .
TP There always has been, and will be, resistance to additional reporting requirements. Regardless of the

grounds offered for that resistance, it is useful to keep in mind the likelihood that much of the basis for such
resistance, and for any sympathy with which such resistance may meet, is at least in (large) part due to the
fact that many efforts to enhance tax collection are met with resistance because of the negative (towards
tax) cognitive and emotional frames that so many people have towards tax. I acknowledge from the outset
that almost no meaningful advances in minimizing the tax gap can be made without accompanying changes
in how we think and feel about taxes—from the man in the street to the lawmakers in the capital. See
discussion at

       Outside of situations where the payor needs to report payments in order to get a
tax deduction for those payments, or where the payor already had in place a system for
reporting payments, third party reporting has typically not been modeled, required,
encouraged, or rewarded; and, not surprisingly, it has not occurred. As a result, numerous
taxable payments go unreported by third parties. Taxpayers, aware that the IRS is
unaware of these receipts, feel no need to alert it to facts that would only add to their own
tax liability; and, as a result, these payments often go completely unreported and untaxed.
       The vast majority of these unreported payments are made by purchasers of
services or goods. The providers of many goods and services typically are aware that the
IRS has no direct knowledge of their gross receipts, and often feel free to simply ignore
those receipts for tax purposes. As a result, sellers in the cash economy go vastly
undertaxed, if not completely untaxed.
       To simply “require” purchasers to report to the IRS all payments made for
services or goods would be both ineffective and potentially suicidal (because of the angry
response that would come from taxpayers and purchasers required to report). As
discussed above, both the sellers and the buyers have built-in incentives to hide the
payments from the IRS—the sellers can avoid paying taxes, and the buyers can typically
get a lower price by enabling the seller’s tax avoidance. In addition, reporting of taxable
payments takes time and effort. Simply put, it represents an additional cost to the
reporter. To actually bring about such reporting, the reporting would have to be made
relatively easy and at least potentially rewarding, and non-reporting would have to be
made more difficult, more likely to be detected, and at least potentially subject to
punishment. Fortunately, this actually could be done.

ii. Making the changes, third party reporting:
To begin at the ideal end, consider what would happen if the laws, norms and
expectations in place were that (1) purchasers of services and goods report all purchase
information to the IRS, (2) sellers report in full all payments received, (3) those sellers
and purchasers who fail to make accurate reports are likely to be detected and punished,
and (4) the IRS routinely matches payment and income reports. All involved in the
purchase/sale would expect the IRS to be fully informed of the payments by the other

party. As a result, all involved would report accurately themselves, since they would
realize that the IRS would find out about the transaction in any event. Individuals on
either side of the transaction would be very hesitant to suggest deviation from the
reporting laws and norms both because such suggestions would be viewed as nothing
other than attempted conspiracy, and because they realized that not only were they likely
to be detected in the end, but they would tend to be viewed as dishonest lawbreakers
rather than as “honest working people trying to make a buck and not give it all to the
        This end is not as unreachable as it may seem.

a. Perhaps the most obvious way to increase reporting of payments for consumer goods
and services would simply be to require such reporting by the consumers. Because of the
recordkeeping and other costs associated with reporting consumer purchases, even the
most draconian reporting requirements would need to exclude some de minimus
amounts. 31
          TPF   FPTP   P   Of more importance, though, is the fact that merely “requiring” such
reporting does not ensure that it will occur. 32 Unless the reporter has some incentive to
                                                      TPF   FPT

make the required report and/or some more realistic potential penalty for not making the
report, simply requiring reporting is likely to do little. The following are possible ways to
make it more likely that any such “required” reporting will be met by compliance with
that “requirement.”

b. Reporting by Financial Institutions: Because reporting by purchasers of goods and
services might be very time-consuming, any chance of reaching the goal of full reporting
would need to include a way to simplify that reporting. Rather than the purchaser
reporting directly to the IRS, banks and other financial institutions might much more
easily report check purchases and credit card purchases directly on their customers’
behalf. In addition, large cash withdrawals could be automatically reported, so that cash
payments do not go undetected. Similarly, check deposits and checks cashed might be

31 At this juncture I would hesitate to define “de minimus” other than to suggest it would probably be
T   T

somewhere between $500 and $10,000 paid in a year.
32 Indeed, such “requirements” have proven unsuccessful in many areas, including household employees.
T   T

directly reported by financial institutions, including banks and other businesses that
regularly cash checks.
        Where purchases are made using credit cards or debit cards, reporting the
information becomes even easier, since the electronic records already exist. 33 Efforts to      TPF   FPT

spread the use of such cards for payments for services and goods could then have
significant rewards. Such efforts might include both making it easier to acquire and use
debit cards, requiring all service providers to accept such cards, and making such
acceptance fairly simple.
        Whether or not more people made more use of electronic payments, there would
be some cost to the financial institutions associated with making spending reports, and
these costs would likely be accompanied by resistance to additional reporting
requirements. If those costs were compensated (likely by a small portion of the revenues
generated by the required reporting), the investment would likely prove a very profitable
one for the US, and the government could at least ensure that it was not a losing one for
the financial institutions.
        There would, initially, be resistance to such reporting, and the resistance would be
significantly more than the resistance would be to other reporting requirements aimed at
detecting criminal activity. This is a result in part of the fact that presently most
Americans do not equate tax evasion with other kinds of criminal activity. I address ways
to change our cognitive and emotional frames around taxes infra.

c. Individuals Signing Up for Full Reporting: A less drastic and less complete alternative
to requiring financial institutions or individuals to report all spending by all customers
would be to allow individuals to volunteer to sign up with the IRS to facilitate such
reporting with respect to all of their present and future accounts. Individuals who are just
starting to become taxpayers (and, of course, many individuals who have long been
taxpayers) often intend to be completely honest about their taxpaying. 34 If told that they
                                                                                    TPF   FPT

33 See Joshua D. Rosenberg, A Helpful and Efficient IRS, Some Simple and Powerful Suggestions,
T   T

Kentucky Law Journal, (2000) at 58-61.
34 I do not mean to suggest that new taxpayers are on the whole more honest than other taxpayers. Instead,
T   T

it is more likely that they are simply more naive. Most people tend to think of themselves as honest and
good people, so most of those just starting to pay tax imagine that they will never cheat, just as most
newlyweds imagine that they will never have an extramarital affair. Those who have not yet paid taxes on

could sign up with the IRS to establish a system whereby all of their receipts and
expenditures were reported directly to the IRS, 35 and in return the IRS would, at their
                                                               TPF    FPT

request, compute their taxes for them and would tax them at a reduced rate on lifetime
income, millions of people might well sign up. 36        TPF    FPT

              Such an incentive system would enable those who signed up to pay a little less tax
( the incentive for signing up) than they otherwise would have if they had not signed up
and had reported all their receipts accurately, so would come at some cost. Indeed, one
might suggest that such a program might attract only the most honest of taxpayers, or
only those who are aware that all of their income will be reported in any event, so that the
actual benefits would be minimized.
              Because individuals typically change careers several times during their lives, the
fact that an individual’s income might be reported by third parties in any event when he is
a 20 year old corporate employee does not mean that future reports will not involve other
professions, including many that are parts of today’s underground economy.
Other benefits would also likely exist. In addition to having access to all financial records
of those who signed up, the IRS would also receive potentially significant information
about all who had dealt and will deal with those who enrolled. In addition to all of the
information about specific payors and payees, the IRS would also have access to data
about different professions, etc. that could prove useful in other analyses, and that
otherwise could not be obtained without costly and intrusive audits. Finally, because no
seller of goods or services would know who had and who had not signed up, the deterrent
effect could be very significant.

d. Extending Whistleblower and Qui Tam Provisions: As I have discussed at length
elsewhere, 37 Whistleblower provisions can provide a very important step in the direction
                 TPF   FPT

anything other than earned income on which there has been withholding have not yet had to deal with any
of the numerous temptations to cheat, just as newlyweds have not yet been tempted to have an extramarital
35 It would take some time and thought to develop such a system. It might have to rely on the use of
T         T

electronic payments or other methods that would enable fairly simple and accurate cash tracing. In addition,
it might take consent of spouses when joint returns are involved. In addition, the consents signed would
have to provide access to and information about any future accounts as well as current ones.
TP The only empirical research I have to support this statement is the responses of a few hundred tax

students to whom I have posed the question.
37 The Psychology of Taxes, supra.
T         T

of rewarding tax honesty and accurate reporting, rather than tax deception, by both
directly rewarding the reporter and by making it more likely that the non-reporter will be
detected and punished.
       Indeed, effective whistleblower laws, combined with broad-based reporting
requirements for purchasers, could go a long way to help establish a norm of tax honesty,
rather than tax evasion. To see how such laws might work, assume that Builder does
$20,000 worth of work for Homeowner. Currently, Homeowner (if she thinks about it)
and Builder (who definitely will think about it) likely expect that Builder will report little
if any of this payment to the IRS, and that Builder in turn will pass on a portion of his
resulting tax savings to Homeowner. Since there is nothing in existence today that would
likely inform the IRS of the payment and of Builder’s taxable income, these expectations
are likely to come to pass. Consider, though, the likely result if the parties to the
transaction both had the expectation that the IRS would learn about the payment from
Homeowner to Builder. It is quite likely that, in anticipation of that eventuality, both
parties would (if required) make sure that they themselves report it. While it may not be
possible to ensure that all such payments are reported, it is likely possible to at least make
it so that many more such transactions are reported and so that both parties are at least
aware of the likelihood, if not the necessity, that the transaction will be reported.
       Whistleblower laws could bring about such results. In this example, if Builder
knew that Homeowner was not only “required” to report such payments but also that
Homeowner might make money by reporting accurately to the IRS, Builder would be
likely to report honestly himself in order to avoid fraud penalties. Homeowner,
understanding this, is likely to take his own reporting obligations more seriously. Rather
than living in our current tax culture, where nonreporting is the understood norm, we
might soon have a tax culture in which both Homeowner and Builder understood that the
IRS would likely be made aware of the details of the transaction, and in which both
parties acted on the assumption that taxes would be paid rather than on the assumption
that taxes would be avoided. Homeowner and Builder might still explicitly conspire to

break the laws, but such explicit conspiracy is less likely to occur than is the
underreporting that is currently the accepted norm.          TP   F        FTP

         It is true that we now have whistleblower awards potentially available for those
who report tax fraud, and that those potential rewards have not had the effect I have
suggested (or any meaningful impact at all). A close look at the failure of the current tax
whistleblower provisions, however, suggests not that such provisions are necessarily
ineffective, but only that they need to be fixed.                                Qui Tam provisions (essentially,
whistleblower provisions) have long been part of the federal law. Prior to 1986, Qui Tam
provisions resembled the current tax whistleblower provisions. Like those provisions,
they were generally considered to be useless. Amendments in 1986 changed the Qui Tam
provisions to their current status as a powerful tool to both generate revenue and change
behavior. 39
           TPF   FPTP   P

         Among the 1986 amendments' most important provisions were ones that: (1)
replaced a provision that allowed the plaintiff to recover "up to 10%" of the amount
received by the government with a guaranteed minimum of 15% of the amount
recovered; (2) allowed the Qui Tam plaintiff to be a party to the action, to participate in
the prosecution, and to object to any government-proposed settlement, whether or not the
government also participates; (3) provided for the payment of reasonable hourly
attorneys' fees by the defendant to the prevailing Qui Tam plaintiff; (4) protected Qui
Tam plaintiffs from retaliatory dismissal; and (5) allowed Qui Tam plaintiffs to bring
certain actions based in part on public information on which the government has failed to
         Like the ineffective pre-1986 Qui Tam provisions, tax whistleblower provisions
have provided no guaranteed minimum for the whistleblower (and, in addition, they cap
rewards at the greater of 10% of the recovery or $100,000); they do not allow individuals

38 Still, if Homeowner could receive a maximum reward of 15% to 30% of the taxes generated by his
T   T

reporting of his own dealings with Builder, an offer by Builder to not report the transaction, and to then
split the tax savings 50-50 might be enticing, even if it meant a more explicit conspiracy. On the other
hand, if Homeowner’s potential reward included either the possibility of a percentage of other taxes
collected from Builder as a result (directly or indirectly) of his reporting, or if a whistleblower could be
entitled to a potentially higher percentage of lower amounts, the provisions might work quite well. For
example, the whistleblower might be entitled to only 20% of tax revenue his report generates if the total tax
revenue exceeds $1,000,000, but 60% of tax revenue generated if the amount is less that $3,000.
39 See Psychology of Tax at 214.
T   T

to participate at all in tax collection activities (so that the receipt of any reward is entirely
dependent on the decisions and efficiency of the Service); they provide for no
reimbursement of costs and attorneys' fees and no job protection for the whistleblower;
and they provide no reward for actions based on tax-significant information that is
already publicly known but that would otherwise go unused.
        The 1986 amendments to the False Claims Act essentially changed the Qui Tam
relator's status from that of an informant who might or might not receive compensation to
that of a very financially interested potential plaintiff. Indeed, under those provisions,
private parties and attorneys can now assert claims on behalf of the federal government in
every area other than tax. Revenue from these actions in the past decade has been
approximately $10,000,000,000. Nonetheless, the most important role of the Qui Tam
plaintiff has turned out to be bringing to light information otherwise not available to or
accessible by the government—the same role played by a whistleblower. This is because
while Qui Tam actions may at times be prosecuted by the individual plaintiffs, the vast
majority of such actions that ultimately succeed are initially filed by the private plaintiffs
but ultimately prosecuted by the government. Making tax whistleblower provisions
similar to the current Qui Tam provisions would likely bring about similarly meaningful
changes in their enforcement.
        The tax whistleblower laws are moving in the right direction. Recent amendments
have provided for guaranteed rather than completely discretionary rewards in some cases.
The new statutory provisions apply only to transactions involving more than $2,000,000
and only to taxpayers earning over $200,000 per year.            TPF    FPT   Of course, for whistleblower
rewards to be most effective, they would have to be not only more likely to be awarded
but also be applicable to ordinary transactions and ordinary amounts of money. Clearly
some minimum amount ought to be required: just as honest employees are not expected
to report co-workers who take home an occasional pen or pencil from work,
whistleblowers ought not be expected to report very small amounts. What that minimum
should be is debatable, but it ought to be substantially less than $2,000,000.

40 Nonetheless, the new office for tax whistleblowers may on its own make changes in implementing the
T   T

current laws for smaller cases.

           Another possible improvement to the new whistleblower provisions might clarify
whether and to what extent one will be rewarded if he has participated in the transaction
giving rise to the tax evasion. In order to ensure that purchasers and sellers will assume
that their transactions will be accurately reported, though, rewards would need to be
available even to those who participate in a reported transaction. If mere knowing
participation in a tax-avoidance transaction (such as an unreported payment from
Homeowner to Builder) disabled the participant from subsequently reporting (and
profiting from) another person’s tax evasion, then, in the above transaction, Builder
might be able to guarantee that Homeowner will not report accurately later by simply
seeing to it that Homeowner does not report accurately right now (and is therefore viewed
as a participant in the tax-avoidance transaction and not able to profit from
whistleblowing). As a result, if mere participation in a transaction were to eliminate from
rewards the person who participated, the impact of the provisions would be significantly
reduced. The “in terrorem” effect of the whistleblower provisions might simply
disappear. While tax attorneys who encourage clients to engage in questionable activities
ought not be able to reap rewards by then turning in their clients, and corporate decision-
makers ought not be able to reap rewards by turning in their corporations for tax
violations resulting from their own decisions, provisions that significantly reduce or
eliminate from rewards any “individual who may have planned and initiated the actions
that led to the underpayment of tax” (as do the current whistleblower provisions) may go
too far.
           While the recent statutory amendments to the tax whistleblower provisions may
be minor, we need to wait to see whether the new Whistleblower Office within the
Treasury Department will administer the whistleblower program in a way that
significantly increases its effectiveness. In time, that office may overcome the hesitancy
and distrust many may now feel about reporting to the IRS and depending on the Service
to properly value and act on the information reported. Qui Tam tax provisions, similar to
those in place under the false claims act, might have an impact that largely overlaps with
effective whistleblower laws.
           At least for now, though, enabling tax Qui Tam actions might have some
additional benefits. Many people might feel a little more comfortable reporting if they

could report to someone other than (or in addition to) the IRS, or if they felt at least a
little more confident that the information they reported would be acted on. Allowing Qui
Tam plaintiffs the ability to bring actions themselves, at least in some cases, might make
them feel somewhat more assure that their information will not be ignored.
                In some cases, Qui Tam provisions could accomplish significantly more than
whistleblower rewards alone. 41 This would likely be the case in large dollar,
                                       TPF   FPT

sophisticated cases rather than in more ordinary cases. Most if not all of the transactions
reported by whistleblowers under the newly enacted amendments are likely to be large,
and potentially complex, transactions, and most of the taxpayers reported are likely to be
able to afford large teams of high-priced lawyers to represent them. Of course, the IRS
has many equally (if not more) brilliant and dedicated attorneys. 42 The difference is that
                                                                                 TPF   FPT

while there is a fixed number of IRS attorneys, and that number each work at fixed
salaries, attorneys for taxpayers are often compensated on an hourly basis. As a result,
taxpayers involved in disputes with the IRS may be able to outman the IRS simply
because they can throw more money at more attorneys. The more tax dollars at stake, the
more the taxpayer is able to spend on attorneys fees. These imbalances would be quickly
undone if attorneys and others seeking to collect taxes were compensated at rates and in
amounts similar to those seeking to avoid them. At least for now, it is doubtful that
government compensation will equal that of private firms. Until and unless those

TP  At bottom, Qui Tam provisions would likely be at least as effective as whistleblower laws in changing

perceptions about taxes in addition to raising revenue. If individuals who know about tax evasion could
bring tax collection actions against the taxpayers who they know are hiding tax liabilities, and if they could
receive a sufficient percentage of any taxes and penalties collected to make it worth their while, those
individuals would be likely to bring that relevant information to light. People would at last have real
financial motivations to make public accurate information about tax misrepresentation, rather than to keep
it undercover. Once people know that someone else with access to information about tax misreporting has a
real financial incentive to report the facts accurately, they will quickly realize that their own previously
assumed economic interest in facilitating tax under-reporting no longer exists. Tax advisors and planners,
and others, would know that they could trust others involved in the planning to be honest and to ensure that
the law was enforced, but that they could no longer trust them to join an implicit (or explicit, as the case
may be) conspiracy to avoid taxes. Each participant in tax-significant planning and transactions is likely to
act honestly. While such honesty may, in the beginning, be only the result of a reassessment of each
participant’s economic self-interest, it is likely to breed more honesty, and eventually to lead to a culture of
honesty rather than of facilitated deception. In little time, such honest reporting would become the norm.

42 In fact, the IRS has had some recent significant successes against well-financed adversaries in tax
T           T

shelter cases.

litigation battles are funded equally, Qui Tam provisions that provide for or allow
significant attorney’s fees can go far towards leveling the playing field.

Problems with whistleblower laws
Inevitably, some will react to the idea of effective whistleblower or Qui Tam laws by
voicing concerns over creating mistrust among citizens. Will angry spouses complain
about their ex’s and will neighbors spy on and despise each other? Given the history of
the false claims act, such will quite likely not be the case. Effective whistleblower
provisions would result not in mistrust among taxpayers, but only in an expectation that
the “default” assumption underlying economic transactions is tax honesty, instead of tax
evasion. Rather than “trust” each other to go along with tax evasion, they would trust
each other to comply with the tax laws. Just as honest employees who can be relied on to
report embezzlement to the employer do not create “mistrust” among employees, honest
citizens who can be relied on to report what is essentially embezzlement against the
government will not create mistrust among taxpayers.
            It is worth noting that private enforcement of public policy is also the norm even
outside of current federal Qui Tam provisions. For example, while tort actions require the
plaintiff to establish a direct connection between her own injury and the defendant's
actions, individual claims are nonetheless based on the notion that the plaintiff is, in
seeking personal redress, enforcing some public policy (e.g., the policy that persons
should exercise reasonable care). 43 Indeed, punitive damages in tort and in other areas
                                         TPF   FPT

where they are allowed serve solely to enhance enforcement of public policy, rather than
to compensate any particular individual for an injury. In the case of tax evasion, the
injury, to everyone, is real. The difference between allowing tax Qui Tam actions and
providing for punitive damages in non-tax cases would be that in addition to ensuring that
the wrongdoer is punished, tax Qui Tam actions would also provide for the bulk of any
money recovered to go to the government rather than to the individual plaintiff.
TP Indeed, requiring the tort plaintiff to prove that the defendant's wrongful conduct caused the injury

actually limits the efficiency of tort laws as enhancers of public policy, but we require it nonetheless
because it tends to ensure that actions will be brought by plaintiffs most able to prove both the defendant's
tortious conduct (because they have been injured by that conduct, so must have been there to see it) and the
nature and extent of the damages caused by that conduct (because they are the ones who suffered that

       In addition, if some Qui Tam actions were prosecuted by individuals, they might
pit an honest employee against an employer, an individual against a large corporation, or
some other tax-seeking "David" against a tax-avoiding "Goliath" instead of always pitting
the taxpayer against what too many perceive to be the giant (and hated) Service. People
might tend to see tax enforcement, and taxes in general, less as the individual against the
government, and more as answers to questions about the responsibilities of individuals to
each other. They might begin to form some identification with the tax collector, and
would no longer identify only with the tax-evader. Communication about taxes would
improve not because the Service would become more intrusive, but because those who
already have access to information would begin to use it to enforce rather than evade the

d. Tax Products and Tax Preparers: Finally, whistleblower rewards might prove useful if
amended to enable reasonable rewards in two tax-specific settings: (1) the development
of tax “products,” and (2) tax preparers. Because of complexity of the Code or of
financial transactions, and because of some uncertainty in the law, tax attorneys and
others may devise tax-related products that enable taxpayers to reduce their tax liability
by significant sums in ways not explicitly prevented by law because not previously
contemplated by lawmakers. Although the Treasury department consults regularly with
many of the top tax attorneys and makes other efforts to become aware of these
“products,” it would seem that any action that might help inform Treasury of these
developments as quickly as possible is to be encouraged. Rewarding information about
such real and potential products, so that those aware of them can choose to benefit by
going to Treasury rather than by marketing them to would-be tax avoiders, might prove
useful. To analogize to private industry, rewarding such reports would be akin to a
business having in place a system that encourages employees to suggest improvements to
the organization rather than one that encourages employees to take their ideas to
competitors or to become competitors themselves.
       A significant percentage of Americans pay others to prepare their taxes. Tax
preparers are worthy of attention not because they as a group are any more or less
trustworthy than anyone else, but because they are such an integral part of the tax

collection system, and because each individual preparer can impact hundreds or
thousands of tax returns. Rewarding accurate reports of misrepresentation or misdeeds by
preparers might be a way to let preparers know that they ought to do what they can to
encourage honest reporting by their clients.

ii. Requiring, Encouraging, and Rewarding accurate self-reporting: While self-reports
will doubtless become more accurate as taxpayers become aware that the government is
likely to obtain accurate information from other sources, more can be done to encourage
early and accurate self reports.

a. Automatic Reporting Sign-ups: As suggested earlier, taxpayers ought to be not only
permitted, but affirmatively encouraged, to enroll with the IRS to ensure that all of their
potentially tax-significant (for themselves and for those they deal with) information is
automatically reported to the IRS. In return for being assured that the Service received all
accurate information, the taxpayers might at least have the IRS calculate their tax liability
for free, and perhaps be rewarded by being taxed at a discounted rate. The number of
people who might sign up for such services, and who would remain enlisted throughout
their numerous careers, might be significant and would likely provide substantial useful

b. Tax Software: Currently the government encourages the use of tax software and of
electronic filing in cases where individuals use commercially produced tax software
programs. It is arguable that, because tax software is today’s equivalent of the paper tax
forms of three decades ago, 44 it is incumbent on the IRS to make such programs freely
                                          TPF   FPT

available. Whether government chooses to require taxpayers to use government-provided
software or whether it continues to allow commercially available programs, government
can (and should) exercise significantly more control over the content of those programs.
                  There is nothing wrong with tax software programs, such as those currently on the
market, that enables pop-ups and reminders about potential additional deductions and tax
savings. These are entirely appropriate. Equally appropriate, though, are pop-ups that

TP     PT   A Helpful and Efficient IRS, supra.

prompt taxpayers to include all items of income. Other friendly pop-up helpers could
remind taxpayers that certain items they wanted to deduct are actually not deductible and
that others are deductible only if they can be properly substantiated. They could explain
the kinds of substantiation required by law, explain that taking the deductions without
having the required records is fraudulent, and reveal the penalties for fraud at appropriate
times during the process. The program could firmly suggest that certain actions or
questions that typically tend to generate untruthful responses may be watched closely,
and it could make its point most forcefully at times by pointing out some of the more
severe and intimidating penalties for noncompliance.
            While gently (but firmly and clearly) pointing out the dangers of attempting to
underpay taxes, the IRS could also input into programs information and affirmative
incentives to enter data correctly. Simply and clearly highlighting any facts, the
knowledge of which tends to improve compliance, could have a substantial positive
            By ensuring that taxpayers are properly guided beginning with the initial tax
preparation and continuously throughout the entire filing process, the IRS could gain
significant access to taxpayers' attention at the time it would do the most good, and the
impact on compliance could be huge. 45        TPF   FPTP   P

c. Reporting Questionable Positions: Because taxpayers are not prescient, we cannot
require that they always take the correct position in ambiguous situations, but we can
require that whatever position they take, they disclose fully and accurately what they are
doing. If a taxpayer takes a position and there is any question about whether that position
is correct, the taxpayer ought to be required to disclose: (1) the issue, (2) the position, (3)
the amount of money at stake, and (4) the number of times this or a similar or related
issue arises for the taxpayer in the taxable year. These disclosure requirements could be
clear and explicit, the disclosures themselves could be made on simple forms, and the
amount of information and analysis required to be disclosed could be minimal.

TP If the idea of impacting commercially-produced software raises first amendment concerns, those

concerns can be dealt with fairly easily. While government cannot dictate what others put into their own
software, it can certainly determine for itself what electronic filings it will accept. If commercial software
providers balked at providing software that government believed was sufficiently tax-friendly, government
itself could simply provide, and require that taxpayers use, its own software programs.

            While tax attorneys, accountants, preparers, and filers might complain of
substantially increased burdens, requiring disclosure in all cases would, for most
taxpayers most of the time, simplify and clarify the process. It would require less difficult
decision-making (about whether or not to report a transaction), although more honesty,
than does the current system.
            Some might suggest that because taxpayers are not to blame for the problem of
substantive indeterminacy, they are "entitled" to report their transactions as they see
them: it is the Service's (or Congress') job, and not the taxpayer's job, to make sure that
the law is clear. From the perspective of one seeking to make the system function better,
however, the "job" of bringing questions to light ought to fall on whoever is in the best
position to know about them. When taxpayers know the facts, have researched issues and
are well aware of ambiguities, reason would suggest, incentives for taxpayers to share,
rather than hide, their knowledge. If a taxpayer has not researched an issue prior to filing
a return, it simply should not be taking an aggressive and arguably wrong position.
Taxpayers should do at least enough research to determine if their positions are
questionable; if so, they need not answer that question, so long as they make the Service
aware of its existence.
            Again, merely “requiring” reporting, even with possible penalties attached to
failure to report, is not enough to bring about needed changes. 46 In order to best ensure
                                                                              TPF   FPT

TP  Without changes in the payoff amounts, even greatly reduced odds of winning the audit lottery would

not dissuade financially rational taxpayers from taking their chances. Consider, for example, the situation
of a corporate taxpayer ("X") that engages in a transaction that has a 50% chance of being found by a court
to generate $2,000,000 of tax liability and a 50% chance of being found to be tax-free. If X stands a 30%
chance of winning the audit lottery and, if caught, has a 50% chance of losing, then in the absence of
penalties, it might value its tax cost of failing to disclose and contesting liability at $700,000 [70% (the
chances of being detected) x 50% (the likelihood that, if detected, it will be found liable for tax) x
$2,000,000 (the tax due if liability is found)]. On the other hand, the tax cost of disclosing the transaction
would be simply 50% of $2,000,000, or $1,000,000.
            In order to make full disclosure worthwhile from a purely economic perspective, X's cost of
nondisclosure would have to be increased, and/or the benefits of disclosure would need to be increased, by
at least $300,000. If this were done solely by way of penalties, those penalties would need to be at least
$428,571 to convince the economically rational taxpayer to disclose (a 70% probability of being caught and
having to pay $428,571 represents an estimated cost of $300,000).
           Even if very substantial penalties for nondisclosure were enacted, problems would persist. In all
likelihood, penalties of a sufficient magnitude to outweigh the benefits of playing the audit lottery would be
attacked as punitive. More importantly such penalties, even if enacted, would rarely if ever be imposed.
Because cases tend to settle, and to settle somewhere in the middle, it could well be that not only the tax
due, but also the penalties, would be compromised. Whether intentionally or not, the Service, like anyone
else, is likely to accept less from those who fight hardest and persist in denying liability than it would from

that taxpayers who played by the rules would do better than those who did not, we would
need a system that both guaranteed that appropriate penalties for nondisclosure would be
enforced and, where possible, provided positive rewards for honest taxpayers. To avoid
the strong pull towards giving better settlements to those who fight hardest, thereby
rewarding exactly the behavior we might wish to discourage, the Service might
encourage early and accurate reporting by providing benefits for those who are most,
rather than least, cooperative.
        One significant benefit the Service could provide to honest taxpayers would be
one that could come at no cost to the Service. For example, assume that X has not yet
acted, but is contemplating the possibility of engaging in a transaction that has a 50%
chance of being found to be tax-free and a 50% chance of being found to be taxable.
Currently, it is clear that the Service is not likely to approve of the transaction in advance
and give up a 50% chance of collecting $2,000,000. Instead, X simply has to decide
whether to engage in the transaction without any certainty about the tax results, and then
must decide whether or not to report the transaction to the IRS. While the case, if
reported fully, is likely to eventually settle, X cannot know that for sure, and it cannot
know the amount for which it will settle until after making its decision. The information
X needs to best decide its course of action will be made available only after that action
has been taken.
        By compromising with taxpayers who are considering transactions with uncertain
tax consequences in advance of the actual transaction, the Service could provide a benefit
that would, especially for risk-averse taxpayers, potentially offset the value of any
possibility of escaping audit and detection. In return for providing accurate information to
the Service, they could obtain accurate, and useful, information from the Service.
Because any controversy is likely to ultimately settle in any event, pre-transaction
compromises would do little more than allow those settlements to be made when they
will do the most good. Because these settlements would necessarily entail compromise
rather than a conclusive determination that a specific transaction is or is not taxable, the

those who disclose and acknowledge liability. In other words, even with added penalties for nondisclosure,
it would probably cost X less to attempt to evade and then contest liability than it would to acknowledge
and disclose it.

Service would not commit itself with respect to future taxpayers, so that current concerns
surrounding the issuance of private letter rulings would disappear.
            Allowing pre-transaction settlements might also do much to improve relationships
between some taxpayers and the Service. Being legally required to disclose while being
financially rewarded for failing to disclose puts taxpayers in a bind that inevitably
generates hostility. If, instead, taxpayers were financially rewarded for doing what they
are actually supposed to do, much of that hostility might disappear. In addition,
encouraging mutually beneficial cooperative problem-solving could add a further positive
element to taxpayer-Service relationships. 47       TPF   FPT

            Unfortunately, merely allowing the Service to settle cases before transactions are
completed would be unlikely to bring about any useful compromises, because in the
typical situation the taxpayer would be anxious to resolve the case so that it could get on
with its business, while the Service would have no pressure driving it towards early
resolution. Such unequal time pressures on the parties to a negotiation are likely to
severely distort bargaining results against the party experiencing the time pressure.
Taxpayers could obtain early resolution only at the cost of what might be a significant
financial sacrifice.
            One way to foster early and fair settlements would be to put some reasonable
constraints on the Service's ability to reject reasonable offers of compromise. Taxpayers
(and, perhaps the Service, in appropriate cases) who make pre-transaction compromise
offers that are not initially accepted might become entitled to reimbursement of the costs
of necessary delay if the case is subsequently concluded for an amount not in excess of
that offer.

TP I have often heard corporate tax attorneys complain that the Service is inaccessible, and have just as

often heard attorneys for the Service complain that some corporate tax attorneys are dishonest. Because the
Service and taxpayers see themselves as being on opposite sides, and because there has been little room for
cooperation between them in the past, these different perspectives are only natural. When people are
required to work against each other, conflict and distorted perceptions are unavoidable. We tend to see the
people we are working against as bad and the people we are working with as good, and we tend to ignore
evidence that arises that might contradict those perceptions. When people work together on a mutual
problem, the opposite tends to happen. Giving the Service and taxpayers an opportunity to work *223
together, by encouraging taxpayers to bring issues out in the open early, and rewarding them with early
resolution rather than penalizing them with uncertainty, delay, and assertion of full tax liability, might be a
big first step in the right direction.

       Another way to negate the impact of differing time pressures on the Service and
taxpayers would be to establish a neutral board of tax experts to decide on appropriate
pre-transaction compromise settlements. These decisions could be made by taking into
account the tax due if liability were found ($2,000,000, in the above case) and the
likelihood of an ultimate finding of liability (50%, above). Decisions by this board could
be made binding by prior agreement of the parties. Alternatively, a decision by the board
could serve as a suggested settlement amount which, if rejected by either party, could
result in the imposition of costs on any party that rejects the decision and subsequently
does less well (by settlement or verdict) than it would have under the board's

5. Changing Cognitive, Emotional and Perceptual Frames around Tax

As suggested earlier, behavioral interventions are most likely to succeed when
accompanied by corresponding and reinforcing cognitive and emotional changes, and
vice versa. Thus, while each kind of change is discussed separately, benefits will be
optimized when they are enacted in tandem.
       Basically, the behavioral reforms discussed above focus on ways to maximize the
useful information that flows into the IRS, on the presumption that more relevant and
accurate information will lead to better tax collection. The other part of the equation,
changing taxpayer frames, focuses primarily on the flow of information from the
government to taxpayers, tax advisors and tax preparers. The major questions are (1)
how the desired information can be conveyed most effectively—how can it be conveyed
most broadly and with the most forceful impact, and (2) exactly what information ought
to be so conveyed in order to have the maximum impact. While I believe that it is
important to address both of these questions at length, the following suggestions at times
address them together in order to suggest possible actions in a way that might be taken
seriously. I do not suggest that my suggestions are the best answers. They are no more
than an attempted beginning exploration of the significant questions.

1. Using The Media

Many people have learned about tax evasion by hearing others proudly tell of their own
real or imagined successful tax evasion. Unlike tax avoidance, tax honesty is something
almost no one talks about. Indeed, few would feel safe from scorn if they even publicly
admitted such behavior. Such straightforward honesty seems at least silly, if not
affirmatively un-American, in part because we have so few visible models of honest
            We might accomplish surprising results simply by replacing the model of good
citizen as tax cheat with one of good citizen as honest taxpayer. Imagine the changes that
might accompany a culture which viewed those who cheat on taxes as being no better
than any other criminals, and in which boasting about tax evasion was as unacceptable as
boasting about embezzlement. If people began to see that the people they respect and
admire do pay their taxes in full, as a society, we would associate paying taxes with good
citizenship rather than with feeling cheated and powerless.

While I have no expertise in public relations, my own idea is for a television campaign 48                 TPF   FPT

featuring a sequence of short spots, each with an American hero or heroine saying    HH

something like “I don't try to steal from starving children, or from old people, or from
sick people, or from the US army. That’s just what trying to cheat on your taxes is. I pay
my taxes. Pay yours. If you hear somebody talking about not paying their taxes, or
cheating on their taxes, let them know it’s a crime. Even better, report the criminal who’s
trying to steal from us all."
            Instead of feeling embarrassed about the fact that we are honest on our taxes, and
instead of feeling like we are being cheated if we pay our taxes, positive models could

TP While government has so far had little impact by way of the media, the reason is not that the media

cannot be an effective policy tool, but because to date the government has simply not used the media
effectively. The solution lies neither in government's abandoning the use of media nor in continuing its
present course, but in learning to use media effectively.
         All of our largest and most successful corporations have found media advertising to be a very
effective way to influence people's thoughts and behavior. Mass media advertising convinces us that we
need things we never knew of before and that only a certain brand of whatever it is will do, allowing
manufacturers both to establish and enlarge the market and increase their own market share. The media
impacts dramatically not only our buying habits, but also the vocations we choose, our recreational
behavior and our dress, and our tendency towards violence.

allow us to begin to feel that paying, rather than evading, taxes, is desirable, and that the
onus lies not on ourselves to cheat, but on all Americans to be honest.
            In addition to positive models for taxpaying, attitudes towards tax would also
benefit from having positive associations toward taxes themselves. There is much that
government does that is good, but our association of taxes with any of these things is
essentially nonexistent. To enhance that association, we could air another series of short
advertisements pointing out some of the productive things the government does. Brief
shots of hospitals, schools, roads, jetfighters, people eating wholesome food, taking safe
drugs, etc., could make us understand some of the thousands of ways that our taxes
contribute to our lives in ways we typically ignore. Instead of seeing taxes only as a
villain, we could have a sense of the positive contribution we are making with our tax
            Finally, in order for any laws, be they whistleblower or Qui Tam provisions or
anything else designed to close the tax gap, to work, people need to know about them.
Informing people that not only is tax compliance a positive good in itself, but also that
helping tax compliance can be exceedingly rewarding financially, is the best way to
ensure that any positive financial incentives enacted will actually work. 49        TPF   FPT

Other avenues

While mass media can be a very effective tool for distribution of information by
government, it ought not be the only such tool. There are several places where
government has access to individuals, and any effective publicity campaign ought to take
advantage of those access points.
            Perhaps the most obvious places where government has easy access to
individuals are schools, and these have long been used to deliver important information.
If nothing else, taxpaying is an essential civic responsibility and ought to be taught as
such. But there is even more that could and ought to be taught. Just last week my own
child came home from school with a directive to his parents to be aware of and to map

TPThe current, nontax, Qui Tam provisions began generating very significant revenue shortly after a “60

Minutes” show revealed the millions of dollars that a few early whistleblowers made.

out all possible emergency exits from our home. Messages from schools directed at
parental behavior relating to tax compliance are at least as justifiable. Indeed, if all the
children at public schools were given information to take home about their parents tax
obligations with respect to childcare providers, compliance in that area might increase
       Government access to individuals goes far beyond schools, of course. Indeed, at
least in cities it would be difficult to find any significant area in which government did
not own and/or control substantial property. While some governments have recently
begun to learn that that they can reap benefits by selling or leasing naming or advertising
rights on such property to commercial enterprises, those unwilling to go that far might at
least using some of those areas to help get out government’s own important, nonpartisan

6. Conclusion

Much is being done to close the tax gap. Much more can be done. In many ways, changes
in taxpaying behavior have been limited because of the frames of both taxpayers and tax
lawmakers. In turn, the frames of taxpayers and tax lawmakers remain biased against tax
honesty in large part because such honest behavior is not the norm in taxes. We have a
choice to either make changes that do little to change norms, but that at least cause little
controversy because they do not challenge those norms, or to take on the task of
significantly changing both norms and behavior.