March 21, 1995
Remarks of Stuart L. Brown Chief Counsel for
the Internal Revenue Service
Information Gathering for Tax Administration in a Global
Economy: What Should the IRS and Taxpayers Expect of Each Other?
The emergence of a global economy is affecting the way all
of us do business. We have seen almost unbelievable political changes around the world
during the past 5 years; these political changes have given a whole new meaning to the
concept of emerging markets. Along with these sweeping political changes, we have seen
major new trade agreements -- such as NAFTA and GATT -designed to reduce the economic
barriers to international trade. In addition to new markets and reduced trade barriers, we
have also seen an increasing number of complex new transactions, particularly financial
transactions such as trading in derivatives. These transactions seem to permit major
shifts of capital, almost instantaneously, across industries, markets, And national
borders, often in ways that can only be fully understood and measured using sophisticated
For both taxpayers and the IRS, these changes mean that the traditional issues of tax
compliance increasingly present challenges of global scope. Today, I'd like to focus on
just one of those challenges: gathering the information necessary to determine the correct
U.S. tax liability of a world-wide business enterprise. As you well know, in recent years,
the U.S. rules that govern this information gathering process in the international context
have engendered both confusion and controversy. In contrast to the relatively stable
domestic field, we have seen a steady flow of both litigation and legislation dealing with
various aspects of the international information gathering process. Despite all the recent
activity, I doubt anyone is totally satisfied with the current situation.
The question then is what should we do now? Do we look forward to continued legislative
activity to address particular areas of concern? Will we have continuing administrative
disputes and litigation to resolve individual cases and issues of controversy? Or, is
there a better way?
I think there is. I believe that some meaningful part of the difficulties we have
experienced in recent years stem from a lack of clearly understood mutual expectations. If
that's right, then we should be able to avoid a meaningful portion of those difficulties
in the future simply by clarifying what the Service can expect of taxpayers when it asks
for information about multinational business activities, and what protocol those taxpayers
can expect from the Service in return.
Later on, I'm going to explain why I think it is possible for us to make progress in
this direction -- why the prospect of reducing information gathering controversies by
clarifying our mutual expectations is not a hopeless "pie in the sky" prospect.
And then I'm going to conclude my remarks by suggesting some general principles that could
serve as a starting point for discussion of what these mutual expectations should be. But
first, in order to provide a better background for that discussion, I want to review
briefly some of the specific areas that have raised particular concerns in the context of
international information gathering in recent years.
II. Specific Areas of Concern
A. Issues Addressed by Recent Legislation. Three of the
specific areas I want to mention have been the subject of recent legislation. Many of you
probably remember the heightened public and Congressional interest in transfer pricing
issues that began in the late 1980's. You may remember, in particular, the series of
hearings on this topic before the House Ways & Means Oversight Subcommittee, chaired
by Congressman Pickle.
One focus of these hearings, and the resulting publicity, was an increased
Congressional awareness of the administrative difficulties faced by the Service in dealing
with section 482 cases. In three successive tax bills, 1989, 1990, and 1993, Congress
responded to these difficulties by enacting some powerful tools to facilitate IRS
information gathering in the context of multinational audits. Let me talk briefly about
three of these new tools:
1. Sections 6038A and 6038C -- Reporting and Penalties. As part of the 1989 and 1990
tax bills, Congress enacted legislation to enhance the reporting obligations both of
foreign controlled corporations (defined as those having 25% foreign ownership) and of
foreign corporations doing business in the United States. These provisions, sections 6038A
and 6038C, generally require affected corporations: (a) to report transactions with
foreign related shareholder; (b) to maintain records needed to determine the correct U.S.
tax liability of transactions with related parties; and (c) to act as agent to receive any
IRS summons for books and records of certain related foreign parties.
The changes to section 6038A and new section 6038C also provided substantial penalties
for noncompliance with the new statutory requirements. The clear Congressional message,
however, was that the penalty was simply a means to ensure that relevant information would
be made available to the Service on a timely basis. Any increased revenue was expected to
come from enhanced voluntary compliance with respect to the underlying transactions -- not
from collecting penalties.
Nevertheless, there was some public concern that the new penalty would be asserted
inappropriately against taxpayers who were attempting, in good faith, to comply with the
statute. Despite this concern, our experience has been that the penalty has been applied
sparingly, because taxpayers have largely complied with the requirements of the statute.
I'd like to think that this positive record of compliance was helped by the fact that the
Service issued final regulations under section 6038A within about 18 months of the
original enactment of the statute.
2. Section 6503(k) -- Designated Summons. The second area of recent legislative
activity I want to mention is the designated summons provision of section 6503(k). This
provision was enacted in 1990, and reflected a Congressional decision that legislation was
needed to deal with an information gathering tactic employed by some taxpayers. These
taxpayers reportedly made a practice of responding quite slowly to IRS requests for
information, and then deciding quite quickly not to extend the statute of limitations.
Section 6503(k) authorizes the Service to issue summonses which toll the running of the
statute of limitations during the period in which judicial enforcement proceedings are
pending, And for either 60 or 120 days thereafter, depending on whether or not the court
orders compliance with the Summons. The purpose of this provision was not primarily to
extend the statute of limitations in a large number of cases; rather, Congress expected
that once taxpayers realized the Service had this tool available, they would provide
requested information on a timely basic.
Here, too, Congressional expectations have been largely realized in practice. In the
more than 4 years Since the enactment of the statute, the Service has issued only 10
designated summonses. The Internal Revenue Manual specifically recognizes that despite the
broad Statutory language, designated summonses are generally issued only to CEP taxpayers.
Moreover, our procedures require pre-issuance review of any designated summons by Counsel,
both at the District and Regional level. Yet despite the conservative use of the new tool,
we are no longer hearing nearly as many reports from our examiners about the type of
behavior that led to enactment of the provision.
3. Section 6662(e) -- Transfer Pricing Documentation and Penalty. The third recent
legislative action affecting international information gathering that I want to mention
today is found in the 1993 amendments to section 6662.
Congress first amended the substantial valuation misstatement penalty in 1990 to cover
transfer pricing cases. The penalty was to apply if either the transfer price was way out
of line (20% penalty if the taxpayer'; prices are less than 50% or more than 200% of the
correct price; 40% penalty if the taxpayer's prices are less than 25% or more than 400% of
the correct price) or there was substantial underreported income ($5,000,000 for the 20%
penalty; $10,000,000 for the 40% penalty).
Even after the 1990 amendments, however, Treasury and Congress remained concerned that
reliance on the traditional rules for after-the-fact penalty determinations were not
sufficient to motivate enhanced voluntary compliance in the transfer pricing area.
Therefore, in the 1993 tax bill, Congress provided that the section 6662(e) penalty for
large transfer pricing misstatements could be excused only if the taxpayer timely produced
contemporaneous documentation showing that it had reasonably applied a transfer pricing
methodology. This legislation was based on a Treasury proposal that was originally
developed in the course of working on the regulations under sections 482 and 6662.
As with the other two provisions, the intention was to change taxpayer behavior -- not
to impose the penalty in a large number of cases. This penalty is, in effect, an effort to
revitalize the self-assessment principle in the context of transfer pricing. Thus,
implementation of these penalty rules was linked with updating the substantive section 482
transfer pricing regulations, which were intended to facilitate taxpayers' application of
the arm's length standard.
Again, however, some taxpayers have expressed concern that the IRS will assert this
penalty in inappropriate cases. While it may be too early to claim much of a track record
here, we have already taken several steps to ensure that this does not occur. First, we
promptly issued proposed and temporary regulations to interpret the statutory
requirements. Although the basic structure of these regulations is likely to remain
unchanged, commentators have suggested a number of clarifications that we are considering
in order to alleviate specific concerns expressed by taxpayers. For example, we are
considering moving to the regulatory text a portion of the preamble that clarified the
scope of a taxpayer's obligations to consider alternative methodologies to the one they
actually applied. In addition, the Commissioner recently announced formation of a transfer
pricing Penalty Oversight Committee, consisting of both National Office and field
personnel, to ensure uniform application of the documentation requirements and uniform
interpretation of the reasonableness standard on a nationwide basis.
B. Issues Not Addressed by Recent Legislation. On balance, then, there has been some
notable improvement in each of the three information gathering issues affected by recent
legislation. However, as these issues have become less contentious, we have seen a number
of others emerge as the focal points of concern. In order to give you a fair picture of
the problems we face across the board, I want to discuss three of these issues as well.
1. IRS Access to Computer Software. The first issue that has recently come to the fore
involves the scope of the Service's authority to examine the computer software used to
prepare a taxpayer's return. While this issue may arise in either a domestic or
international context, it is perhaps most prominent with respect to the Software programs
that taxpayers use to calculate their foreign tax credits and related matters, including
the allocation of interest expense and computation of Subpart F income.
In appropriate situations, a review of this software -including both the actual
computer code and all related documentation -- may be necessary in order for the Service
to identify, analyze, and verify the legal analysis and/or the factual characterizations
that are the basis for the taxpayer's computation of its taxable income. Accordingly, the
Service may seek access to the software pursuant to its powers under section 7602.
As you may be aware, this issue is currently pending in one summons enforcement case.
And, while I would not comment on that particular situation, we think it is clear as a
general matter that the Service has the authority to verify the accuracy of taxpayers'
returns. Where that verification could be enhanced by an examination of the software used
to prepare the return -- for example, to determine whether the software is consistent with
Service position on the application of priority rules among foreign tax credit baskets --
we think the Service is entitled to that information. We simply do not believe taxpayers
can expect the Service to accept calculations that emerge from a black box as being
off-limits to reasonable scrutiny.
At the same time, however, we are aware that software producers have expressed concerns
that the software is valuable proprietary information and could be misused or
inadvertently disclosed by the IRS. I want to assure you that we understand and respect
those concerns. We are aware of our obligation to use this software only for the purposes
authorized by the Internal Revenue Code, and we intend to comply fully with that
obligation. Beyond that, we are open to discussing with any interested stockholders how
their disclosure concerns could be addressed in a manner consistent with the needs of tax
2. Scope of Attorney-Client Privilege and Work Product Doctrine in Transfer Pricing
Cases. A second old issue that is coming up in a new form is the scope of the attorney
client privilege and the attorney work product doctrine in the context of transfer pricing
As I mentioned previously, the recent changes to section 6662(e), which require
contemporaneous documentation of the taxpayer's transfer pricing analysis, were designed
to revitalize the self-assessment principle in the transfer pricing context. This means
that taxpayers are being asked to analyze how their prices comply with the arm's length
standard prior to filing a tax return, and to capture that analysis in contemporaneous
documentation that the Service can later review.
Despite the clear purpose of this legislation, we understand that some taxpayers are
being advised to retain legal counsel to prepare the required transfer pricing analysis.
We are told that taxpayers have been advised to retain counsel for this purpose (rather
than doing the work in-house or through other types of outside experts) because, in
addition to whatever substantive expertise counsel may provide, the analyses they prepare
will permit a taxpayer to Shield inconvenient information behind the attorney-client
privilege and the attorney work product doctrine.
To the extent this practice is actually developing, it would seem fundamentally at odds
with the whole thrust of the recent statutory changes in this area, and indeed, with the
basic philosophy of the new section 482 regulations themselves. Virtually everything
Congress has done in recent years to address transfer pricing problems has centered on
making sure that the Service has reasonable access to the necessary information. The
Section 482 regulations likewise reflect a fair and flexible approach that also depends on
the availability of relevant data.
We think there are serious doubts that claims of privilege in these circumstances will
be sustained by the courts in a broad range of cases -- particularly when the transfer
pricing studies are prepared for the purpose of filing the taxpayer's return, and when
some (but not all) of the studies are provided to the Service in support of the position
reflected on the return. I assure you that we are prepared to vigorously contest over
broad assertions of privilege in this area. But put aside the issue of whether claims of
privilege will be sustained or rejected by the courts in any given case. My point today is
that, as we enter a new era of transfer pricing practice, I believe taxpayers should ask
themselves whether, as a matter of corporate policy, this is way they want to approach
their information gathering obligations in the future.
3. Summons for Third-Party Comparable Data. The last longstanding issue that has
received renewed attention in recent months is the question of when the IRS will use its
summons power to obtain transfer pricing information from a third-party engaged in
transactions comparable to those of the taxpayer under audit. While we firmly believe that
section 7602 gives us the power to Summons this third-party comparable data, we have, as a
matter of administrative practice, been reluctant to exercise the full extent of that
power. Instead, we have traditionally relied on information available from public sources
and on the voluntary cooperation of third-parties.
In considering whether our policy for the future should be different from the past, we
have identified three distinct interests that must be considered:
- First, the IRS needs information from third-party comparables in order to apply the
arm's length standard;
- Second, the taxpayer under audit is entitled to know what information the IRS is using
to make an adjustment to the transfer prices reported on the return; and
- Third, the third-party has legitimate interests in protecting the confidentiality of
sensitive commercial or technical information and in not being subject to unreasonable
burden or expense.
As we reassess our policy, we are consulting with a number of interested parties,
including TEA. In fact, to follow-up on a question that was raised at your annual IRS
liaison meeting last fall, I understand that a group of your members, together with Tim
McCormally and Mary Lou Fahey from the TEI staff, recently met with the Service to discuss
this issue. I realize that TEI does not necessarily agree with our views on the underlying
legal merits, and I particularly appreciate your willingness to engage in a dialogue about
the issue to see if we can develop a workable resolution.
Based on this meeting, I understand that there is already a fair amount of common
ground. For example, one of the principal concerns expressed by TEI was that the taxpayer
under audit have access to whatever third-party data is being considered by the Service.
As a matter of principle, we agree that the taxpayer should see that data at an
appropriate point in the audit -though we may need to do more work to identify precisely
when that point is reached.
We also agree with TEI that the third-party has legitimate concerns, but we're not sure
that they should be given overriding weight in all cases. Perhaps we will need to consider
the balance case-by-case, rather than with an across the board rule. For example, where
the data is old, the prospect of commercial harm is limited, and the burden of production
is reasonable, the balance might shift in favor of the Service's needs for the data.
Moreover, we may be able to agree on procedures that will minimize either the harm from
disclosure or the burden of production on the third-party. In any event, we are actively
studying the issue, and will continue to consult with stockholders as we move toward a
III. Building Mutual Expectations
I hope this brief survey of current issues demonstrates at
least the first part of my original thesis: that information - gathering issues in the
context of international tax compliance represent an increasingly important focus for tax
administration, with a high potential for disagreement. Unless we develop a reasonable
understanding of what taxpayers and the IRS can expect from each other, there will be
continuing confusion and controversy, more litigation, and possibly more legislation. Even
the issues that have had recent legislative attention have not yet settled into the type
of comfortable routine that is typical of most purely domestic information gathering
But why do I think we can do any better? Isn't this type of controversy inevitable when
there is so much at stake? Or Should we expect that the system will settle down, in time,
perhaps when we've had as much experience in the international context as we've had over
the past 80 years in the domestic arena.
I don't think we need to wait that long. I say that based on our experience in the APA
program. In that context, we deal with exactly the same type of highly factual,
multi-million dollar transfer pricing issues that we face on audit. Yet, somehow,
information gathering in the APA process has worked remarkably well.
Taxpayers who seek an APA know they will be required to produce the necessary
information on a timely basis -- and they do. The Service participants understand that
requests for unnecessary information impose unfair burden and expense on taxpayers -- and
they work with the taxpayers to limit the scope of what is requested. Moreover, this
experience has been true not only in the early years of the APA program when it was
largely centered in the National Office, but it has continued to the present day, with
significantly expanded field participation -- participation by exactly the same
international examiners and agents who are involved in audits. It's not simply a matter of
one process being focused on the future while the other looks to the past. We are finding
an increasing number of situations in which taxpayers are willing to apply the methodology
developed in the course of an APA request to resolve their liability for past years as
Being an optimist, I think we should assume that we can extend our experience in the
APA process to other aspects of international information gathering. I believe that what
we need to make that leap is a common understanding of what to expect from each other --
just the sort of mutual understanding that has characterized the APA process. With a view
to building those mutual expectations, let me Suggest a few basic principles from the
point of view of each side.
From the Service perspective, we should expect that:
- Taxpayers will maintain sufficient records to establish the accuracy of the income
reported on U.S. tax returns;
- When requested by the Service, taxpayers will provide those records on a timely basis,
and in a form that makes them reasonably usable; and
- IRS requests for information will be interpreted reasonably, according to the type of
information requested and the purpose for which the request is made.
From the other side of the table, taxpayers should expect that:
- The Service will provide reasonable guidance as to what records must be maintained,
either by published regulations, rulings or procedures or through District Director
agreements or similar taxpayer-specific advice;
- IRS requests for information will be sufficiently specific to guide the taxpayer's
search and production, will take into account the burden of producing the records, and
will be explained or clarified as necessary to ensure the taxpayer understands the
intended scope of the request; and
- The Service will challenge tactics designed to deny it legitimate access to relevant
information, so that taxpayers who use such tactics will not be able to gain a competitive
advantage relative to taxpayers who provide requested information properly.
I want to be clear that I offer these principles as a starting point for dialogue,
rather than as clear picture of what the final answer should include. While I said I'm an
optimist, I'm also enough of a realist to know that resolving the issues I've talked about
today about won't be as simple as making a list of rules for what constitutes proper
information gathering etiquette.
At the same time, I think there is some reason to hope that we might be able to make
progress today to a degree that would not have been possible just a few years ago. And
what's the alternative -- years of controversy? Given the choice, I think its worth a try
to see if we can develop an international information-gathering practice that is as
sensible, efficient and effective on audit as it is in the APA process.
Previous | Next
1995 IRS News Releases | News Releases Main | Home