Statement of Nina E. Olson, National Taxpayer Advocate,
Internal Revenue Service

Testimony Before the Subcommittee on Oversight
of the House Committee on Ways and Means

Hearing on the IRS National Taxpayer Advocate Annual Report and the IRS Oversight Board Annual Report

February 28, 2002

Thank you for providing me the opportunity to appear before you today to discuss the National Taxpayer Advocate’s 2001 Annual Report to Congress.  The Report represents the work of many dedicated Taxpayer Advocate Service employees, and it is my distinct honor to come before you today to discuss their work.  I am truly grateful for and proud of their efforts.

Today marks the completion of my first year of service as the National Taxpayer Advocate (NTA), and it has been a very interesting year, indeed.  The National Taxpayer Advocate is fortunate, in that both the position and the elements of her two Annual Reports to Congress are established and described in the Internal Revenue Code.  Section 7803(c) clearly establishes that the Annual Reports are considered a direct communication between the National Taxpayer Advocate and Congress.  In order to preserve the NTA’s independence, Congress mandated that the Reports be delivered to it prior to delivery to the Secretary of the Treasury or the IRS Commissioner.

While the focus of my remarks today is tax simplification, I would like to make a few preliminary comments about the Annual Report’s methodology.  As you know, when I first assumed this position, I met with many of the members of this subcommittee and the Senate Finance Committee in order to learn, more specifically, what Congress expected from the Annual Report to Congress.  From these discussions I learned that you view the Annual Report as providing certain information that will help you resolve taxpayer problems either through legislation or by oversight of the Internal Revenue Service.  You view the Taxpayer Advocate Service (TAS) as a credible source of this information, given our unique role within the IRS as an independent and impartial advocate for taxpayers who are trying to resolve their tax problems.

The independence, impartiality and confidentiality that the Taxpayer Advocate Service offers is critical to our ability to deliver this information to you.  Without these assurances, taxpayers will not come to TAS to discuss their problems.  In my upcoming June 30^th Report to Congress, I will describe in detail the efforts we have made to enhance independence, impartiality, and confidentiality in TAS’ operations.  

With respect to the Annual Report to Congress, I am pleased to report to you today that I will soon be recruiting and hiring two Senior Advisors to the National Taxpayer Advocate.  One Senior Advisor will serve as an advisor for legal matters pertaining to the Annual Report to Congress, including assisting my employees with development of legislative proposals.

The second Senior Advisor will oversee all research activities and establish a research agenda for the Taxpayer Advocate Service, including the Annual Report to Congress.  I hope to fill these positions in the next four months.  Next year’s Annual Report will be enhanced by these additions to the NTA’s staff. 

I view the information provided in the Annual Report as a continuum.  First, we identify taxpayer problems.  Our focus is the administration of the tax laws, not the underlying policy considerations.  This year, we have developed two lists of “Top 23” taxpayer problems – one derived directly from the TAS database of cases and one that reflects TAS employees’ judgement, as informed by their conversations and outreach with individual and business taxpayers, practitioners, and other IRS employees.  Some of these problems surface again in our analysis of the ten most litigated issues.

Many of the top taxpayer problems can be addressed administratively by the Service.  Indeed, in many instances the Service has detailed, long-term plans for addressing these problems.  Some solutions require issuance of guidance, or a different allocation of resources, or even greater resources.  But some problems cannot be fully resolved unless there is legislative change.

In making our key legislative recommendations, we narrowed our focus to issues that have the potential to achieve simplification or reduce taxpayer or administrative burden with respect to a significant number of taxpayers.  We have also focused on proposals that are “actionable” – that is, with the exception of one proposal, they do not require major reform of the Code and can be enacted, in one form or another, immediately. 

For each recommendation, we describe the problem, illustrate it with examples derived from TAS cases (some of which were referred by Congressional offices), and provide a description of current law and an explanation of our proposed legislative change.  Although others have suggested some version of these proposals, our recommendations reflect the viewpoint, experience, and pragmatism of TAS employees who wrestle with the underlying problems daily in their case inventories.

The Constituency for Simplification

During the past year, I have spoken at many taxpayer, tax practitioner, and IRS employee meetings about some of our legislative proposals and the complexity of the Internal Revenue Code.  All too often, I have heard people say that tax simplification will never occur because there is “no constituency for simplification.”  I am not sure what this phrase actually means. 

If the lack of constituency refers to the fact that individual taxpayers do not have special interest groups speaking on their behalf or are not flooding Congressional switchboards demanding tax simplification, the statement may be true.  But if, for example, we consider the number of taxpayers positively affected by developing a uniform definition of a “child” for purposes of the Internal Revenue Code, the statement is false.  Making uniform the provisions that relate to family status would have simplified tax computations for taxpayers filing 44 million returns for tax year 2001.  In a universe comprised of approximately 130 million individual returns, this is a significant constituency.

Family Status Issues

Family status is most taxpayers’ point of entry into the tax system.  And yet, Treasury Secretary O’Neill has noted, only within the confines of the Internal Revenue Code would one find five different definitions of a “child.”[1]  On a Form 1040, before a taxpayer even enters one dollar of wage or interest income, he or she must determine whether his or her child qualifies as a dependent or entitles the taxpayer to claim head of household status or be considered not married.[2]   These provisions apply different tests for eligibility with respect to the same child.  Further along in the form, the taxpayer must look up other definitions or limitations for the child tax credit, the child and dependent care credit, the earned income credit, and various education tax incentives.

At the end of the day, the taxpayer or his or her preparer is woefully confused about a subject that is fundamental to one’s economic and tax existence.  Each element of variance increases the likelihood of an honest mistake.  In some instances, the differences provide opportunities for deceit.  In all instances, the differences increase the administrative burden on the IRS to fairly implement the tax laws.  Further, the IRS must conduct intrusive inquiries into the lives of taxpayers in order to monitor compliance with these non-uniform provisions.

The Taxpayer Advocate Service’s recommendation is really very simple.  However, as with all simplification, the details require careful attention.  We propose that for each of six family status provisions we adopt a common definition of “qualifying child.”  Our proposed definition focuses on a type of familial relationship about which we can be reasonably certain, once certain tests are met, that we have correctly awarded a tax benefit, at least with regard to the “child” element.  Thus, we propose a uniform relationship, place of abode, and age test.  Once a taxpayer has established that he or she has a child (biological, adoptive, or step) or a descendant, we need only determine if that child (or descendant) lived with the taxpayer for more than half of the tax year and if that child (or descendant) is of the appropriate age.

We suggest that this definition be placed in the definitional sections of the Code so that it can be easily cross-referenced in the current tax provisions.  As new family status provisions are enacted, this definition will become the norm.

This definition has several advantages over the current state of the law.  First, both relationship and age are relatively easy for a taxpayer to demonstrate in an audit situation.  Second, it is easier for the taxpayer to prove that the child lived with him or her for more than half the year than it is for the taxpayer to prove that he or she provided more than half the support of the child or more than half the cost of maintaining a home in which the taxpayer and the child lived for more than half the year.  Finally, by limiting these tests to the proposed family relationship grouping (parent/child/descendant), we significantly lessen the chances that the “wrong” taxpayer would receive the tax benefits.

Our proposal differs from the Joint Committee on Taxation’s proposal that the uniform definition of a qualifying child completely replace the support and household maintenance tests.  We do not propose substituting this relationship/abode/age test for all family relationships.  We believe that there are other family and household arrangements that should qualify for the various family status benefits.  Yet those family and household arrangements require the Service to make additional inquiry into the household circumstances in order to ensure that the special tax benefits are awarded to the appropriate party.

We also propose to extend a version of the recently amended EITC “tie-breaker” rule to the uniform definition of a qualifying child.  That is, where two or more taxpayers having a qualifying relationship with the child live in the same household, the parental relationship would trump others.  A grandmother, mother, and child may live together, and the mother may not want to claim the child as dependent or for head of household status.  That is, the mother wishes to “opt-out” of the uniform definition of a qualifying child safe harbor.  In these situations, as under current EITC law, the mother simply does not claim the child on her return and the grandmother will be able to claim her grandchild.  In other household arrangements outside the relationship safe harbor, the cohabitator would still have to satisfy the current statutory tests, including the appropriate support or household maintenance tests.

Our sensitivity to non-nuclear family arrangements is best evidenced in our proposed expansion of the definition of “foster child”.  Under present law, we have at least three different definitions of foster child – one for dependency exemptions, one for head of household status, and one for the Earned Income Tax Credit.  We propose establishing a uniform foster child definition between these provisions.  The foster child must meet either a relationship or custodial test and must also live with the taxpayer for more than half the year. 

In order to provide more certainty sooner in the administrative process, we propose a rather narrow relationship test.  We define the term “child” as an individual who bears on of the following relationships to the taxpayer:

•        A son/daughter or descendent (e.g. grandchild or great-grandchild);

•        A stepson/stepdaughter or descendent (e.g., step-grandchild or step-great-grandchild);

•        An adopted child; or

•        An eligible foster child.

We also propose that the taxpayer could demonstrate the “foster” relationship by proving that the child was placed in his or her home by an authorized placement agency as under current law or by showing custodianship through some other means.  We contemplate that the law would permit a taxpayer to claim a foster child when the taxpayer could prove that some other federal, state, or local governmental agency (or its agent) had made a determination that the child lived with the taxpayer.  Depending on the circumstances, this “certification” could take the form of food stamp certification, of school placement, or of an award of temporary custody by a family court. 

The key feature of this proposal is that the IRS would not have to make intrusive inquiries.  Rather, it is relying on a set of verified third-party statements that have an acceptable degree of reliability.  This approach – accepting reliable third-party documentation -- should be extended administratively to all family status audits, including those involving the Earned Income Tax Credit.  Such an audit approach removes the government from having to pry into the most intimate affairs of its citizens without due cause.  I have stated elsewhere how humiliating such an audit can be.  That these audits are routinely visited upon the poorest and least represented portion of our population represents a fundamental flaw in the design of the law as well as in our tax administration system.  Middle class and wealthy taxpayers would not tolerate such intrusions.

Individual Alternative Minimum Tax

Increasingly, middle class taxpayers face a different type of problem – the individual alternative minimum tax (the AMT).  A minimum tax was first introduced into the Internal Revenue Code in 1969 as a means to tax high income taxpayers who had gross income and yet paid no taxes because of various provisions.  The first minimum tax listed 14 preferences that were to be added back in to taxable income.  In 1978, the minimum tax was expanded to include additional deductions, exemptions, and credits; all typically claimed by middle class taxpayers. 

The design of the minimum tax has not achieved its original goals.  The number of high-gross income earners who pay no tax has increased slightly since 1978.  The Alternative Minimum Tax does enhance revenue, largely by subjecting increasing numbers of middle-income taxpayers to its requirements.  These unintended targets are drawn into the alternative tax system because they have large families, live in high-tax cities or states, or earn their income in a variety of ways other than direct compensation, such as stock options.  Today, 1.5 million taxpayers pay the AMT.  The Joint Committee projects that 35.5 million taxpayers will be subject to the AMT by the year 2010.[3]

As a tax administrator, I view these numbers with concern.   Many of the taxpayers do not even know that an alternative minimum tax system exists.  Thus, in 2010 the Service could be faced with sending out 35.5 million letters to taxpayers who believed they filed a perfectly accurate tax return and who will learn that they did not. 

To avoid this situation, we propose three alternatives to the recommendation that the individual AMT be repealed.  I personally favor the alternative proposal of establishing a gross income threshold, below which a taxpayer will not be subject to the AMT.  This proposal has the advantage of providing relief to those taxpayers who must dedicate 12 hours to simply determining whether they are subject to the tax at all.  Under the gross income proposal, a taxpayer need only look to a line on his or her return and check it against the threshold for his or her filing status.  The taxpayer will learn from the face of his or her return whether he or she is subject to the AMT.  This proposal also provides relief to middle income taxpayers who would otherwise be subject to the AMT because they exercised incentive stock options.  If their gross income was below the threshold, the exercise of these options would not be subject to the AMT.

Worker Classification

One of the most difficult areas of law within which to strike a balance between competing interests is worker classification.  This difficulty is in large part attributable to the variety of workers and type of work that would be affected by any guidelines.

I believe that both the tax law and the tax administration system can do a far better job addressing this issue.  Most businesses and their workers are capable of reaching agreement on employment arrangements by themselves.  There are, however, groups of workers who have limited ability to understand and pay self-employment tax.

One such group of workers is known as home-based service workers.  These workers provide home care and services for persons who are disabled or elderly and who need assistance in daily living activities.  The workers themselves are often low income.  Usually these workers are paid with some combination of state and federal funds.  Most commentators agree that, under common law, the workers are employees of the service recipients.  And yet the service recipients are usually not in the position of issuing W-2 forms or 1099-MISC statements.  In some instances, states or localities undertake the issuance of 1099 forms; some states and localities actually withhold FICA tax from the workers’ pay and issue them W-2 forms.

In the case of home-based service workers, we have proposed deeming the payor agency to be the “employer” solely for purposes of tax withholding and income reporting.  Alternatively, these workers could be deemed independent contractors but subject to a flat withholding rate sufficient to pay self-employment tax.  Either approach addresses the critical failures of the current system – where workers receive a 1099 form and have no tax withheld; where they are unable to pay a lump sum of quarterly or year-end taxes; and where the Service ends up placing the workers in “currently not collectible” status because the workers cannot afford to pay a pyramiding tax liability.

Removing the burden of tax reporting and withholding from the service recipients relieves them of the need to file a separate Schedule H, Household Employment Taxes.  This is appropriate because most service beneficiaries do not have a filing requirement.

Taxpayer Rights

The National Taxpayer Advocate’s taxpayer rights proposals include provisions relating to partial payment installment agreements; the return of levy proceeds under IRC section 6343(b) and (d); the realignment of underpayment interest rates simultaneous with the repeal of the late payment penalty; and a first-time penalty waiver (also known as the “one time stupid act” waiver).  Each of these proposals is discussed in detail in the Annual Report.

I would like to identify, here, two themes that run through these taxpayer rights proposals.  The first is that it is very difficult, in practice, in a tax system that relies on voluntary compliance to draft laws and develop procedures that differentiate between habitual or chronic offenders and taxpayers who simply make a mistake.  If the law is written so broadly that IRS personnel have significant discretion in implementation, taxpayers may experience disparate results and even face abuse of that discretion.  On the other hand, if we do not permit IRS personnel to exercise common sense and good judgement – and give them the tools to help honest taxpayers who have made a mistake get back into compliance with the system – taxpayers will feel that they are caught within an “Alice in Wonderland” or Kafka-esque system.

The second consideration is how to remedy situations where the IRS has done something wrong.  In many instances, it takes a while for either a taxpayer or the IRS to figure out that the IRS was wrong in taking a certain collection action.  The usual period for protesting such actions may have passed – in the case of levies, that period is nine months from the date of levy.  In these instances, should the Commissioner be permitted to grant relief?  Our proposal regarding the return of levy proceeds is our first attempt to address this issue; we were unable to develop a proposal that was broader in application. 

Joint and Several Liability

The proposal to eliminate joint and several liability on jointly filed individual income tax returns grew from the awareness that “innocent spouse” relief under IRC section 6015 threatens to become an exception that swallows the rule.  Almost 1 percent of the Service’s workforce in 2001 processed innocent spouse cases.  I am skeptical of the IRS projections that show the number of full-time equivalents (FTEs) dedicated to innocent spouse will drop significantly over the next two years.  As current case law becomes part of the mainstream and future case law develops under this provision, I believe more taxpayers will request relief.

It is for these reasons that the Taxpayer Advocate Service took the bold step of proposing that taxpayers receive all the benefits of a joint return without incurring joint and several liability for any tax debt attributable to that return.  We developed a prototype Form 1040 M on which married taxpayers can allocate between themselves their items of income, deduction, adjustments, credits, and payments.  If such an election is made, it will be dispositive for any later adjustments relating to those items.  Under this proposal, “innocent spouse” relief will no longer be available to married taxpayers.  However, taxpayers will still be able to challenge even an allocated liability under the common law defense of duress.

It is our experience that married couples in fact do allocate items between themselves, even if not consciously.  The advantage of our proposal is that the allocation occurs at the front end of the filing process, when the information is fresh in everyone’s minds, rather than years after the filing, when documentation is missing and the married couple has separated.  We also acknowledge that it may be simpler to require the allocation on joint returns, rather than permitting an election to allocate.  Since we were addressing this problem from an administrator’s point of view, we proposed the narrower solution to the administrative problem.  We believe that requiring married couples to allocate is a policy decision and beyond the scope of our report.

We recognize that this proposal is the one key legislative recommendation that cannot be enacted overnight.  It amounts to a significant change to our filing system.  We intend to explore the issue of taxation of married couples from several different perspectives over the next few years.  We will review state community property laws and their impact on federal tax law as well as the treatment of tenancy by the entireties property.  Each of these topics raise the truly thorny question of whether there should be, in this arena, a subset of property rights that are defined by the Internal Revenue Code and that trump state property law for purposes of the Code.  We are taking a long-term view and hope to address a different aspect each year, which will allow for a full development and discussion of the issues.

The Causes of Tax Complexity

Clearly, no one in Congress or in the IRS sat down and said, “let’s try to make the Code so complicated that no one will ever be able to figure out even basic provisions like family status by himself.”  Tax complexity creeps up on us – we try to eliminate a perceived abuse (as with the minimum tax in 1969), or carve out relief for one special set of circumstances (as with innocent spouse).  We keep adding exceptions, limitations, and rules as other inequities reveal themselves.  Certainly revenue considerations play a role.  Sometimes we actually think we’ve solved a problem – as with the dependency exemption between divorced or separated parents under IRC section 152(e) in 1984 – only to find that forces outside the federal tax universe – here, state courts’ interpretation of their domestic relations jurisdiction – have foiled all our best efforts.

Tax simplification is anything but simple.  It requires discipline and persistence.  It really requires us to admit that we tried a solution that seemed the best at the time, but that it turned out to need some more tinkering.  Perhaps over time, as with the AMT, we will discover that our first solution missed the mark completely.  We need to keep an open mind, view these provisions as controlled experiments, and learn from our misjudgements.  Above all, we must have a firm resolve that while we are simplifying current law we do not enact new laws that further complicate the Code. 

The National Taxpayer Advocate’s Annual Report to Congress is one source of information about complexity and simplification.  We have limited our proposals to provisions about which we have experience and which we believe are “actionable” and will significantly simplify matters for many taxpayers.  I hope you find this Report helpful.  On behalf of all the Taxpayer Advocate Service employees, I thank you for the opportunity to report on these matters, both in writing and before you in person today.

[1] O’Neill Claims Simplification, Stimulus Top Priorities, 2002 TNT 36-1

[2] IRC § 7703(b)

[3] U.S. Congress, Joint Committee on Taxation, Estimated Effects of the Conference Agreement for H.R. 1836[1](JCX-51-01) p.8.

 MOST SERIOUS PROBLEMS
ENCOUNTERED BY TAXPAYERS

1 Access to Customer Service Toll-Free Telephone Service

2 Multiple Definitions Of “Qualifying Child”

3 Determining Earned Income Tax Credit (EITC) Eligibility

4 Answers to Questions on Customer Service Toll-Free Lines

5 Documenting Earned Income Tax Credit (EITC) Eligibility

6 Refund Inquiries

7 Earned Income Tax Credit (EITC) Examinations

8 Understanding Estimated Tax Payments

9 Explanations on Math Error Notices

10 Processing Claims for Refund

11 Recertification for Earned Income Tax Credit (EITC)

12 Computing Income Tax Using Schedule D (Capital Gains & Losses)

13 Awareness and Understanding Federal Tax Deposits Requirements

14 Obtaining Employer Identification Numbers (EINs)

15 Misapplied Payments

16 Lack of Access to Free Tax Preparation for Low-Income Taxpayers

17 Processing Offer in Compromise Applications

18 Computing the Alternative Minimum Tax (AMT)

19 Determination and Notification of Revised Tax Liability

20 Cost of Electronic Filing for Low-Income Taxpayers

21 Automated Underreporter (AUR) Tax Assessments

22 Status of Innocent Spouse Claims

23 Delay in Receiving Requested Documents