These are not the first interesting situations to be presented to the world. Two months ago I shared a step-sibling puzzle, after having explained the changes in a prophetically-named posting, Redefining Children (at least in the Tax World).
Here is one hypothetical presented by the NAEA:
Twin nine-year old children of deceased parents, who live with their adult cousin for the entire year and are fully supported and cared for by the cousin, cannot be claimed as dependents by the cousin. Under the new rules, this cousin cannot claim the two children as qualifying relatives because the children meet the definition of a qualifying child with respect to one another. The problem does not exist if there is only one such child living with a cousin, so if each twin were to be taken in by a different cousin, they could be claimed as qualifying relative dependentsNeither twin can be a qualifying child of the adult cousin. Why? Neither twin is a child, descendant, sibling, step-sibling, nephew, niece, or child of a step-sibling of the adult cousin. Thus, neither twin meets the relationship test for qualifying child status, even though the adult cousin meets the other three tests of abode, age, and support for each twin. Each twin has the same principal place of abode as the adult cousin, each meets the age requirements by being under 19, and each has not provided more than one-half of his or her own support.
Can each twin be a qualifying relative of the adult cousin? An individual is a qualifying relative if the individual satisfies four tests as to the taxpayer seeking the dependency exemption: relationship, gross income, support, and non-qualifying child. The relationship test is met because each twin is someone who is not the adult cousin's spouse and who has the same principal place of abode as does the adult cousin. The gross income test presumably is met because the facts of the hypothetical suggest that they have none, being fully supported by the adult cousin. The support test is met because the adult cousin provides more than half, in this instance all, of the support of each twin. What about the non-qualifying child test?
The non-qualifying child test requires that the individual for whom the taxpayer seeks a dependency exemption deduction not be the qualifying child of the taxpayer or of any other taxpayer. It already has been demonstrated that neither twin is the qualifying child of the adult cousin. Is either twin the qualifying child of another taxpayer? According to the statutory language, no. There are no other taxpayers in the picture aside from the adult cousin.
The problem, however, is that the IRS, in its publications, changes the language of the statute. To quote from the NAEA letter to the Commissioner: "In IRS publications this has been translated into language such as the qualifying child of anyone else or the qualifying child of another person." This is a HUGE difference. If the test is that the person not be the qualifying child of anyone else, the twins are not the qualifying relative of the adult cousin because they are qualifying children of each other. Each twin is a sibling of the other, thus satisfying the relationship test for qualifying child. Each twin meets the other three tests (abode, age, support) because each twin has the same principal place of abode as the other twin, each meets the age requirements by being under 19, and each has not provided more than one-half of his or her own support.
In effect, the NAEA letter is asking the Commissioner, "What happened? On what grounds did someone drafting a publication change the word taxpayer to "anyone else or another person"? Not every person is a taxpayer. Not every person who is an "anyone else" is a taxpayer. Under Code section 7701(a)14), a taxpayer is "any person subject to any internal revenue tax." In contrast, section 7701(a)(1) defines "person" as "construed to mean and include an individual, a trust, estate, partnership, association, company, or corporation." Person and taxpayer are two different concepts, and using one term to mean the other is unwise.
Not all the problems, though, are of IRS making. Consider this hypothetical from the NAEA:
Twin nineteen-year old brothers live together in their home and attend school full-time. Their parents are deceased. The brothers do not provide more than half of their own support. Although they have part time jobs and earn about $5,000 annually, their principal support comes from their aunts and uncles who together contribute about $25,000 per brother towards their household and college expenses. The aunts and uncles do not live with the brothers. Each brother meets the definition of a qualifying child with respect to the other. Putting the dependency rules together with this, if each twin is able to claim the other as a dependent, it means that the other one cannot because a dependent cannot have dependents. However, since neither can be claimed, it means they can have dependents. This loop continues endlessly – we now have the qualifying child paradox.Let's call the brothers A and B. A is the qualifying child of B. Why? A is the brother of B, A has the same principal place of abode as does B, A meets the age test by being a full-time student under 24, and A does not provide more than half of his own support. B is the qualifying child of A. Why? B is the brother of A, B has the same principal place of abode as does A, B meets the age test by being a full-time student under 24, and B does not provide more than half of his own support. However, even though A is the qualifying child of B, A cannot be the dependent of B because B is a qualifying child of A, and thus would be the dependent of A but for the fact that because A is the qualifying child of B, B would be a dependent of A. It is a classic paradox. Though there are many special rules in section 152, none addresses this puzzle.
However, one of the problems described by the NAEA is not a problem. Consider the NAEA's hypothetical:
Mom, dad, Alice (14), and Joe (22) live in the family house. Mom and dad file a joint return with an AGI of $400,000. Since Alice is a qualifying child of mom and dad, they could claim her as a dependent but would receive no tax benefit as their personal exemptions are phased out and the child tax credit would not be available to them. Joe is not a full-time student and his only income is a W-2 with $15,000 in wages. Under §152, Alice is a qualifying child of Joe, so he claims her as a dependent and thus gets the child tax credit and yes, even the earned income tax credit. Assuming Joe had no tax withheld, he goes from a balance due of $683 to a refund of $3,158.I agree that Alice is the qualifying child of Mom and Dad. She is their child. She has the same principal place of abode as they do. She is under 19. She does not provide more than half of her own support. Alice also appears to be the qualifying child of Joe. She is his sibling. She has the same principal place of abode as he does. She is under 19. She does not provide more than half of her own support. But in this instance the Code provides a rule to break the impasse. Under section 152(c)(4)(A), an individual who may be claimed as a qualifying child by two or more taxpayers is treated as the qualifying child of the individual's parent if one of the taxpayers claiming the individual as a qualifying child is the individual's parent. The fact that the amount of the exemption for the parents is zero because their AGI is high enough to trigger total phase-out of the exemption amount does not change the definition of qualifying child.
Technical amendments to the 2004 legislation that changed the dependency definitions are awaiting action by Congress, but these amendments do not address the problems raised by the NAEA. The ABA Section of Taxation has identified yet another problem for which it suggests additional technical amendments.
There are lessons to be learned from this analysis. As is the case with computer programs, the more complex the provision, the higher the chances for a crash. The faster the writing, the greater the chance of error. The more authors, the higher the number of inconsistencies and paradoxes. The less review and testing, the faster it is rushed to market, the larger the number of snags.
The underlying problem is that the definitions in section 152 are being used for too many purposes. It's not that there ought to be multiple definitions. It's that there should be far fewer purposes. Strip the tax law of the wide array of credits, deductions, special provisions, and other attempts to create a different tax law for each taxpayer, and the complexity is reduced. Rather than an earned income tax credit, why not eliminate income taxes on people with adjusted gross income under the poverty level? Rather than trying to figure out who claims who as a dependent, why not assign each person an exemption amount that the person can choose to use on his or her own return, or transfer or sell to one other person (presumably a taxpayer who could make use of it), similar to the manner in which pollution credits are traded? It surely would be easier than the current approach, which has generated thousands of cases and hundreds of thousands of audit adjustments over the years.
In the meantime, let's hope the IRS responds quickly and sensibly to the NAEA's letter. And I might ask those folks to write some exam questions for my students!