Note: The Tax Court has migrated to a different operational internet platform. As of last Friday, any opinions (if any) issued by the Court since it closed its old platform in late November, are not accessible. Further, older opinions are also not accessible, unless another website (such as leagle.com or casetext.com) captured a copy before the old platform closed. That is why I am unable to provide a link to this week's case. If any reader has public link to the opinion I would be grateful.
A fundamental concept I teach my tax students is the idea of control. Taxpayers who engage in schemes where they ostensibly never touch a payment but in reality control its disposition often cannot escape taxation. In Brett John Ball v. Commissioner, T.C. Memo. 2020-152 (Nov. 10, 2020) (Judge Halpern), the taxpayer caused his self-directed IRA to distribute money to a wholly owned LLC, then caused the LLC to issue short-term loans to real estate entities. When the loans were repaid, Mr. Ball re-deposited the money into the IRA. The taxpayer had some decent arguments on why he should not have to report the IRA distributions as income. Judge Halpern rejected those arguments, teaching us a lesson about how much control is too much control.
Law: The Importance of Control
The Supreme Court stated the proposition plainly in Helvering v. Horst, 311 U.S. 112, 118 (1940): “The power to dispose of income is the equivalent of ownership of it.” The issue of control is a recurring one in tax law.
For example, a taxpayer’s control over disposition of income is important when deciding assignment of income issues. In Horst, a father owned corporate bonds that had interest coupons attached. He tore off the interest coupons and gave them to his son who received the interest payments as they became due. The Supreme Court held that the payments made to the son were gross income to the father in large part because of the control the father exercised in directing the interest income to his son. Amen to that.
The concept of control also underlies the doctrine of constructive receipt. Generally, cash method taxpayers must report only that income which they actually receive during the yearly accounting period. §451(a). The doctrine of constructive receipt is an exception to that general rule. The thrust of the doctrine is that if a taxpayer (1) has a right to money and (2) the payor was ready, willing, and able to make the payment, then (3) the taxpayer must report that amount as income when the failure to receive the money is simply the exercise of the taxpayer's own control. In contrast, "income is not constructively received if the taxpayer's control of its receipt is subject to substantial limitations or restrictions.” Treas. Reg. 1.451-2(a)
A third example where control plays an important role is the claim of right doctrine. North Am. Oil Consolidated v. Burnet, 286 U.S. 417, 424 (1932)(“If a taxpayer receives earnings under a claim of right and without restriction as to its disposition, he has received income which he is required to return, even though it may still be claimed that he is not entitled to retain the money, and even though he may still be adjudged liable to restore its equivalent.”). To avoid application of this doctrine a taxpayer must show that the receipt of the income was accompanied by a fixed obligation---fixed meaning beyond the taxpayer’s control---and the taxpayer had made provisions to comply with that fixed obligation. Thus, when a taxpayer received payments but a portion of those payments were obligated to be returned as illegal kickbacks, the Tax Court held the kickback money was not income to the taxpayer. Shaara v. Commissioner, T.C. Memo.1980–247. For a good discussion see Nordberg v. Commissioner of Internal Revenue, 79 T.C. 655 (1982).
Control is not an on/off concept. It’s a continuum concept. One sees that in the grantor trust rules in §§671-679. Those sections provide that certain powers a grantor retains over trust property will cause the grantor to be deemed to be the owner for income tax purposes of some or all of the trust property. Grantor trusts are a well-used tool in the estate planning toolbox because what may be defective for income tax purposes is perfectly effective to remove the assets from the grantor’s estate. It all depends on the degree of control. For a short lucid discussion, see Roger Russell, Inside the Intentionally Defective Grantor Trust, Accounting Today (June 30, 2020).
Degrees of control has also been important in the case law on taxation of IRA distributions. Let’s take a quick look.
Law: SEP-IRA Distributions
Simplified Employee Pension Plans (SEP) are a special type of Individual Retirement Accounts (IRA) that can be created by self-employed taxpayers under §408(k). One of the main requirements for all types of IRAs is that the taxpayer must give up direct control of their funds to a custodian who holds the funds in a fiduciary capacity. Custodians may be banks or investment companies such as J.P Morgan, Vanguard, Fidelity, or the like. Here’s one web-site’s list of top-rated custodians.
Taxpayers do retain control over how the retirement funds are invested. That control, however, is subject to (a) applicable statutory limitations, both in §408 and other statutes (e.g. §4975), and (b) contractual limitations created by the chosen custodian.
Distribution rules for SEP-IRAs are similar to those for traditional IRAs, notably: (1) distributions must be reported as gross income governed by the annuity rules in §72. §408(d)(1); (2) distributions will have a zero basis because the contributions are deductible. Treas. Reg. 1.408-4(a); and (3) SEP-IRAs are subject to the same early distribution penalty rules as traditional IRAs. Treas. Reg. 1.408-1(c)(6). Basically that means taxpayers who take a distribution before age 59.5 pay both tax on the earnings and a 10% penalty, unless the distribution is for a qualified purpose.
Taxpayers who seek to maximize their control over their IRA investment choices are attracted to what are popularly known as a “self-directed” IRAs. I find no statutory basis for the term---perhaps I am just missing something and a more knowledgeable reader can clarify in a comment. But the term does not strike me as accurate because all IRA’s are self-directed to the degree permitted by statute and custodial contracts. From what I can tell the term “self-directed IRA” is kind of a code phrase to identify IRAs held by very permissive custodians, sometimes so permissive that the SEC has this webpage warning taxpayers about them.
When a taxpayer wants to invest their IRA in an asset that the IRA custodian either will not or cannot invest in directly, taxpayers play a dangerous game whereby they withdraw funds from the IRA, purchase the asset they want, then put it back into the IRA. The game is to get an asset under the IRA tax shield while dodging the distribution rules and early distribution penalties.
Sometimes taxpayers win the game. In McGaugh v. Commissioner, T.C. Memo 2016-28, aff'd 860 F.3d 1014 (7th Cir. 2017), Mr. McGaugh wanted his IRA custodian (Merrill Lynch) to buy stock in a certain non-publicly traded company. Merrill Lynch refused to make a direct purchase of the stock. Mr. McGaugh worked around the refusal by directing Merrill Lynch to wire $50,000 to the company whose stock he wanted, and then directed the company to send the stock certificate back to Merrill Lynch. While he thus exercised control over the selection of the investment, the Tax Court found that this scheme did not create a taxable distribution to Mr. McGaugh because he was, in effect, simply acting as a conduit to effectuate an asset purchase that Merrill Lynch could or would not do directly. Judge Gustafson used this analogy to the claim of right doctrine: “We accept as sound law the rule that a taxpayer need not treat as income moneys which he did not receive under a claim of right, which were not his to keep, and which he was required to transmit to someone else as a mere conduit.” Op. at 10.
The IRS pointed out, sensibly enough, that Mr. McGaugh was indeed exercising considerable control over the use of the funds. To this Judge Gustafson replied: “The owner of an IRA is entitled to direct the investment of the funds without forfeiting the tax benefits of an IRA. Even acknowledging that Mr. McGaugh pulled all the strings, it remains true that the funds the IRA released went straight to the investment and resulted in the stock shares' being issued straight to the IRA.” Op. at 9.
Sometimes, however, taxpayers overplay their hand. The same day that Judge Gustafson held for the taxpayer in McGaugh, Judge Buch held for the IRS in Vandenbosch v. Commissioner, T.C. Memo. 2016-29 (2016). There, Dr. V. lent $125,000 to one Mr. Carver, who was the officer of a company that was seeking capital to build a power plant in Columbia. The Promissory Note identified the lender as “Mark J. Vandenbosch, SEP IRA.” And Dr. V. funded the loan with a $125,000 distribution from his SEP-IRA. But the distribution was not made from the SEP-IRA to Mr. Carver. It was made into Dr. V’s personal bank account. Further, the Note’s detailed repayment provisions required the money to be paid simply to Dr. Vandenbosch and not his SEP-IRA custodian, Edward Jones. It gave his personal residence as the place for payment to be sent. Finally, the Note was never transferred to Edward Jones but continued to be held by Dr. V.
On these facts Judge Buch had no trouble finding that Dr. V. needed to have reported the $125,000 distribution as income. Unlike Mr. McGaugh, Dr. V. was not acting as a conduit to facilitate the transfer of an asset into his SEP-IRA. He had basically taken a distribution from his SEP-IRA to fund an outside investment.
Today’s case falls between the facts of McGaugh and the facts of Vandenbosch. Let’s see.
In 2012 Mr. Ball formed an LLC, with the imaginative name of “The Ball Investment Account, LLC.” He was the 100% owner. The LLC created a checking account.
In June 2012 Mr. Ball directed his SEP-IRA custodian, J.P. Morgan, to move $170,000 from his SEP-IRA into the LLC’s checking account. The LLC immediately wired the money to a title company to fund a real estate loan. In April 2013 the title company repaid the loan with a check made out to the SEP-IRA and Mr. Ball so deposited the check.
In July 2012 Mr. Ball did a similar transaction, moving just under $40,000 from the SEP-IRA into the LLC account, then using that to fund a real estate loan. The second loan was repaid in various installments in 2012 and 2013. Each repayment check was made out to the SEP-IRA and so deposited.
Oh, and during 2012 Mr. Ball was younger than 59.5 years old.
Lesson: When Control Crosses The Line
Control is a continuum concept. Yeah, I said that before but I like the alliteration.
Mr. Ball argued that his case was like McGaugh because the money distributed from his SEP-IRA went immediately to fund investments whose earnings were then directed back into his SEP-IRA. He noted that, unlike Dr. V., the IRA distribution did not go into his personal checking account but instead went into his imaginatively-named LLC and it was immediately wired to the borrowers. The LLC account was just a conduit to facilitate the loan.
Judge Halpern was not impressed. He noted that just because the LLC was a separate legal entity did not make it a conduit: “Petitioner controlled Ball LLC, and nothing in the record convinces us that he did not have unfettered control over the [money distributed from his IRA].” What was critical for Judge Halpern’s decision was that while Mr. Ball used the distribution ostensibly for the benefit of his IRA account, as evidenced by the repayments going directly into the account, the initial distribution was under Mr. Ball’s complete control. Judge Halpern writes: “Yes, petitioner caused Ball LLC to lend the distributions nominally for the benefit of ‘The Ball SEP Account’, but he could just as well have made the loans in Ball LLC’s name or in his own name.”
The “too much control” came on the distribution side of the transaction. Mr. Ball's case was different from McGaugh because he did not direct the IRA to send the money to the borrowers. Judge Halpern notes that the IRA custodian had no knowledge of or control over how the funds would be used. The IRA simply put money into an account totally controlled by Mr. Ball. That was too much control. Worse, the loans themselves never became assets of the SEP-IRA. Judge Halpern writes that the IRA custodian had no documents associated with either of the two loans.
Comment: An Asset Test?
To have come within the “conduit” rule of McGaugh, Mr. Ball needed to have directed his IRA custodian to send the funds directly to the borrower, either by check or by deposit into the borrower’s bank account. That would eliminate the problem of too much control on the distribution side. His failure to do that is what did him in here.
However, I think Mr. Ball’s scheme has a deeper problem, a problem shared by Dr. Vandenbosch but not Mr. McGaugh: an asset problem. The loans never became assets of the IRA. In contrast, Mr. McGaugh not only acted as a conduit but also had the resulting investment asset (stock in a privately held company) titled in the name of his IRA. That is, he moved the asset into his IRA. Neither Dr. V. nor Mr. Ball did that. Instead, they both created an asset (a loan) outside the IRA and then just redeposited the earnings into the IRA.
So for Mr. Ball’s scheme to work, I would think he would need to have the loan repayment obligation run directly to the IRA custodian. There may be some practical problems with that, however, such as the IRA custodian not agreeing to sign the loan documents. Failing that, the next best move in the game would be for Mr. Ball to follow Odysseus and the Sirens. He would need to create a loan document that legally tied his hands and required him to put the loan repayments back into his IRA. It might be a provision that required all repayments to be made in the name of the IRA. Or it might be something else.
The conduit rule in McGaugh seems to work when the taxpayer seeks to get an asset into an IRA. In that situation, the law will ignore what otherwise looks like a distribution because the distribution was actually just a way for the IRA to buy the asset. But when the IRA never owns the asset, then I wonder: How can the conduit rule make a distribution intended to fund an investment outside the IRA anything other than a distribution?
Perhaps we will learn the answer to that question in a future Lesson From The Tax Court.
Bryan Camp is the George H. Mahon Professor of Law at Texas Tech University School of Law. He invites readers to return each Monday for a new Lesson From The Tax Court.