In the past, I’ve written extensively about penalties – touching on topics of fairness, equity, and whether penalties are effective in promoting taxpayer compliance. See my 2014 Most Serious Problem and my 2008 study: A Framework for Reforming the Penalty Regime. Today, I want to focus on a procedural requirement the IRS must follow in order to assess a penalty. In 1998, Congress added Section 6751 to the Internal Revenue Code (IRC), and subsection (b) provides that no penalty “shall be assessed unless the initial determination of such assessment is personally approved (in writing) by the immediate supervisor of the employee making the penalty determination.” This provision protects a taxpayer’s right to a fair and just tax system by ensuring penalty decisions are appropriate in light of the facts and circumstances. There is an exception for penalties automatically calculated through electronic means. The IRS has interpreted this exception to include any penalties calculated through its Automated Under Reporter (AUR) program, which matches income reported on a taxpayer’s return with income reported to the IRS by third-party payors.

In this blog, I’d like to discuss one of my past legislative recommendations, which answers the question: should an accuracy-related penalty imposed on the basis of negligence be calculated through electronic means and exempt from the supervisory approval requirement? I also want to delve into some recent court decisions discussing two key questions regarding IRC § 6751(b): When does this supervisory approval have to occur? And, can a taxpayer challenge the IRS’s failure to obtain supervisory approval when the tax has not yet been assessed? Although my legislative recommendation makes the case that accuracy-related penalties based on negligence should always require supervisory approval, the U.S. Tax Court and the Court of Appeals for the Second Circuit are currently split on when the supervisory approval has to occur and when a failure to obtain it may be challenged.

Should accuracy-related penalties based on negligence be automatically calculated through electronic means and be exempt from the supervisory approval requirement?

The IRS maintains that penalties calculated through its AUR program are automatically calculated through electronic means and thus do not require supervisory approval under IRC § 6751(b). However, in determining whether to assert the accuracy-related penalty based on negligence, the IRS should examine whether the taxpayer’s actions constituted a reasonable attempt to comply with the tax laws, which can be demonstrated by the taxpayer’s facts and circumstances. Under IRC § 6664(c)(1), the negligence penalty does not apply to any portion of an underpayment if the taxpayer had reasonable cause and acted in good faith. By using an automated process to assert these penalties and not having a supervisor review the determinations, the IRS does not consider the facts and circumstances of a case until the taxpayer contacts the IRS to challenge the proposed penalty. Taxpayers who did make reasonable attempts to comply and acted in good faith must take extra, burdensome steps to rid themselves of arbitrary penalties.

I would argue that the IRS should never rely solely on its electronic systems to automatically calculate a penalty based on negligence without the involvement of an employee. For example, the IRS should not assume that a taxpayer’s failure to include third-party information returns for two consecutive years means the taxpayer didn’t exercise ordinary and reasonable care in the preparation of the return – which is how the AUR system was programmed to assert the penalty based on negligence. In 2013, TAS performed a study to determine whether accuracy-related penalties increased compliance and found that Schedule C filers receiving accuracy-related penalties by default assessment or who appealed the penalties actually had worse compliance for years thereafter. Thus, it is entirely possible that by assessing penalties automatically, the IRS is signaling that it believes taxpayers are bad actors and taxpayers increase their future noncompliance to conform to that belief. I call this the “Oh yeah? Well, I show you” response.

Even if the IRS maintains that it can program certain facts into its systems to identify negligence, at a bare minimum, these determinations should be reviewed by a supervisor. This provides a check on penalty determinations calculated by the AUR, where the programming cannot take into account a taxpayer’s unique facts and circumstances. As I recommended to Congress, the Code should require managerial approval prior to assessment of the accuracy-related penalty imposed on the portion of underpayment attributable to negligence or disregard of rules or regulations under IRC § 6662(b)(1).

When does the supervisory approval required by IRC § 6751(b) have to occur and when may a taxpayer challenge the IRS’s failure to obtain supervisory approval?

Because IRC § 6751(b) requires supervisory approval of the “initial determination of such assessment” before the IRS can assess a penalty, there is a question as to whether this approval may occur at any time up until assessment.  However, as detailed in the dissent in Graev v. Commissioner, allowing the approval to occur at any time leads to a troubling result when the IRS no longer has the discretion to change the assessment and the taxpayer can’t challenge the IRS’s failure to comply with it in court.

In Graev v. Commissioner, the IRS disallowed a charitable deduction for the donation of a façade easement, and the revenue agent’s manager approved a 40 percent gross valuation misstatement penalty under IRC § 6662(h). IRS Counsel subsequently recommended the IRS assert, in the alternative, the 20 percent accuracy-related penalty under IRC § 6662(a), which was included on the notice of deficiency, but not submitted for supervisory approval. The taxpayers argued the IRS could not assess the 20 percent penalty because it failed to comply with the IRC § 6751(b)(1) requirement for supervisory approval.

Focusing on the plain language of the statute, a majority of the U.S. Tax Court held that it was premature to conclude the IRS had failed to comply with the supervisory approval requirement because the penalty had not yet been assessed. The written approval of the initial determination of the assessment could occur at any time before the assessment is made. In this case, assessment could not happen until the Tax Court’s decision became final and unappealable.

According to the dissent in Graev, “[t]he fact that a rule is cast as a bar on ‘assessment’ does not preclude pre-assessment consideration of compliance with that rule.”  The dissent held that part of the IRS’s burden of production under IRC § 7491(c) in penalty deficiency cases is showing compliance with IRC § 6751(b). Moreover, the statute requires approval by a revenue agent’s supervisor at a time when the supervisor still has the ability to approve or disapprove the penalty. Such approval would be meaningless once the taxpayer petitions the Tax Court because it is the Tax Court that now determines the amount of the liability that will be assessed

After Graev was decided, the Second Circuit held in Chai v. Commissioner that:

(1)    IRC § 6751(b)(1) requires a supervisor to approve an IRS employee’s penalty determination before the IRS first asserts penalties by issuing a notice of deficiency (or filing an answer or amended answer), and
(2)    The IRS has the burden to establish that it complied with IRC § 6751(b)(1) in deficiency cases under IRC § 7491(c).

The Second Circuit concluded that IRC § 6751(b)(1) was ambiguous because, quoting the dissent in Graev, “one cannot ‘determine’ an ‘assessment.’” The court considered the legislative history, which indicated the statute was intended to discourage IRS agents from threatening unjustified penalties in an effort to encourage taxpayers to settle. It found the Tax Court’s review of penalty determinations does not prevent this problem because taxpayers can be pressured to settle before the Tax Court gives its decision. Further, once the Tax Court issues an opinion, the supervisor no longer has discretion to give or withhold approval of the penalty because it is final. For IRC § 6751(b)(1) to have any effect, supervisory approval must be obtained before the IRS issues a notice of deficiency (or asserts penalties in court).

Following Chai, the IRS requested, and the Tax Court agreed, to vacate the Graev decision because Graev was appealable to the Second Circuit. However, the Tax Court continues to follow Graev in cases not appealable to the Second Circuit.

I am troubled by the IRS’s above position, articulated in the Tax Court’s holding in Graev, because it seems to contravene the intent of the statute and takes away the ability for a taxpayer to challenge the supervisory approval requirement in court.  As the dissent in Graev pointed out, once the opinion is final, the IRS no longer has the discretion to change the penalty, so the supervisory approval serves no purpose.  Furthermore, if it is determined after the assessment that the IRS had not complied with the supervisory approval requirement, what venue would the taxpayer have for challenging that failure?  Interestingly, in a footnote, the Graev decision suggests a post assessment Collection Due Process (CDP) hearing might provide an option for a taxpayer because the Appeals Officer must verify that the requirements of applicable law or administrative procedure have been met.  However, I wonder if this is a viable option if the Tax Court has decided the case given the fact that under section 6215(a) once the Tax Court’s decision becomes final the IRS shall assess the deficiency suggesting that the IRS would not have the authority to abate the penalty.

The supervisory approval requirement is an important part of the taxpayer’s right to a fair and just tax system. The IRS’s current interpretation allows it to sidestep considering the taxpayer’s facts and circumstances in those situations where they are most important – where the IRS asserts negligence based on an automatic calculation. Under Graev, there is little incentive for the IRS to comply with the requirement if the taxpayer cannot challenge the IRS’s failure to comply in a deficiency proceeding in Tax Court. Although the IRS could solve the first problem by administratively adopting my legislative recommendation, the second issue will likely continue to be played out in the courts unless Congress clarifies the law. Stay tuned for the Most Litigated Issues section of my upcoming Annual Report to Congress, which will discuss these issues further.