How do stylistic drafting choices affect the administration and substance of the tax law? One might suppose that mere questions of style do not matter at all, as long as the provisions operate as intended. Shu-Yi Oei and Leigh Osofsky’s new work challenges this assumption, and sheds new light on the implications of these stylistic choices for the tax system. When it comes to drafting of tax statutes, it turns out, form may in fact matter.
The work begins by reviewing the process for the production of tax statutes, and the role of various cooks in the tax law kitchen, including House and Senate members, their legislative counsel, the House Committee on Ways and Means, the Senate Committee on Finance, the IRS and Treasury, the JCT, and outside interest groups and lobbyists. The work then provides a topography of the basic drafting choices these different actors face.
A tax provision can be written as a specific rule, or as a general rule followed by exceptions (for example, the general inclusion of income under Section 61 and the subsequent exclusions). In some cases, rules can be virtually swallowed by the subsequent exceptions (for example, the general rule allowing deductibility of interest under Section 163 and the significant exceptions that follow). Alternatively, a provision can make significant use of defined terms, and then shift the substantive law to the definitions. Drafters also choose between freestanding provisions, which are self-contained, or rules that reference definitions or exceptions in other sections of the Code (for example, Section 108 is definition-inclusive, whereas Section 351 references the “control” definition in Section 368). Finally, all of these decision points may intersect, and the combination of choices can result in particularly convoluted drafting.
Do these drafting choices matter? The authors argue that they do, even if the substantive law is not affected by these stylistic choices. In particular, the authors argue that the language of the Code can affect how easy it is to interpret, and convoluted language can increase compliance costs for taxpayers and administrative costs for the government. Language choices may also ultimately affect judicial interpretations and the application of interpretive canons. These choices can also create unintended presumptions, and influence policy debates by framing certain rules as “normative” or socially dominant and others as irregular exceptions. Finally, the authors suggest that an “unwieldy and verbose code” can foster a perception that the Code is “broken.”
How then, are these consequential drafting choices made? To answer this question, the authors conducted interviews with government officials who have participated in the drafting of tax statutes in different capacities. Their results are fascinating (even if unsurprising at times) and should challenge long-held assumptions about the nature of tax law and the Code, and ultimately the prospects for successful tax reform.
First, most of the respondents indicated that the provisions are not drafted for the “ordinary taxpayer,” but rather for an audience of Treasury and IRS officials, a peer group of tax professionals, and tax preparation software developers. The predominant view among respondents is that drafting choices do not particularly matter, as long as the policy goals of Congress are faithfully effectuated through the substantive operation of the statute. For the same reason, members of Congress are less likely to focus on statutory text, and instead to defer to the assessments of their counsel. Respondents also described the effect of the accretion of statutory language over time, and drafting was often undertaken with a goal of making the minimal number of changes and avoiding upsetting settled rules and interpretations. Furthermore, in the legislative process it is easier to reach agreement on small changes rather than on major rewrites. Respondents also described changes in the control of drafting over time, and a shift from centralized control by House and Senate legislative counsel to more “amorphous control” with input from JCT, Senate Finance, Ways and Means, and outside groups. In this new era, the authors suggest that a broad range of actors can make actual substantive decisions through stylistic drafting choices.
The work concludes by suggesting implications of the study and its findings. First, the authors suggest that their findings pose a challenge to interpretation through textualism, since even the drafters themselves (and certainly members of Congress) may have been motivated more by broader policy purposes than specific textual choices. The authors also suggest that these drafting choices have distributional implications, and create a mismatch between a tax system written for experts but that imposes primary tax reporting responsibility on individual taxpayers. And finally, the findings have implications for public understanding of the legislative process, and suggests a disconnect between the “myth and reality” of congressional engagement with statutory text.
The Article is enlightening and important, and builds upon a growing body of empirical work on the legislative drafting process and its implications for interpretation, delegation, and legislative policy more broadly. In particular, the data uncovered by the authors’ interviews reveals a process for writing tax laws that is messier and more contingent than might have been previously appreciated.
In the background of this work is an essential question: Who should the tax code be written for. The authors implicitly assume that the rules should be written with taxpayers in mind, and present strong arguments why this should be the case. It’s not clear, however, exactly how much simpler the code could be made. When considering prospects for tax code simplification, Boris Bittker observed that simplifying the language of the code could also lead to complexity and uncertainty, and that “by its very nature, a tax on income must take account of an almost infinite spectrum of business, investment, and personal events and transactions.” The authors make a strong argument that drafting choices matter and better drafting could result in a more accessible and legible code. As the authors also imply, however, the potential gains from improved stylistic choices may be limited by the inherent challenges of accurately reflecting fundamentally complex rules. (Bittker similarly points to the advantages of “stylistic discipline”—particularly for “mass” provisions affecting many taxpayers—but does not describe the scope of the problem nor the stakes that Oei and Osofsky identify.)
One of the most compelling—and possibly troubling—implications of the work is the practical obstacles to significant tax reform particularly if legislators and their professional drafters favor surgical rewrites of particular code provisions that minimize the “brain damage.” If the tax code is a creaky ship, encrusted with the barnacle of prior legislation, enacting significant policy changes through narrow rewrites of specific provisions is almost certain to create new frictions elsewhere in the Code. For example, the authors allude to the problems resulting from Congress’ decision to import the old “specified services” definition from §1202 for the exception from the new 20% pass-through deduction. Similarly, Congress reformed the corporate tax through the simplest drafting change possible—a modification of §11 to tax all corporations at a flat 21% rate—but failed to consider how this change would place new pressures on anti-abuse rules elsewhere in the Code such as § 531 (the accumulated earnings tax) or § 541 (the personal holding company rules).
Oei and Osofsky’s new work is an invaluable (and sobering) account that sheds new light on the drafting choices in writing the tax law and their outsized implications.