MLR Online
Inside Regulatory Interpretation: A Research Note

Christopher J. Walker*

We now live in a regulatory world, where the bulk of federal lawmaking takes place at the bureaucratic level. Gone are the days when statutes and common law predominated. Instead, federal agencies—through rulemaking, adjudication, and other regulatory action—have arguably become the primary lawmakers, with Congress delegating to its bureaucratic agents vast swaths of lawmaking power, the President attempting to exercise some control over this massive regulatory apparatus, and courts struggling to constrain agency lawmaking within statutory and constitutional bounds.

This story is not new. Over two decades ago, for instance, Professor Lawson lamented the rise of the administrative state and traced it back to at least the New Deal.[1] But law schools are just now starting to catch up by requiring courses in regulation and administrative law, often in the first year.[2] Despite the publication of thousands of law review articles and judicial opinions on the interpretation of the Constitution, statutes, contracts, and other legal texts, to date little attention has been paid to the theory or practice of regulatory interpretation. Indeed, Professor Stack’s 2012 article Interpreting Regulations,[3] the subject of this Research Note, is the seminal piece on the topic.

In light of the importance of regulatory interpretation today, it is not surprising that the American Bar Association recognized Interpreting Regulations as the best work of administrative law scholarship published in 2012.[4] This recognition is well deserved. In Interpreting Regulations, Stack provides the first comprehensive approach to regulatory interpretation and situates this approach within the larger literature on legal interpretation. His theory of regulatory interpretation is simple yet pioneering: “a regulation should be read in light of its purposes, with the regulation’s text and the statement of basis and purpose constituting the privileged interpretive sources.”[5]

To provide some context, unlike the drafters of statutes and most other legal texts, federal agencies are required by law to publish a statement of basis and purpose as part of the final agency rule.[6] Although the Administrative Procedure Act (APA) requires only “a concise and general statement of [the rule’s] basis and purpose,”[7] judicial doctrine has evolved to require agencies to provide a comprehensive account of the various purposes and justifications for the regulatory action.[8] In other words, in practice these statements are neither concise nor general, but highly detailed and specific. They often respond to particular points raised during the public comment period and provide explanations for choosing one regulatory approach over another.[9]

Moreover, perhaps motivated by constitutional separation-of-powers concerns, courts review regulations differently than other legal texts. Instead of “determin[ing] if there is any conceivable basis for upholding [regulations], as courts do in constitutional review of legislation,” Stack explains that courts must “ask whether the agency articulated grounds in its statement of basis and purpose on which the regulations may be upheld.”[10] Combining these administrative law principles, it naturally follows—at least in theory—that the statement of basis and purpose that is promulgated along with a final agency rule can and should play a critical role in interpreting that rule.

This Research Note looks inside regulatory interpretation to explore the empirical foundation for Stack’s novel approach to regulatory interpretation. In 2013, the author of this Research Note conducted a 195-question survey of 128 federal agency rule drafters at seven executive departments (Agriculture, Commerce, Energy, Homeland Security, Health and Human Services, Housing and Urban Development, and Transportation) and two independent agencies (the Federal Communications Commission and the Federal Reserve). Many of the results from that empirical study have been reported elsewhere.[11] Four questions, however, were designed to assess Stack’s theory of regulatory interpretation from the perspective of the agency officials who draft these statements of basis and purpose.[12] Part I presents the findings on those questions, which largely support Stack’s theory. Part II explains how the other findings from the study bear on regulatory interpretation.

I. Agency Views on Statements of Basis and Purpose as Sources for Regulatory Interpretation

Especially in light of how intuitive Stack’s theory of regulatory interpretation appears, one might imagine that federal agencies draft these statements of basis and purpose with the express aim of guiding the courts, the public, and the regulated entities in interpreting final agency rules. And one might imagine that agencies expect courts to use those statements when interpreting regulations. To explore these assumptions, the agency rule drafters were asked four separate questions, the responses to which are depicted in Figure 1.[13]

As Figure 1 illustrates, the rule drafters surveyed generally agreed that statements of basis and purpose are drafted with an eye toward judicial interpretation and that courts should rely on them as interpretive tools, but perhaps not as much as one would expect, at least as a descriptive matter. When asked if “[a]gencies should draft the statements of basis and purpose accompanying their rules in part to guide courts in interpreting those rules,” about two thirds agreed (40%) or strongly agreed (29%), and another 24% somewhat agreed. No one strongly disagreed, but 8% disagreed.[14] If converted to a composite score, the score would be 3.9 on a 5.0 scale (4.0 = agree).[15]

Figure 1
The Role of Statements of Basis and Purpose in Regulatory Interpretation


When asked the same question descriptively (“actually do draft”) as opposed to normatively (“should draft”), the agency respondents were a bit less confident: About three in five rule drafters surveyed strongly agreed (22%) or agreed (39%), and another 27% somewhat agreed. Again, none strongly disagreed, but 12% disagreed—for a composite score of 3.7 (3.0 = somewhat agree; 4.0 = agree).[16] Moreover, the number of respondents who indicated they did not know (and thus are not included in these percentages) nearly doubled from six to eleven.

In other words, about nine in ten rule drafters surveyed at least somewhat agreed with Stack’s theory that agencies should draft (92%) and actually do draft (88%) statements of basis and purpose in part to guide courts in interpreting those rules. But when disaggregated by strength of agreement, less than a third strongly agreed. A couple unsolicited comments may shed some additional light. For instance, one respondent observed that “[g]enerally, agencies should draft statements of basis and purpose for explaining to the public the rule.”[17] And another remarked: “I think agencies routinely blow this opportunity.”[18]

Turning to the second pair of questions about judicial use of statements of basis and purpose, the results are perhaps even more interesting. When asked if “[c]ourts should use statements of basis and purpose when interpreting those rules,” all but one respondent agreed to some degree: 31% strongly agreed, 47% agreed, and 21% somewhat agreed.[19] This amounts to a composite score of 4.1 (4.0 = agree), the highest of the questions reported in Figure 1. One rule drafter’s comment in response to this question may reflect a broader consensus of the rule drafters surveyed:

All our rules contain a “purpose” section, and I assume that is what you’re referring to. And naturally a court should consider them, as they’re part of the rule. If it was just a statement that appeared somewhere other than the CFR, however, I think the courts would use those less.[20]

When asked the same question descriptively (“courts actually do use”) as opposed to normatively (“should use”), however, the agency respondents were far less confident: only about half strongly agreed (12%) or agreed (43%), with another 40% somewhat agreeing and 6% disagreeing (no one strongly).[21] This results in a composite score of 3.6 (3.0 = somewhat agree; 4.0 = agree). Moreover, the number of respondents who indicated that they did not know—and thus are not included in these percentages—nearly tripled from 10 to 29 respondents. In other words, a third of the rule drafters surveyed candidly admitted that they have no idea whether courts actually use these statements when interpreting agency regulations.

What should we make of these findings with respect to Stack’s theory on regulatory interpretation? It seems like the rule drafters surveyed agree that agencies should draft these statements with an eye toward helping courts interpret regulations and that courts should use such statements when interpreting the regulations. But they are less confident about whether courts actually do use statements of basis and purpose when interpreting regulations. This finding seems consistent with Stack’s own view (at least as of 2012) that “courts have not developed a consistent approach to regulatory interpretation under [administrative law deference] doctrines or elsewhere.”[22] Indeed, this disconnect between agency intent and judicial practice reinforces Stack’s call for courts to develop a more comprehensive approach to regulatory interpretation that reflects the actual mechanics of agency rulemaking.

Unfortunately, if citation to Interpreting Regulations is any indication of a change in judicial approach, no such change is evident—at least not yet. According to Westlaw KeyCite, Interpreting Regulations has been cited more than forty times since its publication in 2012, but never in a judicial opinion. Perhaps that is due to a lack of judicial exposure to this important article, as it has only been cited in two legal briefs. Hopefully this Research Note will help urge courts—and the government and private lawyers who litigate these issues[23]—to adopt a more comprehensive approach to interpreting regulations. The responses from the agency rule drafters surveyed provide pretty compelling support for Stack’s theory that courts should use statements of basis and purpose when interpreting regulations because agencies draft them with that purpose in mind and hope courts use them accordingly.

II. Purposivism, Textualism, and Regulatory Interpretation

Stack goes to great lengths to ground his theory of regulatory interpretation in Hart and Sack’s Legal Process School and thus as a purposivist theory of interpretation.[24] So much so that Professor Jellum, in reviewing the article, concluded that “[a]t bottom, Professor Stack’s article offers a new, but familiar, method for interpreting regulations; one that purposivists will certainly embrace, but one that textualists will likely rebuff.”[25] Jellum’s conclusion about textualism is debatable. But this textualism-purposivism classification may take on added significance if interpreters care about utilizing the same interpretive methodology as the drafters of the legal text under review. In particular, half of agency rule drafters who responded to the survey self-identified as textualist (35% moderate, 15% strong), whereas about a quarter identified as purposivist (19% moderate, 3% strong)—with the remainder stating they did not know or indicating “other.”[26]

Although Stack is firm in labeling his theory purposivist in light of its grounding in Hart and Sacks, it is not at all clear that textualists should take issue with the theory (though they would obviously dispute the label). Indeed, Stack is not so definitive in the article itself, noting that “[a]t the level of specification given thus far, the theory fits under both mantles [of textualism and purposivism], and provides an example of their common ground.”[27] More specifically, Stack first starts with the regulatory text both as a privileged source for discerning the regulation’s meaning and as a source of constraint on the interpretation.[28] This is consistent with textualism. Second, because the statement of basis and purpose is required by law and an integrated part of the rule, Professor Stack argues that such statement too is a privileged source of discerning meaning and purpose—akin to the privileged nature of a formally enacted statement of purpose in legislation.[29]

In subsequent writings, moreover, Stack has underscored the constraining nature of relying on such statement in regulatory interpretation:

At a basic level, this purposive orientation treats an agency’s public and authoritative justifications for its regulations as more than an elaborate and costly nuisance necessary to survive judicial review—they also create commitments that continue to guide the regulations’ meaning, a goal that drafters of the APA envisioned for these statements. To be sure, this approach inhibits agency flexibility. It disallows agency constructions that are permitted by the regulation’s text but inconsistent with the agency’s original, public justifications. But one party’s flexibility is another’s unpredictability. Holding the agency to constructions of its regulations that are consistent with the agency’s own explanatory justifications translates directly into greater notice of the regulation’s meaning.[30]

Accordingly, Stack makes a strong case that “both regulatory text and the regulation’s statement of basis and purpose count as part of the ‘text’ on which a textualist should center her interpretive inquiry.”[31]

Framed this way, it seems that even Justice Scalia should agree with this approach to regulatory interpretation. After all, Justice Scalia has explained that “[p]erhaps no interpretive fault is more common than the failure to follow the whole-text canon, which calls on the judicial interpreter to consider the entire text, in view of its structure and of the physical and logical relation of its many parts.”[32] Because the statement of basis and purpose is legally part of the regulation itself—as Stack convincingly argues—and its role in regulatory interpretation finds strong empirical support from the agency rule drafters surveyed in this author’s study, it seems reasonable to conclude as a matter of textualism that the regulatory language at issue must be read in view of the regulation’s statement of basis and purpose.[33]

Indeed, in a recent article entitled Regulatory Textualism, Professor Nou proposes a textualist approach to regulatory interpretation.[34] Just like Stack’s purposivist theory, Nou’s textualist approach instructs that “judges should first consider the preamble’s provision-by-provision analysis of the regulation, which frequently responds to public comments raising potential ambiguities in the proposed regulation.”[35] This is because the preamble—which is another term for the statement of basis and purpose[36]—“is the best evidence of the regulatory text’s public meaning because it results from the agency’s back-and-forth with external commenters and political monitors.”[37] Drawing on insights from positive political theory, Nou then outlines the hierarchy of additional sources from which to derive the public meaning of regulatory text.[38] An analysis of Nou’s overall theory of regulatory interpretation exceeds the scope of this Research Note, but her theory’s grounding in textualism—with the primary, privileged source of textual meaning being the statement of basis and purpose—underscores that any purported textualism-purposivism dichotomy in Stack’s theory is arguably a false one.[39]

That textualists and purposivists should both embrace Stack’s theory of regulatory interpretation does not mean we have arrived at one coherent theory of everything in regulatory interpretation—a theory that has eluded us in the context of statutory interpretation and elsewhere. As Stack notes, the interpretive camps will still part ways when “the text of the regulation and the statement of basis and purpose have been well mined by the court and neither sheds light on the interpretive question posed.”[40] At that point, the interpreter must resort to other interpretive tools. The textualist and the purposivist may well disagree on which tools should be prioritized to resolve any remaining ambiguity. The textualist may well place more weight on semantic canons, with the purposivist focusing more on the purposes and policies that seem to drive the regulation (and statute underlying the regulation).

If the interpreter were concerned with what the agency rule drafters actually used in drafting the regulation, it would be wise to not follow the canons indiscriminately and instead look inside regulatory interpretation. As explored in much greater detail elsewhere, the agency rule drafters surveyed for this study did not embrace all of the canons equally.[41] For instance, they reported general awareness and use of the ordinary meaning canon, the whole act rule, consistent usage canon, noscitur a sociis (associated words canon), and ejusdem generis (residual clause canon).[42] Legislative history was also among the interpretive tools most reported by the agency respondents as used when drafting rules.[43] But they reported less usage and reliance on expressio unius (negative implication canon), the presumption against superfluities (surplusage canon), and most of the substantive canons—just to mention a few.[44]

In other words, even if courts, litigants, and scholars fully embraced Stack’s theory, we would still be left with many questions and debates about how to interpret regulations. But those debates are not new ones, as they have been commonplace in legal interpretation dating back to at least Professor Llewellyn’s famous cannoning of the canons.[45] To the extent, however, we care about whether interpretive tools are grounded in the empirical realities of agency rulemaking, much more investigation inside regulatory interpretation needs to be done.


Despite the important theoretical work Professor Stack has done in Interpreting Regulations, it seems courts (and litigants) have yet to embrace an approach to regulatory interpretation that treats statements of basis and purpose as a “privileged source of interpretation.”[46] The views of federal agency rule drafters presented herein provide compelling support for such an approach to regulatory interpretation and should further encourage courts to move in that direction. Stack’s theory deserves more scholarly attention and judicial adoption from purposivists and textualists alike. We can then turn to perhaps more difficult questions about which other interpretive tools should be kept or discarded in the regulatory interpretation toolkit in light of how federal agencies actually draft rules in the modern administrative state.

* Assistant Professor of Law, Michael E. Moritz College of Law, The Ohio State University.

[1]. Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231 (1994). But see Jerry L. Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law (2012) (tracing the history of the regulatory state from the Founding to the Gilded Age).

[2]. Dakota S. Rudesill, Christopher J. Walker & Daniel P. Tokaji, A Program in Legislation, 65 J. Legal Educ. 70, 72–78 (2015) (detailing the rise of the first-year legislation-and-regulation course).

[3]. Kevin M. Stack, Interpreting Regulations, 111 Mich. L. Rev. 355 (2012).

[4]. Award of Scholarship in Administrative Law: Recipients, Am. Bar Ass’n, []. Technically, Professor Stack shared this honor with Jerry Mashaw, for his terrific book Creating the Administrative Constitution. Id. Professor Jellum also features Interpreting Regulations in The Journal of Things We Like (Lots). Linda Jellum, A Textualist Approach to Purposivism in the Regulatory Arena, JOTWELL (Apr. 10, 2013), [].

[5]. Stack, supra note 3, at 361–62.

[6]. 5 U.S.C. § 553(c) (2014) (“After consideration of the relevant matter presented [during the notice and comment period], the agency shall incorporate in the rules adopted a concise general statement of their basis and purpose.”).

[7]. Id.

[8]. Stack, supra note 3, at 378–79, 380–81 (discussing relevant judicial doctrines and their implications for agency rulemaking).

[9]. Id. at 392–93 (“[A]gencies today issue statements of basis and purpose that are far from mere preambles; they are extremely detailed rationales for, and explanations of, their regulations.”).

[10]. Id. at 361 (citing Kevin M. Stack, The Constitutional Foundation of Chenery, 116 Yale L.J. 952, 960–71 (2007)); see also Christopher J. Walker, The Ordinary Remand Rule and the Judicial Toolbox for Agency Dialogue, 82 Geo. Wash. L. Rev. 1553, 1561–65 (2014) (discussing the evolution of the Chenery doctrine and accompanying separation-of-powers concerns).

[11]. Christopher J. Walker, Inside Agency Statutory Interpretation, 67 Stan. L. Rev. 999 (2015) [hereinafter Walker, Inside Agency Statutory Interpretation]; see also Christopher J. Walker, Chevron Inside the Regulatory State: An Empirical Assessment, 83 Fordham L. Rev. 703 (2014) (exploring further the findings on the use of administrative law doctrines to shape agency interpretive behavior). The study is modeled on a similar study that Professors Gluck and Bressman conducted on congressional drafting. Abbe R. Gluck & Lisa Schultz Bressman, Statutory Interpretation from the Inside—An Empirical Study of Congressional Drafting, Delegation, and the Canons: Part I, 65 Stan. L. Rev. 901 (2013). The study methodology and its limitations are set forth in Walker, Inside Agency Statutory Interpretation, supra, at 1013–16, and will not be repeated here. A copy of the survey is reproduced as an appendix to that prior work, id. at 1068–79, and this Research Note uses the abbreviation “Q_” to reference the question number being discussed, with the number of respondents to that question indicated as “(n=_).” Moreover, the survey allowed the respondents to make additional comments on most questions, and the dataset includes 345 such comments. This Research Note references such comments by question number and then the order in which the comments appear in the dataset.

[12]. Professor Stack graciously provided feedback on these questions, which was incorporated prior to survey administration.

[13]. These four questions asked the agency rule drafters to “[p]lease evaluate the following statements”:

“Agencies should draft the statements of basis and purpose accompanying their rules in part to guide courts in interpreting those rules,” Q34(a) (n=91) (nk=6);

“Agencies actually do draft the statements of basis and purpose accompanying their rules in part to guide courts in interpreting those rules,” Q34(b) (n=85) (nk=11);

“Courts should use statements of basis and purpose when interpreting those rules,” Q34(c) (n=87) (nk=10); and

“Courts actually do use statements of basis and purpose when interpreting those rules,” Q34(d) (n=68) (nk=29).

The options were “strongly agree,” “agree,” “agree somewhat,” “disagree,” “strongly disagree,” “I don’t know,” and “other (explain).” A varying number of rule drafters indicated that they did not know for particular statements, as reported in the “no knowledge” (nk) number. Accordingly, the number of respondents considered and percentage calculations in Figure 1 do not include those responses. Nor, due to difficulty in coding, does the number of respondents include the 1-2 respondents who marked “other.” One of those respondents provided a reason for other: “ran out of time.” Q34 cmt. 1. This question was the second to last one in the survey.

[14]. Q34(a) (n=91) (nk=6).

[15]. Composite scores are calculated by giving five points for “strongly agree,” four points for “agree,” three points for “somewhat agree,” two points for “disagree,” and one point for “strongly disagree.” The aggregate is then divided by the total number of respondents for the question, excluding those who indicated they did not know and those who gave “other” responses.

[16]. Q34(b) (n=85) (nk=11).

[17]. Q34 cmt. 3.

[18]. Q34 cmt. 2.

[19]. Q34(c) (n=87) (nk=10).

[20]. Q34 cmt. 4; cf. Stack, supra note 3, at 361 (“As a result of these doctrines, the text of a regulation and its statement of basis and purpose stand in a unique relationship: together, they constitute the act of regulation, an act that is not complete without either element of this couplet.”).

[21]. Q34(d) (n=68) (nk=29).

[22]. Stack, supra note 3, at 359.

[23]. Cf. Nicholas R. Parrillo, Leviathan and Interpretive Revolution: The Administrative State, the Judiciary, and the Rise of Legislative History, 1890-1950, 123 Yale L.J. 266, 281–83 (2013) (tracing the rise of judicial reliance on legislative history in statutory interpretation in the 1930s and 1940s to the increased use and citation of legislative history in briefing by federal government attorneys).

[24]. See, e.g., Stack, supra note 3, at 383–91 (revisiting in depth Henry M. Hart, Jr. & Albert M. Sacks, The Legal Process (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994)).

[25]. Jellum, supra note 4.

[26]. Q35 (n=98); see also Walker, Inside Agency Statutory Interpretation, supra note 11, at 1016–18 (detailing background and experience of the survey respondents). As Stack notes elsewhere, the survey did not define these classifications, so we should be careful to not read too much into the respondents’ self-classification as textualist or purposivist. See Kevin M. Stack, Purposivism in the Executive Branch: How Agencies Interpret Statutes, 109 Nw. U. L. Rev. 871, 898 n.113 (2015). Much more empirical investigation needs to be done.

[27]. Stack, supra note 3, at 406.

[28]. Id. at 392 (advocating interpretation “to the extent permitted by its text”); id. at 407 (noting that when a text states purposes, “a textualist will attend to those purposes as part of her commitment to discerning the meaning of the text”).

[29]. Id. at 362–63.

[30]. Kevin M. Stack, How To Interpret a Regulation: First Principles, RegBlog (Feb. 11, 2013), []; accord Stack, supra note 3, at 407–10, 414–16.

[31]. Stack, supra note 3, at 407.

[32]. Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 167 (2012).

[33]. This does not mean that the current textualists on the Court have embraced Stack’s theory of regulatory interpretation. A number of Supreme Court decisions apply the plain text of the rule without resorting to the statement of basis and purpose, whereas others—including Talk America, Inc., v. Michigan Bell Telephone Co., 131 S. Ct. 2254, 2261–62 (2011), and Coeur Alaska, Inc. v. Southeast Alaska Conservation Council, 557 U.S. 261, 287–91 (2009)—expressly rely on the statement. See Stack, supra note 3, at 372–74 (discussing cases). Moreover, one could imagine Justice Scalia dismissing the theory on formalist grounds by rejecting the argument that a statement of basis and purpose is part of the regulatory text but more akin to legislative history. Stack, however, seems to have the better argument on that point.

[34]. Jennifer Nou, Regulatory Textualism, 65 Duke L.J. 81 (2015).

[35]. Id. at 85.

[36]. See id. at 85 n.12 (“To maintain consistency with other scholars’ terminology, this Article will also define ‘preamble’ as the agency’s statement of basis and purpose.”).

[37]. Id. at 85.

[38]. Id. at 116–27.

[39]. To be sure, Nou’s approach takes issue with some of the more purposivist leanings in Stack’s approach. For instance, “regulatory textualism rejects reliance on the broad statements of purpose often found in preambles in favor of more specific explanatory provisions,” as these broader statements “often admit of multiple purposes or simply mirror the language of the statute in ways that do not shed any independent interpretive light.” Id. at 120. This disagreement, however, does not seem to go to the heart of whether the interpretive theory is textualist or purposivist, but how textualists and purposivists would differ in interpreting the text of an agency’s statement of basis and purpose—similar to debates that take place when interpreting legislatively enacted statements of purpose.

[40]. Stack, supra note 3, at 407.

[41]. See Walker, Inside Agency Statutory Interpretation, supra note 11, at 1020 fig.2 (comparing the agency rule drafters’ reported use of over twenty interpretive tools).

[42]. For more on the findings related to the sematic canons, see id. at 1022–31.

[43]. For more on the findings related to legislative history, see id. at 1034–48.

[44]. For more on the findings related to the substantive canons, see id. at 1031–34.

[45]. Karl N. Llewellyn, Remarks on the Theory of Appellate Decision and the Rules or Canons About How Statutes Are to Be Construed, 3 Vand. L. Rev. 395, 401 (1950) (“[T]here are two opposing canons on almost every point.”).

[46]. Stack, supra note 3, at 420.

Download as PDF