No Surprises? Texas Medical Association v. HHS: Part II
Summary: A recent district court decision, Texas Medical Association v. HHS, Dkt No. 6:21-CV-425 (E.D. Tex. Feb. 23, 2022), accessible on WESTLAW at Medicare & Medicaid Guide ¶307259 (C.C.H.), 2022 WL 597141, has troubling implications both for agencies’ traditional authority to create rebuttable presumptions and for the scope of Administrative Procedure Act section 553’s “good cause” exception permitting agencies to bypass notice and comment rulemaking. The decision invalidated a key provision of a regulation implementing a significant federal statute, the No Surprises Act, by reading the Act in a narrow textualist manner without acknowledging the contextual considerations set forth by regulators. This is the second in a series of three blogposts exploring the decision. The first post described the statute, the interim final regulations, and the district court decision. This post offers a critique of the district court’s decision regarding interpretation of the relevant statutory multi-factor test. The third post will critique to district court’s refusal to accept the relevant agencies’ “good cause” justification for bypassing notice and comment rulemaking.
What’s Wrong With Weighting Statutory Factors and Establishing a Rebuttable Presumption?
In Texas Medical Association, the Court read the listing of factors in the No Surprises Act (42 U.S.C. §300gg-111(c)(5)(C)) as precluding the Departments from implementing the statute by establishing any presumptions or any weighting of the factors for the certified IDR entities to apply. Such a ruling may have widespread implications. The United States Code almost certainly contains numerous statutorily-mandated multi-factor balancing tests. A search of “factors w/5 consider!” in WESTLAW’s USC Unannotated database returned 786 results. Should such statutes be read to preclude agencies from weighting the factors and establishing presumptions for adjudicators, whether inside or outside the agency, to employ?
Agency Weighting of Factors
There is nothing inherent in a multi-factor test to suggest that each factor has equal weight, or that there are no categories of cases in which some factors may weigh more heavily than others. For example, the statutory codification of the “fair use” doctrine in copyright law sets forth a four factor test which requires that each factor “shall” be considered. 17 U.S.C. §107. Yet courts routinely identify the fourth factor as the most important.
The question of the weighting of the factors set forth in a statutorily-mandated multi-factor test is a substantive legal decision. Giving greater definition to multi-factor tests can be categorized as either a legislative or adjudicatory function. Ordinarily, either legislative bodies themselves, or entities to whom the legislative body has delegated quasi-legislative functions, possess authority to specify or give direction to adjudicators who must consider multiple factors in resolving disputes. Of course, weighting the factors on a case-by-case basis, and thus developing a common law doctrine governing factors’ relative weights, is a task regularly performed by adjudicatory bodies in the course of applying a statutorily-mandated multi-factor test.
Customarily, quasi-legislative and quasi-adjudicative functions are combined in one administrative agency, obviating the need to determine the relative roles of entities exercising quasi-legislative or quasi-adjudicative functions in making such decisions that give further definition to multi-factor tests. In the No Surprises Act, Congress appears to have departed somewhat from the norm, seemingly splitting quasi-legislative and quasi-judicial functions between the Departments and certified IDR entities. So either the Departments charged with implementing the complex provisions of the No Surprises Act or the certified IDR entities, or both, could lay claim to the power to develop “substantive law” regarding the weighting of the factors listed in 42 U.S.C. §300gg-111(c)(5)(C).
Nothing in the statutory text suggests that Congress resolved the issue of the locus of law-generating authority concerning the section 300gg-111(c)(5)(C)’s multi-factor test. The primary role of the certified IDR entities seems to be providing a forum for impartial adjudication, namely application of the applicable legal standards to the facts surrounding a particular reimbursement rate dispute. Congress appears to have left the decision regarding the relative law-generating responsibilities of the Departments and certified IDR entities to the Departments themselves, which Congress charged with implementing this complex statute. At a minimum, silence on the subject should be viewed as conferring the authority to give further precision to substantive law, including establishing general weighting or use of factors, to the Departments.
The basis for such a presumption is even more powerful with respect to the No Surprises Act, given Congress’ assignment of adjudicatory functions to multiple non-governmental entities, namely the certified IDR entities. Courts should presume that law-generating responsibilities lie with politically-accountable governmental entities, rather than unaccountable private entities. That presumption is supported by the constraints on delegating quasi-legislative power to private entities. Carter v. Carter Coal Co., 298 U.S. 238, 311 (1936); Association of American Railroads v. United States Department of Transportation, 721 F.3d 666, 670-71, 673-74 (D.C. Cir. 2013) (involving decisional authority exercised by Amtrak and by private arbitrators), vacated on other grounds, Department of Transportation v. Association of American Railroads, 575 U.S. 43 (2015). The presumption is justified all the more under the No Surprises Act, where certified IDR entities’ decisions are not even subject to substantive judicial review, 42 U.S.C. §300gg-111(c)(5)(E) (precluding judicial review except for the grounds set forth in the Federal Arbitration Act), and where only the Departments are in a position to coordinate the adjudicatory decision-making by multiple independent certified IDR entities.
The “split enforcement” model dividing regulatory and adjudicatory power between two separate governmental bodies, is not unknown. Even in that context, when the adjudicator is a public rather than a private body, the Supreme Court has recognized that the primary law-generating responsibility is held by the regulatory agency rather than the adjudicatory agency, thus according Chevron deference to interpretations promulgated by the regulatory agency. Martin v. Occupational Health & Safety Review Comm’n, 499 U.S. 144, 151-55 (1991); George Robert Johnson, Jr., The Split-Enforcement Model: Some Conclusions From The OSHA And MSHA Experiences, 39 ADMIN. L. REV. 315, 329-32 (1987).
Moreover, section 300gg-111(c)(5)(C)’s structure suggests that the QPA either should be considered the major factor or should serve as the focus of the certified IDR entity’s choice between reimbursement rates proposed by the opposing parties. Even if the factors were all enumerated in one list, the first item might reasonably be considered most important simply by virtue of its placement. Arguably, legislatures tend to list items in descending order. See generally, 2A SUTHERLAND STATUTORY CONSTRUCTION § 47:19 (7th ed.) (“legislatures naturally enumerate objects in descending order”)[available on WESTLAW]. It would have been reasonable for the Departments to interpret a list in such a manner, in the absence of anything suggesting the contrary. Granted perhaps the strictness of such an inference could be questioned in modern usage, see Mark Nichol, 5 Rules for Using Logic to Order Lists.
Other more sophisticated approaches to specifying the relative weight of the mandatory considerations would also be quite reasonable. The Departments could, quite reasonably, assess the relative weight of the factors in light of section 300gg-111(c)(5)(C)’s purpose or the overall goals of the No Surprises Act, among them “reducing prices that may have been inflated due to the practice of surprise billing prior to the No Surprises Act.” 86 Fed. Reg. at 55996; see, 86 Fed. Reg. 55997 (explaining that a provider’s level of expertise or training should not ipso facto suggest the reimbursement rate should exceed the QPA, “as this would lead to an increase in prices without a valid reason and does not align with the goals of the No Surprises Act.”) In fact, as the Departments explained, they did just that. Id. at 55996. Another quite reasonable approach would have been to accord greater weight to a factor that reflects the “market price” for compensation, namely the in-network reimbursement the insurer provides to its own providers,” than other factors that do not reflect the market price. The Departments appeared to have relied on such reasoning as well. Id.
But, as noted above, the factors do not merely appear on an undifferentiated list. Section 300gg-111(c)(5)(C) first lists the QPA by itself, and then in the following provision refers to five other factors, as the Departments noted. Id. An agency could certainly reasonably determine that such a structure suggests that the first factor is distinct from the remaining factors, and may, in most cases, be considered the primary or focal factor. Indeed, the remaining factors could be viewed as distinct (but possibly overlapping) reasons why the “market price” is inadequate.
The QPA may be inadequate because of the market power of one of the participants. 42 U.S.C. §300gg-111(c)(5)(C)(ii)(II). It may be inadequate because the particular service is not typical of the type of service provided, either because it required a physician of unusual skill or experience, required a teaching hospital or the like, or was made more complicated by the particular circumstances of the patient. 42 U.S.C. §300gg-111(c)(5)(C)(ii)(I), (III), & (IV). Or a rate that diverged from the QPA might have been negotiated between the contesting parties had each acted in good faith in negotiating a reimbursement rate. 42 U.S.C. §300gg-111(c)(5)(C)(ii)(V).
On such an account, the statutory list of factors does not reflect an expectation of a freewheeling, unstructured, all-things-considered balancing of presumably commensurable variables. Rather, it reflects an expectation that certified IDR entities ascertain whether there are one or more substantial reasons for believing the QPA to be inadequate with respect to a particular reimbursement dispute, and, if so, to determine the most appropriate rate given the one or more deficiencies it has identified. The Departments indeed appear to have interpreted the statute in that very way, 86 Fed. Reg. at 55997-98, and it is entirely reasonable for them to have done so.
Perhaps weighting the factors lies within the Department’s authority, but what about establishing presumptions?
Creating presumptions should also be categorized as either a legislative and adjudicatory function. Presumptions generally embody considerations of “policy, fairness and probability.” Edward W. Cleary, Presuming And Pleading: An Essay On Juristic Immaturity, 12 STANFORD L. REV. 5, at 11-13 (1959); see, EDWARD W. CLEARY, ET AL. MCCORMACK ON EVIDENCE §343 at 968-69 (3d. ed. 1984)(later editions available). In other words, presumptions may be created because of a strong connection between the facts found and the fact to be inferred (i.e., probability), or may reflect concerns involving imbalance to access to proof (i.e., fairness).
In common law adjudication, one might expect the courts to make such judgments, and establish presumptions, in adjudicating cases. But agencies should also have the power to create presumptions on the same basis, as an aspect of giving substance to the applicable legal standard — process and substance cannot easily be separated. Indeed, as we shall see below, the power to create presumptions may be essential to the agency’s responsibility to ensure consistent adjudicatory decisions across multiple individual adjudicators.
But, as suggested above, presumptions also often reflect “social or economic policy” preferences that favor one result over another. See, Juristic Immaturity, supra, at 11-12; MCCORMACK ON EVIDENCE, at §343, 968-69. In such circumstances, when a statute is to be implemented by an agency, the agency should have the power to adopt presumptions to embody those policy preferences, even if the responsibility for adjudication cases is assigned to independent ALJs or to private arbiters. In that setting, policy preferences should clearly be made by agencies, not adjudicators. As long as the presumptions are rebuttable, such policy preferences seem entirely consistent with statutory text and the type of policy judgment administrative agencies are expected to make.
The IDR process is essentially a mass adjudication context in which numerous individual arbitrators making decisions may render determinations that conflict with each other. It is entirely reasonable for the Department not to rely upon the certified IDR entities to ensure consistency among arbitrators and instead to take steps themselves to ensure that decision-making is more consistent. Specifying the manner in which factors are to be considered and creating rebuttable presumptions provide means of enhancing arbitral consistency.
And that is precisely one of the justifications the Departments proffered both for instituting the QPA presumption and providing direction regarding the manner in which the non-QPA statutory factors are to be treated. As noted earlier, the Departments explained “[a]nchoring the determination of the out-of-network rate to the QPA will increase the predictability of IDR outcomes.” 86 Fed. Reg. at 55996. Moreover, consistency is not merely a goal, but has salutary instrumental effects as well. In particular, adjudicatory consistency increases the likelihood that pre-arbitration negotiations between insurers and providers/facilities will prove successful. The Departments noted such a salutary benefit as well. Id.; accord, id. at 55985, 56061.
ALJ’s enjoy decisional independence. Yet even with respect to ALJs agencies can impose conceptual analytical frameworks and substantive rules to direct decision-making in ways that limit adjudicator discretion and increase the adjudicatory consistency. Heckler v. Campbell, 461 U.S. 458 (1983), is the classic example of judicial approval of such an approach. The Social Security disability statute requires the Secretary to determine whether sufficient jobs exist in the national economy for an applicant to be gainfully employed given that individual’s unique combination of physical and/or mental impairments and “age, education, and work experience.” 42 U.S.C. §423(d)(2)(A). In Heckler v. Campbell, the Supreme Court permitted the Social Security Administration to require ALJ’s to use “a grid,” an agency-developed “matrix of the four factors identified by Congress,” id. at 461, to make such a determination. Id. at 467-70. As the Court would later observe in Bowen v. Yuckert, 482 U.S. 137 (1987), the “grid” contributed “to the uniformity and efficiency of disability determinations.” Id. at 153. For a fuller discussion of the grid, see JON C. DUBIN, SOCIAL SECURITY DISABILITY LAW AND THE AMERICAN LABOR MARKET 61-71 (2021).
Courts have permitted agencies to institute presumptions, so long as they are consistent with the underlying statute being implemented and rational, i.e., that there is a sound connection between the proved and inferred facts. Thus, in United Scenic Artists, v. N.L.R.B., 762 F.2d 1027 (D.C. Cir. 1985), citing a string of U.S. Supreme Court precedents, the D.C. Circuit noted that “[p]resumptions may, of course, be established both by legislative bodies and by administrative agencies.” Further, the validity of such presumptions “depends as a general rule upon a rational nexus between the proven facts and the presumed facts.” This is so even when the statute requires findings. Thus, the FCC’s rebuttable presumptions with regard to competition in a particular franchise area, based on nationwide data, was upheld in National Association of Telecommunications Officers and Advisors v. Federal Communications Commission, 862 F.3d 18 (D.C. Cir. 2017). The Court explained:
“we hold the Commission’s use of a rebuttable presumption to comply with the statutory requirement that it make a finding on the state of competition in each franchise area is a permissible construction of the statutory requirement that the Commission “find[ ]” “effective competition” before terminating rate regulation.”
Id. at 25. Indeed, the Court invoked Chevron: “Because the Congress has not spoken directly to the question whether the Commission may use a rebuttable presumption in lieu of case-by-case findings of fact, we analyze the Commission’s decision under Chevron step two.” Id.
The presumption the Supreme Court upheld in Bowen v. Yuckert, a conclusive presumption, is particularly instructive. The statutorily-mandated multi-factor test for disability, declares that an applicant “shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy.” Id. at 140 (emphasis added). Thus, a finding of disability would appear, from the plain text of the statute, to require consideration of factors other than a claimant’s impairments, namely the claimant’s age, education, and work experience.
As a part of a five-step sequential process imposed upon ALJs in making disability determinations, the Social Security Administrator had adopted a severity regulation, step two in the five-step process. The severity regulation mandated denial of a disability claim solely on the basis of the applicant’s physical and/or mental impairments if the applicant’s impairment or combination of impairments was not “severe.” Id. at 140-41. At the next step, step three, an applicant’s disability claim would be granted, without further consideration of other factors, if the physical/mental impairment or impairments were particularly severe, i.e., where “the impairment is equivalent to one of a number of listed impairments that the Secretary acknowledges are so severe as to preclude substantial gainful activity.” Id. at 141. The court upheld the severity regulation, and suggested that the step three metric was permissible as well. The Court noted:
The severity regulation increases the efficiency and reliability of the evaluation process by identifying at an early stage those claimants whose medical impairments are so slight that it is unlikely they would be found to be disabled even if their age, education, and experience were taken into account. Similarly, step three streamlines the decision process by identifying those claimants whose medical impairments are so severe that it is likely they would be found disabled regardless of their vocational background.
Id. at 153. The Court upheld the step two presumption (the “severity” regulation) despite evidence that, as applied, the severity regulation mandated a finding of no disability in a significant number of cases in which claimants could actually prove disability, when age, education, and work experience were considered. But, as the concurring Justices explained, such errors were not ones inherent in the agency’s use of a presumption, a conclusive presumption no less, but in the agency’s misapplication of it. Id. at 158 (O’Connor, J., concurring).
In short, not only does the QPA presumption give the Departments greater control over out-of-network reimbursement rates, keeping further refinement of the framework in the hands of accountable government officials rather than private arbitrators, it also enhances the prospects that the results of arbitrations will be consistent with each other. Further, it increases the likelihood that pre-arbitration negotiations succeed, precisely because the results of the arbitration in many cases will be quite predictable. Arbitrators will be left to handle cases where factors other than the QPA justify a significant departure from the QPA.
What’s wrong with that? There’s certainly nothing in the relevant statutory text, or anything else, to suggest that Congress intended to preclude such an approach.
Conclusion and the Upcoming Post
The Departments acted well within the discretion typically accorded agencies in specifying the weight of the statutorily enumerated factors certified IDRs must consider in choosing between competing proposed reimbursement rates. Similarly, the Departments acted within the customary range of discretion accorded agencies in concluding that the QPA should be the focal point of the reimbursement rate inquiry, and in establishing a rebuttable presumption that in the absence of any evidence that the QPA was an inappropriate reimbursement rate, the QPA was presumptively the optimal rate. Nothing in the statutory text suggests that Congress meant to deprive the Department of such discretion. And the Departments fully explained their reasoning on all three counts.
In the final contribution to this series I will tackle the District Court’s conclusion that the Departments lacked “good cause” for promulgating the IDR process rules as interim final rules before completing the “notice and comment” process.
 There are also, of course, numerous judicially-created multi-factor balancing tests. See, Lawrence Solum, Legal Theory Lexicon: Balancing Tests, Legal Theory Blog (AUGUST 23, 2009) (“[b]alancing tests are ubiquitous in American law”); See T. Alexander Alienikoff, Constitutional Law in the Age of Balancing, 96 YALE L.J. 943, 943-44, 963-72 (1987) (exploring and evaluating use of balancing in constitutional jurisprudence).
 The search captures provisions that do not establish mandatory multi-factor tests, as well as those that do. For examples of those that do, see 49 USCA §5112(1)(I) (factors to be considered by states and tribes prescribing highway routing for hazardous materials transport); 54 U.S.C. §101913(5) (factors to consider in the award of concession contracts, including “primary” and “secondary” factors); 19 U.S.C. §1673d(b)(4)(A) (considerations in making a finding related to antidumping duties); 49 U.S.C. §40122 (factors for the arbitration board to consider in resolving disputes between the FAA Administrator and unions representing FAA employees); 49 U.S.C. § 30172 (criteria for making whistleblower awards). Another example, which is not codified (but is in the notes to the United State Code Annotated), specifies three factors for the Secretary of Health and Human Services to consider in “establishing standards relating to the exception for health center entity arrangements” from antikickback penalties,” Pub. L. 108-173, 117 Stat. 2287, tit. IV, § 431(b)(2003),.
And, of course, mandatory multi-factor tests may be phrased in that my search parameters would not capture.
 Take for example the ten-factor traditional test for determining whether an agent is a servant or an independent contractor. RESTATEMENT (SECOND) OF AGENCY §220(2) (1958). Does each of the ten factors have equal weight, or is the first listed, “the extent of control which, by the agreement, the master may exercise over the details of the work” of prime importance? Id., §220(1) comm. d. (“although control or right to control the physical conduct of the person giving service is important and in many situations is determinative, the control or right to control needed to establish the relation of master and servant may be very attenuated”).
Another multi-factor test is used to assess the voluntariness of a confession, Schneckloth v. Bustamonte, 412 U.S. 218, 223-27 (1973)(seven factors); Haynes v. Washington, 373 U.S. 503, 513-15 (1963); see, Miranda v. Arizona, 384 U.S. 436, 506-08 (1966) (Harlan, J., dissenting). Does each factor have equal weight? The Court simplified the voluntariness test by establishing a presumption for a wide swath of cases, those in which a Miranda warning had not been given.
 See Harper & Row Publishers v. Nation Enterprises, 471 U.S. 539, 556 (1985); Triangle Publ’ns v. Knight-Rider Newspapers, 626 F.2d 1171, 1175, 1177 (1980) (noting that the fourth statutory factor is “widely accepted to be the most important”); see also Religious Tech. Ctr. v. Netcom On-Line Com. Servs. 923 F.Supp. 1231, 1248 (N.D. Cal. 1995) (noting that though all statutory factors must be assessed, the fourth is “central”).
 See 73A CORPUS JURIS SECUNDUM Public Administrative Law and Procedure § 295 (“In an administrative adjudicatory proceeding, an agency or the legislature may provide that proof of one fact shall create a presumption of another fact.”)[available on WESTLAW].
 Certainly this is the assumption underlying Chevron v. Natural Resources Defense Council, 467 U.S. 837, 865 (1984). There, the Court said:
“In these cases the Administrator’s interpretation represents a reasonable accommodation of manifestly competing interests and is entitled to deference . . . Congress intended to accommodate both interests, but did not do so itself on the level of specificity presented by these cases. Perhaps that body consciously desired the Administrator to strike the balance at this level, thinking that those with great expertise and charged with responsibility for administering the provision would be in a better position to do so; perhaps it simply did not consider the question at this level; and perhaps Congress was unable to forge a coalition on either side of the question, and those on each side decided to take their chances with the scheme devised by the agency. For judicial purposes, it matters not which of these things occurred.”
Id. at 185 (emphasis added).
 See, Department of Transportation v. Association of American Railroads, 575 U.S. 43, 62 (2015)(Alito, J., concurring) (“it is hard to imagine how delegating ‘binding’ tie-breaking authority to a private arbitrator to resolve a dispute between Amtrak and the FRA could be constitutional[;][n]o private arbitrator can promulgate binding metrics and standards for the railroad industry.”)
 A seminal article that highlights the various ways in which Congress requires federal agencies to share a “regulatory space” is Jody Freeman and Jim Rossi, Agency Coordination in Shared Regulatory Space, 125 HARV. L. REV. 1131 (2012). The split enforcement model is only one such arrangement.
 Accord, Donovan v. A. Amorello & Sons, Inc., 761 F.2d 61, 63-66 (1st Cir. 1985)(Breyer, J.); Marshall v. Sun Petroleum Products, 620 F.2d 1176, 1184 (3rd Cir. 1980)(“the Review Commission was designed strictly as an independent adjudicator, with no rulemaking authority other than for procedural hearings, no direct policy role in administering the Act . . .”).
 In the Sutherland Treatise the comment is offered in the context of applying ejusdem generis. In particular, the inference regarding statutory drafting is used to suggest that matters of a “higher order” than those listed are not encompassed in the statutory enumeration. While the context is different in Texas Medical Association, the observation regarding legislative draftsmanship is relevant nevertheless.
Indeed, the human mind tend to best remember the first and last items in a series. See, The Decision Lab, Why do we better remember items at the beginning or end of a list?: The Serial Position Effect, explained; Dr. Saul McLeod, Serial Position Effect, Simply Psychology (2008).
 Albeit in a different context, the matters that most concerned the framers of the U.S. Constitution appear early in the enumeration of the federal government’s powers, namely the powers to tax, spend, borrow, and regulate commerce.
 Lists are often arranged alphabetically, chronologically, or sequentially. However, there is no reason to conclude that Congress ordered the section 300gg-111(c)(5)(C) factors in any of those ways.
 The Department also quite reasonably considered section 300gg-111’s overall structure. In their view, the detailed specification of the manner in which the QPA is to be calculated, particularly in contrast to the paucity of statutory guidance regarding consideration of the other enumerated factors, and the inclusion of audit provisions to ensure the insurers properly calculate the QPA suggested that “an accurate and clear calculation of the QPA is integral to . . . to the certified IDR entity’s determination of the out-of-network rate,” and should be the starting point for the inquiry. 86 Fed. Reg. at 55996.
 This is a typical problem for administrative agencies. See, Bernard W. Bell, Replacing Bureaucrats with Automated Sorcerers?, 150 DÆDALUS 89, 94 (2021)(“[t]he critical internal challenge for government bureaucracies is synchronizing line-level decision-makers, both with the intended agency policy and with each other”).
 As the Court observed in Yuckert: “We have recognized that other aspects of the Secretary’s sequential evaluation process contribute to the uniformity and efficiency of disability determinations. Heckler v. Campbell. The need for such an evaluation process is particularly acute because the Secretary decides more than 2 million claims for disability benefits each year, of which more than 200,000 are reviewed by administrative law judges.” (Citations omitted.)
 As Dubin notes, the grid, which has not been updated since the 1970’s is woefully out of date, and its use has become a challenge.
 Accord, Commonwealth of Massachusetts v. U.S., 856 F.2d 378 (1st Cir. 1988). The Nuclear Regulatory Commission (“NRC”) codified a presumption it had initially adopted in an adjudication. The rule created a rebuttable presumption that state and local officials would generally follow the applicable nuclear facility’s emergency plan in the event of a radiological emergency. Id. at 381-82. Petitioners challenged reliance on the presumption in circumstances where states and localities have previously refused to participate in utility company’s emergency planning for the relevant facility. The Court held that “the NRC rule [is not] objectionable because it is a ‘presumption.’ Agencies are permitted to adopt and apply presumptions if the proven facts and the inferred facts are rationally connected.” Id. at 383.
 Similarly, in a case that involved a presumption established in common-law fashion by the NLRB through its resolution of adjudication, the Court said: “[l]ike a statutory presumption or one established by regulation, the validity [of the Board’s presumption regarding the permissibility of no-solicitation rules], perhaps in a varying degree, depends upon the rationality between what is proved and what is inferred.” Republic Aviation Corp. v. NLRB, 324 U.S. 793, 804-805 (1945).
 See, Fischer v. F.C.C., 417 F.2d 551 (D.C. Cir. 1969). In Fischer, the FCC rejected applications submitted by two different applicants for a radio broadcast license based on a presumption established by the FCC in a policy statement. The policy statement established the rebuttable presumption that where an applicant’s signal would penetrate the geographic boundaries of a much larger community than the one to which the broadcast license is assigned, “the applicant ‘realistically proposes to serve that larger community rather than [the] specified (smaller) community.’” Id. at 553. The Court upheld the presumption as reasonable.
 The severity regulation provided quite explicitly:
“If you do not have any impairment or combination of impairments which significantly limits your physical or mental ability to do basic work activities, we will find that you do not have a severe impairment and are, therefore, not disabled. We will not consider your age, education, and work experience.”