(dissenting) — These regulations prohibiting most workplace smoking were not promulgated in accor*441dance with requirements of the Administrative Procedure Act (APA), RCW 34.05, or the Washington Industrial Safety and Health Act (WISHA), RCW 49.17. Neither the agency’s concise explanatory statement nor the administrative record itself demonstrates the rational rule-making process envisioned by the APA. Moreover the record fails to demonstrate the agency properly considered feasibility or that it based its decision on the best available evidence, as required by WISHA. On these grounds I would reverse and remand.
Procedurally, the APA requires a regulatory agency to issue rational regulations developed through a reasoned process, utilizing correct scientific factors while eschewing incorrect ones. Neah Bay Chamber of Commerce v. Department of Fisheries, 119 Wn.2d 464, 473-74, 832 P.2d 1310 (1992). Substantially, WISHA requires the regulations be at once feasible and predicated upon the best available evidence. RCW 49.17.050(4). Together the APA and WISHA statutes mandate regulations based upon the best available scientific evidence, while prohibiting those prompted by personal bias or political persuasion.
Unfortunately the specter that these regulations were promulgated on less than the required reasoned, scientific basis permeates this record. Even when Department of Labor and Industries (L&I) Director Mark Brown first announced the rules he proudly provided what the department termed “sound bites” such as “another blow to Joe Camel” and “How high does the body count have to go?,” suggesting an emotionalized hyperbole more indicative of a political decision than reasoned, scientific inquiry.15 AR 58: 34,470.16 The Joint Administrative Rules Review Committee of the Washington State Legislature apparently shared this concern, finding these regulations “not within the intent of the Legislature as expressed by the statute which *442the rules implement, and that the rules have not been adopted in accordance with all applicable provisions of law.” Clerk’s Papers at 123.
Regulations Not Issued in Compliance With APA
The APA requires regulations be “the product of a rational decision-maker.” RCW 34.05.570(2)(c) (1992). Judicial review of regulations requires the court to examine their reasonableness but not substitute its judgment for that of the agency. Neah Bay, 119 Wn.2d at 473. APA review measures the regulation against a three-part test: (1) is the agency’s explanation of its own rule clear; (2) has the agency used the appropriate statutory framework, using correct factors and eschewing incorrect ones; and (3) could a decision-maker reach the conclusion reached by the agency, taking the foregoing into account, by some reasonable process? Id. at 474. The standard for judicial review is not intended to be a smoke screen for affirmation, but requires “ ‘thorough, probing, in-depth review.’ ” Id. at 470 (quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 415, 91 S. Ct. 814, 823, 28 L. Ed. 2d 136 (1971), abrogated on other grounds by Califano v. Sanders, 430 U.S. 99, 105, 97 S. Ct. 980, 984, 51 L. Ed. 2d 192 (1977)).
Our review necessarily begins with the agency’s own explanation of its rule. Neah Bay, 119 Wn.2d at 474. The agency explanation is presented in the form of a legislatively mandated concise explanatory statement (CES) prepared by the department.17 “The purpose of such a statement is (1) to assure the agency actually considered all arguments made and (2) to facilitate court review.” Anderson, Leech & Morse, Inc. v. Liquor Control Bd., 89 Wn.2d 688, 693, 575 P.2d 221 (1978). If the CES on its face does not substantially comply with the statute and thus allow the court to conduct *443its review, the regulation so promulgated is invalid for failure to comply with this statutory mandate.18
The complete agency CES in relevant part provides:
There is a substantial body of evidence indicating that environmental tobacco smoke has a causal connection to lung cancer in non-smoking adults. The Environmental Protection Agency has concluded that environmental tobacco smoke is a Class A carcinogen. By definition, a Class A carcinogen is one where there is sufficient evidence from epidemiologic studies to support a causal connection between exposure to the agents and cancer. The conclusions of the Environmental Protection Agency are consistent with reports issued by the Surgeon General of the United States, the National Research Council, and other studies. Additionally, studies have documented a causal link between environmental tobacco smoke and cardiovascular disease.
The presence of environmental tobacco smoke in office workplaces is unnecessary to the work being performed in these workplaces. Therefore, the exposure of office workers to the hazards of lung cancer and cardiovascular disease is entirely unnecessary.
AR 58: 34,449. We begin by testing this CES under the announced criteria of the APA.
Obviously the department places primary reliance on the Environmental Protection Agency (EPA) report’s classification of secondhand smoke as a Class A carcinogen; however, it fails to recognize the important arguments which both challenge this conclusion and render it immaterial. The CES does not evidence the department’s reasoning whereby it accepts thé EPA report as reliable and valid, nor does it place the presence of an alleged carcinogenous element in the context of a predictable and substantial risk to persons *444in the workplace, as Class A carcinogens also include nickel, an element that can be found in mother’s milk, and chromium, a substance found in all tap water.19 We do not ban these substances absent a further showing of practical harm in the environment of actual use; however, the CES does not, on its face, provide that link.
Another central failing of the CES is its failure to evidence critical consideration of the scientific reliability and validity of the EPA report upon which the entire regulation purports to stand. Since there are serious arguments advanced in this record that the EPA report is unscientific and unreliable, it is incumbent that the CES at least identify these arguments so that the court may be assured on judicial review that the agency afforded these objections proper consideration. But it does not.20
The CES states the EPA report is consistent with other unspecified reports and studies; however, we are not told what the consistencies are, nor what the inconsistencies may be. Nor, most importantly, are we told why we should accept these reports as true while rejecting others. These were also questions raised by appellants and not addressed in the CES.
We are told that environmental tobacco smoke (ETS) is “unnecessary to the work being performed” (AR 58: 34,449) and that “[tjherefore” the exposure of workers to hazards of ETS is “entirely unnecessary.” Id. This does state the agency view on feasibility. It tells us the director considers employee smoking preferences illegitimate, from which he concludes any interest in smoking held by these workers is not worthy of regulatory consideration. Whether this reasoning meets the statutory standard for feasibility will be considered later.
*445Failure to recognize important arguments conflicts with the most basic statutory requirements of a proper CES in violation of RCW 34.05.355, repealed by Laws of 1995, ch. 403, § 305 (hereinafter RCW 34.05.355 (1992)). Moreover, the rule as explained by the CES is most problematic under the third prong of the three-part analytical test set out in Neah Bay because it does not provide a clear explanation of the thought process behind the rule, thus denying assurance on judicial review that the rule resulted from a reasoned process. See Neah Bay, 119 Wn.2d at 474.
The majority’s overlooking these shortcomings of the CES might be justified if the administrative record did provide the means to review the department’s actions. Yet, as Director Mark Brown testified, no document in the 45,000 pages of the administrative record articulates any explanation for the department’s decision to regulate ETS in the manner in which it did. Report of Proceedings (RP) at 146-47.
As explained in Neah Bay, our review of the administrative record must enable us to determine the factors employed by the agency and the quality of its reasoning to test its statutory compliance; we do not simply seek evidence to sustain the agency’s conclusion. Neah Bay, 119 Wn.2d at 474. However this administrative record fails to provide any assistance as to the agency’s reasoning process. See also Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto Ins. Co., 463 U.S. 29, 50, 103 S. Ct. 2856, 77 L. Ed. 2d 443 (1983) (agency actions must be upheld on the basis articulated by the agency itself, not post hoc rationalizations of its counsel).21
The majority seeks to distinguish Neah Bay by stating *446that in that case there was no administrative record to supplement. Majority at 419-20. True, but that is not the point. The purpose of the administrative record is to provide a reviewing court with evidence that the agency had in fact engaged in the necessary rational process required by the statute. Neah Bay, 119 Wn.2d at 471 (“ ‘agency must examine the relevant data and articulate a satisfactory explanation for its action including a “rational connection between the facts found and the choice made” ’ ”) (quoting with approval State Farm, 463 U.S. at 43). Whether the record is nonexistent or is simply a catchall collection of 15 boxes is irrelevant. Post hoc rationales and testimony cannot supplement (nor lawfully supplant) that which is not there to begin with. If there is nothing in the administrative record that the department can identify which shows even a modicum of support for its claim that these regulations were the product of reason on the part of the regulators, then the regulation must fail without regard to post hoc rationalizations. Neah Bay, 119 Wn.2d at 473-74.22
It is equally the court’s role to “examine the actions of the agency to ensure that they were only undertaken after proper consideration of appropriate facts.” Neah Bay, 119 Wn.2d at 475. This we must do.
Central to the regulation, and our review, is the EPA report. The department issued these regulations primarily purporting to base them on this report, however, admitted it did no independent research or statistical analysis, nor *447even a critical review of the reliability and accuracy of this critical source document. RP at 116-18; 388-90. In fact, the agency employee charged with responsibility to review the EPA report testified he did not even have the necessary skills to even evaluate the EPA analysis. RP at 482-83; 573. Reliance based on blind faith or prejudicial predisposition is inconsistent with the department’s statutory obligation to engage in a rational decision-making process because it is not rational. See AFL-CIO v. OSHA, 965 P.2d 962, 984-85 (11th Cir. 1992) (agency may use outside consultants in formulating regulation, but may not merely adopt outside recommendations, and, instead, should set forth its own analysis and findings justifying regulations).
The majority excuses this abdication by noting the department made a rational decision to not “reinvent the wheel.” Majority at 426. However, there is a difference between “reinventing the wheel” and inspecting the wheel to make sure it is round, not square. And there is a difference between relying on a comprehensive and critical examination of all the data, and adopting the recommendations of a selected report simply because it affirms the predetermined result the reviewer seeks. Moreover, nothing in this record demonstrates the reasoning process behind the department’s decision to conclude this report is “the best available evidence” as required by WISHA. RCW 49.17.050(4).
Recent judicial review of the EPA report found much to criticize:
In this case, EPA publicly committed to a conclusion before research had begun; excluded industry by violating the Act’s procedural requirements; adjusted established procedure and scientific norms to validate the Agency’s public conclusion, and aggressively utilized the Act’s authority to disseminate findings to establish a de facto regulatory scheme intended to restrict Plaintiffs’ products and to influence public opinion. In conducting the ETS Risk Assessment, EPA disregarded information and made findings on selective information; did not disseminate significant epidemiologic information; deviated from its Risk Assessment Guidelines; failed to disclose important findings and reasoning; and left significant ques*448tions without answers. EPA’s conduct left substantial holes in the administrative record. While so doing, EPA produced limited evidence, then claimed the weight of the Agency’s research evidence demonstrated ETS causes cancer.
Flue-Cured Tobacco Coop. Stabilization Corp. v. United States Envtl. Protection Agency, 4 F. Supp. 2d 435, 465-66 (M.D.N.C. 1998) (footnote omitted).23
Even the actual record before the agency undermines confidence in this reliability of the report. See, e.g., Maurice LeVois, Ph.D., The EPA Report on Environmental Tobacco Smoke Does Not Provide a Scientific Basis for Regulating Smoking in the Workplace (Dec. 14, 1993) (AR 48: 25,143-154) (EPA report concentrated on spousal studies which supported its view while ignoring workplace studies that were contrary); Maxwell W. Layard, Ph.D., Comments on the EPA Review Draft “Respiratory Effects of Passive Smoking: Lung Cancer and Other Disorders” (June 30, 1992) (AR 41: 22,623-647) (statistical analysis in EPA report unsupportable—studies show only a statistically nonsignificant result). As one law review analysis notes, the EPA simply ignored the data on workplace studies, which overwhelmingly showed no statistically significant overall lung cancer risk to nonsmokers from ETS. John C. Fox, An Assessment of the Current Legal Climate Concerning Smoking in the Workplace, 13 St. Louis U. Pub. L. Rev. 591, 629 (1993). Nor is there any evidence in this record demonstrating the director recognized or even questioned whether the EPA manipulated the data in violation of its own guidelines “to artificially double the likelihood of rul*449ing out chance as a possible explanation” of increased risk of lung cancer. Id. at 630. Faith is not reason.24 I fear the majority has too much of the former.
Perhaps another even more telling example of how the department’s lack of reason, if not predisposition, potentially infected the final regulations can be seen in the impact on the department by the Surgeon General’s 1986 report. See U.S. Dep’t of Health & Human Servs., The Health Consequences of Involuntary Smoking: a report of the Surgeon General (1986). This report is one of the documents mentioned in the CES. It is also one of the reports that L&I Director Brown purported to have relied upon. RP at 112; see also Majority at 429 (citing report). Yet the copy of the report in the record submitted by the department is edited to omit the sections most relevant to the question pertinent to the regulation. Missing is much of the section of the report on “Policies Restricting Smoking in Public Places and the Workplace,” including nearly the entire section entitled, “Review of Current Evidence on Impact,” as well as all of the section on “Recommendations for Future Research,” and the report’s “Conclusions.” AR 52: 32,434-83. Why these omissions? What reason justifies these omissions? Where is the evidence demonstrative of a rational review of the data, or even an *450actual review of the relevant data? How could the director reasonably rely upon this report to justify the regulation when the critically relevant material is edited out? The majority doesn’t tell us.
The record also contains numerous studies which contradict the conclusion that ETS is a hazard in the workplace. See, e.g., Geoffrey C. Kabat & Ernst L. Wynder, Lung Cancer in Nonsmokers, 53 Cancer 1214, 1219 (1984) (AR 41: 22,669-74) (plausibility of role of passive inhalation in lung cancer can be questioned on several grounds); EN. Lee et al., Relationship of passive smoking to risk of lung cancer and other smoking-associated diseases, 54 Br. J. Cancer 97, 101 (1986) (AR 41: 22,718; 22,722) (recent review of data has concluded that overall there is no reliable scientific evidence of a causal relationship between passive smoking and lung cancer); Anna H. Wu, Ph.D. et al., Smoking and Other Risk Factors for Lung Cancer in Women, 74 J. Nat’l Cancer Inst. 747, 748 (Apr. 1985) (AR 41: 22,742; 22,743); Lawrence Garfinkel, Time Trends in Lung Cancer Mortality Among Nonsmokers and a Note on Passive Smoking, 66 J. Nat’l Cancer Inst. 1061, 1065 (1981) (AR 41: 22,802, 22,806). But why did the department refuse to credit studies which contradict the EPA report’s conclusions? See RP at 571-75. Did the department seek an answer, or a result? The CES is simply silent on why the EPA report is credited while its critics are ignored. The court must look for reasons, but none can be found in this administrative record.
Even the materials cited by the majority are not clearly indicative of a correlation between passive smoke and cancer. For example, Ross C. Brownson et al., cited by Majority at 430, wrote that “[i]n general, there was no elevated lung cancer risk associated with passive smoke exposure in the workplace.” Ross C. Brownson et al., Passive Smoking and Lung Cancer in Nonsmoking Women, 82 Am. J. Pub. Health 1525, 1527 (1992) (AR 57: 34,255; 34,257). The report by the Washington Department of Health and Fred Hutchinson Cancer Research Center Cancer Prevention Research *451Program (1990 Hutchinson Cancer Research Ctr. Report), cited by Majority at 428, calculates a total death rate from passive smoking of 89.5 deaths a year (including spousal, family, and restaurant exposures), yet, admits it bases this number upon a “crude” analysis which in turn is predicated on assumptions and estimates. See Washington Dep’t of Health & Fred Hutchinson Cancer Research Ctr, Smoking in Washington State: Status of the Problem and Recommendations for Intervention from the Tobacco-Related Cancers Technical Committee 72 (Apr. 1990) (AR 53: 32,524). The Glantz and Parmley report, cited by Majority at 431 n.10, is also a study of the relationship between smoking at home and the risk of heart disease. Stanton A. Glantz & William W. Parmley, Passive Smoking and Heart Disease: Epidemiology, Physiology, and Biochemistry, 83 Circulation 1, 1 (Jan. 1991) (AR 10: 13,001). The NIOSH report, cited by the Majority at 428, and by Director Brown in his testimony, RP at 112, is entitled, Environmental Tobacco Smoke in the Workplace. U.S. Dep’t of Health & Human Servs., Nat’l Inst, for Occupational Safety & Health, Environmental Tobacco Smoke in the Workplace: Lung Cancer and Other Health Effects (June 1991) (AR 49: 30,057). This is an interesting title since the overwhelming data come from studies of household smoking. AR 49: 30,075.25
Notwithstanding, the majority broadly asserts, “It is difficult to even characterize such evidence [regarding the dangers of workplace ETS] as ‘conflicting,’ let alone question the Department for resolving any conflict in the manner it did.” Majority at 430-31. Not only is this statement simply incredible in the face of the overwhelming evidence *452presented in this record, it abdicates our obligation of review as “a disagreement among experts is irrelevant to the standards set forth in the APA . . . .” Neah Bay, 119 Wn.2d at 475. Rather, we “must make certain that judgment was in fact exercised properly and fairly.” Id. (emphasis added). The majority substitutes blind deference for substantive review. In so doing it abdicates the judicial responsibility to protect our fellow citizens from irrational bureaucratic excess. I would hold L&I to the requirements of the statute which gave it existence and confer upon it rule-making authority.26
Even accepting the department’s post hoc rationalizations, it is clear that the regulations enunciated were not the result of a rational decision-making process, and thus not in compliance with the APA.27
Regulations Violate WISHA
WISHA requires a health and safety standard be set “which most adequately assures, to the extent feasible, on the basis of the best available evidence, that no employee will suffer material impairment of health or functional capacity even if such employee has regular exposure to the hazard . . . .” RCW 49.17.050(4). We begin by recognizing if the department’s reliance on the flawed EPA report is a product of prejudice, questionable analysis, and/or relies upon data manipulated in violation of the EPA’s own guidelines, the regulation is hardly predicated “on the best available evidence.” Id.
This court should be guided by the United States *453Supreme Court’s interpretation of identical language in the federal Occupational and Safety Health Act (OSH Act).28 That Court construed language which required the regulatory agency to set a health standard that “ ‘most adequately assures, to the extent feasible, on the basis of the best available evidence’ ” that no material harm would result, to mean the agency should consider differences in degrees of significance rather than simply a total elimination of all risks. Industrial Union Dep’t, AFL-CIO v. American Petroleum Inst., 448 U.S. 607, 643 n.48, 100 S. Ct. 2844, 65 L. Ed. 2d 1010 (1980) (plurality opinion) (hereinafter Benzene) (quoting § 6(b)(5) of the federal OSH Act).29 But what risk does workplace smoke visit on the worker? In what concentration or for what duration does the risk become significant? Are the chances of contracting cancer from ETS less than being struck by lightning on a sunny day, or is the slightest whiff enough to send us choking to the emergency room? Inquiring minds want to know. Proper judicial review requires assurance that the agency has *454investigated the questions and concluded that the burden imposed by the regulation is justified by the risk which it eliminates.30
In Benzene the Court held that prior to the issuance of any safety standard, a determination must be made that the standard is reasonably necessary and appropriate to remedy a significant risk of material health impairment. Benzene, 448 U.S. at 639. As the Court explained, the idea behind the regulations was not to create a risk-free workplace, but a safe one, and that “a workplace can hardly be considered ‘unsafe’ unless it threatens the workers with a significant risk of harm.” Id. at 642. Thus, the Court concluded, “at a bare minimum” new standards could not be issued absent a finding of “a significant risk of harm and therefore a probability of significant benefits . . . .” Id. at 644. To hold otherwise, said the Court, would give an agency the power “to impose enormous costs that might produce little, if any, discernible benefit.” Id. at 645.
Incredibly, the majority concludes that L&I did comply with Benzene. Majority at 425-27. But Benzene says that while it does not impose a “mathematical straitjacket,” the agency “does have an obligation to find that a significant risk is present before it can characterize a place of employment as ‘unsafe.’ ” Benzene, 448 U.S. at 655 (emphasis added). But here there is no such finding. As L&I Manager John Peard (who prepared the regulations) testified, the idea that ETS poses a significant risk is not in written form anywhere in the administrative record nor in any of the written materials he passed to then L&I Director Mark Brown. RP at 385; see also RP at 244 (testimony of L&I WISHA Coordinator Stephen Cant). L&I also made no effort to seek out or quantify any risk to Washington workers from ETS. RP at 399-400; 580-81. It simply relied upon questionable data that supported a conclusion which it ac*455cepted without question. RP at 323; 482-83; 573-74. It was exactly this lapse in effort that the Supreme Court found unforgivable in Benzene. Benzene, 448 U.S. at 646; see also id. at 637 (noting that OSHA had impermissibly relied upon its assumption that “some benefits” would result from a reduction in exposure to Benzene).
The majority also notes the difference between the necessary “by-product” at issue in Benzene and the “unnecessary” ETS. Majority at 424. The department seems to forgo any analysis of actual risk based on the CES claim that smoking is “unnecessary.” However, that is not the test. See, e.g., American Textile Mfrs. Inst. v. Donovan, 452 U.S. 490, 513 n.32, 101 S. Ct. 2478, 69 L. Ed. 2d 185 (1981) (all standards must be addressed to “significant risks” of material health impairment); AFL-CIO, 965 F.2d at 975 (agency “has a responsibility to quantify or explain, at least to some reasonable degree, the risk posed by each toxic substance regulated” (emphasis omitted)).
The majority is also oblivious to the economic and social costs of this regulation.31 In fact the costs are potentially enormous. According to the state, a typical smoking lounge to accommodate 10 smokers “could be built for as little as $1,500.” Resp’ts’ Br. at 66 (emphasis added). Recognizing there are more than 130,000 private firms in Washington (U.S. Department of Commerce, Statistical Abstract of the United States 1998, at 550 (Oct. 1998))—and counting only private employers and further assuming that each firm has only one indoor work environment, whatever the firm’s size—the cost of implementing this regulation statewide must exceed $195,000,000. Although the true economic impact of these regulations was the subject of much *456testimony before the department, see, e.g., AR 64: 42,421-23; 42,660; 42,714; 42,766, the department simply dismissed all cost analysis cavalierly issuing a statement instructing “[e]mployers who wish to accommodate their smoking employees will have to calculate costs factors [themselves] .... The cost could be nothing for employers with ‘smoke free’ policies for their buildings.” AR 58: 34,457 (emphasis added).
More fundamental, however, is the loss of freedom to our workers and their employers to freely bargain working conditions to meet their preferences.
How would our majority review a regulation which forbids men to wear ties or women scarves so as to entirely eliminate the remote prospect that a tie might be caught in a copy machine? In a sense one could say the dollar cost is nil; however, the cost measured by the loss in personal freedom to dress as one pleases would be great. Is not that also a cost worthy of consideration?
We do things not only because we must but because we will. We purchase things not only because they are absolutely “necessary” but also because they give us pleasure. Such is the blessing of a free society. But when the director or the majority says “there is no cost,” what they mean is “there is no cost to them.”32 The cost is borne by those whose individual choices are precluded by the coercive power of the state. In Washington more than 800,000 smoke. AR 58: 34,462. Under preexisting law it is not within the agency’s purview to impose that cost absent a reasoned analysis based upon scientific fact.
If the requirements of complying with this rule are in excess of proven benefits the rule violates WISHA. But how can we review that prospect when the relevant analysis was not made by the director? As the Supreme Court has *457explained, an agency is not empowered to establish regulations that “provide absolutely risk-free workplaces whenever it is technologically feasible to do so . . . . [the Federal OSH Act] was intended to require the elimination, as far as feasible, of significant risks of harm.” Benzene, 448 U.S. at 641. There the Court invalidated a ban on dermal contact with Benzene, which was implemented on the basis of a no-risk policy. Id. at 662. Moreover, where is the evidence that the department even considered, much less appropriately rejected, less onerous regulations. Such is also an aspect of feasibility. See American Textile Mfrs., 452 U.S. at 513 n.32. What about limiting ETS exposure instead of a total ban? What about segregating smokers and nonsmokers? Would a less onerous regulation achieve an acceptable level of risk? What is the risk? And what is acceptable? Neither the director nor the majority tells us. I posit absent evidence of thoughtful consideration these regulations do not meet the statutory standard.
Conclusion
The department’s regulations raise the specter that rather than being the product of a rational process, they may well reflect what one scholar has described, as “the contemporary anti-tobacco frenzy.” Rivka Widerman, Tobacco is a Dirty Weed. Have We Ever Liked It? A Look at Nineteenth Century Anti-Cigarette Legislation, 38 Loy. L. Rev. 387, 389 (1992).33 The department chose to issue only the regulations regarding ETS, and to ignore the many *458other possible sources of health problems resulting from lowered indoor air quality. AR 58: 34,453. Yet reported statistics suggest ETS is the source of indoor air problems only in a minute number of cases. Gray Robertson, “Indoor Pollution: Sources, Effects and Mitigation Strategies” in Environmental Tobacco Smoke, Proceedings of the International Symposium at McGill University 1989, at 333 (Donald J. Ecobichon & Joseph M. Wu eds., 1990).
The department chose to ignore the conclusion of two-thirds of the legislative members of the Joint Administrative Rules Review Committee that the regulations violated the legislature’s intent and were not adopted in accordance with all applicable provisions of law.
The department chose to rely entirely upon a highly flawed study, to ignore all contrary evidence, and to refrain from conducting its own analysis of the data it used.
The department chose not to enter findings that ETS actually poses a substantial risk to the health of the workers of the State of Washington, much less quantify such in even the most rudimentary way.
*459The department chose not to consider the feasibility or economic impact of its regulations.
These are the choices that the majority affirms. Such blanket authority to impose regulations upon even a known carcinogen, which ETS is not, grants to the department pervasive power to impose enormous costs that might produce little, if any, discernible benefit. Such is not the purpose of either the APA or WISHA, but rather the evil against which those statutes guard when properly enforced. See Benzene, 448 U.S. at 645.
As the Supreme Court has warned, “ ‘[ejxpert discretion is the lifeblood of the administrative process, but “unless we make the requirements for administrative action strict and demanding, expertise, the strength of modern government, can become a monster which rules with no practical limits on its discretion.” ’ ” State Farm, 463 U.S. at 48 (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 167, 83 S. Ct. 239, 245, 9 L. Ed. 2d 207 (1962) (quoting New York v. United States, 342 U.S. 882, 884, 72 S. Ct. 152, 153, 96 L. Ed. 662 (1952) (Douglas, J., dissenting))). I would recognize these regulations are invalid because they do not comply, either procedurally or substantially, with state law, and remand to the department with instructions to consider any future ETS regulations in accordance with the requisites of the APA and WISHA.
Dolliver, J. Pro Tern., concurs with Sanders, J.
Perhaps appropriately, the L&I newsletter announcement of the regulations also cited a letter “urging Brown to run for governor.” AR 58: 34,471.
AR refers to the administrative record (L&I’s rule-making file) in this case. See Majority at 417 n.3.
“At the time it files an adopted rule ... an agency shall place into the rule-making file ... a concise explanatory statement about the rule, identifying (a) the agency’s reasons for adopting the rule . . . RCW 34.05.355(l)(a), repealed by Laws of 1995, ch. 403, § 305 (hereinafter RCW 34.05.355 (1992)) (emphasis added).
The majority notes that the APA at that time did require the department to issue a concise statement of the principal reasons for overruling the considerations urged against its adoption at the request of any interested person, a request not made hy appellants. Majority at 421 (citing ROW 34.05.355(2) (1992)). This, however, does not relieve the agency of its independent burden to comply with the statute to meet its purpose. The failure of an “interested person” to object does not waive the duty of the agency to follow the law.
John C. Fox, An Assessment of the Current Legal Climate Concerning Smoking in the Workplace, 13 St. Louis U. Pub. L. Rev. 591, 628 (1993). (Author’s note: Manuscript of this article may be found at AR 48: 25,155-226.)
As the United States District Court has opined, “[t]he court is faced with the ugly possibility that EPA adopted a methodology for each chapter, without explanation, based on the outcome sought in that chapter.” Flue-Cured Tobacco Coop. Stabilization Corp. v. United States Envtl. Protection Agency, 4 F. Supp. 2d 435, 456 (M.D.N.C. 1998).
I note that the majority questions the relevance of federal administrative law cases, stating that “[ajppellants strain to import a federal standard of review into this state APA case.” Majority at 418. But it is our Legislature that has mandated this course of action. As we have held, the Legislature “specifically provided that the APA be interpreted in a manner consistent with other states, model acts, and federal decisions.” Neah Bay, 119 Wn.2d at 470 (citing RCW 34.05.001) (emphasis added). The majority seeks to justify its decision to ignore federal precedents our Legislature has mandated us to follow by quoting a law review article that bemoans the absence of specific language from the state APA. *446Majority at 419 (quoting William R. Andersen, The 1988 Washington Administrative Procedure Act—An Introduction, 64 Wash. L. Rev. 781, 803-04 (1989)). However the very language whose noninelusions so trouble Professor Andersen is also not present in the federal APA. See id. at 803 n.125. Therefore, there is no reason why the absence of that language should impact the following of the Legislature’s command to respect federal precedent.
The majority writes, “a distinguishing fact in Neah Bay cannot be stressed enough: there the Department of Fisheries had failed to present any administrative record as required by the APA.” Majority at 420. However, our holding in Neah Bay was not dependent upon that fact, as we did not address the full impact of a missing administrative record but merely concluded that a “lack of a rule-making file may itself constitute a sufficient reason to invalidate a regulation.” Neah Bay, 119 Wn.2d at 476 (emphasis added).
The majority notes that this decision is on appeal, but then adds that even if the decision is correct, “the Department rule-making built atop it four years earlier would not . . . collapse like a house of cards. Regulations are not that tenuous.” Majority at 427. Why not? It was the department’s statutory duty to rely upon the “best available evidence” and to engage in a rational decision-making process. The ÉPA report was no more reliable then than it is now. The only difference is the director did not conduct any inquiry to determine its reliability, but simply accepted it without articulated basis; whereas now others have performed the critical review which the Director was legally required to perform in the first place before imposing this regulation upon hundreds of thousands of our fellow workers and employers at great cost.
The EPA report included no workplace studies, and instead relied upon reports of nonsmoking spouses of smokers. The majority dismisses this distinction by quoting the trial court’s conclusion that workplace risks would at least be equal to, if not greater than, the spousal studies. Majority at 430 (citing RP at 822-23). However, the conclusions of a judge acting to review an administrative record are no better than the record before him. This is a question for science, not law. See, e.g., Maurice LeVois, Ph.D., The EPA Report on Environmental Tobacco Smoke Does Not Provide a Scientific Basis for Regulating Smoking in the Workplace (Dec. 14, 1993) (AR 48: 25,145-47) (outlining fundamental flaws in using spousal data to estimate lung cancer risks); John C. Fox, supra note 5, at 629 (OSHA has made clear that spousal data are of limited applicability in assessing actual risk in occupational exposure conditions). Also, there is much dispute as to even whether the data on family smoking are accurate. See, e.g., EN. Lee, Environmental Tobacco Smoke and Mortality (1992) (AR 41: 22,678); Linda C. Koo et al., Measurements of Passive Smoking and Estimates of Lung Cancer Risk Among Non-Smoking Chinese Females, 39 Int’l J. Cancer 162, 162 (1987) (AR 41: 22,706); Anna H. Wu, Ph.D., et al., Smoking and Other Risk Factors for Lung Cancer in Women, 74 J. Nat’l Cancer Inst. 747, 748 (Apr. 1985) (AR 41: 22,742, 22,743).
The National Institute for Occupational Safety and Health’s (NIOSH’s) own bent can be seen in its analysis of one of the studies it relied upon. After noting the many flaws of the study’s methodology and miniscule size, including a question as to whether the “nonsmokers” involved had ever smoked, NIOSH concludes, nonetheless, the report “did suggest an increased risk for adenocarcinoma among nonsmokers exposed to ETS.” U.S. Dbp’t of Health & Human Servs., Nat’l Inst, for Occupational Safety & Health, Environmental Tobacco Smoke in the Workplace: Lung Cancer and Other Health Effects 1, 5 (June 1991) (AR 49: 30,070).
Contrary to the majority’s assertion at 439-40 n.14, there is no effort here to retroactively apply a change in the APA to this regulation. Nor does application of the rule of law as stated by this court in its 1992 decision in Neah Bay to an agency regulation issued in 1994 imply retroactivity.
The United States Department of Labor, unlike L&I, has engaged in the proper regulatory process, and has, so far, declined to issue workplace regulations based upon the EPA report or the findings of other governmental agencies. App. 2 to Appellants’ Reply Br., Action on Smoking and Health v. OSHA, No. 92-1661 (D.C. Cir. 1994) (Br. for the Secretary of Labor at 32 n.24) (OSHA may not simply rely on EPA report but must make its own significant risk and feasibility determinations).
While the Supreme Court’s interpretation of such language is not binding it is nonetheless worthy of respect, especially where, as here, adequate state precedent is absent. See Xieng v. Peoples Nat’l Bank, 120 Wn.2d 512, 531, 844 P.2d 389 (1993).
As the majority notes, department regulations may exceed OSHA regulations. Majority at 423 (citing RCW 49.17.050(2)). However, this is irrelevant for two reasons: (1) OSHA has yet to issue ETS regulations because of its decision not to rely on outside agency data, such as the EPA report, see supra note 27; (2) the issue is not whether L&I may exceed OSHA regulations, hut the meaning of identical statutory language.
The majority notes that the rationale behind the Benzene holding was endorsed by only four Justices. Majority at 423. The majority also reads the Supreme Court’s later reference to Benzene to merely indicate that OSHA itself has adopted the test. Id. (citing American Textile Mfrs. Inst. v. Donovan, 452 U.S. 490, 505 n.25, 101 S. Ct. 2478, 69 L. Ed. 2d 185 (1981)). However American Textile Manufacturers holds otherwise. There the Court makes clear it is analyzing whether OSHA has complied “with this Court’s decision in [Benzene].” American Textile Mfrs., 452 U.S. at 505 n.25. Other federal authority also supports the conclusion that the “plurality” view is the binding one. See, e.g., AFL-CIO v. OSHA, 965 P.2d 962, 972 n.13 (11th Cir. 1992) (“courts of appeals have generally considered that the plurality opinion in Benzene was implicitly adopted by a majority of the Court in. American Textile Mfrs. Inst. v. Donovan."). (Citations omitted.) While we are not hound by Benzene, I see no reason why we should not give similar language a similar construction. The Benzene lead opinion is certainly more persuasive, both because of its authoritative nature and its excellent reasoning, than the Benzene dissent which our majority prefers. See Majority at 432.
Dr. Morton Lippman, chairman of the Science Advisory Board committee that reviewed the EPA report, when asked about the danger of ETS replied that the “questioner probably incurred a greater risk driving through Washington traffic to ask his question than he would incur in a lifetime’s exposure to ETS.” John C. Fox, supra note 19, at 628.
While the majority describes these costs as minimal, Majority at 433-34, and adds that this regulation can be complied with cost-free simply by banning smoking, id. at 433 n.12, it also points out a regulation that banned smoking “would present a different case than this one,” id. at 433, while appearing to simultaneously accept a de facto ban as perfectly appropriate. See id. at 433 n.12 (“One struggles to imagine how a working population composed of a vast majority of nonsmokers would somehow be demoralized by the absence from their workplaces of a substance than can kill them.”). The distinction the majority attempts to draw is of degree, not kind, with little difference in degree.
Antismoking regulations have a disproportionate effect upon the poor and African-Americans. Rivka Widerman, Tobacco is a Dirty Weed. Have We Ever Liked It? A Look at Nineteenth Century Anti-Cigarette Legislation, 38 Loy. L. Rev. 387, 421-22 (1992); John C. Fox, supra note 19, at 625 n.186. In fact, policies limiting workplace smoking may well trigger “ ‘disparate impact’ discrimination claims.” Id.
Antismoking regulations inspired by ideological zeal have an inglorious history.
The first European caught smoking was imprisoned by the Inquisition. Wider-man, at 396. Yet Native Americans, who had been using tobacco since the Fifth Century, did not share in this prejudice. Id. at 395. In 1604 James I (James VI of Scotland) authored the first attack on environmental tobacco smoke. Jacob Sullum, Foe Your Own Good: The Anti-Smoking Crusade and the Tyranny of Public Health 143 (1998). In Russia, lips were slit for smoking and noses torn away for using snuff, while in Turkey smoking was a capital offense, prompting mass executions by torture which served only to intensify the desire for that which had been forbidden. Id. at 20-21.
In the 1930s and early 1940s the world’s strongest antismoking movement was championed by state policy in National Socialist Germany. This is not surprising considering the prevailing emphasis on preservation of the race and Hitler’s *458suggestion that had he not given up smoking Nazism might never have succeeded. Robert N. Proctor, The Anti-Tobacco Campaign of the Nazis: a little known aspect of public health in Germany, 1933-45, 313 Beit. Med. J. 1450, 1450 (1996). Yet even such -will was unable to triumph over smoking by propaganda alone as smoking rates actually rose dramatically in the six years after Hitler came to power. Only in the late 1930s and early 1940s was the German state able to reduce smoking though workplace regulations and other extensive bans. Id.
Washington has had its own flirtation with the heady smoke of tobacco prohibition. In Í893, the manufacture, sale, and purchase of cigarettes and cigarette paraphernalia were prohibited. Laws of 1893, ch. LI. This was the first state prohibition on cigarettes enacted in America. Cassandra Tate, Cigarette Wars: The Triumph of “The Little White Slaver” app. at 159 (1999). In 1895 and 1901, extensive regulations on manufacture and sale were enacted. Laws of 1895, ch. LXX; Laws of 1901, ch. CXXII. In 1907 a new ban on sale and manufacturing was passed. Laws of 1907, ch. 148, § 1. Apparently unsatisfied, two years later the legislature outlawed mere possession of cigarettes or cigarette papers. Laws of 1909, ch. 249, § 284. However, as is frequently the case with youthful experimentation, the legislature grew up and repealed this total ban in its next session, limiting the tobacco prohibition to those under 21, Laws of 1911, ch. 133, § 1(4). Like other states, Widerman, supra, at 389, Washington has decided to switch rather than fight, and cash in on the cigarette trade by taxation. RCW 82.24; Laws of 1935, ch. 180, § 83. Interestingly, although statutes regulating the sale or manufacturing of cigarettes were in place from 1895-1907 and from 1911 on, the legislature did not actually repeal the 1893 prohibition on the sale, purchase, and manufacturing of cigarettes until 1925. Laws of 1925, Ex. Sess., ch. 17, § 1.