(dissenting):
While the massive record in this case and the extensive majority opinion indicate that a tremendous amount of very thoughtful study and consideration has been given to the many issues involved, it is my view that a number of bases were not properly touched. Hence the agency is not “home free.” The case should accordingly be remanded: (1) to correct the improper use of consultants, (2) to issue a proper Notice of Rulemaking on the 50 ug/m3 standard and conduct a proper hearing thereon with right of cross-examination, (3) to produce substantial evidence, subject to cross-examination, that properly supports the feasibility of the standard for “Other Industries” which was not done in The Short Report, (4) *183to strike the medical removal protection system because it directly violates the prohibition of the act against “superseding] or in any manner affect[ing] any workmen’s compensation law,” (5) to hold further hearings on the feasibility of the standard imposed because it is not presently supported by substantial evidence and violates the holding in the recent Benzene Case, Industrial Union Department, AFL-CIO v. American Petroleum Institute, 448 U.S. 607, 100 S.Ct. 2844, 65 L.Ed.2d 1010 (1980) by relying on impermissible presumptions, (6) to employ proper methods as outlined hereafter in determining the feasibility of the lead standard, and (7) to make a cost-benefit analysis to determine if the evidence supports a finding of significant benefits derived from economic feasibility.
I. IMPROPER USE OF CONSULTANTS
Subsequent to the close of the record in this rulemaking, and prior to the promulgation of the final standard, OSHA contracted with outside consultants to perform an evaluation, presumably only of record evidence,1 on two topics. First, the agency asked David Burton and David Burton Associates (DBA) to review the voluminous record and evaluate the feasibility of a permissible air-lead exposure standard of 50 ug/m3. The rule as proposed in the notice,2 and virtually all of the record evidence, referred only to the feasibility of a permissible exposure limit of 100 ug/m3. Burton and DBA had previously prepared reports for OSHA which had been introduced into the record, and had testified as expert witnesses during the public hearings in support of the economic and technological feasibility of the 100 ug/nf proposed level. After the record was closed the agency also contracted with the Center for Policy Analysis and Nicholas Ashford to analyze the scientific and medical correlation between air-lead and blood-lead levels. Ashford had submitted a preliminary report on this correlation during the record period as an expert witness for OSHA. The lengthy analyses prepared by the consultants in fulfillment of their contractual responsibilities have never been released to the parties or the public, despite the fact that as far as the report on the economic and technological feasibility of 50 ug/irt3 is concerned, it is the only in depth evidence on the topic in existence. The delegation of this task to these biased witnesses, and the failure to introduce the reports into evidence, constitute prejudicial error which requires the remand of the case to the agency on this point. As Chief Justice Hughes remarked in Morgan v. United States, 298 U.S. 468, 480, 56 S.Ct. 906, 911, 80 L.Ed. 1288 (1936) (Morgan I), “[n]othing can be treated as evidence which is not introduced as such.” In the absence of this report there is insufficient evidence to support the finding.
The majority is correct in the first part of its section addressing the use of outside consultants, in concluding that they may be hired to aid the agency perform many of its tasks. See 29 U.S.C. § 656(c) (1976). However, fundamental requirements of fairness and due process in administrative law compel that these outside consultants to whom the agency delegates its obligation to evaluate the evidence must be unbiased and neutral in their evaluation of the record. Just as the actual decision-maker is to be unbiased, so must those to whom such duty is *184delegated. No court should condone allowing paid consultants to legally change their hats from expert witnesses subject to cross-examination during the hearings, to “agency staff” hired after the close of hearings to evaluate the credibility of their own testimony and others. Their positions as prior witnesses, because of implicit bias, disqualifies them from passing on the probative value of their testimony as a witness visá-vis that of the other witnesses. By having testified as advocates during the hearing on one side of a highly contested issue, their later intra-agency tasks taint the end product.
In addition to the fact that OSHA utilized biased staff, this practice is prejudicial because the consultants were not subject to cross-examination on the secret reports they subsequently developed. According to the agency’s own regulation, “fairness may require an opportunity for cross-examination on crucial issues.” 29 C.F.R. § 1911.15(a)(3) (1979). In addition, the regulations mandate that “[t]he presiding officer shall provide an opportunity for cross-examination on crucial issues.” 29 C.F.R. § 1911.15(b)(2) (1979). Subjecting these consultants to cross-examination initially, and then giving them free reign to evaluate and weigh all contrary testimony, is in effect giving them free and unbridled rebuttal without the benefits that might flow from cross-examination. And OSHA cannot contend that the substance of these secret reports does not relate to “crucial issues” because they go to the very core of the standard.
Various other points in this section of the majority opinion must be addressed. First, the majority condones OSHA’s use of the consultants because the Lead Industries Association (LIA) allegedly failed to prove material prejudice by identifying “hard data or new legal arguments which are contained only in the allegedly improper ex parte communications and on which OSHA demonstrably relied in setting the standard.” Majority Opinion at 1213. The majority is asking of the petitioners the impossible, when all they were allowed to examine were the contracts executed for hiring the consultants, and some ambiguous lists prepared pursuant to Vaughn v. Rosen, 484 F.2d 820 (D.C.Cir.1973), cert. denied, 415 U.S. 977, 94 S.Ct. 1564, 39 L.Ed.2d 873 (1974) in response to a request under the Freedom of Information Act. However, by examining the dearth of evidence on the 50 ug/m3 standard, one easily realizes that the DBA report must have had some relevant information in it for it to provide the agency’s justification for promulgating the lead standard at the 50 ug/m3 level.
Secondly, the majority relies on the Second Circuit’s opinion denying the Lead Industry Association’s Freedom of Information Act request for release of the secret studies, as evidence that these reports were intra-agency reports which need not be disclosed. Lead Industries Ass’n, Inc. v. OSHA, 610 F.2d 70 (2d Cir. 1979). That opinion has no persuasive value for the instant case. That records may not be required to be disclosed under the Freedom of Information Act does not decide whether substantial evidence exists to support a standard adopted in legislative rulemaking proceedings. Also, the legal standard for exemption of intra-agency memoranda bears no relationship to the legal problems existing here, relative to tainted consultants, and lack of proper cross-examination. Those issues were not directly before Judge Friendly in the Second Circuit case.
Finally, the cases cited by the majority fail to support its end conclusion upholding OSHA’s action. For instance, in Doe v. Hampton, 566 F.2d 265 (D.C.Cir.1977) this Court reviewed an appeal from an administrative review of the firing of a civil servant. In the administrative review, albeit labelled an “adjudicatory” review, id. at 269, the agency communicated with a Dr. Eck for his medical opinion on the record evidence about the terminated employee. This Court held that this was not a mere “assembling of the files”, or an “appropriate use of assistants”. Id. at 276. Instead, it was an additional medical evaluation of record evidence upon which the Appellant had already had a full and fair opportunity to comment. Id. at 277. The doctor’s report went to the essence of the validity of the decision. As such, proper procedure *185required that the parties have an opportunity to review the report and to comment upon it. The Court concluded that the delay that would have occurred in allowing the further comments is a cost that would “be more than counterbalanced by the benefits derived from the appearance of a fairer adjudication of the merits.” Id.
The majority cites Doe v. Hampton for the proposition that before ex parte comments will be excluded, a court must find material prejudice. Indeed, Doe v. Hampton concluded that no legal prejudice was exhibited because the doctor’s report was merely cumulative of other evidence. However, the instant case exhibits clear and material prejudice in that the conclusions in the secret reports were not cumulative of the record evidence. They may have included extra-record evidence, and they certainly go to the very essence of the newly promulgated lead standard at the 50 ug/m3 level.
The instant case is also much like United States Lines v. FMC, 584 F.2d 519 (D.C.Cir.1978), also cited by the majority. Assuming as I do that the prior witnesses were not proper and unbiased agency staff when they were hired as consultants to aid the decisionmaker, I must disagree with the majority’s conclusion that United States Lines, Inc., supra, is inapplicable. See Majority Opinion at 1214. It is very applicable. As we stated there:
“The agency’s secrecy as to the ex parte communications is particularly troublesome in this case. For what we do know about the course of the agency’s decision-making suggests that these communications were vital to the agency decision. This necessarily calls into question whether the justifications put forth by the agency in its decision were in fact its motivating force.”
Id. at 541. Recognizing the role of the court in performing a searching and careful review of the record as mandated by the Supreme Court in Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971), we held in United States Lines, supra, that we were precluded from doing so because of the secret communications. Even had the contents of the ex parte communications been revealed on judicial review, United States Lines holds that there would still have been a deprival of adversarial comment on them. 584 F.2d at 542. Finally, we cited Vermont Yankee Nuclear Power Corp. v. National Resources Defense Council, 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978) as supporting this conclusion.
Nor does Vermont Yankee provide a basis for agency procedures or practices which effectively foreclose judicial review where, as here, such review is provided for by statute. Nothing in that decision calls into question the well established principle, found in the Administrative Procedure Act and in the decisions of the Supreme Court, that the court is required to conduct a “searching and careful” inquiry to determine whether agency action is arbitrary or capricious, or, in appropriate cases, supported by substantial evidence. See 5 U.S.C. § 706 (1976); Citizens to Preserve Overton Park, Inc. v. Volpe, supra note 40, 401 U.S. at 415-417, 91 S.Ct. 814. Indeed, the Vermont Yankee decision remanded the case to the Court of Appeals for just such an inquiry.
United States Lines v. FMC, 584 F.2d at 542 n.63 (D.C.Cir.1978).
My conclusion on the issue of the subsequent use of prior witnesses as consultants may be summarized with two points: first, the agency has established a system for hearings and cross-examination that it is bound to follow. When it introduces experts as witnesses, subject to all the procedural rules, and then turns around and incorporates them into its staff to review the same or similar issues, allowing them freely to contradict, and in secret, to oppose evidence given by others, the agency is effectively not complying with its own rules. In addition, the consultants’ secret reports are colored by their prior role as advocates. Second, this practice by OSHA caused material prejudice, because there is virtually nothing in the record regarding the issue of the feasibility of the 50 ug/m3 standard except for these detailed, yet secret, reports. This Court is thereby precluded from performing a careful and searching *186review of the record to determine if the standard is supported by substantial evidence.3
II. THE NOTICE OF RULEMAKING
The agency’s failure to properly notify the public that it was considering such a stringent permissible exposure limit (PEL) as 50 ug/m3 is the root of its troubles in this rulemaking. By instead resting on vague statements in the notice of proposed rule-making, and perhaps the clairvoyance of the participating witnesses, OSHA received almost no evidence on the technological and economic feasibility of this ultimate permissible exposure limit. Because of this fatal flaw, the case should be remanded to the agency for further rulemaking, in which proper notice would be afforded the public to ensure its meaningful participation.
The law on this issue is clear. The notice of proposed rulemaking must adequately inform interested parties of the action to be taken. American Iron & Steel Institute v. OSHA, 577 F.2d 825, 830 (3d Cir. 1978), cert. granted, 448 U.S. 909, 100 S.Ct. 3054, 65 L.Ed.2d 1139 (1980). The Fifth Circuit required that proposed rules, in their initial form, must “give [ ] sufficient notice to the interested parties of the Secretary’s intentions.” Taylor Diving & Salvage Co. v. Dept. of Labor, 599 F.2d 622, 626 (5th Cir. 1979). As noted by the majority, the final standard need not be identical to the proposed rule, International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 632 (D.C.Cir.1973), but it must be a “logical outgrowth” of the provisions found in the standard originally proposed. Taylor Diving & Salvage Co. v. Dept. of Labor, 599 F.2d at 626. Finally, the First Circuit held that the “procedural rules were meant to ensure meaningful public participation in agency proceedings, not to be a straight-jacket for agencies.” BASF Wyandotte v. Costle, 598 F.2d 637, 642 (1st Cir. 1979). As stated by this Court, the proposed rule “should be sufficiently descriptive of the ‘subjects and issues involved’ so that interested parties may offer informed criticism and comments.” Ethyl Corp. v. EPA, 541 F.2d 1, 48 (D.C.Cir.1976) (en banc), cert. denied, 426 U.S. 941, 96 S.Ct. 2663, 49 L.Ed.2d 394.
Here the agency “proposed [a] permissible exposure limit [of] 100 ug/m3 ... “4 It never expressly stated it was considering or might consider the 50 ug/m3 standard that it eventually promulgated. Very little evidence was submitted by the industry petitioners or the agency on any level but the 100 ug/m3, or higher. OSHA concedes that no evidence whatsoever was introduced on the economic feasibility of complying with the 50 ug/m3 level. Because of this, and after noting the thousands of pages introduced by all the parties, supporting both a stringent or lenient rule, and finding no evidence on the 50 ug/m3 level, it is a plain absurdity to conclude that the parties were sufficiently informed to permit their meaningful participation in discussing the possibility that the proposed permissible exposure limit of 100 ug/m3 would be reduced 50%.
A review of the vague statements in the notice of proposed rulemaking also points out that the 50 ug/m3 PEL is not a “logical outgrowth” of the proposed rulemaking at 100 ug/m3. See Majority Opinion at 1216. Whether a 100 ug/m3 PEL would provide an appropriate margin of safety, considering the uncertainty of OSHA’s scientific models and medical evidence, is a very open-ended question. Further, whether subclinical effects should be considered is also misleading since subclinical effects undoubtedly can appear at the 100 ug/m3 level in addition to lower exposure levels. This does not lead to a logical outgrowth that the proposed PEL should be cut in half. Finally, the extra consideration given to *187women of childbearing age most logically leads to the conclusion that special precautions, in the form of respirators or alternative work assignments, might be taken for these susceptible individuals, not that a definite level at 50 ug/m3 would be set.
However, one could accept the most liberal interpretations of these statements in the notice of proposed rulemaking had they indeed produced any significant amount of evidence at the 50 ug/m3 level. But no such evidence evolved. That no such evidence was offered by anyone during the extended hearings, especially considering the intensity of the participants’ adversarial positions, constitutes the best support for concluding that the 50 ug/m3 level was not a “logical outgrowth” of the Notice of the Rulemaking proposal of a “100 ug/m3 ... level.”
Finally, OSHA and the majority rely on a feeble and totally irresponsible rationale for upholding the adequacy of this notice. They assert and conclude that since the LIA and other industry petitioners took the position throughout the proceeding that the proposed 100 ug/m3 level was infeasible, then, a fortiori, they would likewise have claimed the 50 ug/m3 level was infeasible. Thus, why bother hearing their comments and evidence on the stricter level, for it would only be the same “in greater volume or [stated] more vociferously.” BASF Wyandotte v. Costle, 598 F.2d at 644.
This reasoning presumes that the evidence of feasibility for 50 and 100 ug/m3 is interchangeable, and there is no way the decisionmaker could reach a different conclusion on the two. This presumes too much. One level may require an assortment of conventional engineering controls, while the 50 ug/m3 level may demand total rebuilding of the workplace or dislocation of the entire industry. The cost may also increase disproportionately. Moreover, without any evidence of the cost and difficulties of complying with the 50 ug/m3, how would the decisionmaker have the vaguest idea of its feasibility? Taking the majority’s reasoning to its logical, albeit extreme, position one would have to conclude that OSHA could have reduced the PEL to 10 ug/m3. By the agency’s contention, the fact that OSHA may arrive at a lower PEL is allegedly forewarned by the notice of proposed rulemaking, and any evidence that the protestants would have introduced would have been the same; infeasibility would still be argued only more “vociferously.” It is obviously absurd to compare the probative effect of evidence of infeasibility for 100, 50 or 10 ug/m3 and state that the only difference is in the vocal force of the advocacy.
Therefore, since reasonable notice was not given, nor evidence received as to the 50 ug/m3 level, and the majority’s defense of such slipshod practice is wholly illogical, I would remand the case to OSHA for a rulemaking based on proper notice to the interested parties and the public. Thereafter, substantial public evidence would be required to support the agency’s conclusion.
III. THE SHORT REPORT
The majority upholds the acceptance of the Short Report over the objection of the American Iron and Steel Institute (AISI). Majority Opinion at 1221-1222. This Report constitutes OSHA’s principal support for the feasibility of the standard for the groups of companies referred to as the “Other Industries.” I dissent from the ruling accepting this Report for two reasons: First, the agency failed to lay a proper foundation for the Report by not identifying the parties who had helped prepare it. Second, the agency caused the cross-examination of the known authors to be denied.
When a technical report is introduced into evidence, its author must be identified. This is necessary to permit comment and cross-examination by the public and interested parties. Identifying the author is essential in order to test the credibility of its format and substance. Yet, in this case, the majority recognizes the correctness of AISI’s contention that OSHA never informed the parties that David Burton and Associates helped prepare the Short Report, at least as to the section on the feasibility of compliance by the AISI member industries. Therefore the agency failed in a material respect to lay a proper foundation *188for receipt of the Report. The Report must accordingly be stricken from the record.
The greatest prejudice, in failing to identify the author of the report, occurs from the fact that OSHA thereby effectively denied the affected industries their right to effective cross-examination of the authors of the Report. It has already been established in part II that cross-examination is essential under the agency's own rules on crucial issues. The majority opinion in section III E concludes that the interpretation of the cross-examination regulation is a matter left up to the agency, and OSHA has “effectively interpreted” the regulation not to require the cross-examination of the authors of the Report. This simply is not true. As recognized by the majority, OSHA believed it had fulfilled its responsibility by introducing David Burton and his colleagues from DBA, but OSHA never identified them as the author of the report. Any true deferral to agency interpretation would require holding that the true authors must be cross-examined. The regulation, stating that there “shall” be cross-examination on crucial issues, is clear, and feasibility of compliance with the 50 ug/m3 standard by the “other industries” is a crucial issue demanding the allowance of cross-examination of these authors. What kind of cross-examination is possible if the advocates do not even know that Burton participated in compiling the Short Report? Burton even denied knowledge of one of the most crucial aspects contained in this Report-economic feasibility for the other “industries.” Thus, his credibility as a witness for the Short Report was eliminated. An administrative proceeding is not a party-game wherein the contestants must divine the identity of the mystery guest in order to win the door prize.
Therefore, because OSHA by its actions precluded effective cross-examination on the Short Report, I would remand the standard for all of the “Other Industries” that the Report addressed. Since the majority has also remanded to OSHA all the standard based on only evidence contained in the Short Report, my position on this issue has little effect. However, the majority remands for reasons different from mine, and since I greatly disagree with its conclusion in III E, the foregoing comments are essential.
IV. MEDICAL REMOVAL PROTECTION
Industry petitioners challenge OSHA’s authority to promulgate the provision in the standard establishing the Medical Removal Protection (MRP) system. See 29 C.F.R. § 1910.1025(d), (j), and (k). This system at its most extreme limit requires the removal from their jobs of all employees exposed to lead levels exceeding the designated air— lead action level, and whose blood-lead levels exceed prescribed amounts over three consecutive months. Under the authority of Section 1910.1025(k)(2), the employer may assign the removed worker to a low exposure job, or reduce his number of exposed hours on the same job to comply with the time weighted average calculated into the standard. If the employer is incapable of doing this, he must lay off the employee while still maintaining his full wage and seniority benefits for up to eighteen months. When the employee’s blood-lead level has been reduced to a prescribed “safe” level, he must be returned to his original position. The union petitioners maintain that the feature of the system, whereby the worker retains his full salary and seniority for up to 18 months whether he works or not, is the sine qua non of the entire standard, for without this job security, employees would fail to comply out of fear of losing their jobs. The industry petitioners argue that OSHA is not authorized to impose such a system and is also precluded by 29 U.S.C. § 653(b)(4). While agreeing with the agency and unions as to the benefits of the proposed system, I must agree with petitioners that it exceeds the authority conferred by the statute. I therefore dissent from the majority on this issue and would hold that the Medical Removal Protection system, insofar as it requires payment of full wages and benefits while the employee is laid off, should be set aside. If Congress supports such a system it should authorize it specifically and indicate that § 653(b)(4) does not prohibit it.
*189The agency in the lead standard has promulgated the most far reaching removal system of any OSHA standard ever reviewed. Therefore, there is little support in the cited case law.5 It is noted, however, that the Fifth Circuit recently held that OSHA is not authorized to regulate job security. Taylor Diving & Salvage Co. v. Dept. of Labor, 599 F.2d 622, 625 (5th Cir. 1979).
More importantly, the OSH Act clearly indicates that the proposed system is unauthorized and contrary to the law or intent of Congress. The majority relies on various open ended and overly general congressional “authorizations” which assertedly permit this regulation. First, OSHA may assure “safe and healthful working conditions . by developing innovative methods, techniques, and approaches for dealing with occupational safety and health problems.” 29 U.S.C. § 651(b)(5) (emphasis added). Second, the term “occupational safety and health standard” is statutorily defined as one which requires “conditions, or the adoption or use of one or more practices, means, methods, operations, or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment.” 29 U.S.C. § 652(8). Third, the Act describes the manner in which the agency may regulate to warn employees of hazards to which they are exposed. In addition, the standards “shall also prescribe suitable protective equipment and control or technological procedures to be used in connection with such hazards ...” 29 U.S.C. § 655(b)(7). Finally, the Secretary shall prescribe “such rule and regulations as he may deem necessary to carry out [his] responsibilities under this chapter ...” 29 U.S.C. § 657(g)(2). By these statements, OSHA maintains that it can require removal of employees for up to eighteen months at full salary and seniority rights.
OSHA and the majority are reading far more authority into this Act than Congress ever intended it to have. The general impression gleaned from these provisions is that Congress was referring to the regulation of conditions within the workplace. The Act uses such terms as “working conditions”, “places of employment”, and “control procedures ... used in connection with [the hazards to which the employee is exposed]”. The latter authorization is stated in the same phrase with “protective clothing”, indicating that the control techniques are intended for application within the working environment. Congress never even hinted that the Secretary, by this language, could demand the medical removal program, which has nothing to do with providing safe “working conditions”. In fact the MRP imposes no conditions whatsoever on the working place.6
The Supreme Court’s recent discussion of the refusal of Congress to enact a “strike with pay” provision in the OSH Act is very instructive on this issue. Whirlpool Corp. v. Marshall, 445 U.S. 1, 100 S.Ct. 883, 63 L.Ed.2d 154 (1980). The majority’s opinion attempts to distinguish or explain away clear statements by the Whirlpool Court, all the while inadvertently demonstrating how applicable it really is. Majority Opinion at *1901233-1234 n.69. Whirlpool deals with a regulation promulgated under the OSH Act allowing an employee “to choose not to perform his assigned task because of a reasonable apprehension of death or serious injury coupled with a reasonable belief that no less drastic alternative is available.” Id. at 4190; 29 C.F.R. § 1977.12, 38 Fed.Reg. 2681, 2683 (1973), as corrected 38 Fed.Reg. 4577 (1973). The Supreme Court upheld the regulation as being consistent with the OSH Act. However, it did so because the employer was not required to pay the employee who walked off the job.
The pertinent discussion in Whirlpool addresses the Daniels bill which had attempted to introduce the strike with pay provision into the OSH Act. The Court gives a detailed analysis of the congressional intent on the subject and concludes that Congress was not addressing the Whirlpool situation.
Congress rejected a provision that did not concern itself at all with conditions posing real and immediate threats of severe injury. The remedy which the rejected provision furnished employees could have been invoked only after 60 days had passed following HEW’s inspection and notification that improperly high levels of toxic substances were present in the workplace. Had that inspection revealed employment conditions posing a threat of imminent and grave harm, the Secretary of Labor would presumably have requested, long before the expiration of the 60-day period, a court injunction pursuant to other provisions of the Daniels bill. Consequently, in rejecting the Daniels bill’s “strike with pay” provision, Congress was not rejecting a legislative provision dealing with the highly perilous and fast-moving situations covered by the regulation now before us.
It is also important to emphasize that what primarily troubled Congress about the Daniels bill’s “strike with pay” provision was its requirement that employees be paid their regular salary after having properly invoked their right to refuse to work under the section. It is instructive that virtually every time the issue of an employee’s right to absent himself from hazardous work was discussed in the legislative debates, it was in the context of the employee’s right to continue to receive his usual compensation.
When it rejected the “strike with pay” concept, therefore, Congress very clearly meant to reject a law unconditionally imposing upon employers an obligation to continue to pay their employees their regular pay checks when they absented themselves from work for reasons of safety.
Id. 100 S.Ct. at 893 (emphasis added; footnotes omitted). The Supreme Court concluded that the regulation under review did not require this pay, and only required that the employer not discriminate against the employees who walked off the job. It thus did not fall within the legislative history surrounding the Daniels bill.
It is difficult to imagine how the Supreme Court could have spoken more clearly to the instant situation. The majority attempts to distinguish the two situations because under the Daniels bill the employee could initiate his own removal whereas under the lead standard the removal may be imposed on both the employer and the employee against their will. However, the Supreme Court was clearly stressing the congressional disapproval of forcing employers to continue compensating employees removed for safety reasons. This concern would hold true regardless of the outside force that brought it about. In addition, like the situation in the Daniels bill and unlike the Whirlpool scenario, the regulation in the instant case also does not concern “imminent dangers” or risk of harm. Therefore, because of the credence given the legislative history surrounding the Daniels bill by the Supreme Court, it is submitted that the majority in the instant case misapplied the legislative history and misread Whirlpool. Whirlpool alone justifies this dissent.
However, to culminate the review of congressional intent which is adverse to the MRP, 29 U.S.C. § 653(b)(4) must be read. This section of the OSH Act precludes the promulgation of such a far reaching MRP as OSHA seeks to establish for the lead industries. Section 653(b)(4) proscribes *191anything in the OSH Act from “superseding] or in any manner affect[ing] any workmen’s compensation law . . ” Id. (Emphasis added). Since an employee may be removed before he is overtly ill or incapable of performing his job under the lead standard, he receives greater protection under it than under any state workmen’s compensation law in existence. In addition, even though he may be disabled because of lead poisoning, he will be removed under the authority of new regulations with full compensation. Since most workmen’s compensation statutes require a diminution of wages before they become effective, there is no way that lead will ever again be the cause of disabling conditions warranting the receipt of workmen’s compensation. Finally, most workmen’s compensation laws only provide benefits which amount to a portion of the employee’s wages, whereas the lead industries regulation, at its maximum, mandates full seniority and the payment of full wages for eighteen months. For these reasons, the medical removal protection system in the lead standard clearly affects and supersedes workmen’s compensation laws in violation of § 653(b)(4).7
For the foregoing reasons, it is apparent that OSHA exceeded its authority by promulgating the MRP with its attendant financial benefits in excess of the workmen’s compensation laws, and in its present form the MRP program should be stricken from the regulations.
V. THE FEASIBILITY STANDARD AND BURDEN OF PROOF
The majority attempts to assemble the legal mechanism for grappling with a very inartfully crafted provision in the OSH Act, 29 U.S.C. § 655(b)(5) (1976).8 The task evokes my sympathy but I disagree with two aspects of their end conclusion because they fly in the face of the Supreme Court’s recent ruling in the Benzene Case, Industrial Union Department, AFL-CIO v. American Petroleum Institute, 448 U.S. 607, 100 S.Ct. 2844, 65 L.Ed.2d 1010 (1980).
First, in order to construct a working model for the public agencies, and courts, the majority reads into § 655(b)(5) a system of “presumptions” which is not provided by Congress. This gratuitous legislation is characterized as “resolving the circularity” problem. See Majority Opinion at 1271.
The feasibility test for a standard that only expects feasible improvements from employers may appear circular. But reasonable construction of such a standard avoids circularity. The cases have apparently treated these standards as creating a general presumption of feasibility for an industry. A company could not simply refuse to pursue engineering or work practice controls by asserting their infeasibility. Rather, it would have to attempt to install controls to the limits of contemporary technical knowledge and of its own financial resources. Judicial review of feasibility would have some meaning, because the court would have to find substantial evidence to justify this presumption-evidence that the technical knowledge to meet the PEL without relying on respirators was available or would likely be available when deadlines arrived, and that enough firms could afford to so meet the PEL that the market structure of the industry would survive.
*192Majority Opinion at 1269-1270 (emphasis in original). In any later judicial review of temporary variance petitions or enforcement proceedings under 29 U.S.C. § 659 (1976), the employer would bear the burden of proving infeasibility for his company specifically, or the industry generally. Id. at 1270.
This system must be questioned because it does not correlate with the Supreme Court’s discussion in American Petroleum Institute. The Supreme Court likewise devised a system of presumptions for “proving the significance of the risk . . where scientific knowledge is imperfect and the precise quantification of risks is therefore impossible.” 100 S.Ct. at 2870. This imperfect knowledge problem compares to OSHA’s problem of proving technological feasibility where the technology does not presently exist, and the agency is authorized to promulgate standards that are technology forcing.
The following was OSHA’s interpretation of its burden of proving “significant risk” from industrial exposure to benzene prior to American Petroleum Institute:
The Agency’s position is that there is substantial evidence in the record to support its conclusion that there is no absolutely safe level for a carcinogen and that, therefore, the burden is properly on industry to prove, apparently beyond a shadow of a doubt, that there is a safe level for benzene exposure. The Agency argues that, because of the uncertainties in this area, any other approach would render it helpless, forcing it to wait for the leukemia deaths that it believes are likely to occur before taking any regulatory action.
Id. 100 S.Ct. at 2869 (emphasis in original, footnotes omitted). Thus, OSHA believed it could shift the burden of proof to the industries after it has met its initial burden. Likewise, OSHA believes it need only provide substantial evidence that some vague technological ideas or concepts exist now for engineering controls that may appear in the future to aid the industries to comply with the PEL without the use of respirators. From that point on, and for the primary smelters this may last up to ten years, the industries bear the burden to prove infeasibility.
However, the Supreme Court did not accept OSHA’s system concerning proof of significant risk:
As we read the statute, the burden was on the Agency to show, on the basis of substantial evidence, that it is at least more likely than not that long-term exposure to 10 ppm of benzene presents a significant risk of material health impairment. Ordinarily, it is the proponent of a rule or order who has the burden of proof in administrative proceedings. See 5 U.S.C. § 556(d). In some cases involving toxic substances, Congress has shifted the burden of proving that a particular substance is safe onto the party opposing the proposed rule. The fact that Congress did not follow this course in enacting OSHA indicates that it intended the Agency to bear the normal burden of establishing the need for a proposed standard.
Id. 100 S.Ct. at 2869 (emphasis added; footnotes omitted). The most OSHA could arrive at was that the lowered benzene PEL was “likely” to have “appreciable” benefits. This was insufficient.
The Benzene and Lead cases are very comparable. Proof of scientific certainty as to the risks of benzene exposure was difficult to compile. But the Supreme Court would not allow the agency to pass this burdensome standard onto the industries unless it was proven “at least more' likely than not” that long-term exposure to 10 ppm of benzene presents a significant risk of material health impairment. Similarly, it is difficult to show that all the necessary technology exists or will exist for the lead industries to comply with the stringent 50 ug/m3 PEL by engineering or work controls without the use of respirators. This is obvious from the fact that little evidence of working models for innovative controls presently exists, and OSHA was forced to provide for lengthy phase-in periods for the industries to devise the technology. OSHA all the while recognized that it may never be developed. Yet, before such a burden is *193passed to the industries, I believe the Supreme Court teaches that OSHA must prove, by substantial evidence, that it is “more likely than not” that the standard is feasible, and the technology will more likely than not exist to enable the industries to comply. Using the Supreme Court’s reasoning, this is the least that OSHA, as proponent of the rule, should have to prove on judicial review before any presumption of feasibility may be said to exist.
“Likely” and “more likely than not” are dissimilar standards. The latter is a far more stringent burden of proof. Since the majority only applied the “likely” standard in evaluating whether the necessary technology would “likely be available when deadlines arrived”, I dissent. On the issue of feasibility, with respect to industries where the technology to achieve compliance does not presently exist, I would remand the case to the agency for further proof in accordance with the appropriate standard and burden of proof.
A second aspect of the majority opinion’s feasibility discussion that prompts my concern appears on page 1272. This discussion involves the presumption issue as applied to those industries where OSHA concedes that respirators will be necessary in some of the operations in order to comply with the 50 ug/m3 PEL.
Insufficient proof of technological feasibility for a few isolated operations within an industry, or even OSHA’s concession that respirators will be necessary in a few such operations, will not undermine this general presumption in favor of feasibility. .. . In any proceeding to obtain relief from an impractical standard for such operations, however, the insufficient proof or conceded lack of proof will reduce the strength of the presumption a firm will have to overcome in justifying its use of respirators.
Majority Opinion at 1272 (emphasis in original). The idea of a “reduction” in the presumption is absurd. If OSHA concedes that respirators will be essential to meet the PEL, and that engineering controls are infeasible for those specific operations, then OSHA never met- its initial burden of proving by substantial evidence that it was “likely” the technology would exist to comply (if we were to apply, the majority’s standard of proof), much less that it was “more likely than not” that the technology would be available. The latter standard, as I have already explained, is the appropriate one for this court to apply. Therefore, because of OSHA’s own concession and lack of proof, no presumption ever came into existence, and the industry should not have any burden of rebutting a presumption of feasibility for those specific operations. To hold otherwise is to force onto the industry an unjustified and unnecessary burden that statutorily belongs to the agency.
The gist of this section, and the Supreme Court’s opinion in American Petroleum Institute, is that the agency has a responsibility to prove the necessity for, and feasibility of the regulations it promulgates. The OSH Act, although difficult to interpret in part, makes this much clear. If industries are to survive in these troubled times, OSHA must at least be sure of what it is requiring of the industries, before forcing them to invest millions of dollars in projects that may prove to be infeasible and meaningless.
VI. FEASIBILITY OF THE LEAD STANDARD
Beyond the problems of the majority’s construction of the feasibility requirement, many of the agency’s methods in determining feasibility are also questionable. Rather than discuss each separate industry, the general difficulties in methodology will be explained. Because these problems are of serious magnitude, and pervade the entire rulemaking, the whole case should be remanded to the agency.
A. “Best Available Evidence” Requirement
The OSH Act, 29 U.S.C. § 655(b)(5) (1976) requires that the standards promulgated by the agency be based upon the “best available evidence.”9 It is conceded by OSHA *194and the majority that very little record evidence exists for the technological feasibility of the 50 ug/m3 air-lead level, and no evidence exists to support the economic feasibility of that level. OSHA never even introduced its own witnesses to testify on these issues, instead it concentrated on the feasibility of the proposed 100 ug/m3 level. Yet, the majority reason that the final lead standard at 50 ug/m3 is based on substantial evidence, and the best available evidence.
The essence of their reasoning is enunciated in footnote 133 of the Majority Opinion at page 1277.
133. LIA argues that the statutory requirement that OSHA act on the “best available evidence,” 29 U.S.C. § 655(b)(5) (1976), means that OSHA had to direct the rulemaking to the 50 ug/m3 standard. But the “best available evidence” rule, as construed by our cotton dust decision, requires OSHA to act immediately to protect workers as best it can, without waiting for scientific certainty. Industrial Union Dept. AFL-CIO v. American Petroleum Institute, supra note 54, 448 U.S. 656, 100 S.Ct. at 2871 (plurality opinion); AFL-CIO v. Marshall, supra note 54, 617 F.2d at 650. We decline to read this phrase as setting an unprecedented evidentiary burden on the agency-to show not only that its evidence was substantial, but also that its evidence was the best it could possibly have presented.
The inadequacy of the record evidence to sustain OSHA’s statutory burden is apparent from a mere reading of the footnote. The majority permits the agency to utilize as supporting evidence of feasibility the evidence introduced to support the proposed PEL at 100 ug/m3 that is far different from the one actually promulgated at 50 ug/m 3.
This conclusion summarizes one of the most serious problems throughout this rule-making-OSHA actually fails to perceive a difference between the 100 ug/m3 and the 50 ug/m 3 levels. Instead, it rests its decision on the vague and unsupportable concept that the same evidence of economic and technological feasibility that might support the 100 level is also adequate, without more, to support the 50 level. Thus, no additional evidence whatsoever need be gathered for the latter. Yet, its reliance for this conclusion of transferability is not based on one shred of evidence within the record. Therefore, if normal judicial standards were applied the court would hold that it was impossible to support the feasibility of the 50 ug/m3 standard.
Moreover, the conclusion in the footnote is patently irresponsible. OSHA bears the initial burden of proving feasibility based on the best available evidence, and substantial evidence must be introduced to support the finding. To hold as does the majority, that OSHA is relieved of the responsibility of compiling even the most fundamental data to support its regulations is ludicrous. The lead standard need not be promulgated with such speed that its evidentiary support may consist of thin air. By requiring some substantial evidence, the court need not require that exact certainty be evidenced for each provision. The only way that OSHA’s final standard may be upheld is to prove that OSHA made a bona fide attempt to gather the best available evidence. The word “available” connotes a responsibility on the part of the agency to gather the necessary and appropriate data. OSHA cannot cursorily gather what may scantily exist on a subject, promulgate a standard for compliance, and pronounce that it is based on the “best available evidence.” That is in substance the state of the public record in this case. OSHA has not introduced evidence as to the feasibility of a 50 ug/m3 PEL, and its neglect to introduce any evidence in support of a standard at that level cannot serve as a substitute for the necessary substantial evidence.
Finally, approval of OSHA’s failure to fulfill its statutory responsibility is also evidenced by the majority’s attempt to rationalize the difficulty in arriving at the cost to the industry for meeting the 100 ug/m3 level. Majority Opinion at 1277. The majority notes that the technology OSHA evaluated is speculative, and the cost figures are difficult to gather from the industry. Thus, “attempting to extend the highly speculative ‘guesstimates’ on the costs of the 100 ug/m 3 standard to account for the *195incremental costs of the lower standard [at 50 ug/m3] would produce very diminishing analytic returns.” Id. The Act, however, mandates that the agency make the attempt, and the majority is indulging in post hoc rationalizations.
Further, the majority concludes that since the consultants evaluating the economic feasibility of the 100 ug/m3 proposed level indulged in double-counting,10 and failed to calculate into their data the ability of an industry to pass through or absorb costs, “there is nothing inherently invalid about OSHA projecting its determination of the feasibility of a 100 ug/m3 PEL to a well phased-in 50 ug/m3 PEL.” Id. However, one has nothing to do with the other. This conclusion in the majority opinion would seem to indicate that if the same consultants performed an evaluation of costs for a 100 ug/m3 and 50 ug/m3 PEL, both with the same alleged problems of double-counting and failure to pass through the costs, and then these two alleged erroneous calculations were factored out, the cost for the 100 or 50 level would essentially be the same. Absolutely no evidence in the record exists to support this conclusion. Therefore, OSHA should not be permitted to slough off its statutory responsibility to produce the best available evidence by criticizing its own consultants’ techniques.
Because OSHA failed to support its findings of feasibility by any evidence in the public record in support of the 50 ug/m3 PEL, I must dissent from the conclusion of the majority to affirm OSHA’s finding in this respect.
B. Technological Feasibility
Without delving into the specifics for each industry, it is essential to express my doubt as to one more aspect of the majority’s overall methodology regarding technological feasibility. The majority opinion places great reliance upon the findings of Dr. Melvin First of the Harvard School of Public Health. OSHA summarized his principles in the Preamble to the Final Standard at 43 Fed.Reg. 54476-78. In essence, Dr. First contends that all operations in an industry may be “controlled” by separating workers “from contact with lead dust or fume by erecting physical barriers between the worker and the contaminant or by the use of exhaust ventilation that creates air currents to sweep airborne dust and fumes away from the breathing zone of workers and draws them out of the workroom.” Id. at 54477. He boldly alleges that this may be accomplished with every operation that is mechanized and automated. Id.; Exhibit 270.
I do not challenge his basic proposal. However, OSHA and the majority seem broadly to accept this First principle, without an individualized analysis, that almost every operation in almost every industry can be fully contained.
This theory is applied as a rationale in support of technological feasibility throughout OSHA’s Preamble. More detail should have been added to the Preamble explaining, for each affected industry, how this First principle can practically be applied. Only in that way will the agency have fulfilled its responsibility explained in the Cotton Dust case.
To facilitate this review of the record, the agency must pinpoint the factual evidence and the policy considerations upon which it relied. This requires explication of the assumptions underlying predictions or extrapolations, and of the basis for its resolution of conflicts and ambiguities. In enforcing these requirements, the court does not reach out to resolve controversies over technical data. Instead, it seeks to ensure public accountability. Explicit explanation for the basis of the agency’s decision not only facilitates proper judicial review but also provides the opportunity for effective peer review, legislative oversight, and public education. This requirement is in the best interest of everyone, including the decisionmakers themselves.
AFL-CIO v. Marshall, 617 F.2d 636, 651-52 (D.C.Cir.1979) (footnotes omitted).
Only by a more “explicit explanation” of how the First principle applies to each in*196dustry will OSHA have fulfilled its responsibility. The case also should be remanded to the agency on this basis.
C. Economic Feasibility
Little more needs to be said about the lack of evidence concerning the economic feasibility of attaining the 50 ug/m3 PEL. All parties and the majority concede that no evidence was introduced on the subject. Since also no rational evidence was introduced on how OSHA had fulfilled its responsibility by utilizing the evidence introduced for the 100 ug/m3 level, this case should be remanded.
In addition, the conclusion of OSHA that its consultants improperly double-counted in arriving at their ultimate cost estimates is subject to substantial question. In particular, OSHA concluded that its own expert, DBA, had “included] in the cost of the new lead standard the costs that intransigent firms have yet to expend in meeting the old and more generous PEL of 200 ug/m3”. Majority Opinion at 1276; 54494/3-54495/1. DBA explained that segregating these costs would be impossible.11 OSHA, however, disputed the inclusion of the costs necessary to reach the 200 ug/m 3 level. Therefore, the agency concluded that DBA’s estimates were “considerably overestimated.”
OSHA’s conclusion is far too simplistic. For an industry to reach the 200 ug/m3 level, as previously required, it may only have needed respirators, or simple engineering controls at a minimal cost. However, for that industry, to reach 100 ug/m3 or 50 ug/m3 from either the 200 ug/m 3 level, or a higher exposure level, it may be necessary to completely rebuild its plant at an outrageous cost. OSHA fails to give consideration to such possibilities. Furthermore, what the agency should have been concerned with is the ultimate cost and burden or effect on the industry, despite the initial starting point for calculating the cost. It is agreed by all that an OSHA standard is infeasible if it causes massive dislocation to the industry, or imperils its existence. AFL-CIO v. Brennan, 530 F.2d 109, 123 (3d Cir. 1971); American Iron & Steel Institute v. OSHA, 577 F.2d 825, 836 (3d Cir. 1978). Should this be any less true just because substantial segments of the industry might not have complied with the 200 ug/m3 standard? Such ruling is open to serious doubt.
But more importantly, OSHA failed to explain why these costs should not be considered, or how they could be separately extracted, and what the end figure would be once they were extracted. It is insufficient to merely criticize one’s own experts, without setting forth proper conclusions and reasons why.12 As noted in the Cotton Dust case:
*197If the constraint of economic feasibility is to have any effect on the agency’s rulemaking, it demands more serious consideration than was given here. The agency is allowed to rely on the best available evidence, but here it simply gives general criticisms of the cottonseed industry’s cost estimates. It failed to offer an alternative estimate of the standard’s impact on this industry. As a result, the agency’s position is too unclear to permit us to complete our reviewing function.
AFL-CIO v. Marshall, 617 F.2d 636, 672-73 (D.C.Cir.1979) (footnotes omitted). This court remanded the economic feasibility issue for the cottonseed industry to OSHA for clarification and reconsideration because of the defect in OSHA’s procedures indicated by the above criticism. Id. at 673. The same result should be ordered here for every industry in which OSHA contends there was “double-counting.”13
D. Cost-Benefit Analysis
Throughout this rulemaking OSHA has claimed that it need not perform a cost-benefit analysis. The Majority Opinion, relying on the Cotton Dust case, AFL-CIO v. Marshall, supra, concurs and fails to even evaluate the agency’s conclusion. However, since the issue is still very much undecided by the Supreme Court after American Petroleum Institute, supra it deserves some comment here.
The complex legal analyses for and against the statutory necessity of a cost-benefit analysis have been discussed at length by various courts. Compare AFL-CIO v. Marshall, supra with American Petroleum Institute v. OSHA, 581 F.2d 493 (5th Cir. 1978). The Supreme Court recently affirmed American Petroleum Institute, but explicitly determined not to decide the necessity of performing a cost-benefit analysis.14 Yet, from the plurality opinion of Justice Stevens, it is difficult to conclude that a cost-benefit analysis is not necessary.15 He maintained that “it seems manifest that Congress intended, at a bare minimum, that the Secretary find a significant risk of harm and therefore a probability of significant benefits before establishing a new standard.” 100 S.Ct. at 2865. When this is coupled with the necessity of proving economic feasibility under 29 U.S.C. § 655(b)(5), a definite form of cost-benefit analysis is mandated.
To the extent that this bare minimum has not been performed in the instant case, it should be remanded to the agency for reconsideration.
The majority opinion does not make any meaningful reply whatsoever to the above points but asserts generally that in view of *198Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978) the court should be “wary of imposing on the agency any procedural or evidentiary constraints beyond those expressly established by Congress.” Majority Opinion at 1203 n. 6. In my judgment the majority places on Vermont Yankee weight it was never meant to hold, and ignores several constraints expressly established by Congress. Under the Occupational Safety and Health Act whenever “a public hearing on . objections” is requested OSHA is required to specify “the occupational safety or health standard to which objection has been filed and a hearing requested and [to specify] a time and place for such hearing.” 29 U.S.C. § 655(b)(3). Then, if the standards deal with toxic materials or harmful physical agents, the Secretary “shall set the standard 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 dealt with by such standard for the period of his working life.” In addition, among other bases he “shall [consider] . . the latest available scientific data in the field, the feasibility of the standards, and experience gained under this and other health and safety laws.” 29 U.S.C. § 655(b)(5). And, if the “rule promulgated by the Secretary differs substantially from an existing national consensus standard, the Secretary shall, at the same time, publish in the Federal Register a statement of the reasons why the rule as adopted will better effectuate the purposes of this act than the national consensus standard.” § 655(b)(8). Further, “whenever the Secretary promulgates any standard, makes any rule under this [act] he shall include a statement of the reasons for such action, which shall be published in the Federal Register.” § 655(e). And other statutory requirements are set forth above. Finally, on “judicial review” of such standard “the determinations of the Secretary shall be conclusive [only] if supported by substantial evidence in the record considered as a whole.” § 655(f) (emphasis added).
These are exacting evidentiary requirements, as the Supreme Court indicated in the Benzene Case, Industrial Union Dep’t, AFL-CIO v. American Petroleum Institute, 448 U.S. 607, 100 S.Ct. 2844, 65 L.Ed.2d 1010 (1980), and wipe out the wide discretion which is permitted in informal rule-making. By failing to give effect to these requirements, the majority have improperly enlarged the discretion which OSHA possesses in this rulemaking proceeding. Actually the discretion which the majority opinion applies to the Rules is more consistent with that permissible under informal rule making than the requirement of the statute for “substantial evidence,” supra. From the foregoing it is apparent that this dissent is consistent with Vermont Yankee: it does not require procedures beyond those required by the applicable statute. 435 U.S. at 525, 98 S.Ct. at 1202. Every complaint here has a sound basis in the OSH Act. We cannot term as “evidence” that which has not been made a part of the public record (I. Improper Use of Consultants); the notice that OSHA contemplated a 100 ug/m3 PEL was statutorily insufficient since the rule promulgated at 50 ug/m 3 was not a natural outgrowth of the notice of the resultant hearing (II. The Notice of Rulemaking); the Short Report cannot be considered as statutory “evidence” because since its author was not identified no proper foundation was laid for its consideration as part of the statutory “record” (III. The Short Report); the financial benefits of the medical removal protection system clearly violate the statutory prohibition against “supersed[ing] . . . any workmen’s compensation law” (IV. Medical Removal Protection); and the failure to demonstrate technological and economic “feasibility” in accordance with the statutory standard, and the shifting of the burden of proof on that issue in violation of the Benzene Case, are clear violations of other specific statutory requirements of the OSH Act (V. The Feasibility Standard and Burden of Proof, VI. Feasibility of the Lead Standard).
*199Holding the agency , to the requirement of “substantial evidence in the record considered as a whole” is not an “improper[] intrusion] into the agency’s decisionmaking process.” Id., 98 S.Ct. at 1202. In fact the remand which the majority orders, and in which I concur, corroborates the findings of numerous deficiencies in the agency’s rule-making. In addition the specific deficiencies pointed out above are corroborated by the failure of the majority to discuss or answer them. Cf. Majority Opinion at 1203 n. 6.
CONCLUSION
For the foregoing reasons I respectfully dissent and would remand the case for the agency to finish the task assigned to it by the statute.
. Without a review of the actual reports submitted by the consultants, which are not available, it cannot be determined whether or not they contain new, extra-record evidence. The majority reads vague statements from the contracts which designate the evaluating responsibilities of DBA and CPA, and concludes that only record evidence was to have been reviewed to formulate their analyses. However, far too little weight is given to the statement in the CPA contract to the effect that the consultant was permitted “to conduct additional research and prepare material supplementary to the above testimony . . ALD 11; Majority Opinion at 1219. This concern only magnifies the insuperable handicap with which the Court is afflicted by being forced to rule on the propriety of these reports, without ever seeing them. This is a fatal defect in the record.
. In this respect the Notice of Rulemaking on the exposure standard stated the issue to be:
1. Whether the proposed permissible exposure limit to lead should be 100 ug/m3: and whether this level incorporates an appropriate margin of safety;
40 Fed.Reg. 45934/1-2 (1975).
. Note 2, supra.
. Even the majority recognizes that the medical removal system in the Cotton Dust case is very “modest” compared to the one under review. Majority Opinion at 1229 n.63; AFL-CIO v. Marshall, 617 F.2d 636 (D.C.Cir.1979).
. Although a bare reading of the statute is sufficient to arrive at this conclusion, the legislative history of the OSH Act discussed by the majority also evidences the lack of congressional intent to authorize this regulation. Congress had discussed a form of “strike with pay” in the Daniels Amendment, and rejected it. H.R.Rep. No. 91-1291, 91st Cong., 2d Sess. 30 (1970). This Amendment was distinguishable from the MRP, but in many respects its effect on the employer, i. e. having to pay for laid off employees, is comparable. And Congress refused to authorize this.
In addition, weight must be given to the fact that only one year prior to the enactment of the OSH Act, Congress enacted the Federal Coal Mine Health and Safety Act of 1969, 30 U.S.C. § 801 et seq., which did expressly authorize a removal program. 30 U.S.C. § 843(b)(3). With Congress’ full knowledge of this, and rejection of the strike with pay provision, it is difficult to conclude that congressional silence on the benefits issue connotes implicit authorization of the financial and seniority benefits provided in the MRP system.
. The majority and OSHA maintain that all OSHA safety standards affect workmen’s compensation laws in some respects; by making the work place safer, fewer employees will require workmen’s compensation. This argument is a reductio ad absurdum which is clearly unwarranted. There is a great difference between ensuring a safe work place, which is a goal written into the OSH Act, see 29 U.S.C. § 651(b)(5), and providing full financial benefits to removed employees. The latter is in direct contradiction to 29 U.S.C. § 653(b)(4), and was specifically forbidden by Congress in the relevant statute upon which OSHA relies.
. This section provides in part:
(5) The Secretary, in promulgating standards dealing with toxic materials or harmful physical agents under this subsection, shall set the standard 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 dealt with by such standard for the period of his working life.
29 U.S.C. § 655(b)(5) (1976).
. See note 8, supra.
. See part VI C, infra.
. DBA explained that:
Ideally, the costs of complying with the new requirements of the standard should be isolated from the requirements already promulgated, but not presently being complied with. For the industries studied, all of the cost estimates associated with compliance, except engineering costs, are estimates of the incremental costs of compliance. Unfortunately, it is not possible to accurately establish the incremental costs associated with engineering controls. The main reasons for this shortcoming are as follows:
(1) The State-of-the-Art of the effectiveness of engineering controls is not developed to the degree necessary to make such judgments. It is thought that the same basic approaches may be used to control exposures to either 200 or 100 ug/m3.
(2) Engineering controls are often built to standard specifications which are not designed to provide a specific control efficiency, but rather the optimum control possible for the particular system.
(3) Exposure levels characterizations of the workplace are not sufficiently complete to provide data necessary (a) to establish complete contributory emissions, or (b) to establish accurately the levels of reduction necessary to meet a 200 ug/m3 level and a 100 ug/m3 level.
(4) The conditions of exposure, and hence the needs for control, vary widely from operation to operation, plant to plant, day to day, hour to hour, and even from employee to employee at the same operation. (Ex. 26, p. 4-10-4-11.)
43 Fed.Reg. 54494/3.
. For instance, OSHA “explained” how it reduced the double-counting for figures for primary smelters. All it did was list the kinds of double-counting it perceived, and stated:
Given the above, reducing the total capital cost estimate for the 100 ug/m3 interim level *197by one-third for all forms of double-counting is reasonable.
43 Fed.Reg. 544981/1. Why is this reasonable? Where did the figure “one-third” come from? These questions remain unanswered and hence the agency’s computation is not supported and cannot be approved.
.OSHA’s blind eye toward cost impacts may be evidenced in its cavalier attempt to compare the primary smelters’ ability to pass through the costs of 1$ per pound of lead produced. See Majority Opinion at 1282. The agency, by characterizing the cost at such a small fraction, treats this almost as a triviality. In fact when 1$ per pound is multiplied by the total number of pounds of lead produced annually, it is a very substantial sum. For example, OSHA concluded that the amount of lead produced by the primary smelters in 1975 was 723,879 short tons, 42 Fed.Reg. 54498/2, 54509/1. This amounts to 1,447,758,000 pounds of lead. At one cent per pound, the total cost to the primary smelters is $14,477,580.00. Fourteen million dollars per year is not a trivial amount to pass through to the customers. Reducing costs so that they deceptively appear trivial is an approach that a government should be loathe to take with its citizens.
. The Supreme Court stated that OSHA must determine as a threshold matter that the “toxic substance in question poses a significant health risk in the workplace and that a new, lower standard is therefore ‘reasonably necessary or appropriate to provide safe or- healthful employment and places of employment.’” 100 S.Ct. at 2850. Because OSHA had failed to do that in American Petroleum Institute, the Court found it was “unnecessary” to determine whether the Fifth Circuit was correct in reading into the statute a cost-benefit analysis requirement.
. Mr. Justice Powell concurred separately with the plurality opinion, and stated that a cost-benefit analysis is essential. See 100 S.Ct. at 2875 (Powell, J., concurring).