dissenting:
The court is faced with something of a bureaucratic mess. In 1977 the Consumer Product Safety Commission, responding to a 1973 petition from an ad hoc group of industry, labor and consumer organizations, promulgated a set of safety standards for the architectural use of glass and other glazing materials. The CPSC regulations required architectural glass to meet certain tensile and thermal performance minima in order to protect consumers from injury resulting from impact with glass architectural elements. Because I believe that the Commission’s denial of O’Keeffe’s petition to amend the safety standards (1) to include ceramic glazing materials within the regulatory definition of “architectural glazing materials,” and (2) to remove an unintended permanent exemption for wired glass used in fire doors, was arbitrary and capricious, and because I believe that the Commission’s denial of O’Keeffe’s petition leaves unregulated several products which pose an unreasonable risk of serious injury to consumers, I respectfully dissent.
*946I. The exemption for fire-door wired glass
Included in the materials covered by the 1977 safety standards was wired glass used in fire doors. At the time, none of the available wired glass products could meet the Commission’s safety standards, but nearly every building code required wired glass panels in fire doors. The Commission, faced with a techno-legal collision between its safety standard, the then-current state of glazing technology, and the requirements of federal, state and local building codes, temporarily exempted wired glass used in fire doors from the standard. 42 Fed.Reg. 1428,1430 (1977).
A trade group representing wired-glass manufacturers petitioned for judicial review of the CPSC standards. The U.S. Court of Appeals for the District of Columbia Circuit found fault with the Commission’s inclusion of fire-door wired glass within the safety standard, finding in the administrative record (1) a “substantial showing” that fire-door wired glass exhibited functional characteristics significantly different from the other materials covered by the standard, and (2) the absence of a “process of reasoned decision-making” through which the Commission decided to include fire-door wired glass within the safety standard. ASG Indus. v. Consumer Prod. Safety Comm’n, 593 F.2d 1323, 1333 (D.C.Cir.1979).1
In addition to its conclusion that the Commission had not satisfactorily demonstrated sufficient functional similarities between fire-door wired glass and other glazing materials to warrant inclusion of fire-door wired glass within the challenged standard, the court also invalidated the Commission’s temporary exemption of fire-door wired glass. The court rejected the temporary exemption— which the Commission had devised in order to permit the glass industry time to develop a fire-door wired glass which would meet the safety standards — because it found the administrative record “utterly devoid” of any factual basis for the Commission’s projection that the limited three-year exemption period was “meaningful and reasonable.” 593 F.2d at 1334.
The Court of Appeals remanded the case to the Commission for reconsideration of both (a) the question whether the functional characteristics of fire-door wired glass merited its inclusion in the safety standard and (b) the factual support for the Commission’s decision to defer application of the standard to fire-door wired glass pending technological developments in the near future which would permit the application of the safety standard to fire-door wired glass. Id. at 1333, 1334-35.
Following the decision in ASG Industries, the Commission abandoned its efforts to promulgate safety standards for fire-door wired glass. In a 1984 amendment to the regulations, the commission deleted the durational limitation on the wired-glass exemption, in effect permanently relieving the manufacturers of fire-door wired glass of any obligation to comply with the safety standards. That amendment allowed the Commission to sidestep its statutory obligation under Section 9(h) of the Consumer Product Safety Act (codified at 15 U.S.C. § 2058(h)) to amend a consumer product safety standard “by rule” pursuant to the notice and comment and the factual findings provisions of that Act. The Commission sought to avoid the procedural requirements for promulgating and amending its rules by stating flatly that the removal of any expiration date on the temporary exemption of fire-door wired glass from the safety standard did not constitute a “material change” in the safety standard. 49 Fed.Reg. 7106, 7107 (1984).
The Commission argued that Section 553 of the Administrative Procedure Act, which requires an agency to afford interested parties notice and the opportunity to comment on a proposed rule did not apply to the deletion of the expiration date because the D.C. Circuit’s ruling in ASG Industries had already invalidated the temporary exemption. The opinion in ASG, however, had expressly qualified its ruling as a “suspension” of the *947safety standard’s application to fire-door wired glass, and had remanded the case to the Commission “for further consideration and explication.” Thus, the ASG ruling gave the Commission several options: On remand, it could conduct further factual investigation to support the temporary exemption, or they could delete the exemption. The Commission instead saw fit to abdicate its rulemak-ing authority, and attempted to evade the procedural requirements of its own organic statute.
The Commission’s contention that the amendment deleting the durational limit on the fire-door wired glass exemption effected no “material change” in the safety standard was both cynical and meritless: When an agency action has “palpable effects” upon the regulated industry and the public in general, that action must first be subjected to examination and comment by the affected parties. National Helium Corp. v. Federal Energy Admin., 569 F.2d 1137, 1146 (Temp.Emer.Ct.App.1977). The effect of the Commission’s deletion of the durational limit on the exemption of fire-door wired glass from the safety standard was obvious: What was intended as a brief exemption of a dangerous product from the safety standards became permanent. Industry and consumers alike would experience “palpable effects.” Wired-glass manufacturers gained a permanent exemption of their product from federal safety standards, and consumers lost the hope that at least at some point in the near future that product would be governed by the Commission’s safety requirements. How it could be argued that such a change was somehow not “material” is beyond my ability to fathom. See, e.g., Duggan v. Bowen, 691 F.Supp. 1487, 1514 (D.D.C.1988) (finding HHS letter ruling limiting certain types of Medicare coverage to have effected “material changes,” thereby requiring agency to comply with APA notice-and-eomment provisions for “rules”).
Having thus won its permanent exemption from the safety standard, the wired glass industry, predictably, chose not to litigate the matter further. In 1992 Petitioner O’Keeffe’s, a manufacturer of glazing materials, petitioned the CPSC to remove the inadvertently — and irresponsibly — permanent exemption of fire-door wired glass from the 1977 safety standards. O’Keeffe’s sought to persuade the Commission that in the intervening decade and a half the glazing industry had developed materials capable of (a) complying with federal, state and local fire codes and (b) meeting the impact-resistance and breakage-pattern requirements of the Commission’s 1977 safety standards. In other words, there were now available transparent building materials which entirely obviated the fire-door wired glass exemption.
The Commission rejected O’Keeffe’s petition for removal of the wired glass exemption. Its reasoning, for lack of a better word, was that there was insufficient evidence in the 1992 record that the installation of wired glass in fire doors posed an unreasonable risk of serious injury. As O’Keeffe’s pointed out, however, such an explanation is unworthy of credit. From the evidence assembled in the 1992 proceedings the Commission estimated that approximately 35 people are injured each year in collisions with the wired glass panels in fire doors. The Commission found further that these injuries “are not superficial,” and that treatment of each wired-glass injury costs around $3,400.
If the Commission properly concluded in 1977 that fire-door wired glass presented an unreasonable risk of injury, and exempted it from the standard only because there then existed no fire-rated non-wired alternative, what possible difference could the number of injuries in the 1992 record make?
The prohibition on the use of wired glass which fails to meet the safety standards except in fire doors may well have reduced the number of injuries resulting from all uses of wired glass. That fact says nothing about the continued risk of injury from fire-door wired glass. The Commission’s argument that the recording of fewer wired-glass injuries today renders unnecessary application of the safety standard to fire-door wired glass is “so implausible that it c[an] not be ascribed to a difference in view or the product of agency expertise.” The Commission’s denial of O’Keeffe’s’ rulemaking petition is therefore subject to judicial reversal as “arbitrary and capricious” within the meaning of the *948Administrative Procedure Act, 5 U.S.C. § 706(2)(A). Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866-67, 77 L.Ed.2d 443 (1983).
The CPSC concluded that the great disparity between the total cost of all fire-door wired glass injuries and the cost of including fire-door wired glass within the 1977 safety standards compelled it to reject O’Keeffe’s’ petition for rulemaking. The CPSC reckoned the latter cost at $36 million, a figure arrived at by multiplying the cost per square foot of wired glass ($36) by the number of square feet of wired glass installed annually in the United States (1 million square feet). The Commission calculated the total cost of all the estimated wired glass injuries by multiplying the estimated number of annual injuries by the average cost of emergency-room treatment for those injuries. 3 E.R. P00625.
The Commission did not include in its injury-cost calculus any expenses other than emergency-room treatment costs. The record does not mention the cost of replacing broken wired-glass panels, the amount of wages lost due to wired-glass injuries, or a monetary reckoning of the human costs in pain and suffering resulting from those injuries. To this extent, therefore, the Commission “entirely failed to consider an important aspect of the problem.” Such a failure renders their decision to deny the petition “arbitrary and capricious.” Motor Vehicle Mfrs. Ass’n, ibid.
It is simply not possible to ascribe the Commission’s refusal to remove the permanent exemption for fire-door wired glass to any reasoned decisionmaking process. That exemption was neither crafted nor desired by the agency. It was the accidental result of the confluence of a judicial decision, which certainly did not contemplate the results here obtained, and pure bureaucratic sloth. The Commission’s failure in 1984 to follow the recommendation of the Court of Appeals in ASG Industries created the problem. The Commission’s reluctance to correct its error appears to derive more from sheer indolence than from rational analysis of the evidence and competing interests.
The Commission’s own brief recites the two traditional grounds for judicial requirement that an agency reconsider its refusal to institute rulemaking. One is where “a significant factual predicate of a prior decision ... has been removed.” WWHT, Inc. v. FCC, 656 F.2d 807, 819 (D.C.Cir.1981). The fire-door wired glass exemption was predicated on the unavailability in 1977 of glazing materials which could meet both the agency breakage pattern and impact-resistance standards and the federal, state and local building codes which required a transparent panel in doors designed to impede the passage of fire. That factual predicate no longer obtains: Such materials have become available.
II. The definition of “architectural glazing materials”
In addition to its petition for removal of the permanent fire-door wired glass exemption, O’Keeffe’s petitioned the CPSC for an amendment expanding the scope of the 1977 safety standards to include glazing materials other than traditional glass. The definitions section of the safety standards limited their coverage to “glass, annealed glass, organic coated glass, tempered glass, organic coated glass, laminated glass, wired glass; or combinations thereof’ when used in hinged or sliding doors, storm doors, and bathtub and shower enclosures. 16 C.F.R. §§ 1201.1(a), 1201.2(a)(ll)(1995). The word “glass” within the meaning the safety standards means “a hard, brittle, amorphous substance produced by fusion, usually consisting of mutually dissolved silica and silicates that also contains soda and lime.” 16 C.F.R. 1201.2(a)(9).
When the definitions were composed in 1977, they described “the known universe of architectural glazing materials.” Pet. Brief at 8. Since then, transparent building materials resistant to heat have been developed from materials other than amorphous silicates. Modern glazing materials may be made from the traditional silicates, or from ceramic ingredients. They look the same, but are not the same: Ceramic glazing materials, though heat-resistant, are much less resistant to breakage. Ceramic “glass” breaks like annealed glass, and when broken shatters in dangerous patterns.
*949In 1989 the distributor of a fire-rated ceramic glazing product called “Firelite” received from the CPSC an informal ruling that Firelite, being made of crystalline, rather than amorphous ingredients, did not need to meet the 1977 safety standards. E.R. A, at 17-18. That letter ruling was, as O’Keeffe’s points out, while technically correct, inconsistent with the original stated purpose of the safety standards. Throughout the regulations, the materials covered are described generically as “architectural glazing materials” precisely because the Commission “intended the standard to cover glass and non-glass materials”. 42 Fed.Reg. 1428, 1428 (1977) (emphasis added). A 1990 glazing industry newsletter complained that the Firelite letter ruling ignored the fact that, while chemically distinct from glass, Firelite served the same purpose as glass, and posed risks of injury similar to those associated with glass, but was now exempt from the federal safety standards. E.R. A, at 28.
There are now on the market a number of glass products which meet the fire-resistance requirements of federal, state and local building codes. Because they are amorphous, they are subject to the 1977 impact and breakage pattern standards. By contrast, the crystalline glazing products now being marketed are excluded from those standards, though they are used like glass and present the same risks of injury as glass.
It was this situation which prompted O’Keeffe’s’ 1992 petition to the CPSC to amend the definitions in the 1977 safety standards. O’Keeffe’s manufactures fire-rated architectural glass under the brand names “Contraflam,” “Pyroswiss,” and “Superlite I.” These products must meet federal safety standards. O’Keeffe’s’ ceramics and wired-glass competitors are, at present, exempt from those same standards.
The Commission refused to amend the definition of architectural glazing materials to include crystalline, or ceramic materials. It claimed it had no basis on which to conclude that ceramic glazing posed an unreasonable risk of injury. O’Keeffe’s has argued that the very facts that (a) ceramic glazings are used in the same way as annealed glass and (b) that at present those ceramics fail to meet the Commission’s own safety standards should be enough to conclude that ceramics present the same risks of injury that noncomplying annealed glass presents. O’Keeffe’s logic is unassailable. O’Keeffe’s presents a simple syllogism:
1. Glass which does not meet the safety standards poses an unreasonable risk of injury.
2. Ceramics look like glass, act like glass, and are installed in the same places as glass.
3. Ceramics do not meet the safety standards.
4. Therefore, ceramics pose an unreasonable risk of injury.
The majority insists that “there is no evidence in the record to support [the] assumptions” that the unregulated ceramic glazings look, act, and are used in the same way as the regulated silicate glazings. The majority ignores several important items of evidence which were in the record at the time the Commission denied O’Keeffe’s petition.
First, there was evidence in the record that ceramic respond to impacts in the same manner as glass: Ceramic glazing materials failed impact tests under the same load as annealed glass. Both materials failed under a load of only 40 foot-pounds. E.R. A, at P00031.2 Second, the Commission itself admitted that glass and transparent ceramics appear identical to the naked eye, see E.R. G at P00620, and indeed conceded that had ceramic glazings existed at the time of the *950drafting of the original definition, they would have been included. Id. Third, there was evidence in the record that one of O’Keeffe’s competitors markets a ceramic glazing intended precisely for use in fire doors. E.R. A at P00020. The majority refuses to assume that non-complying ceramics which break like non-complying glass will cause the same injuries when shattered. It requires no great leap of logic to conclude that ceramic glazing panels in fire doors pose precisely the same risks of injury as annealed glass in fire doors.
The Commission was not satisfied with the undisputed evidence that if ceramic glazing materials were installed like glass, they would shatter and injure like glass. The Commission stubbornly proclaimed that “information is not now available to establish that ... noncomplying transparent ceramics in any of the products subject to the standard present an unreasonable risk of injury.” CPSC Decision, E.R. K, at P00767.
But to the extent the Commission required a litany of documented injuries from human collisions with ceramic glazing materials, the Commission misapprehended applicable law. There is no requirement in the CPSA that before promulgating a safety rule the Commission “develop a precise body count of actual injuries” caused or likely to be caused by the product under consideration. Forester v. Consumer Prod. Safety Comm’n, 559 F.2d 774, 788 (D.C.Cir.1977) (internal quotation marks omitted). As Congress recognized, “[wjhen your intelligence tells you that something will create an injury and that it seems conceptually clear that an injury will occur, it is primitive to wait until a number of people have lost their lives, or sacrificed their limbs before we attempt to prevent those accidents.”3
To this extent, then, the Commission offered an explanation for its decision to deny O’Keeffe’s petition that runs counter to the evidence before it. Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Ins. Co., 463 U.S. 29, 42, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983); Hawaii Helicopter Operators Ass’n v. Federal Aviation Admin., 51 F.3d 212, 214, 215 (9th Cir.1995). There was evidence that transparent ceramics were being used in fire doors, and there was evidence that those ceramics behaved exactly like annealed glass.
Conclusion
The Commission’s factual and political determinations appear to lack adequate basis in the record. The facts found by the Commission and the legislative considerations which it took into account could not by themselves lead a reasonable person to make the judgment the Commission has made. Natural Resources Defense Council v. S.E.C., 606 F.2d 1031, 1053 (D.C.Cir.1979).
An agency may, in rare circumstances, be forced by a reviewing court to institute rule-making if a significant factual predicate of a prior decision on the subject has been removed. American Horse Protection Ass’n v. Lyng, 812 F.2d 1, 5 (D.C.Cir.1987) (citing Geller v. FCC, 610 F.2d 973 (D.C.Cir.1979)). Given the Commission’s behavior following the decision in ASG Industries, it does not appear that it can be trusted to follow judicial instructions to reexamine the factual support for its decisions. The court should take the unusual measure of requiring the Commission to engage in rulemaking to consider removing the fire-door wired glass exemption from the safety standards and to consider redefining “architectural glazing materials” to include ceramic glazing products.
For the foregoing reasons, I dissent.
. The D.C. Circuit did, however, find that the Commission had met the statutory requirements for inclusion within the scope of the challenged safety standard of wired glass in uses other than in fire doors. ASG Industries, 593 F.2d at 1336-37. For that reason the court did not require the Commission to make any additional findings with respect to non-fire-door uses of wired glass.
. Petitioner engaged engineers to conduct impact and breakage pattern tests on wire glass, annealed glass, ceramic glazing materials, and Petitioner's treated glass products, using the Commission’s own test methods. E.R. A, at P00031. The Commission tests architectural glazing materials under the test methods prescribed by the American National Standards Institute ("ANSI”). 16 C.F.R. § 1201.4(a). The material is set in a frame, and a leather punching bag filled with lead shot is suspended from a string attached to a stand. The punching bag, or "impactor,” is aimed at the center of the subject material and swung towards the glass. 16 C.F.R. § 1201.4. Schematic diagrams of the test apparatus appear at 16 C.F.R. Pt. 1201, Subpt. A, Figs. 1 to 5; see also 41 Fed.Reg. 6183-84 (1976).
. S.Rep. No. 91-237, 91st Cong., 1st Sess. 2-3 (1969) (quoting testimony of Arnold Elkin, Chairman of the National Commission on Product Safety, in support of the Consumer Product Safety Commission’s authority under the Federal Hazardous Safety Act to issue regulations where a product presents an unreasonable risk of personal injury or illness — a standard nearly identical to the standard under which the same Commission may regulate products under the CPSA) (Quoted in Forester, 559 F.2d at 788-89).