Southland Mower Company v. Consumer Product Safety Commission, John O. Hayward v. Consumer Product Safety Commission

619 F.2d 499, 1980 U.S. App. LEXIS 16476

• Court: Court of Appeals for the Fifth Circuit
• Decided: June 19, 1980
• Docket: 79-1469, 79-3193
• Status: Published

Opinion

GEE, Circuit Judge:

Approximately 77,000 people are injured each year in the United States by contacting the blades of walk-behind power mowers. 1 Of these injuries, an estimated 9,900 involve the amputation of at least one finger or toe, 11,400 involve fractures, 2,400 involve avulsions (the tearing of flesh or a body part), 2,300 involve contusions, and 51,400 involve lacerations. The annual economic cost inflicted by the 77,000 yearly blade-contact injuries has been estimated to be about $253 million. This figure does not include monetary compensation for pain and suffering or for the lost use of amputated fingers and toes. 2

To reduce these blade-contact injuries, the Consumer Product Safety Commission (“CPSC” or “the Commission”) promulgated 3 a Safety Standard for Walk-Behind Power Lawn Mowers, 16 C.F.R. Part 1205 (1979), 44 Fed. Reg. 9990-10031 (Feb. 15, 1979), pursuant to section 7 of the Consumer Product Safety Act (“CPSA” or “the Act”), 15 U.S.C. § 2056 (1976). 4 In the present case we consider petitions by the Outdoor Power Equipment Institute (“OPEI”), manufacturers of power lawn mowers, 5 and an interested consumer to review 6 the Safety Standard for Walk-Behind Power Lawn Mowers.

The standard consists of three principal provisions: a requirement that rotary walk-behind power mowers pass a foot-probe test, 16 C.F.R. § 1205.4, 44 Fed. Reg. 10025-26, a requirement that rotary machines have a blade-control system that will stop the mower blade within three seconds after the operator’s hands leave their normal operating position, 16 C.F.R. § 1205.5(a), 44 Fed. Reg. 10029, and a requirement, applicable to both rotary and reel-type mowers, that the product have a label of specified design to warn of the danger of blade contact, 16 C.F.R. § 1205.6, 44 Fed. Reg. 10029-30. The standards also contain additional directives that are intended to increase the effectiveness of the primary regulations. Thus, because the foot-probe provision can be satisfied by shielding the blade area, the standard mandates tests to assure that shields have a certain minimum strength, 16 C.F.R. § 1205.4(a)(2), 44 Fed. Reg. 10026-28, and that a shielded mower can traverse obstructions. 16 C.F.R. § 1205.4(a)(3), 44 Fed. Reg. 10028-29. The standard also stipulates that shields that move to permit attachment of auxiliary equipment must either automatically return to their normal position when the supplemental equipment is not attached or prevent blade operation unless the shield is manually returned to its normal position when the added equipment is not used. 16 C.F.R. § 1205.4(c). Similarly, the three-second blade-stop requirement is supported by ancillary instructions that mowers employing engine cutoff to halt the blade have power restart mechanisms, 16 C.F.R. § 1205.5(a)(iv), and that all mowers have a control that must be activated before the blade can resume operation in order to prevent the blade from accidentally restarting. 16 C.F.R. § 1205.5(a)(2).

OPEI challenges the legality of the standard by contending that it includes nonconsumer products within its scope of regulation, as well as unique consumer products not proven to present the same hazards as typical consumer lawn mowers. OPEI also argues that substantial evidence on the record as a whole 7 does not support the Commission’s determination that the foot-probe and shielding requirements “are reasonably necessary to reduce or eliminate an unreasonable risk of injury” 8 associated with walk-behind power lawn mowers. In addition, OPEI attacks the blade-stop requirement on the grounds that it is a design restriction rather than a performance requirement, imposes criteria that can only be satisfied by technology that is currently unsafe and unreliable, and mandates a three-second blade-stopping time that is not justified by substantial record evidence. OPEI further asserts that the labeling requirement is invalid because its issuance is not authorized by section 27(e) of the Act, 15 U.S.C. § 2076(e), and it lacks the substantial evidentiary support necessary for it to be promulgated under section 7(a)(2) of the Act, 15 U.S.C. § 2056(a)(2). Finally, OPEI urges that the standard’s effective date is unfeasibly early and that, in any event, substantial record evidence fails to establish that the benefits of the standard bear a reasonable relationship to its costs and demonstrate that the rule is reasonably necessary and in the public interest. Consumer advocate Hayward, on the other hand, claims that the blade-stopping time chosen by the Commission is too slow and that the effective date imposes unnecessary delay. We shall examine each of these objections to the standard in the following discussion.

Scope of the Standard

The CPSC may regulate only “consumer products.” 15 U.S.C. §§ 2052(a)(1), 2058(a)(1). 9 See ASG Industries, Inc. v. Consumer Product Safety Commission, 593 F.2d 1323, 1327 (D.C.Cir.1979); Consumer Product Safety Commission v. Anaconda Co., 593 F.2d 1314, 1317 (D.C.Cir.1979); Kaiser Aluminum & Chemical Corp. v. Consumer Product Safety Commission, 574 F.2d 178, 179-81 (3d Cir.), cert. denied, 439 U.S. 881, 99 S.Ct. 218, 58 L.Ed.2d 193 (1978). As defined by the Act, a consumer product is an article, or component thereof,

produced or distributed (i) for sale to a consumer for use in or around a permanent or temporary household or residence, a school, in recreation, or otherwise, or (ii) for the personal use, consumption or enjoyment of a consumer in or around a permanent or temporary household or residence, a school, in recreation, or otherwise, but . . . not . . . (A) any article which is not customarily produced or distributed for sale to, for use or consumption by, or enjoyment of, a consumer .

15 U.S.C. § 2052(a)(1).

OPEI asserts that the Commission exceeded its statutory authority by attempting to regulate lawn mowers that are not consumer products and by failing to exclude “unique” products, such as high-wheel and three- or five-wheel mowers, that differ significantly from the typical mower used by consumers and that allegedly were not shown by substantial evidence to need regulation in order to eliminate or reduce an unreasonable risk of harm. We find, however, that the safety regulation’s coverage is not impermissibly broad. The standard expressly states that: “Except as provided in paragraph (c) of this section, all walk-behind rotary and reel-type power lawn mowers manufactured or imported on or after the effective date of the standard are subject to the requirements of this standard if they are ‘consumer products’ . . . [as defined in section 3(a)(1) of the CPSA, 15 U.S.C. § 2052(a)(1)].” 16 C.F.R. § 1205.-1(b)(1) (emphasis added). Thus, if a lawn mower is not customarily produced or distributed for sale to, or use or consumption by, or enjoyment of, a consumer, it is ab initio not covered by the standard.

In addition, the standard excludes 10 from regulation mowers that are 30 or more inches wide, weigh 200 or more pounds, and, if engine powered rather than electric, have an engine with eight or more horsepower. 16 C.F.R. § 1205.1(c)(1). Even if the marketing and usage patterns of these larger machines do not automatically place them outside the standard’s coverage of consumer products, they are nevertheless excused from its requirements in recognition of the greater burden that compliance would impose on manufacturers of these units and of their less frequent use by consumers. 11 The

standard’s statement of scope establishes that, in prescribing lawn mower safety measures, the CPSC acted within its authority and sought to regulate only consumer products.

OPEI’s claim that substantial record evidence does not justify application of the standard to “uniquely” designed consumer lawn mowers, 12 such as high-wheel, three-wheel and five-wheel mowers, mowers with nonmetal blades, and air-supported mowers, is also without merit. Sufficient evidence clearly establishes that, as a general category of consumer products, walk-behind power mowers are sufficiently hazardous to justify their regulation by a consumer product safety standard. The Commission ruled that three-wheel, five-wheel, high-wheel, and air-supported rotary mowers were included in the standard because: “The number of wheels or the use of an air cushion instead of wheels would appear to have little relevance to the likelihood of contact with the rotating lawn mower blade. These mowers could present the same risks of blade contact as the common four-wheel rotary mower.” 44 Fed. Reg. 9990, 9997-98 (Feb. 15, 1979). It further decided that mowers with semi-rigid or rigid blades made of nonmetal material, such as plastic or rubber, fell within the standard’s coverage. The Commission reasoned that

[N]o convincing evidence has been presented ... to show that these blades present a different risk of injury than metal blades. If these blades are rigid or semi-rigid, they appear to present an unreasonable risk of amputation, laceration, fracture, or avulsion in the same manner as metal blades. Blades typically rotate at a top speed of approximately 200 mph and a hand or foot struck by a semi-rigid or rigid blade made of material other than metal, that also cuts grass, can be expected to cause the same types of injuries as a metal blade.

Id. at 9998.

The Commission may treat “a range of similar products as a single product class” if they “exhibit . . . sufficient similarity of functional and risk characteristics.” ASG Industries, Inc. v. Consumer Product Safety Commission, 593 F.2d 1323, 1330 (D.C.Cir.1979), cert. denied, 444 U.S. 864, 100 S.Ct. 133, 62 L.Ed.2d 87 (1979) (emphasis added). The task of evaluating function and risk factors to determine product classification for purposes of inclusion within a safety standard “is committed primarily to the judgment of the Commission.” Id. See 15 U.S.C. § 2058(c)(1). 13 However, the Commission does have “a responsibility to justify application of the standard to a product or product use that, according to the facts elicited in the course of rulemaking, exhibits significantly dissimilar functional or risk characteristics when compared with the other products covered by the standard.” ASG Industries, Inc. v. Consumer Product Safety Commission, 593 F.2d at 1330. 14 In the face of apparent functional or risk characteristics that distinguish one group of products from another, justification for general regulation may be provided by a reasoned decision either "(1) that the product presents no significantly dissimilar functional or risk characteristics pertinent to the objectives of the standard; or (2) that despite significant differences, application of the standard to the product remains ‘reasonably necessary to prevent or reduce an unreasonable risk of injury associated with such product . . . ” Id. (emphasis added).

In the present case no evidence was introduced to show that the so-called “unique design” or “specialty” mowers were significantly dissimilar in function or risk characteristics from the typical walk-behind power lawn mower studied by the Commission. As the Commission determined, these mowers have rotating, hardcutting blades that produce the hazards addressed by the safety standard. Thus, the unique design machines are part of the general product category of walk-behind power lawn mowers, and in the absence of any evidence pointing to features differentiating their risk or function characteristics for purposes of protecting consumers from blade-contact injuries, the Commission properly included them in the standard. See Bunny Bear, Inc. v. Peterson, 473 F.2d 1002, 1007 (1st Cir. 1973). 15

Foot-Probe and Shielding Requirements

The standard mandates that walk-behind power rotary mowers pass a foot-probe test designed to assure that the machine guards the operator’s feet against injuries caused by contact with the moving blade. The test requires that a probe simulating a human foot be inserted along the rear 120 degrees of the mower and at the discharge chute without coming into contact with the blade when inserted. 16 C.F.R. § 1205.4, 44 Fed. Reg. 10025-26. See also Fed. Reg. 10001-10002. Mowers meet the foot-probe test by having shields that prevent the probe from entering the blade’s path. 16 C.F.R. § 1205(aXl), 44 Fed. Reg. 10025.

OPEI does not deny that a foot-probe test for the rear area of the mower is reasonably necessary to reduce injuries. 16 Rather, it asserts that application of the test to the discharge chute is not supported by substantial record evidence. It alleges that the injury data does not show that foot injuries occur at that location and that it would be theoretically impossible for an operator to suffer a foot injury at the discharge chute while holding the “deadman’s” blade-control 17 switch on the mower handle. 18

The Act requires that safety standards be supported by “substantial evidence on the record as a whole.” 15 U.S.C. § 2060(c). 19 The foot-probe provision can be sustained only if the record contains “ ‘such relevant evidence as a reasonable mind might accept as adequate to support a conclusion’ ” that an unreasonable risk of foot injury exists from blade-contact at the discharge chute, that the foot-probe test will ameliorate it, and that the benefits of this proposed reform make it reasonable in light of the burdens it imposes on product manufacturers and consumers. Aqua Slide ‘N’ Dive v. Consumer Product Safety Commission, 569 F.2d 831, 838-40 (5th Cir. 1978) [hereinafter cited as Aqua Slide] (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 216, 83 L.Ed. 126 (1938)).

The determination of whether an unreasonable risk of discharge-chute injury exists involves “a balancing test like that familiar in tort law: The regulation may issue if the severity of the injury that may result from the product, factored by the likelihood of the injury, offsets the harm the regulation imposes upon manufacturers • and consumers.” Aqua Slide, 569 F.2d at 839 (quoting Forester v. Consumer Product Safety Commission, 559 F.2d 774, 789 (D.C. Cir. 1977) (defining “unreasonable risk” in case brought under the Federal Hazardous Substances Act, 15 U.S.C. §§ 1261-74 (1976), and referring to “similar language” and legislative history of CPSA for support)). 20 See H.R.Rep.No.1153, 92d Cong., 2d Sess. 33 (1972); 21 Note, The Consumer Product Safety Commission: In Search of a Regulatory Pattern, 12 Colum.J.L. & Soc. Prob. 393, 401-12 (1976); Note, The Consumer Product Safety Act: Risk Classification & Products Liability, 8 Ind.L.Rev. 846, 849-52 (1975) (citing cases and materials discussing meaning of “unreasonable risk”). Thus, under the unreasonable risk balancing test, even a very remote possibility that a product would inflict an extremely severe injury could pose an “unreasonable risk of injury” if the proposed safety standard promised to reduce the risk effectively without unduly increasing the product’s price or decreasing its availability or usefulness. Aqua Slide, 569 F.2d at 839-40. 22 Conversely, if the potential injury is less severe, its occurrence must be proven more likely in order to render the risk unreasonable and the safety standard warranted.

In the present case, the discharge-chute probe is intended to reduce the risk of such injuries as amputation of toes, fractures of bones in the feet or toes, avulsions, deep lacerations, and contusions. While the seriousness of these injuries cannot be gainsaid, it does not rise to the level of gravity that would render almost any risk, however remote, unreasonable if the risk could be reduced effectively by the proposed regulation. 23 Substantial evidence that such injury is significantly likely to occur is therefore necessary to sustain this portion of the lawn mower safety standard.

Our examination of the record has failed to reveal substantial evidence that injury at the discharge chute was sufficiently probable that it made the risk addressed by the foot probe of this area unreasonable. In a study of 36 blade-contact foot injuries conducted for the CPSC by the National Electronic Injury Surveillance System (NEISS), 24 one injury occurred when the operator inserted his foot into the blade path at the discharge chute while holding the mower handle. This injury represented almost three percent of the blade-contact foot injuries in the sample. However, the study did not involve a random sample, and it is not possible to extrapolate the percentage of total blade-contact injuries represented by discharge-chute incidents involving the operator’s feet from the limited information furnished in the record. In any event, trustworthy statistical inferences cannot be drawn from a single incident of discharge-chute injury. Without reliable evidence of the likely number of injuries that would be addressed by application of the foot-probe test to the discharge chute, we are unable to agree that this provision is reasonably necessary to reduce or prevent an unreasonable risk of injury. See D. D. Bean & Sons Co. v. Consumer Product Safety Commission, 574 F.2d 643, 650-51 (1st Cir. 1978) (hereinafter D. D. Bean) (holding that absence of relevant injury data associated with particular hazards renders requirements of safety standard addressed to them invalid and observing that “a single injury ... is not substantial evidence of an ‘unreasonable risk of injury.’ ”). It must be remembered that “[t]he statutory term ‘unreasonable risk’ presupposes that a real, and not a speculative, risk be found to exist and that the Commission bear the burden of demonstrating the existence of such a risk before proceeding to regulate.” Id. at 651.

Our conclusion that substantial evidence fails to justify this provision is not altered by the fact that the industry’s voluntary standard, ANSI B71.1-1972 § 1968, and ANSI 71.1b-1977 § 11.8, requires probing of the discharge chute. See 44 Fed. Reg. at 10003. A private industry safety standard cannot, by itself, provide sufficient support for a Commission regulation. “ ‘While such private standards may tend to show the reasonableness of similar Commission standards, they do not prove the need for such provisions.’ ” Aqua Slide, 569 F.2d at 844 (quoting Forester v. Consumer Product Safety Commission, 559 F.2d at 793 (emphasis in original)). We therefore vacate that part of the standard requiring the discharge-chute area of power lawn mowers to pass a foot-probe test.

OPEI also attacks certain aspects of the shielding requirements that supplement the foot-probe test as devoid of substantial evi-dentiary support. The standard directs that shielding must pass a shield-strength test in order to assure that shields maintain their structural integrity and remain attached under conditions of use and provide the intended protection. 16 C.F.R. § 1205.-4(a)(2). 25 Similarly, the standard requires that shielded mowers undergo an obstruction test, 16 C.F.R. § 1205.4(b)(2), which simulates surface irregularities that the mower may encounter in normal use. 26 The test is designed to ascertain that the shielding will not interfere with mower performance by catching on obstructions and stopping the mower. It is also intended to prevent the shields themselves from contributing to lawn mower hazards, either by lifting the mower excessively when it meets an obstruction and thus exposing the blade or by suddenly halting the mower and causing it to lift and/or the operator to stumble on it.

The Commission estimates that the foot-probe/shielding requirements, including the supporting strength and obstruction tests, will reduce blade-contact foot injuries by 13,000 incidents each year. It did not apportion this injury reduction among the respective foot-probe shielding requirements on the ground that this was impossible because they are interrelated. The Commission estimates that the shield provisions will cost about $4 per mower, primarily for redesigning shields. Little expense is believed to result from the shield-strength test, since it is similar to a requirement in the existing voluntary industry standard, ASNI B71.1 — 1972 §§ 11.5, 11.15; ANSI B71.1b-1977, §§ 11.5, 13.2.1, which receives almost universal compliance. 27

OPEI readily admits that shielding is necessary to protect operators against blade-contact foot injuries at the rear of the mower. It contends, however, that the shield-strength test is invalid because few injuries have been shown to be caused by inadequately strong shields. OPEI also argues that the possibility that users might

remove the protective shielding if it interfered with mower utility by catching on surface irregularities cannot justify the obstruction test and asserts that the test has not been proven necessary to guard against injuries cause by the mower’s sudden stopping when it catches on obstacles.

We find that OPEI’s approach to the shield-strength test is misconceived. It is true that were this requirement viewed in isolation as intended to address a risk of injury from mower shields falling off or cracking and exposing the operator’s foot to the blade, independent of the standard’s requirement that mowers be equipped with foot-probe shields, the record would not contain sufficient evidence that such injuries are so numerous that they support the regulation. 28 But the shield-strength test is mandated as an ancillary feature of the foot-probe and shielding requirements that in turn are concededly necessary to reduce an unreasonable risk of operator blade-contact foot injuries. The shield-strength provision is not to be understood as a discrete measure addressing a distinct type of operator foot injury but as part of the Commission’s effort to make the shielding remedy itself effective and safe. Since an unreasonable risk .of operator foot injury from blade contact has been established, the shields are reasonably necessary to prevent access to the blade, and the “curative effect” of the shield-strength requirement in preventing blade exposure from inadequate shielding is “patent,” we “do not think that the Commission had to cite empirical data in support of its finding that the particular [shield strength] requirement [was] likely to reduce the risk of injury.” D. D. Bean, 574 F.2d at 649. 29

Our opinion in Aqua Slide, 569 F.2d at 842, lends support to the determination that the shield-strength test can- be justified as logically related to the necessary shielding remedy and as providing a clearly effective remedy for a patent risk of injury from broken shields without a separate “body count” of incidents involving structurally unsound shields. In Aqua Slide we observed that the Commission’s burden to produce empirical evidence showing that a proposed warning sign was a reasonably necessary means of reducing an unreasonable risk “may be especially light” when the warning signs were aimed at first-time adult' and teenage users, making it logically more likely that the signs would be heeded and thus effective in reducing injuries. See also id. at 845 (Wisdom, J., concurring: “ ‘untested theory’ that some people will react and be affected by warning signs . needs little, if any, empirical support to meet the substantial evidence standard. . . . ” when “the message is [included] [for] . . first-time adult and teenage [testers]”).

Moreover, the shield-strength test is a particularly appropriate exercise of Commission authority to implement remedies for unreasonable risks because the hazards that will accompany structurally unsound shielding will not become widely manifest until shielding is generally required for consumer power lawn mowers by the standard. When part of a standard is directed at making sure that required safety measures provide their intended level of protection, and it is clear that if they do not “ ‘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.’ . . . [N]o precise statistical showing is required.” Forester v. Consumer Product Safety Commission, 559 F.2d 774, 789 (D.C.Cir.1977) (quoting 1969 Senate Report to the Federal Hazardous Substances Act, S.Rep.No.71,237, 71st Cong., 1st Sess. 2-3 (1969)). We therefore uphold the shield-strength test provision of the standard.

For similar reasons we reject OPEI’s challenge to the obstruction test requirement and find that it is reasonably necessary to guard against intentional consumer defeat of the shielding safety device and to prevent shielding from interfering with mower utility. Although OPEI does not dispute the likelihood that consumers will remove shields if they interfere with mower performance 30 or contest the feasibility of the obstruction test, it does contend that the possibility that consumers may remove protective shielding if it hampers mower utility by catching on surface obstructions does not present the kind of unreasonable risk of injury that the Commission has authority to regulate. In essence, OPEI argues that the risk of injury from consumer defeat of safety shielding is not “unreasonable” because . consumers would have chosen to incur the risk, and their judgment must be respected.

However, Congress intended for injuries resulting from foreseeable misuse of a product to be counted in assessing risk. Aqua Slide, 569 F.2d at 841. See S.Rep.No.92-749, 92d Cong., 2d Sess. 14, 92 Cong.Rec. 36198 (1972) (remarks of Sen. Moss) (risk of injury “associated with” consumer products, 15 U.S.C. § 2052(a)(3), to be regulated by CPSC includes risks of injury resulting from “exposure to or reasonably foreseeable misuse of a consumer product”); Kimber, Federal Consumer Product Safety Act § 94 at 109 (1975). Cf. Pacific Legal Foundation v. Department of Transportation, 593 F.2d 1338, 1345 (D.C.Cir.1979) (Secretary of Transportation required to consider probable public reaction to passive automobile restraints, including possibility of attempts to deactivate them, in promulgating safety standard under National Traffic & Motor Vehicle Safety Act of 1966, 15 U.S.C. §§ 1381 et seq. (1976)). This principle, and not the tort liability concept of “assumption of risk,” governs the Commission’s authority to treat consumers’ foreseeable action of removing safety shields as creating an unreasonable risk of injury and to issue rules addressing that danger. See United States v. General Motors Corp., 518 F.2d 420, 434-35 (D.C.Cir.1975) (“[determinations of fault or liability relevant to the award of damages . . . are not controlling on the interpretation of the” meaning of product “defect” in prophylactic defect notification legislation of National Traffic & Motor Vehicle Act of 1966). Of course, a fully informed choice on the part of consumers to employ a dangerous product may provide information that is relevant to the Commission’s assessment of the reasonableness of a risk of injury. For example, consumers’ decisions to use sharp knives may pose a reasonable risk of injury because duller knives, while safe, would be useless for cutting purposes, and the Commission could reasonably find that consumers have accurate information of the severity and likelihood of injury posed by sharp knives. See Aqua Slide, 569 F.2d at 839; S.Rep.No.92-749, supra at 6-7. In the present case, however, there is no evidence that consumers accurately appreciate the nature of the risk of blade-contact injuries and that their presumed willingness to defeat protective measures is reasonable.

The record contains substantial evidence, in the form of comment letters from both consumers and consulting engineers, that shields on some mowers interfered with the machine’s mobility and consequently were removed by their owners, thus eliminating any safety benefits they were intended to provide. Since a risk of blade-contact injury is clearly created by removing the protective shields and the risk can be effectively reduced by eliminating the annoyance imposed by shields that interfere with mower movement, it was proper for the standard to include an obstruction test to reduce the risk of this foreseeable misuse of lawn mowers.

The obstruction test is valid also as a measure designed to assure that shielding is feasible as a method of reducing the risk, see Aqua Slide, 569 F.2d at 839; Forester v. Consumer Product Safety Commission, 559 F.2d at 789 n.21, and does not inordinately interfere with mower utility. In promulgating a safety standard, the Commission must consider the “probable effect of such rule upon the utility ... of such products . . .” 15 U.S.C. § 2058(c)(1)(C) (emphasis added). See Aqua Slide, 569 F.2d at 839; H.R.Rep.No.1153, 92d Cong., 2d Sess. 33 (1972). In order to fulfill its statutory responsibility to determine that the requirements of a safety standard will not unreasonably reduce product usefulness, the Commission must have authority to establish performance criteria for its chosen remedies. Therefore, the obstruction test, as a component- of the foot-probe/shielding requirements, is reasonably necessary to ensure that mowers satisfy the standard in ways that do not decrease the product’s utility. Accordingly, we uphold the obstruction-test provision.

Blade-Control System

The second key element of the standard requires a blade-control system that (1) will prevent the blade from rotating unless the operator activates a control, (2) allows the blade to be driven only if the operator remains in continuous contact with the “deadman’s” control, and (3) causes the blade to stop moving within three seconds after the deadman’s control is released. 16 C.F.R. § 1205.5. In addition, the standard directs that engine-powered mowers with only manual starting controls must have a blade-control system that stops the blade without turning off the engine. This requirement is intended to reduce the temptation for consumers to disable the blade-control mechanism in order to avoid the inconvenience of having to restart the engine manually each time the deadman’s switch is released. Also, all walk-behind rotary power mowers must have a separate restart mechanism as well as an initial ignition device in order to prevent inadvertent starting of the blade by accidental activation of the primary ignition control. The controls must be located in an “operator control zone” at the rear of the mower so that during normal use, including starting efforts, the operator generally will be behind the mower handle. Id.

The blade-control system is intended to protect the operator against blade-contact injuries to both hands and feet by stopping the blade before the operator can contact it after he or she leaves the normal operating position and thus releases the deadman’s control. 31 The Commission estimates that the blade-control provisions will eliminate approximately 46,500 operator blade-contact injuries a year. This figure represents approximately 60 percent of all blade-contact injuries and nearly 80 percent of all injuries claimed to be reduced by the standard. As OPEI acknowledges, the blade-contact requirements thus are the “centerpiece” of the Commission’s strategy for reducing blade-control injuries.

Perhaps as befits its importance in the regulatory scheme promulgated by the Commission, the blade-control system is vigorously attacked by both OPEI and consumer proponents of stricter safety requirements. OPEI asserts that the blade-control system provision is expressed as a design requirement, rather than as a performance requirement, in violation of the Act. It argues that a number of alternative requirements are available that are less design restrictive and more performance oriented than the blade-stop criterion and that the Commission therefore erred in adopting the blade-stop approach. OPEI further contends that the standard is unreasonable because it requires the use of mechanisms that allegedly are not safe or reliable. And, OPEI claims, the three-second stopping time is unreasonably short. In contrast, consumer advocate John 0. Hayward maintains that the three-second blade-stopping time is too lax and that substantial evidence demonstrates that only a two-second or shorter blade-stopping time is justified.

The CPSA directs that a safety standard’s provisions “shall, whenever feasible, be expressed in terms of performance requirements.” 15 U.S.C. § 2056(a). The statutory preference 32 for performance requirements is rooted in the belief that this mode of regulation stimulates product innovation, promotes the search for cost-effective means of compliance, and fosters competition by offering consumers a range of choices in the marketplace, while design-restrictive rules tend to freeze technology, stifle research aimed at better and cheaper compliance measures, and deprive consumers of the opportunity to choose among competing designs. See S.Rep.No.92-835, 92d Cong., 2d Sess. 30 (1972), U.S.Code Cong. & Admin.News 1972, p. 4573; H.R. Rep.92-1153, 92d Cong., 2d Sess. 189 (1972).

Although only a limited number of designs can satisfy the blade-stop provision, we find this part of the standard is nonetheless a performance requirement. While the standard mandates that mower blades stop within a specified time period, it does not dictate a specific means of fulfilling this condition. Manufacturers are neither formally nor practically restricted to employing a particular design, since two existing mechanisms, a blade-disengagement system employing a brake-clutch device and an engine-stop system, are capable of passing the blade-stop test. See, e. g., D. D. Bean, 574 F.2d at 652. Moreover, the existence of many different designs for brake-clutch apparatus preserves manufacturer discretion in building mowers. And manufacturers remain free to develop new forms of technology that will meet the blade-stop criterion. The statutory direction that safety standards are to be framed as performance requirements when feasible does not mean that the Commission must develop standards that are so general that they can be' satisfied by every possible means for achieving an acceptable level of safety. Indeed, a performance requirement may be quite design restrictive in its practical import. See, e. g., Pacific Legal Foundation v. Department of Transportation, 593 F.2d 1338, 1340-41 & n.6 (D.C.Cir.), cert. denied, 444 U.S. 830, 100 S.Ct. 57, 62 L.Ed.2d 38 (1979) (upholding, under National Traffic & Motor Vehicle Safety Act of 1966, safety requirement that court characterized as “performance standard” and that could be satisfied by only two currently available systems, automatic passive seat belts and air bags); Paccar, Inc. v. National Highway Traffic Safety Administration, 573 F.2d 632, 639 & n.26 (9th Cir.), cert. denied, 439 U.S. 862, 99 S.Ct. 184, 58 L.Ed.2d 172 (1978) (evaluating “performance requirement” issued under NTMVSA that as a practical matter required employment of a particular device). Chrysler Corp. v. Department of Transportation, 472 F.2d 659, 664 & n.4 (6th Cir. 1972) (holding that performance requirement could establish minimum conditions that could be met by few systems since industry not foreclosed from develop ing other ways of complying). The blade-stop provisions therefore qualify as performance requirements.

We also find that the Commission did not act improperly in failing to formulate a safety standard that would have allowed lawn mowers to be equipped with safety devices based on concepts of blade inaccessibility or blade harmlessness 33 as alternatives to blade-stop safety mechanisms. The CPSC did consider blade harmlessness and blade inaccessibility requirements but did not develop and adopt these alternative measures as part of the standard because it found that they were not feasible means of reducing or eliminating the risk of blade-contact injury at the time the standard was issued. The record contains substantial evidence that the alternative requirements now demanded by OPEI were justifiably rejected as flawed. The Commission entertained suggestions for tests of blade harmlessness in the hope that safer blades might prove an effective means of reducing blade-contact injuries. However, experts in evaluating industrial injury experiments found that the proposed tests of blade harmlessness measured only the damage that could occur to an inanimate test object when it encountered a revolving blade and did not adequately indicate whether a part of the human body was likely to be-injured by a moving blade. 34 The. Commission therefore correctly concluded that'the standard could not discriminate between safe and unsafe mowers on the basis of a blade-harmlessness criterion.

Similarly, proposed tests of blade inaccessibility were found wanting. The Commission carefully evaluated suggested blade inaccessibility requirements and found that the probes for assessing hand accessibility upon which this alternative measure would rely did not adequately simulate the various positions and movements of the human hand. It therefore concluded that the test did not enable researchers to determine whether blade access mechanisms would function to protect the operator. 35 The foot probes urged by the industry to test blade inaccessibility were essentially the same as the standard’s rear' periphery foot-probe test. However, the Commission found that reliance upon a foot-probe test without a blade-stop requirement would not protect consumers against foot injuries occurring when an operator returned to a running mower after having left it to perform a task, such as removing debris from its path, and inadvertently placed a foot under the blade housing at the machine’s side or front.

After studying the proposed alternative requirements, the Commission concluded that research time and money exceeding the Commission’s resources 36 would be necessary in order to develop adequate tests, if they could be designed, to submit the alternative safety standards for public comment, and to issue a final standard that included the additional safety provisions. In its judgment, delaying the promulgation.of the lawn mower safety standard until such an effort was completed was not justifiable because the heavy annual toll of serious injuries inflicted by mowers could be reduced immediately by implementation of the blade-stop requirement. Both sound policy and substantial record evidence support the Commission’s decision.

Because the proposed alternative requirements were deficient at the time this standard was prepared, we need not decide whether a proposed standard is “reasonably necessary to reduce or eliminate an unreasonable risk of injury associated with [a consumer] product,” 15 U.S.C. § 2056(a) (emphasis added), if other safety measures, equally effective and feasible, exist but are not permitted to satisfy a CPSC standard. 37 Cf. Chrysler Corp. v. Department of Transportation, 472 F.2d 659, 674 (6th Cir. 1972) (agency’s choice of air bags over active seat belt restraints as “practicable” and meeting “the need for motor safety” must be upheld when neither method “is clearly superior to the other in every respect”). We hold only that on the record in this case the standard’s blade-stop directive is properly framed as a performance requiremént and that the standard permissibly mandates this method of protection without also including other conceivable safety measures.

We turn now to the issue of whether substantial evidence supports the selection of three seconds as the time limit within which blades must stop. The Commission based the three-second blade-stop time limit primarily upon four time-motion stu dies of operator-blade access time. These experiments were designed to measure the interval between the moment the operator released the deadman’s control and the instant he or she reached the mower blade. One study of operator-blade access data collected at the University of Iowa in 1971 and analyzed by the National Bureau of Standards in 1975 showed that operators reached the blade hazard point after moving directly to it from the mower’s rear in times ranging from 2.0 to 3.5 seconds. 38 A second operator-blade access study, using 100 subjects ranging in age from 16-62 with a mean age of 35, was conducted by Consumer’s Union at Eckerd College in St. Petersburg, Florida, on March 5-7, 1975. The participants were tested using a reaction-time device designed to simulate a walk-behind power lawn mower. Each participant underwent 25 trials, for a total of 2,500 time-motion incidents. A December 7, 1976, summary of this Eckerd College Study reported that operator access time ranged from 1.66 to 4.90 seconds 39 for normal, as opposed to intentionally fast, movements. This summary cited data from trials using 12 and 15 subjects, aged 18-21. The Final Report on the Eckerd College Study, issued in May 1977, shows operator access time for the 100 participants as ranging from .6 to 3.3 seconds. The discrepancy is not explained. A third blade-access study using five subjects was conducted by the Office of the Medical Director (OMD) of the CPSC. In this experiment, direct, “casual” approach time ranged from 2.35 to 3.3 seconds, with an average time of 2.81 seconds. 40 OMD also recorded intentionally fast direct access times. These ranged from 1.26 to 2.5 seconds, with an average access time of 1.79 seconds. 41 In addition, OMD tested subjects’ access times when they approached the discharge chute blade area by going around the opposite side and front of the mower. The access time in the indirect, casual approach test ranged from 3.64 to 4.70 seconds, with an average time of 4.25 seconds. 42 Operator-access time in the indirect, hurried approach test fell between 2.48 and 3.68 seconds, with an average of 2.98 seconds. 43 The record contains only preliminary findings from the fourth empirical examination of operator blade-access times. This study was undertaken by the National Bureau of Standards. “[A] superficial examination of the limited observational data” was reported to show an average blade-access time of under three seconds.

In setting the blade-stop time, the Commission considered not only the time in which an operator could reach the blade after releasing the deadman’s control but also the incremental cost of successively faster blade-stop times. The record contains substantial evidence that the cost of blade-stop mechanisms varies inversely with the length of time in which the device stops the blade, 44 so that a three-second blade-stop requirement will be cheaper to implement than a one- or two-second time limit.

We find that the three-second blade-stop requirement is not too lax, contrary to petitioner Hayward’s claim. The three-second measure will protect consumers against many, although certainly not all, blade-contact injuries. While the Commission may not rely upon mere “common sense” or speculation to establish the existence of an unreasonable risk of injury, it may exercise considerable discretion in determining an appropriate remedy. The Commission was entitled to consider the incremental cost of requiring a shorter blade-stop time in rejecting a one- or two-second blade-stop solution. The standard need not guarantee protection for all consumers; it is sufficient that it promises greater safety for consumers and is reasonably necessary to reduce the risk of blade-contact injuries.

Correspondingly, we find no merit in OPEI’s contention that the requirement of a three-second stopping time is unreasonably demanding. OPEI strenuously urges that the empirical studies cannot support a three-second blade-stop limit because they unrealistically fail to account for such psychological factors as a person’s fear of a noisily operating machine and resulting reluctance to approach it. OPEI argues that these conditions would tend to make the operator stay away from the blade area for a longer time than was recorded in the artificial tests and perhaps until all noise and vibration from a moving blade ceases. OPEI also argues that the three-second requirement is not reasonable because a blade that is not moving under power would be decelerating and thus would inflict less damage than a fully powered blade.

The Commission properly rejected these contentions. The record contained neither suggestions for devising a test to measure psychological factors nor evidence indicating that in more “realistic” circumstances operator access to blades would be slower than registered in the time-motion studies. No deceleration curve information was cited to substantiate the claim that between three and seven seconds after power shutoff a blade would be coasting slowly, and no available test can accurately determine the blade velocity at which a risk of injury is reasonable.

We are also unpersuaded by OPEI’s contention that current technology is inadequate safely to achieve a three-second blade-stop time. OPEI asserts that this requirement can only be met by using brake-clutch systems, which it alleges are unreliable and unsafe. The record does contain some evidence of unsatisfactory blade-clutch performance. For example, one lawn mower manufacturer provided test data on 19 mower units using five different brake-clutch designs that revealed that none of the mowers functioned successfully for 250 hours of testing, the period suggested as necessary to establish machine-life reliability. However, the manufacturer of the brake-clutch mechanisms employed in this test analyzed these results and concluded that the problem arose because the mowers’ engines ran at a lower speed than engines for which the clutch was designed. The clutch manufacturer explained that while its standard clutches would not perform satisfactorily on low-speed engines, most lawn mowers had engines with faster speeds. It also expressed confidence that its standard clutches could be modified to perform well on low-speed engines. Moreover, other evidence in the record suggested that different types of clutches from those used in this experiment have been used successfully with low-speed motors.

Extensive evidence that available brake-clutch technology can produce lawn mowers capable of three-second blade stops overcomes the doubt expressed by some commentators as to their ability to provide a safe and reliable three-second blade-stop mechanism. One clutch manufacturer introduced empirical data showing that its brake-clutch mechanism was durable and reliable. In actual mowing tests, three of its units successfully completed over 250 field test hours, and in laboratory testing, several mower units completed 250 hours of operation without blade-stop failure and without increased wear on the engine or crankshaft. The General Services Administration subjected another brake-clutch design to fairly extensive field testing in 1977. The GSA used 50 mowers under varied conditions in ten locations for four months and reported no brake-clutch failures during the trial. This test was continued in 1978, and it was reported that the mowers were continuing to perform well, with complete reliability and safety. While some commentators criticized current mechanisms for being vulnerable to clutch slippage under heavy mowing conditions, the GSA report did not suggest that clutch slippage was a major source of dissatisfaction.

Most convincing proof that safe and reliable three-second blade-stop mowers are currently feasible is the fact that, at the time the standard was issued, at least two mower manufacturers were currently producing and marketing mowers that had a brake-clutch mechanism complying with the standard. As one of these manufacturers declared, that such mowers are offered for sale demonstrates their manufacturer’s belief in the safety and reliability of this type of brake-clutch mechanism. In addition, the evidence established that a brake clutch is basically a simple device that is widely used with safety and reliability in the production of automobiles and other types of machinery, such as chainsaws. The technology associated with brake clutches is therefore highly developed, and comments presented to the Commission made clear that potential difficulties with the particular application of various types of brake-clutch devices to power lawn mowers can be identified and resolved with present technical knowledge.

This evidence provides substantial support for the Commission’s judgment that technology is available to design, produce, and assemble brake-clutch power lawn mowers that are unlikely to fail in an unsafe manner. While the precise means of stopping the blade within the mandated time must be developed for each model of mower and requires engineering analysis and testing of a lawn mower’s entire system to assure that the brake clutch is compatible with other components of the mower, the record establishes that existing technology can be refined and applied to power lawn mowers to bring them into successful compliance with the requirements of the standard. The three-second blade-stop time is therefore valid. See, e. g., ASG Industries, Inc. v. Consumer Product Safety Commission, 593 F.2d at 1333-34 (CPSC has “authority to employ technology forcing standards” so that, even when “the required technology [to remedy a risk] is not now available,” the Commission may “predicate a finding of unreasonable risk on the projection of technological advance [if the agency has] some basis in its records . supporting the projection as meaningful and reasonable, as contrasted with mere speculative desire”). Cf. Pacific Legal Foundation v. Department of Transportation, 593 F.2d at 1344 & n.44 (Motor Vehicle Safety Act of 1966 “authorizes safety standards that push the automobile industry beyond present engineering capabilities”); Chrysler Corp. v. Department of Transportation, 472 F.2d 659, 673 (6th Cir. 1972) (“the Agency is empowered to issue safety standards which require improvements in existing technology or which require the development of new technology, and it is not limited to issuing standards based solely on devices already fully developed”).

Labeling Requirement

The safety standard directs that both reel-type and rotary-power walk-behind lawn mowers carry a label of specified design. The label, which is to be placed on the blade housing or other blade shielding, states “DANGER, KEEP HANDS and FEET AWAY.” It also contains a pictorial representation of a red, blade-like object cutting into the forefinger of an extended, black hand. 16 C.F.R. § 1205.6. The portrait of the blade and hand does not depict the hand as bleeding or dismembered.

The Commission issued the section 1205.6 labeling provision under the purported authority of sections 7 and 27(e) of the CPSA, 15 U.S.C. §§ 2056(a)(2), 2076(e). In order for the label to have been properly issued under section 7, there must be substantial evidence that it is reasonably necessary to prevent or reduce an unreasonable risk of injury. 45 Section 27(e) authorization for the labeling requirement does not require this finding. However, this section of the Act does allow the Commission to direct manufacturers to provide information thát conveys performance and technical data necessary to carry out the purposes of the Act, including the protection of the public against unreasonable risk of injury associated with a consumer product. 46 OPEI contends that section 27(e) does not provide authority for the issuance of the type of warning label mandated by the standard and that substantial record evidence does not support requiring the warning label pursuant to section 7 of the Act.

We agree with OPEI that section 27(e) does not authorize promulgation of the warning label requirement. Notification of the hazard of blade-contact injury by means of a graphic design and the exclamation, “Danger, Keep Hands and Feet Away,” does not fit within the “plain, obvious, and rational meaning” of “performance ■ and technical data ” 47 that the Commission may order manufacturers to provide to prospective purchasers and to a product’s first consumer under section 27(e). See Old Colony Railroad v. Commissioner, 284 U.S. 552, 560, 52 S.Ct. 211, 213, 76 L.Ed. 484 (1932) (in interpreting statutory language, “the plain, obvious and rational meaning of a statute is to be preferred to any curious, narrow, hidden sense”).

However, we are convinced that the Commission acted permissibly under section 7 in ordering manufacturers to furnish mowers with warning labels. The record clearly establishes that an unreasonable risk of blade-contact injury exists. The warning label directly addresses this risk of injury by warning operators and bystanders to keep hands and feet from the area in which the blade is accessible. It is true that the standard’s shielding and blade-stop requirements also address the risk of blade-contact injury, but there is no claim that these prophylactic measures will be fail-safe. Thus, the Commission properly concluded that there was a need for labeling as a safety measure.

The reasonableness of the labeling requirement is also supported by the record. The Commission systematically tested various warning labels, and those with designs similar to the label required by the standard were found effective in communicating the risk of danger from moving blades. Significantly, the Commission found that the labeling requirement would impose little, if any, additional cost. The industry’s voluntary standard already requires a warning label, and the mandatory label would be no more burdensome. Its contents are not shocking or gruesomely explicit and would not pose an unwarranted deterrent to potential purchasers of lawn mowers. See Forester v. Consumer Product Safety Commission, 559 F.2d at 793 (“private standards may tend to show the reasonableness of similar Commission standards,” though by themselves they do not prove a need for them) (emphasis added)). Cf. Aqua Slide, 569 F.2d at 841-43.

The labeling requirement upheld here is very different from that we found invalid in Aqua Slide. In that case the risk of paralysis from sliding headfirst into the water on a swimming pool slide was one in ten million. The Commission did not test its proposed warnings signs, and their efficacy in reducing the risk was slight, since it was unlikely that readers would understand or attend to the warning’s long message. 569 F.2d at 841. Focusing upon these circumstances, we noted that “the crucial question” in “weighing the ‘reasonable necessity’ for the signs . . . [was] whether the [anticipated] benefit has a reasonable relationship to the disadvantages the sign requirement imposes.” Id. at 842. In Aqua Slide the prime disadvantage was “the warning’s effect on the availability” of the consumer product. The swimming pool slides regulated by the standard in that case were luxury items subject to highly elastic market demand. Id. Indeed, slide sales dropped 42 percent from 1973 to 1974, and half of that decrease “could be attributed just to uncertainty about what the standard would say.” Id. Because the Commission had not tested the signs, “it provided little evidence of whether the signs were so explicit and shocking in their portrayal of the risk of paralysis as to constitute an unwarranted deterrent to the marketing of slides, and, hence, their availability to users.” Id.

Our own examination of the record in Aqua Slide disclosed some evidence that the signs, which did not indicate that the risk of paralysis was only one in ten million, were so alarming that they might well impose high costs on the industry and on potential purchasers of swimming pool slides. Id. Since the proof that the signs could reduce the risk of injury was weak, an adequate analysis of buyer reaction to the alarming signs had not been performed, and the Commission’s analysis of the economic impact of the entire standard was otherwise unreliable, we ruled that substantial evidence did not support the crucial finding that the “relationship between the advantages and disadvantages of the signs is reasonable” and that consequently the reasonable necessity of the warning signs had not been demonstrated. 569 F.2d at 841. In the present case, an unreasonable risk of blade-contact injury has been shown, and the warning label clearly does not threaten to impose costs on lawn mower manufacturers and consumers like those that were at issue in Aqua Slide. On this record, we hold that the labeling requirement is reasonably necessary to reduce an unreasonable risk of injury.

Finding of Reasonable Necessity and Public Interest

In the preceding discussion, we have examined the petitioners’ challenges to specific provisions of the standard and have found several of the complaints unfounded. However, OPEI also contends that the standard as a whole is not supported by substantial record evidence. It claims that the Commission based its determination that the standard was “reasonably necessary” and would produce a net benefit to society upon a document, “Economic Impact of Blade Contact Requirements for Power Mowers” (“the economic report”), which was unreliable because its methodology was fatally flawed, and its findings had never been exposed to public scrutiny in the administrative rulemaking process. Petitioner Hayward also criticizes the Commission’s evaluation of the standard’s net social benefit. He argues that the CPSC undervalued the safety benefits of the standard by erroneously failing to place a monetary value on the pain and suffering inflicted by the injuries that the rule was expected to reduce.

The Commission seeks to counter OPEI’s broad attack on the standard by defending the final economic report’s data and methodology and by asserting that the report was, in essence, merely a revision of earlier cost-benefit analyses of shielding and blade-stop requirements in light of additional evidence introduced into the record by OPEI. The Commission further contends that it adequately considered pain and suffering in evaluating the standard’s benefits and that it was not required to quantify these aspects of the cost of lawn mower accidents.

We have carefully scrutinized the record and find the substantial evidence supports the conclusion that the safety benefits expected from the standard bear a reasonable relationship to its costs and make the standard reasonably necessary and in the public interest. The cost-benefit analysis contained in the final economic report and adopted by the standard, 44 Fed.Reg. at 10020-21, 10030, is not methodologically flawed. The Commission estimated that the regulations would raise the retail price of a complying lawn mower $35, costing the consumer $4.40 per year over the projected eight-year life of the mower. Total yearly compliance costs were believed to be $189 million for 5.4 million mower units (1978 production estimate). Blade-contact injuries were calculated to cost $253 million annually, exclusive of pain and suffering. Since, as we have noted, there are approximately 77,000 blade-contact injuries from walk-behind power mowers each year, each injury costs about $3,300, without counting the cost of pain and suffering. Currently there are some 40 million mowers in use by consumers, so that a consumer has about one chance in 500 (V520) of incurring an injury costing $3,300, exclusive of pain and suffering. The standard’s injury cost associated with each mower without the safety features is thus $6.35 per year. The Commission anticipated that implementation of the standard would reduce this injury cost by 83 percent, for an annual savings of $5.30 per mower, exclusive of the savings of pain and suffering costs. Because the standard would result in a net benefit of $.90, a mower meeting the standard’s safety requirements would represent a worthwhile investment for the consumer, and the standard’s implementation is in the best interests of society.

OPEI questions this analysis by asserting that, given an eight-year mower life, full compliance with the standard will take eight years because only one-eighth of all mowers will be replaced each year. As a result, only one-eighth of the total 83 percent estimated injury cost reduction attributed to the standard will be achieved after the standard’s first year, with an additional one-eighth of the projected savings being achieved for each succeeding year until full compliance is had. From this observation, OPEI reasons that a complying mower will reduce injury costs by a mere $.70 after one year and not by the $5.30 projected by the Commission in justifying the standard. OPEI derived the $.70 per mower savings figure by reducing both the number of annual injuries and their total costs by one-eighth in determining a $5.65 per mower injury costs after one year, rather than the Commission’s estimated annual injury cost of $6.35 per mower. OPEI then subtracted the $5.65 from the $6.35 to ascertain the savings in per mower injury cost after one year. On the basis of this calculation, OPEI claims that after one year a consumer will have paid $4.40 to reduce his mower injury costs by only $.70 and that the standard thus imposes an unreasonable expense on the consumer.

OPEI’s attack on the cost-benefit analysis underpinning the standard is misconceived. It fails to recognize that only purchasers of complying mowers will pay the $4.40 annual cost for the standard’s safety features and that these owners will receive the full benefit of the standard’s anticipated effectiveness in reducing their injury costs for each year of the mower’s life. OPEI’s cost-benefit assessment thereby erroneously substitutes a figure representing the injury savings per mower in use, whether complying or not, for the amount of injury savings for owners of complying mowers, which is the relevant calculation. The correct analysis employed by the Commission to support the standard demonstrates its reasonable necessity in reducing the risk of blade-contact accidents. Because the economic findings establish that the standard is in the public interest even absent reliance upon the costs of pain and suffering, we need not determine whether the Commission must assign a dollar value to these aspects of the cost of accidents when precise calculation of their cost is not crucial for determining whether a standard is reasonably necessary and in the public interest. The methodology of the standard’s supporting findings as to the costs and benefits of its requirements is sustained.

We also find that the evidence in the 1979 final economic report supporting the standard’s cost effectiveness is reliable. In this report the Commission ultimately found that the standard would cost $189 million annually to implement, would prevent $211 million in yearly injury costs when full compliance was achieved, and would save $26.4 million after one year, assuming that the costs of accidents remained constant. The Commission’s final economic analysis of the standard used data from fiscal year 1977 to compute an annual blade-contact injury cost of $253 million, exclusive of pain and suffering costs, caused by an estimated 77,-000 injuries.

OPEI asserts that the figures from the final economic report lack the indicia of reliability because they differ from estimates of annual injury costs in a previous economic report and because the final economic report was completed after the period for public comment had closed and therefore was not exposed to such public scrutiny as would bolster confidence in its accuracy. The petitioner thus seeks to bring the present case within the holding of Aqua Slide. There we decided that when the only record evidence pertaining to the economic impact of a safety standard was a Commission report that was never exposed to public comment and was allegedly unreliable, it did not provide substantial evidence that the standard was reasonably necessary and in the public interest. 569 F.2d at 842-43. See Note, The Consumer Product Safety Commission: In Search of a Regulatory Pattern, 12 Colum.J.L. & Soc.Prob., 393, 394 n.10 (1976) (public testing of Commission findings often impeded because Commission’s research frequently is not complete until immediately before the standard is promulgated).

As proof of the final economic report’s unreliability, OPEI points to alleged discrepancies between the economic data in the final report and in previous cost-benefit studies. OPEI claims that in a February 1977 economic analysis entitled “Preliminary Assessment of the Possible Economic Effects of the Draft Safety Standard for Power Lawn Mowers,” the annual injury costs of all power mower accidents was reported to be $93 million, exclusive of pain and suffering, while in the 1979 final economic report the yearly cost associated with walk-behind mower blade-contact injuries alone was $253 million. We have searched the 35,000-page record, guided by the index approved by all parties in the case, and have been unable to locate the alleged earli er inconsistent estimate of lawn mower injury costs. 48

In any event, our examination of the final economic report and the Stanford Research Institute (SRI) study 49 upon which it purports to place primary reliance for its cost estimates convinces us that the standard’s economic findings, derived from the final economic report, do have the requisite “indicia of reliability” and have not been impermissibly shielded from public scrutiny. We do not base our holding that the standard’s economic analysis is reliable on a weighing of diverse technical data, for such an evaluation is beyond the function and capacity of this or any other court. See Aqua Slide, 569 F.2d at 838. Rather, our decision is founded on the judgment that the “Commission ‘carried out [its] essentially legislative task in a manner reasonable under the state of the record before [it].’ ” Id. (quoting Florida Peach Growers Association v. United States Department of Labor, 489 F.2d 120, 129 (5th Cir. 1974)).

In making its economic findings, the Commission estimated that there were approximately 77,000 blade-contact power mower injuries each year. It obtained this figure from 1977 National Electronic Injury Surveillance System data, which is public information. The Commission’s analysis of the NEISS data, establishing its understanding of the number of annual injuries caused by blade contact, was placed on the record and made available to public critique in a preliminary staff report that was eventually incorporated into the final economic report. See W. D. Barr, CPSC Memorandum, “Estimates of Power Blade Contact Injuries & Proposed Blade Contact Effectiveness” (Oct. 6, 1978). Thus, the petitioners and the public had an opportunity to present their views on the threshold finding of the number of annual injuries addressed by the standard, and this finding therefore enjoys the presumption of reliability afforded by public examination. Aqua Slide, 569 F.2d at 842; see BASF Wyandotte v. Cos-tie, 598 F.2d 637, 642 (1st Cir. 1979); South Terminal Corp. v. EPA, 504 F.2d 646, 658-59 (1st Cir. 1974).

In addition, the Commission explained in its final economic report that it had modified its earlier calculations of the yearly costs attributable to blade-contact injuries because it agreed with criticism of its previous injury-costing approaches presented in the Stanford Research Institute (SRI) report. SRI had observed that the injury-costing methods suggested by the National Safety Council and the National Highway Traffic Safety Administration and evidently used initially by the CPSC understated injury costs by failing to account for such factors as the victims’ ages, the rate of hospitalization from lawn mower injuries, and the severity of the injuries. Accordingly, the Commission raised its estimates of injury costs to accommodate these considerations. That the Commission learned from the comments on its proposals and adjusted its conclusions accordingly increases, rather than undermines, our confidence in the reliability of the economic analysis justifying the standard. The changes made by the Commission in response to submissions from the public do “not automatically generate a new opportunity for comment.” International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 632 (D.C.Cir.1973). “A contrary rule would lead to the absurdity that in rule-making ... the agency can learn from the comments on its proposals only at the peril of starting a new procedural round of commentary.” Id. at 632 n.51. See also South Terminal Corp. v. EPA, 504 F.2d at 659.

The Commission also explained in the final economic report that it had revised upward earlier projections of the costs of injuries and the savings offered by the standard’s safety measures because consumer prices for medical care had risen 8.3 percent from April 1977 to April 1978. It therefore adjusted SRI’s injury cost figures to accommodate this price increase. Thus, the Commission not only carefully explained the rationale for its evaluation of the standard’s costs and benefits and submitted much of the material used in the final economic report to public scrutiny in its preliminary form, but it also significantly based its final assessment of the cost effectiveness of the safety requirements on data from interested parties, most notably OPEI. 50 These attributes of the standard’s cost-benefit analysis cloak it in a sufficient mantle of reliability for us to find that substantial record evidence supports the Commission’s finding that the rule is in the public interest.

Effective Date

Finally, petitioners OPEI and Hayward challenge the standard’s December 31, 1981, effective date. OPEI argues that this date should be postponed because it is not feasible in light of an alleged current lack of reliability in brake-clutch devices. Hayward, in contrast, asserts that the standard should become binding on December 31, 1980, on the ground that brake-clutch mechanisms have already been proven reliable, and the availability of the necessary technology makes earlier implementation reasonable.

The Act provides that normally a new standard shall take effect 180 days after it is issued by the CPSC. 15 U.S.C. § 2058(d)(1). However, “for good cause shown,” the Commission may exercise its broad discretion to balance consumer needs and manufacturer costs and capabilities in establishing a standard’s effective date. In postponing implementation of a standard, the Commission must publish its reasons for finding that delay is in the public interest. Id. See ASG Industries, Inc. v. Consumer Product Safety Commission, 593 F.2d at 1335; Pacific Legal Foundation v. Department of Transportation, 593 F.2d at 1342-48.

The Commission fully articulated its reasons for affording manufacturers a 34-month lead time to comply with the standard. 44 Fed.Reg. 9998-10000. It found that a 21-30 month preparation period was necessary in order to prevent major bottlenecks in the production of safety-related components for the lawn mower industry and to allow smaller firms, lacking engineering sophistication, time to obtain from their suppliers safety-related components and engineering advice as to how to incorporate the required devices into their designs. It also found that the delayed implementation date it selected would reduce the retail prices of the safety features and that a greater variety of mowers would be available to consumers if the longer lead time were established. In rejecting pressures for a still later date, the Commission determined that the power mower industry could make the necessary changes in 31 months so that the December 31, 1981, date afforded leeway to enable manufacturers to overcome any unexpected difficulties that they might encounter. The Commission’s findings were well founded in record evidence. Therefore, in deciding that the standard should become effective on December 31, 1981, the Commission appropriately balanced the need for the standard’s accident-reducing benefits against the potential adverse economic impact on manufacturers and consumers that a shorter effective date would impose. Its judgment in this matter must be respected.

In summary, the standard’s scope and its requirements that mowers pass a rear foot-probe test, shield-strength and obstruction tests, satisfy a three-second blade-stop criterion, and carry a prescribed warning label are upheld. The Commission’s conclusion that the standard is reasonably necessary and in the public interest is also valid. The standard’s requirement that mowers pass a discharge chute foot-probe test is not justified by substantial evidence on the record as a whole and is therefore vacated.

AFFIRMED IN PART, VACATED IN PART.

1 . 44 Fed. Reg. 9990, 10030 (1979). The estimates are based on 1977 data.

2 . Id.

3 . The gestation period for the safety standard was long and complex. The administrative process was initiated on August 15, 1973, when, pursuant to § 10 of the CPSA, 15 U.S.C. § 2059, the Outdoor Power Equipment Institute, Inc. (OPEI) petitioned the CPSC to begin a proceeding to develop a consumer product safety standard addressing the hazards of power lawn mowers and asked the Commission to adopt a voluntary standard, ANSI B71.1-1972, “Safety Specifications for Power Lawn Mowers, Lawn & Garden Tractors, & Lawn Tractors,” approved by the American National Standards Institute, Inc. as the proposed consumer product safety standard. On November 16, 1973, the Commission, after considering information about injuries associated with power lawn mowers, granted that portion of OPEI’s petition that requested a proceeding to develop the power lawn mower safety standard. The Commission denied OPEI’s request to publish ANSI B71.1-1972, with amendments, as a proposed consumer product safety standard, however.

Instead, the Commission solicited offers to develop a standard pursuant to § 7(b) of the CPSA, 15 U.S.C. § 2056(b). Subsequently, the Commission selected Consumers Union of United States, Inc. (CU) to develop the safety standard. See 39 Fed. Reg. 37803 (1974). As the offeror, CU gave representatives of industry, consumers, and other interests the opportunity to participate in developing the standard. It submitted the resulting proposal to the Commission on July 17, 1975. The recommended standard comprehensively addressed all types of lawn mowers and lawn mower injuries and contained requirements relating to blade-contact and thrown-object injuries, as well as injuries resulting from lawn mowers’ slipping, rolling, overturning, or failing to steer or brake, injuries caused by burns from direct contact with exposed heated surfaces of mowers or from fires ignited by lawn mower ignition fluids, and injuries caused by electric shock from electrically powered lawn mowers or electric ignition systems.

After analyzing the recommended CU standard, on May 5, 1977, the Commission published a proposed comprehensive power lawn mower safety standard for public comment. 42 Fed. Reg. 23052 (1977). The proposal elicited more than 100 initial comments, and the Commission solicited and received further comments on these already submitted comments. 42 Fed. Reg. 34892 (1977). On June 7, 1978, the Commission published a notice that it would issue requirements addressing injuries from blade contact with walk-behind power mowers before issuing separate standards dealing with injuries associated with thrown objects, fuel and electrical hazards, and riding mowers. 43 Fed. Reg. 24697 (1978). In November 1978, the Commission requested additional comments on the safety and reliability of brake-clutch mechanisms. 43 Fed. Reg. 51638 (1978). On February 26, 1979, Part 1205-Safety Standard for Walk-Behind Power Lawn Mowers, applying only to blade-contact injuries from walk-behind power lawn mowers, was issued, to become effective December 31, 1981.

4 . 15 U.S.C. § 2056(a) (1976) provides:

(a) The Commission may by rule, in accordance with this section and section 2058 of .this title, promulgate consumer product safety standards. A consumer product safety standard shall consist of one or more of any of the following types of requirements:

(1) Requirements as to performance, composition, contents, design, construction, finish, or packaging of a consumer product.

(2) Requirements that a consumer product be marked with or accompanied by clear and adequate warnings or instructions, or requirements respecting the form of warnings or instructions.

Any requirement of such a standard shall be reasonably necessary to prevent or reduce an unreasonable risk of injury associated with such product. The requirements of such a standard (other than requirements relating to labeling, warnings, or instructions) shall, whenever feasible, be expressed in terms of performance requirements.

5 . For convenience, we shall refer to the lawn mower industry petitioners collectively as OPEI.

6 . Review of a safety standard by this court is authorized by § 11 of the CPSA, 15 U.S.C. § 2060 (1976).

7 . In authorizing judicial review of consumer product safety standards, the Act provides that “[t]he consumer product safety rule shall not be affirmed unless the Commission’s findings under section 2058(c) of this title are supported by substantial evidence on the record taken as a whole.” 15 U.S.C. § 2060(c).

8 . 15 U.S.C. § 2058(c)(2)(A) directs that:

The Commission shall not promulgate a consumer product safety rule unless it finds (and includes such finding in the rule)—

(A) that the rule (including its effective date) is reasonably necessary to eliminate or reduce an unreasonable risk of injury associated with such product. .

Every part of the standard must meet the “reasonably necessary” criterion. Aqua Slide 'N’ Dive Corp. v. Consumer Product Safety Commission, 569 F.2d 831, 838 (5th Cir. 1978) (hereinafter Aqua Slide ).

9 . 15 U.S.C. § 2052(a)(1) provides in pertinent part:

(a) For purposes of this chapter:

(1) The term “consumer product” means any article, or component part thereof, produced or distributed (i) for sale to a consumer for use in or around a permanent or temporary household or residence, a school, in recreation, or otherwise, or (ii) for the personal use, consumption or enjoyment of a consumer in or around a permanent or temporary household or residence, a school, in recreation, or otherwise; but such term does not include—

(A) any article which is not customarily produced or distributed for sale to, or use or consumption by, or enjoyment of, a consumer

15 U.S.C. § 2058(a)(1) provides:

(a)(1) Within 60 days after the publication under section 2056(c), (e)(1), or (f) of this title or section 2057 of this title of a proposed consumer product safety rule respecting a risk of injury associated with a consumer product, the Commission shall—

(A) promulgate a consumer product safety rule respecting the risk of injury associated with such product if it makes the findings required under subsection (c) of this section, or

(B) withdraw by rule the applicable notice of proceeding if it determines that such rule is not (i) reasonably necessary to eliminate or reduce an unreasonable risk of injury associated with the product, or (ii) in the public interest;

except that the Commission may extend such 60-day period for good cause shown (if it publishes its reasons therefor in the Federal Register).

10 . As noted, supra at 2, the standard also exempts reel-type mowers from the performance requirements but subjects them to the labeling provision of § 1205.1(c)(2). The Commission excused reel-type mowers from the performance requirements because it found that

the economic and injury data presently available to it do not show that the risk of injury associated with reel-type mowers justifies applying the standard to reel-type mowers. Reel-type mowers constitute less than 1% of the walk-behind mower market, and thus consumer exposure to that product is not as great. In addition, the injury data associated with reel-type mowers that are available to the Commission indicate that injuries with these mowers comprise less than 1% of the injuries associated with walk-behind mowers (NEISS Power Mower Baseline Study, July-September, 1977).

44 Fed. Reg. 9990, 9997.

11 . In explaining the maximum blade width, horsepower, and weight exemption, the Commission indicated that it believed that these larger mowers fell

generally within the definition of the term “consumer product.” However, they were not intended by the Commission to be included within the scope of this standard because of their relatively infrequent use by consumers . . [and because t]he requirements of the standard may affect these larger mowers, and their manufacturers and users, differently

and more harshly, since the larger mowers were mostly produced by smaller firms in limited numbers, thus concentrating the cost of compliance over fewer units. 44 Fed. Reg. 9990, 9998 (1979).

A CPSC staff memorandum, “Market Divisions in Walk-Behind Mowers” (Sept. 21, 1978), contains a trade magazine’s review of walk-behind power mower models that identified the typical mower produced for household use as having 18-22 inch cutting width, 3-4 horsepower engine, and weight of 75-100 pounds. The Commission’s study of 97 injury reports in which mower cutting width was specified revealed one injury from a 25-inch mower and another injury from a 26-inch mower but none from mowers over 30 inches wide. To be a “consumer product” under the Act,

a product must be customarily — not just occasionally — produced or distributed for the use of consumers. Jurisdiction does not require a showing that a majority of product sales are to consumers, but there must be a significant marketing of the product as a distinct article of commerce for sale to consumers or for the use of consumers before the product may be considered as “customarily” produced or distributed in that manner.

ASG Industries, Inc. v. Consumer Product Safety Commission, 593 F.2d 1323, 1328 (D.C.Cir.), cert. denied, 444 U.S. 864, 100 S.Ct. 133, 62 L.Ed.2d 87 (1979). The exemption for larger mowers therefore may overlap with the standard’s general limitation of coverage to “consumer products.” However, the large-mower exclusion does not expand the standard’s statement of basic coverage and thus does not support OPEI’s contention that the regulations apply to nonconsumer products.

12 . OPEI does not contend that these “unique design” machines are not consumer products but rather argues that they are consumer products for which there is no substantial evidence that they present an unreasonable risk of injury-

13 . This section provides in relevant part that:

(1) Prior to promulgating a consumer product safety rule, the Commission shall consider, and shall make appropriate findings for inclusion in such rule with respect to—

(B) the approximate number of consumer products, or types or classes thereof, subject to such rule

14 . In ASG Industries, unlike the present case, manufacturers of wired glass who challenged inclusion of their product in a CPSC safety standard regulating architectural glazing mate rials, 16 C.F.R. § 1201 (1978), introduced extensive evidence that “the functional and risk characteristics of wired glass are dissimilar from those of other glazing materials covered by the safety standard.” 593 F.2d at 1330. The safety standard at issue in ASG Industries prescribed impact performance requirements for glazing materials used in certain architectural products, including various types of doors. The manufacturers established by record evidence that when wired glass fractures it produces shards with right rather than acute angles and therefore presents a lesser risk of injury. They also explained that the wire embedded in the glass increases the product’s visibility, thus reducing the risk of inadvertent collision, and that wired glass has unique functional advantages when used in fire doors and other fire-retardent barriers because it is the only transparent construction material that does not shatter, ignite, or produce dangerous fumes when exposed to the extreme temperatures produced by a major fire. The court ruled that the enhanced visibility of wired glass, lessening the danger of accidental collision, represented a significantly different risk characteristic compared to other glazing materials and that its advantages in fire barriers represented a significantly different functional characteristic compared to other products covered by the standard. It therefore decided that it was necessary for the Commission to justify its inclusion in the standard, id. at 1330-33, and remanded the standard to permit the Commission to determine whether the differential risk and functional characteristics of wired glass were relevant to the standard’s objectives and, if so, whether application of the standard to wired glass was nevertheless reasonably necessary to prevent or reduce an unreasonable risk of harm from the product. The evidentiary showing triggering the Commission’s duty to justify inclusion of a significantly different product is entirely lacking in the present case, See text, infra at 507.

15 . In Bunny Bear manufacturers of crib mattresses sought exclusion of their product from a flammability standard issued by the Secretary of Commerce under the Flammable Fabrics Act, 15 U.S.C. §§ 1191-1202. Each flammability standard must be

based on findings that such standard . is needed to adequately protect the public against unreasonable risk of the occurrence of fire leading to death, injury, or significant property damage, is reasonable, technologically practicable, and appropriate, [and] is limited to products which have been determined to present such unreasonable risks

15 U.S.C. § 1193(b) (emphasis added). The challenged standard required mattresses, including crib mattresses, to pass a cigarette test in which a burning cigarette was placed on the mattress to determine whether it would ignite. 473 F.2d at 1004. The crib mattress manufacturers argued that their products should not be covered by the flammability standard because “infants and young children ... do not smoke,” and therefore their bedding was not subject to the same risk of fire as were adult mattresses. The court applied a standard of review that focused on the “reasonableness of the conclusions drawn from the evidence before the Secretary,” which differed “little, if at all, from the substantial evidence review” applicable to CPSA cases. Id. at 1006. It ruled that the record supported the inclusion of crib mattresses within the standard adopted for all mattresses. The First Circuit reasoned that the Secretary was entitled to infer that crib mattresses presented the same risk of fire as adult mattresses because parents who smoked might accidentally drop a lighted cigarette on the crib mattress while tending a child. The court specifically noted that “[w]hether the test is ‘arbitrary, capricious’ or ‘substantial evidence,’ agencies are not precluded from drawing reasonable inferences from facts commonly known.” Id. at 1007. Inclusion of crib mattresses was proper, the court held, in part because:

[T]he Secretary could take account of the crib manufacturers’ failure to present any evidence warranting exclusion. The burden to justify inclusion of a product within a general safety standard is initially upon the Secretary. But at some point . . the burden shifts to those seeking exclusion to come forward with their own evidence. If they do so, the Secretary may then have to develop further evidence in response — or otherwise risk our determination that inclusion is unsupported. The Secretary’s initial burden, however, is not limitless — especially when dealing with inclusion of a product falling within a common product-line, as to the latter of which there is extensive evidence supporting application of the standard. If from what is before him, he plausibly opts for inclusion of a particular product, it is not unreasonable to require affected manufacturers to point out with particularity those features which make special treatment necessary.

Id.

16 . Both OPEI and the Commission agree that the evidence establishes that the great majority of operator foot injuries occur at the rear of the mower.

17 . A deadman control refers to a device on the mower handle that requires continuous pressure to sustain rotation of the mower blades. Only if the operator can reach the discharge chute with his foot while holding the mower handle can he suffer blade-contact injuries at the discharge chute by mowers meeting the blade-stop requirement. Discharge-chute foot injuries suffered after the operator has released the mower handle and deadman’s controls are addressed in the standard’s blade-stop provision.

18 . The rear 120 degrees of the mower and the discharge chute were selected for the foot-probe test because the Commission found that they are “areas where foot contact injuries are known to occur while the operator is holding the [mower] handle. Foot contact injuries that occur while the operator is not holding the handle will be addressed by the blade control . . . .” 44 Fed. Reg. 9993.

19 . See n.7, supra.

20 . The Federal Hazardous Substances Act provides in part that “[a]n article may be determined to present a mechanical hazard if, in normal use or when subjected to reasonably foreseeable damage or abuse, its design or manufacture presents an unreasonable risk of personal injury or illness . . 15 U.S.C. § 1261(s) (1976) (emphasis added).

21 . The House Committee Report on the CPSA stressed that:

[T]he Commission’s authority to promulgate standards under this bill is limited to instances where the hazard associated with a consumer product presents an unreasonable risk of death, injury, or serious or frequent illness.

[I]t is generally expected that the determination of unreasonable hazard will involve the Commission in balancing the probability that risk will result in harm and the gravity of such harm against the effect on the product’s utility, cost and availability to the consumer. An unreasonable hazard is clearly one which can be prevented or reduced without affecting the product’s utility, cost or availability; or one which the effect on the product’s utility, cost or availability is outweighed by the need to protect the public from the hazard associated with the product.

22 . For example, in Aqua Slide this court ruled that the severity of paraplegic injury from swimming pool slides was so great that a one in ten million risk of such injury, which is less than the risk that an average person will be killed by lightning, would be an “unreasonable risk” if the proposed safety standard “actually promised to reduce the risk without unduly hampering the availability of the slides or decreasing their utility . 569 F.2d at 840. The court found, however, that the necessary showing of the standard’s effectiveness had not been made and that the burdens the regulation would impose had not been adequately evaluated. Id. at 840-44.

23 . Cf. Aqua Slide, 569 F.2d at 840 (remote risk of paraplegia could be unreasonable under CPSA); Bunny Bear, Inc. v. Peterson, 473 F.2d at 1006-07 (harm of burns or smoke inhalation, including death, from crib fires was sufficiently great to justify requirement that crib mattress pass a “cigarette test” by not igniting when touched with cigarette, when possibility existed that adult attending child might drop lighted cigarette on mattress).

24 . NEISS collects data from selected hospitals and reports them to the CPSC. The system, which has been operational since July 1, 1972, was designed to develop statistically valid, rationally representative product-related injury data. It employs a computer-based network of 119 statistically selected hospital emergency rooms located throughout the country. 15 U.S.C. §§ 2054, 2055. See “Draft Hazard Analysis of Power Mower Related Injuries & Analysis of Proposed Power Mower Standard,” U.S. CPSC, Bureau of Epidemiology (March 1977), at pp. 3-5. The Commission also obtains injury information from in-depth investigations (IDI’s) of particular accidents reported through the NEISS network. IDEs conducted by the Commission, unlike basic NEISS accident data, are not statistically representative of all injuries in a particular product category. IDEs do, however, provide details concerning the sequence of events involved in the injury not available from NEISS surveillance information. Id. at 5. To overcome this deficiency, the Commission weighted IDI cases involving specified types of injuries to derive an adjusted IDI sample that conforms to NEISS data.

25 . This provision mandates that any shield in the areas to be foot probed shall not permanently separate, crack, or deform when subjected for 10 seconds to a 50-pound static tensile force uniformly distributed over not less than half the shield’s length.

26 . The obstruction test consists of moving the mower back and forth at a set speed over a test surface that has specified depressions and raised obstacles. The mower may not stop as a result of contacting a raised obstacle nor may more than one wheel at a time lift from the surface. In addition, the shield may not enter the path of the blade.

27 . OPEI asserts that its voluntary standard requires mowers to pass a shield strength test to assure their structural strength and that therefore § 1205.4(a)(2) is not reasonably necessary to reduce or eliminate an unreasonable risk of injury. The mere existence of a similar voluntary standard does not undermine the Commission’s authority to make the shield strength test mandatory. At present, shielding is not required on all mowers. The Commission may, by means of the safety standard, seek to increase the likelihood that when the new shielding requirements become effective and all mowers are equipped with shielding, there will be uniform compliance with tests intended to assure the shield’s structural integrity. See 15 U.S.C. § 2051(b)(3) (“The purposes of this chapter are ... to develop uniform safety standards for consumer products” (emphasis added); Forester v. Consumer Product Safety Commission, 559 F.2d at 793 (assuming that existence of similar voluntary standard does not prevent Commission from making it part of mandatory safety rule).

28 . The record contains an in-depth CPSC investigation of 102 blade-contact injuries that revealed only two incidents in which the operator’s foot was hurt by the blade at the rear of a mower that had a broken or missing shield. It also contains a few letters from consumers complaining of shields that did not remain attached to their'mowers.

29 . In D. D. Bean the First Circuit upheld a safety standard’s design requirement for matchbooks, including a “reverse friction” requirement by which the friction element was to be located on the back of the matchbook. The Commission’s computerized injury surveillance system and in-depth studies had established that a significant number of people were injured annually by matchbooks that accidentally ignited when a match struck a friction strip located on the front cover. 574 F.2d at 649. The record contained “no detailed empirical data to support the conclusion that reverse friction was safer,” but the court nevertheless held that “the Commission might rationally so conclude since ignition in the case of an open matchbook is obviously made more difficult when the friction is located on the reverse side.” Id. It also ruled that the design requirements, notably that “the cover remained closed and that no stray friction material be found anywhere in a matchbook,” were valid, even absent empirical data that they were likely to reduce the risk of injury, because these provisions were “logically likely to reduce the risk.” Id.

In contrast, the court overturned performance requirements that were intended to guard against delayed ignition, split splints, and afterglow or reappearance of a flame. Id. at 650. It found no evidence that these conditions either existed “to any appreciable extent among currently manufactured matches” or caused “any very significant number of injuries.” Id. In discussing the dearth of evidentiary support for the performance requirements, the court noted that the American Safety for Testing & Materials comment had stated that:

There was no factual evidence presented to indicate that [reappearance of a visible flame] had caused any injuries and no mention was made of the defect in the risk analysis [of matchbook hazards]. However, common sense seems to indicate that a hazard might exist if a flame reappears after a visible flame had apparently been extinguished.

Id. (emphasis added).

The court observed that the “common sense” conclusion that if a condition existed it might present a hazard was “not evidence that the reappearance of a visible flame is a condition that exists with any appreciable number of matches, thus requiring correction” and did not establish the existence of an unreasonable risk of injury that is a predicate of Commission regulation. Id. at 650-51.

The present case differs significantly from this aspect of the D. D. Bean situation in that the shield-strength requirement, unlike the performance requirements in D. D. Bean, is an ancillary provision intended to assure the efficacy of an admittedly reasonable remedy for an acknowledged safety hazard. As is clear from the portions of the D. D. Bean opinion quoted in text, supra, D. D. Bean does not .stand for the proposition that the Commission must produce empirical evidence for every step of its reasoning and regulatory process, even if a remedy for a hazard is “patent.”

30 . The experience of massive consumer resistance to ignition lock-seat belt systems is instructive as to the possibility that consumers will defeat safety devices they find inconvenient to use despite the loss of safety benefits such avoidance entails. See Pacific Legal Foundation v. Department of Transportation, 593 F.2d 1338, 1340-42 (D.C.Cir.1979); “Safety Belt Interlock System Usage Survey” (August 1976) (reporting high percentage of drivers defeated seat belt system).

31 . For example, when the operator releases the deadman’s control and approaches the blade area to clear the path of the mower, unclog a blocked discharge chute, adjust the wheels, or empty a grass catcher, the blade will stop rotating within three seconds. According to the Commission, by the time the operator reached the danger zone the blade would have stopped, or significantly slowed, thus preventing or minimizing injury.

32 . While the statute favors performance requirements when feasible, it does not prohibit design standards when they are appropriate. 15 U.S.C. § 2056(a)(1) provides in part that:

The Commission may by rule, in accordance with this section and section 2058 of this title, promulgate consumer product safety standards. A consumer product safety standard shall consist of one or more of any of the following types of requirements:

(1) Requirements as to performance, composition, contents, design, construction, finish, or packaging of a consumer product,

(emphasis added).

33 . As their names imply, safety mechanisms based on “blade inaccessibility” are designed to prevent contact with the moving blade by making it virtually impossible for a consumer to reach the blade with hands or feet by removing all controls from the blade area and eliminating the discharge chute. Devices based on “blade harmlessness” would employ grass-cutting means other than the standard rigid or semi-rigid blades that would not inflict serious injury if touched while in operation. It should be noted that the present standard already exempts such innovative, blade-harmless mowers as those using monofilament “blades” from the standard’s safety requirements. 16 C.F.R. § 1205.3(a)(1); 44 Fed.Reg. at 9993.

34 . For example, the Commission rejected a suggestion in the Consumer’s Union recommended standard that the rule include a “blade harmlessness” test as an alternative to a blade-stop provision. CU had proposed that blade harmlessness be measured by inserting a vinyl-wrapped dowel into the revolving blade’s path. If the vinyl tape were uncut, the risk of injury would be considered acceptable. The Commission dropped CU’s blade-harmlessness alternative from the proposed standard that it published because the Commission had no evidence that a vinyl-wrapped dowel was a reasonable facsimile for a human hand.

Another commentator urged that blade harmlessness be tested by replacing the vinyl coating on the CU probe with surgical tubing or by using chicken legs as the probe. Here, too, no evidence indicated that surgical tubing or chicken legs were reasonable approximations of human body parts and would satisfactorily demonstrate the potential for injury to the mower operator. OPEI suggested that blade harmlessness probes be covered with relatively soft material that, if cut, would establish an unacceptable risk of injury from the moving blade. These recommendations, too, were rejected by the Commission experts as similarly inadequate for purposes of estimating blade-contact injury to the human hand on the grounds that the proposed probe coverings did not successfully simulate human skin or register when such nonlaceration injuries as fractures and avulsions would occur.

35 . OPEI recommended using a hand-probe test for blade inaccessibility that involved attempting to insert two cupped hands under the lower edge of the mower housing. The Commission rejected this test because no evidence had been presented to show that the shape of the probes adequately simulated the range of dimensions and motions of the human hand. Although the cupped probe might not slip under the mower, a more extended hand might do so. The Commission concluded that, while blade inaccessibility could be a desirable approach to explore, extensive research into the types of hand movements users would employ and into the dimensions and maneuverability of probes nec essary to simulate these movements was necessary before a viable regulation based on that concept could be developed.

36 . The Commission estimated that the development of the alternative safety requirements would take approximately five Commission staff years and cost $500,000 in research funds.

37 . We do reject, however, OPEI’s claim that the CPSC has a broad obligation to research and develop all possible safety measures. The Commission may, within reason, exercise its discretion in allocating agency resources to various projects. Interested persons who would seek the development of alternative safety proposals may, of course, attempt to persuade the Commission to explore their suggestions. But failing to engage the Commission in such an enterprise, they remain free to develop their suggestions themselves and to introduce them into the record as evidence that the Commission must at least fully consider in promulgating its safety standards in order for the standard to be supported by substantial evidence. Thus, in the present case, the lawn mower industry sought to meet the CPSC’s objections to its proposals of alternative requirements by financing and partly staffing a major research program. The manufacturers’ efforts were designed to investigate known patterns of hand-contact injuries, to devise comprehensive probing requirements, utilizing an anthropomorphically sound hand probe based on those injury studies, and to compare the effectiveness of the resulting performance criteria with the success of the agency’s blade-stop requirement. Technical development of this program was completed by June 11, 1979; by August 24, 1979, the industry had finished studies comparing the efficacy of its proposed requirements and the standard’s blade-stop rule. As a result of this work, on September 12, 1979, OPE1 filed a petition to amend the safety standard for walk-behind power lawn mowers pursuant to 15 U.S.C. § 2059. This industry development of additional evidence in order to expand the means of satisfying an otherwise reasonable safety standard represents a proper allocation of responsibility for the full development of safety standards.

38 . In eight trials, as analyzed in 1975, operators reached the blade in the following times (in seconds): 2, 3, 3, 2.5, 3.5, 2, 2.5, 3. (An earlier analysis performed by the NBS in 1974 and revised by the 1975 analysis had found that the access time ranged from 2.5 to 7.0 seconds. More accurate measurement methods used in the later study appear to account for the discrepancy.) It was recognized that the Iowa study was of limited value, however, because it employed machines that did not conform to the industry’s voluntary standard in that they had shorter discharge chutes than the ANSI code now permits for current models. As a result, these blade-access times may have been shorter than access to the blades of current mower models. It should be noted, however, that figures derived from the other studies are in reasonable agreement with those of the University of Iowa test and that the test was not without some evidentiary value. .

39 . The data table attached to the summary shows that the longest blade-access time was 5.65 seconds, not 4.90 seconds. The discrepancy is not explained.

40 . Blade-access times for the direct, casual approach test were: for subject A, 2.89 and 2.75 seconds; for subject B, 3.14 and 2.95 seconds; for subject C, 2.43 seconds; for subject D, 2.60 and 2.35 seconds; and for subject E, 2.90 and 3.30 seconds.

41 . Blade-access times for the direct, intentionally fast approach test were: for subject A, 1.26 and 1.59 seconds; for subject B, 2.21 and 2.20 seconds; for subject C, 1.43 and 1.35 seconds; for subject D, 1.80 and 1.49 seconds; and for subject E, 2.52 and 2.05 seconds.

42 . In this test, subject A had access times of 3.64 and 3.88 seconds; subject B, 4.52 and 4.18 seconds; subject C, 4.39 seconds; subject D, no recorded time; and subject E, 4.42 and 4.70 seconds.

43 . Here access times were 2.60 and 2.48 for subject A; none recorded for subject B; 3.45 and 3.02 for subject C; 2.63 for subject D; and 3.68 for subject E.

44 . A report by the Stanford Research Institute (SRI), “An Economic Analysis of the Proposed Consumers Union Power Mower Safety Standard” (April 1975), stated that the projected cost increases for walk-behind power mowers attributable to the blade-stop requirement would be $90-115 if the stop time were 3.0^.9 seconds but only $75-100 per mower if the stop time were 5.0-6.0 seconds. In another 1977 study, SRI estimated that a 3-second blade-stop requirement would cost $16.50 more per mower than a 5-second test for manual start rotary mowers and $10.50 more per mower for electric start mowers. The Consumers Union Economic Support Study estimated roughly that the marginal per mower cost of a mechanism would be $10, while a 5-second blade-stop requirement would permit a $2 per mower savings over a 3-second blade clutch. OPEI submitted comparative incremental retail costs in September 1977 that stated that, for gas-powered rotary mowers with electric start mechanisms, a 7-second stop requirement would cost $5 per mower without “easy restart,” and a 3-second brake-clutch device would cost $13.50 per mower. For gas-powered rotary mowers with manual starts, OPEI’s cost estimate was $10 for a mower meeting a 7-second stop requirement, $15 for a mower having 5-second engine kill with easy restart, and $34.50 for a 3-second brake-clutch system.

45 . See n.4, supra.

46 . Section 27(e) of the Act, 15 U.S.C. § 2076(e), provides:

The Commission may by rule require any manufacturer of consumer products to provide to the Commission such performance and technical data related to performance and safety as may be required to carry out the purposes of this chapter, and to give such notification of such performance and technical data at the time of original purchase to prospective purchasers and to the first purchaser of such product for purposes other than resale, as it determines necessary to carry out the purposes of this chapter.

47 . Performance and technical data that manufacturers may be ordered to furnish consumers under section 27(e) include information pertaining to the proper engine-blade rotation speed of the mower.

Our holding that the warning label does not communicate performance and technical data is not based on the fact that the warning uses symbols as well as words but rather is grounded in the type of information conveyed. Cf. regulations at 49 C.F.R. § 575.1 (1978) (implementing similar performance and technical data provisions of National Traffic & Motor Vehicle Safety Act, requiring manufacturer to disclose, e. g., vehicle-stopping distance and tire-quality grading standards).

48 . The citation to the record in OPEI’s briefs directed us to Exhibit 1645, which OPEI indicated was a report entitled “Preliminary Assessment of the Possible Economic Effects of the Draft Safety Standard for Power Lawn Mowers & Preliminary Report” (February 1977). Exhibit 1645 does not contain such a report, nor could we locate it elsewhere in the record. Exhibit 1646 did contain all but the first two pages of an economic report entitled “Draft Hazard Analysis of Power Mower Related Injuries and Analysis of Proposed Power Mower Standard” (March 1977). That report did not ascribe a $93 million figure to the cost of injuries. We also examined several other reports, including the December 30, 1976, Bat-telle report, “Research Report on Economic Impact Analysis of Proposed Safety Standard for Power Lawn Mowers to Bureau of Economic Analysis, Consumer Product Safety Commission,” and Consumers Union’s economic analysis of its recommended standard, “Economic Impact of Proposed Lawn Mower Safety Standard” (July 1975), without locating the inconsistent data.

49 . The SRI study was commissioned and submitted into the administrative record by OPEL

50 . See n.49, supra.