OSHRC Docket Nos. 3973; 4397; 4501; 4853; 5327; 7122; 7910; 7920

Occupational Safety and Health Review Commission

August 24, 1976


Before BARNAKO, Chairman; MORAN and CLEARY, Commissioners.


Altero D'Agostini, Regional Solicitor, USDOL

William Kloepfer, Assoc. Regional Solicitor, USDOL

Norman Winston, Assoc. Regional Solicitor USDOL

Charles M. Chadd and Thomas Nyhan, for the employer

A. E. Lawson, Asst. General Counsel and Francis P. Grimes, Staff Representative, Safety & Health Department, United Steelworkers of America, for the employees

Timothy M. Biddle, (FIA) & (AMI), Amicus Curiae

Raymond J. Durn, (FIA) & (AMI), Amicus Curiae

Jeffrey L. Gibbs and George H. Cohen, (USW), Amicus Curiae

George Miron, (ABMA), Amicus Curiae

Thomas A. Daly, National Soft Drink Association (NSDA), Amicus Curiae

Baruch A. Fellner, Counsel for Regional Litigation, USDOL




BARNAKO, Chairman:

These consolidated cases arose upon citations alleging violations of Complainant's (Labor) noise standard n1 at plants in which Respondent (Continental) manufactures metal cans. n2 The citations alleged that Continental violated the standard by failing to institute feasible engineering controls to reduce the noise levels in these plants below the limits specified in the standard. n3

- - - - - - - - - - - - - - [*2] - - - -Footnotes- - - - - - - - - - - - - - - - - -

n1 The standard, 29 C.F.R. 1910.95(b)(1) provides:

When employees are subjected to sound exceeding those listed in Table G-16, feasible administrative or engineering controls shall be utilized. If such controls fail to reduce sound levels within the levels of Table G-16, personal protective equipment shall be provided and used to reduce sound levels within the levels of the table.

Table G-16 -- Permissible Noise Exposure

Duration per day, hours

Sound level dBA slow response











    1 1/2






          1/4 or less


n2 The various docket numbers relate to plants in the following locations: 3973 (Los Angeles, California), 4397 (Pittsburg, California), 4501 (San Jose, California), 4853 (Pascagoula, Mississippi), 5327 (Worthington, Ohio), 7122 (Bedford Heights, Ohio), 7910 (Cincinnati, Ohio), 7920 (Columbus, Ohio). At the time of the hearing, only the three California cases had been consolidated, and the hearing concerned only them. It was thereafter stipulated that the record developed at the hearing would also govern the disposition of the other cases.

n3 It was stipulated that Continental's employees who worked in areas of excessive noise used personal protective equipment which effectively reduced their exposure to below the limits specified in Table G-16 of the standard. Also, Labor does not allege that administrative controls to reduce the noise levels are feasible. Accordingly, Continental was only in violation of the standard if it failed to implement feasible engineering controls.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The record established, and Continental concedes, that excessive noise levels exist in Continental's plants and that reduction of the noise could be achieved by engineering means. Labor contends that this is sufficient to show that Continental failed to implement feasible engineering controls, and therefore violated the standard. Continental argues that economic factors must be considered in determining whether engineering controls are feasible. Continental contends that if the standard is interpreted, as Labor suggests, to require engineering controls to be implemented regardless of their cost n4 and the benefits they produce, it is arbitrary and capricious, and therefore invalid. n5

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n4 Labor would allow economic consideration to enter into a determination of feasibility only if the cost of such controls would "seriously jeopardize the financial condition of the company." United States Dept, of Labor, Occupational Safety and Health Administration, Field Operations Manual, Sec. X-G6c2. Continental does not claim that the cost of the engineering controls involved in this case would jeopardize its financial health.

n5 Amicus curiae briefs in support of Continental's arguments were filed by the American Boiler Manufacturers Association, the Alliance of Metal Working Industries, the Forging Industry Association, and the National Soft Drink Association. An amicus brief in support of Labor was filed by the Industrial Union Department, AFL-CIO.

In addition to the substantive arguments raised, the issue of whether the Commission has authority to rule on the validity of standards has been extensively briefed. The Commission has held that it has such authority. Tobacco River Lumber Co., 17 OSAHRC 235, BNA 3 OSHC 1059, CCH OSHD para. 19,565 (1975); accord, Atlantic & Gulf Stevedores, Inc. v. OSHRC, 534 F.2d 541 (3rd Cir., March 26, 1976). In the latter case, the Court noted that such authority was a necessary and appropriate adjunct to the Commission's adjudicatory function under the Act.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Judge Robert N. Burchmore held that the standard, as interpreted by Labor, was invalid. He therefore vacated the citations. In another decision issued today, we have upheld the validity of the standard. We therefore reject the Judge's contrary holding. That holding, however, was premised on acceptance of Labor's interpretation of the standard. We do not agree with Labor's interpretation. See Brennan v. OSHRC (Ron M. Fiegen, Inc.) 513 F.2d 713,715 (8th Cir. 1975); Brennan v. Gilles & Cotting, Inc., 504 F.2d 1255, 1262 (4th Cir. 1974). We hold that the standard only requires the implementation of those engineering controls which are economically, as well as technically feasible. We find that the record fails to establish the economic feasibility of engineering controls. Accordingly, we adopt the Judge's order vacating the citations.

I. The Evidence

Continental's plants cited in this case use a number of machines to perform various functions in converting sheet metal into finished cans. Each machine generates a certain amount [*5] of noise, much of which results from metal striking metal during the various manufacturing steps. As a result, at most locations in the production areas of the plants, noise levels exceed 90 dBA. As employees work 8 hour or longer shifts, the limits of Table G-16 of 29 C.F.R. 1910.95 (hereinafter, the "G-16 limits") are correspondingly exceeded. n6 The record establishes that the noise levels at employee work stations in the vicinity of the cited machines were as follows: n7

Los Angeles Plant

Feeder end of double die presses on lines 4, 10,

99-100 dBA

and 11.

Stacker end of double die presses on lines 4, 10,

97-98 dBA

and 11.

Operator's station at scroll shear on line 13.

98-99 dBA

Pittsburg Plant

Feeder end of single die presses on lines 20 and

96-97 dBA


Operator's station at gang die press #5.

94-100 dBA

Maintainer's station at gang die press #5.

97 dBA

San Jose Plant

Operator's station between slitter and body-

96-100 dBA

maker on line #4.

Operator's station between slitter and body-

102 dBA

maker on line #12.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n6 In the general case, whether the G-16 limits are exceeded depends both on the noise level and the time of exposure. Since exposure times in this case are eight hours or longer, the required noise level to satisfy the standard is 90 dBA.

n7 The machines for which noise levels are listed represent those mentioned in the citations. These machines, however, represent only some of the total number of machines in the three plants. We take the levels as representative of noise levels in the other plants in view of the stipulation (note 2, supra).


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

A. Engineering Controls

The parties agreed that the noise levels could be reduced and that the proper engineering approach involves the installation of enclosures around the machines in order to absorb the noise before it reaches the employees.

Shortly after the standard became effective in 1971, Continental established a program aimed at developing engineering controls to reduce noise at its can manufacturing plants. There are 79 plants in Continental's Metal Division, eight of which are represented in this case. All plants in the division have similar machinery and similar noise problems, although not all types of machines are present in each plant. Early in its program, Continental determined that enclosures were the only realistic engineering method to achieve significant noise reductions. It therefore began to design, develop, and test prototype enclosures in order to determined the feasibility of this approach.

The plants in the Metal Division contain approximately 6600 separate machines. About 6300 of these were of 50 types, and were classified by Continental into 19 families of basically [*7] similar machines. Continental analyzed the noise generated by the various families, and then arranged them into a priority list, with the order determined by the severity of the noise problem in terms of noise levels and number of employees affected throughout the division. Continental decided to attack the noise problem by first developing enclosures for the highest priority machines and then progressing down the list.

Of the three highest priority families, on one includes the bodymakers cited at the San Jose plant. Another is punch presses, including the double die presses at the Los Angeles plant and the single die presses at the Pittsburg plant. All 19 families were represented at the three California plants, but the citations encompassed only five families.

Continental's director of engineering services, Mr. Prendergast, is in charge of the department conducting the noise reduction program. He testified regarding the progress the program had achieved to date with respect to the bodymaker and punch press families. For the bodymaker, it was decided that a total enclosure would be necessary. Such an enclosure was designed, fabricated, installed, and tested for several [*8] months in a can plant. The enclosure reduced the noise to an acceptable level, but created problems of access, visibility, and operator positioning. The enclosure was, at the time of the hearing, being redesigned to meet these difficulties.

A partial enclosure had been designed and tested on one of the double die presses. A noise level of 85 dBA was achieved when the press was first installed and operated in a room by itself. As time went on, however, the enclosure lost some of its effectiveness, and the noise level rose to 89 dBA. However, the ambient noise exceeded 90 dBA when the press was operated with other machines. It was therefore concluded that the partial enclosure was inadequate, and Continental decided to design a total enclosure for the double die presses. A total enclosure had been designed and tested for a similar machine in a different division with encouraging results.

Continental's experience with the partial enclosure on the double die press emphasized the nature of the problem of reducing noise levels to less than 90 dBA in all locations. The noise level at the operator's station of a particular machine results, not only from noise generated by that machine, [*9] but from noise generated by all machines in the immediate area. If the noise generated by a particular machine is reduced to less than 90 dBA, the ambient noise level can still exceed 90 dBA n8 when other machines are also operating. In general Continental found that noise levels in the vicinity of a particular machine are 5 dBA greater than the noise level generated by that machine alone. Reduction in noise levels to G-16 limits could therefore only be achieved by quieting all the machines in a plant and not simply those mentioned in the citations. Mr. Prendergast was of the opinion that it was technically feasible to reduce noise levels to less than 90 dBA in all 79 plants in the Metal Division if all 6300 machines were either totally or partially enclosed.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

N8 The decibel scale for measuring noise is logarithmic rather than arithmetical. Each 3 dBA increase in noise level actually represents a doubling of the air pressure at the listener's ear. Thus, if a person is equidistant from two machines, and the noise level from each machine alone would be 85 dBA, the resultant noise level would be 88 dBA. If the person was equidistant from four such machines, the resultant noise level would be 91 dBA.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

To the date of the hearing, Continental had spent approximately $400,000 on its enclosure development program for the Metal Division. It estimated that $1,230,000 would be required to complete the development and testing program for all 19 families. Fabrication and installation of enclosures around all 6300 machines was estimated to cost $32,000,000. The annual cost of maintaining the enclosures was estimated to be $175,000.

Completion of the development program for only the five families of machines cited at the California plants would cost approximately $132,000. Fabrication and instalation of enclosures around only those machines would cost $362,000. Annual maintenance costs would be $2,500. Continental's manager of training and safety for metal operations, Mr. Campbell, estimated that such a program would bring the exposure of 125 employees in the Los Angeles and Pittsburg plants within G-16 limits, but that in the San Jose plant, it was questionable whether G-16 limits would be achieved for any employees. n9

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - [*11] -

n9 Apparently, the principal difference between the plants is that, in Los Angeles and Pittsburg, the cited machines represent most of the total machines, while the same is not true in San Jose. Thus in San Jose, machines in families which were not cited produce a greater percentage of the total noise, and enclosing only the cited machines will not achieve as great a reduction in the noise levels as at the other plants.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

By way of contrast, Continental estimated the annual cost of its current personal hearing protection program to be $100,000 for the entire Metal Division. The annual cost of the hearing protection program for the 125 employees in the California plants whose exposure would be reduced to within G-16 limits by enclosure of the cited machines was estimated to be $650. These costs include replacement of lost and deteriorated ear protectors, and audiometric testing of persons newly hired or returning from layoffs.

In view of the costs involved in enclosing all 6300 machines, and considering that an alternative form of protection through the use of personal protective equipment was [*12] available at much lower cost, Mr. Prendergast and Mr. Campbell were of the opinion that reduction of the noise by engineering controls was not feasible.

Labor's expert witness, Dr. Salmon, conducted a noise survey of the machines listed in the citations. As a general proposition, he suggested that maximum noise reduction could be achieved by "as complete an enclosure as possible" consistent with the requirement to have access to the machines for production and maintenance purposes. When intermittent access is required, complete enclosures could be used and access provided by hinged panels, or by panels attached with readily removable fasteners. Where constant access is required, partial enclosures could be used. The interior of the enclosures should be treated with sound absorbing and damping materials, and baffles to absorb sound should be installed in appropriate locations. In order to obtain the maximum possible attenuation for any machine, particular attention would have to be paid to the fitting of the panels, to sound absorbing treatment at any opening remaining in the enclosures, and to the proper selection and use of baffles and sound absorbing materials.

Although the [*13] details of the enclosures would vary for each type of machine, Dr. Salmon was of the opinion that, for all machines, a reduction in noise of 5 to 15 dBA could be obtained. He thought that a 5 dBA reduction could be easily obtained by casual application of the techniques he suggested. In order to obtain a 10 dBA reduction, greater attention to these principles would be required. To obtain a further reduction, he stated that "you would really have to start to fight." The maximum 15 dBA reduction could only be obtained by optimizing every aspect of the enclosure design.

B. Personal Protective Equipment

As noted above, it was stipulated that Continental's employees were adequately protected from the excessive noise through the wearing of personal protective equipment. There was nevertheless considerable testimony regarding the utility of such equipment to protect against hearing loss. Most of this testimony was ultimately consolidated into the following two stipulations:

Personal ear protective devices, i.e., ear plugs and ear muffs, are available that will reduce noise levels below 90 dBA and will thereby effectively protect the employees' hearing from harmful noise levels [*14] of the magnitude shown by the evidence in this case provided that such devices are properly maintained, properly fitted, properly seated and worn. n10

There are employees who dislike earplugs and their dislike of the earplugs may be because they think they are uncomfortable or because of any other personal reason they may have, and such employees sometimes will refrain from wearing the earplugs or earmuffs, even though they are told to do so by their employer. And this creates a necessity in any hearing protection program for the employer to exercise supervision and his supervision may or may not be successful depending on how well it is conducted and how much attention is paid to it and how responsible the employees are.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n10 The record shows that noise attenuation of up to 27 dBA can be obtained with earplugs, and up to 35 dBA with earmuffs.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Continental presented an expert witness, De. Hildyard, who had extensive experience in implementing hearing conservation programs through the use of personal ear protectors. [*15] He was of the opinion that such equipment provided an effective means of protecting persons exposed to excessive noise from hearing loss. However, he testified to certain drawbacks inherent in this means of protection. In addition to the problem of some workers failing to wear the equipment for personal reasons, some individuals suffer hearing loss due to improperly fitted or improperly worn equipment. In analyzing records of 100,000 employees, Dr. Hildyard found that less than 2% suffered any hearing loss for this reason. n11

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n11 Dr. Hildyard testified that between 1% and 2% of the employees he tested showed some hearing loss, and in some cases this was attributable to causes unrelated to their employment.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Hildyard also testified that some individuals could not wear ear plugs due to deformed or unusually shaped ears, or to ear infections. Also some individuals are allergic to earmuffs, and their use can result in dermatitis (skin infection).

II. Interpretation of the Standard

The standard requires that [*16] engineering (and administrative) controls be utilized as the primary means of protecting employees against excessive noise. The standard, however, only requires such controls to the extent they are "feasible." The dispute between the parties basically reduces to a dispute over the proper interpretation of this word as it is used in the standard.

A. Must G-16 Limits Be Achieved?

Continental first argues that contols are per se not feasible unless they would reduce the noise levels to within G-16 limits. n12 It argues that, because of the use of personal protective equipment, its employees are now exposed to noise levels below G-16 limits, and that reduction of noise from current levels to lower levels which still exceed 90 dBA will not obviate the need for its employees to wear personal protective equipment. Thus, according to Continental, such controls are not feasible because they will essentially result in the expenditure of a large sum of money with no significant change in the situation which now exists.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n12 As noted above, Continental's expert witness, Mr. Prendergast, expressed the opinion that G-16 limits could be achieved throughout the Metal Division if enclosures are installed on all 6300 machines. At the hearing, however, Labor attempted to limit the inquiry to the five cited families in the California plants. The evidence shows that installation of enclosures around only these machines would achieve G-16 limits in some, but not all locations in these plants. Accordingly, Continental argues that the engineering controls sought by the Secretary are not feasible because at least some employees would still be exposed to excessive noise.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The first difficulty with this argument is that it is inconsistent with the literal language of the standard. The standard requires the implementation of feasible engineering or administrative controls, and then goes on to provide for the use of personal protective equipment "if such controls fail to reduce sound levels within the levels of Table G-16". The standard thus contemplates that there will be some situations where engineering or administrative controls are to be considered feasible even though they fail to reduce the noise below G-16 limits. The record in this case reveals some of the rationale underlying this requirements.

The evidence shows that employees may suffer hearing loss even if an effective hearing conservation program is administered. Some employees will not always wear ear protectors, and others will not receive full protection from them due to improper fitting or use. For employees who do not receive the full benefit possible from personal ear protectors, any significant reduction n13 in the ambient noise levels provides a benefit. Furthermore, even though a particular [*18] reduction in noise levels may not totally obviate the need for personal protective equipment, it can allow the G-16 limits to be achieved by the use of such equipment for a shorter time. To this extent, it can result in less employee resistance to the use of such equipment, and increase the effectiveness of a hearing conservation program in which the use of such equipment is required.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n13 In determining how great a reduction is significant, the logarithmic nature of the decibel scale must be considered. (Note 8, supra). Since a reduction of 3 dBA represents a halfing of the air pressure, such a reduction is clearly significant.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Another difficulty with Continental's argument is that it overlooks the realities involved in noise reduction by engineering means. As Dr. Salmon's testimony and Continental's experience show, engineering controls will not always produce discrete, predictable reductions in noise levels. It may be clear that a substantial reduction in noise can be achieved, but the full reduction that may [*19] ultimately result cannot be determined until a certain amount of development and testing has already occurred. The feasibility of some controls may not be determinable until other controls have been implemented. It would therefore be unrealistic and would ignore the nature of the problem to hold that the ultimate reduction that controls can achieve must be proven before the implementation of any controls is required.

This is not, however, to say that whether G-16 levels can be achieved is irrelevant for all purposes. Obviously, as these levels are approached, it becomes easier to predict whether further controls can in fact achieve full compliance. Also, as they are approached, the detrimental effect to employees not adequately protected by personal protective equipment is less than at noise levels substantially in excess of G-16 limits. Thus, whether the G-16 limits can in fact be achieved does become a relevant factor in determining whether particular engineering controls are feasible when the noise levels are only slightly in excess of these limits.

B. Must Controls Be Economically Feasible?

The basic controversy between the parties concerns whether the word "feasible" [*20] as used in the standard should be interpreted to mean only "technically possible," as Labor contends, or whether it includes a component of economic feasibility, as claimed by Continental.

The record establishes that the technology exists to achieve a significant reduction of the noise in Continental's plants. Labor contends that this is sufficient to establish a violation of the standard, and that our inquiry need go no further. Continental contends, however, that in determining whether controls are feasible, the costs of such controls must be considered and weighed against the benefits which will be produced. Either interpretation is arguably within the literal meaning of "feasible," and the standard itself does not further define the term. Nor does the history and background of the standard provide any insight into its meaning. n14 The standard must therefore be interpreted to effectuate the Congressional purposes underlying the Act. Brennan v. Gilles & Cotting, Inc., supra, 504 F.2d at 1260, Brennan v. OSHRC (Gerosa, Inc.) 491 F.2d 1340, 1343 (2d Cir. 1974).

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n14 The standard was originally promulgated by Labor under the Walsh-Healey Act (41 U.S.C. 35 et seq). The Walsh-Healey Act pertains to the procurement of supplies by the United States government. The safety and health standards issued under the Walsh-Healey Act were intended to implement the requirement in 41 U.S.C. 35(e) that none of the items procured by the government pursuant to the Walsh-Healey Act should be manufactured "under working conditions which are unsanitary or hazardous or dangerous to the health and safety of employees engaged in the performance of said contract." 41 C.F.R. 50-204.1.

A review of the history of the promulgation of the standard under the Walsh-Healey Act is found in Turner Co., supra, n. 9. When the Occupational Safety and Health Act was enacted, Congress authorized Labor to promulgate and enforce the Walsh-Healey standards, as well as other established Federal standards, without resort to further rulemaking proceedings, 29 U.S.C. 655(a). Furthermore, Congress specifically provided that the Walsh-Healey standards, as well as certain other established federal standards, should be deemed to be occupational safety and health standards under the Act. 29 U.S.C. 653(b)(2).

We have not been cited to, and have not found, any indication of the intended meaning of "feasible" when the standard was adopted under the Walsh-Healey Act. Nor is there any indication in the legislative history of the Occupational Safety and Health Act that Congress, when it provided for the adoption of a large number of established federal standards, considered the interpretation of this particular standard.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The Act provides that, when promulgating standards for toxic substances or harmful physical agents, Labor should consider, inter alia, the "feasibility of the standards." n15 As used in the Act, "feasibility" contemplates economic as well as technological feasibility. AFL-CIO, v. Brennan, 530 F.2d 109, 122-123 (3rd Cir. 1975); Industrial Union Department, AFL-CIO v. Hodgson, 499 F.2d 467 (D.C. Cir. 1974). In the latter case, the court observed:

There can be no question that OSHA represents a decision to require safeguards for the health of employees even if such measures substantially increase production costs. This is not, however, the same thing as saying that Congress intended to require immediate implementation of all protective measures technologically achievable without regard for their economic impact. To the contrary, it would comport with common usage to say that a standard that is prohibitively expensive is not "feasible." 499 F.2d at 477 (footnote omitted).

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n15 29 U.S.C. 655(b)(5). This requirement pertains to standards promulgated by Labor pursuant to notice and hearing, as provided in Section 6(b) of the Act, 29 U.S.C. 655(b). As noted in note 14, supra, the standard here at issue was an established federal standard and was promulgated pursuant to Section 6(a), 29 U.S.C. 655(a).


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Although Congress intended that standards be economically feasible, the history of this standard reveals that no attention to this factor was necessarily paid in its promulgation. The standard was first promulgated pursuant to the Walsh-Healey Act, which contained no requirement that economic feasibility be considered. It was then, under this Act, made applicable to all employers throughout the country pursuant to procedures which precluded further consideration of its economic feasibility. See note 14, supra. Thus, were we to hold that economic feasibility is irrelevant in enforcing the standard, the consequence would be that the standard would be applied without any consideration ever having been given to its economic consequences, either on a national basis or on an industry-by-industry basis. Such a result would be inconsistent with the intent of Congress as expressed by the court in Industrial Union Department, AFL-CIO v. Hodgson, supra.

It is true that Labor would allow consideration of economic factors if the cost of implementing controls would seriously jeopardize the financial health [*23] of a company. See note 4, supra. We do not, however, think this approach is reasonable. It ignores the nature of the hazard because it does not distinguish between hazards. Thus the hazard is not life threatening in this case and although it will produce serious loss of hearing in some cases we must consider the fact that the harm in other cases will be little if any hearing loss. The situation, therefore, is distinguishable from life threatening hazards such as those posed by carcinogenic substances. See: Society of the Plastics Industries Inc. v. OSHA, 509 F.2d 1301 (2d Cir. 1975); Industrial Union Department, AFL-CIO v. Hodgson, supra; Synthetic Organic Chemical Mfrs. Ass'n v. Brennan, 503, F.2d 1155 (3d Cir. 1974).

The argument also ignores the benefits to be gained since it requires employers to expend funds for each and every health hazard whether life threatening or not without limit so long as the expenditures for each hazard can be borne without putting the employers' financial condition in jeopardy. Clearly, employers have finite resources available for use to abate health hazards. And just as clearly if they are to be made to spend without limit [*24] for abatement of this hazard their financial ability to abate other hazards, including life threatening hazards, is reduced. We can conceive that in some if not many cases it will be reduced to the point that they will be put in financial jeopardy when faced with the problem of abating a life threatening hazard. Such a result does not comport with the purposes of the Act because in our view resources should be allocated on a priority basis to obtain the benefits that may be achieved by eliminating life threatening hazards first and lesser hazards seeond. And by saying "second" we do not mean second in time but rather second in priority for allocation of resources. In this regard Congress recognized the need for establishing priorities in eliminating workplace hazards. Thus it mandated Labor to set priorities on an urgency of the need basis. 29 U.S.C. 655(g).

Accordingly, we conclude that the standard should be interpreted to require those engineering and administrative controls which are economically, as well as technically feasible. Controls may be economically feasible even though they are expensive and increase production costs. See Arkansas-Best Freight Systems, Inc., [*25] 529 F.2d 649, 653 (8th Cir., Jan. 29, 1976); Industrial Union Department, AFL-CIO v. Hodgson, supra, 499 F.2d at 477. But they will not be required without regard to the costs which must be incurred and the benefits they will achieve. In determining whether controls are economically feasible, all the relevant cost and benefit factors must be weighed.

C. Benefits of Controls for Noise

As discussed in Section II.A supra, any significant reduction in ambient noise levels which are above G-16 limits will tend to provide some protection to the hearing of employees. But we cannot overlook the fact that protection against most of the detrimental effects of excessive noise can be achieved through the use of a relatively reliable and convenient type of personal protective equipment. Additionally, the effects of excessive noise are gradual rather than sudden, and a certain amount of exposure to continuous noise levels as high as 115 dBA is permitted by the standard itself.

Other factors to be considered in the benefits which noise reduction will achieve include the number of employees exposed to excessive noise, and the net reduction in their exposure which engineering or administrative [*26] controls can reasonably be expected to produce. As to the last noted factor we would emphasize that the benefits to be obtained will vary according to the magnitude of the noise levels existing prior to the imposition of controls. Cost factors which must be considered include both the direct costs of installing and maintaining controls, and indirect costs due to any loss of productivity or efficiency resulting from their implementation. This list is not meant to be all-inclusive. We cannot here establish specific guidelines which will be applicable to all cases. The ultimate determination in each individual case must be made by weighing all the relevant factors in light of the objectives of the Act.

III. Analysis of the Evidence

The burden of proving that engineering or administrative controls are feasible lies with Labor. Continental [*27] did present evidence relative to economic feasibility. Its evidence showed that reduction of noise to within G-16 limits could reasonably be expected at most, if not all locations at all 79 plants in its Metal Division through the expenditure of approximately 33 million dollars. n17 Additionally, maintenance costs would approximate $175,000 per year, as compared to a $100,000 annual cost of its current hearing conservation program utilizing personal protective equipment. Continental's two expert witnesses, Mr. Prendergast and Mr. Campbell, expressed the opinion that engineering controls were not economically feasible. There was no contrary expert opinion testimony.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n16 In Atlantic & Gulf Stevedores, Inc. v. OSHRC, supra, the Court indicated that an employer could generally defend against an alleged violation of a standard by proving that the standard was not economically feasible as to him. The standard here at issue, unlike the one before the Court, explicitly requires only those engineering and administrative controls which are feasible. Accordingly, proof of feasibility is an element of the charge, and must be borne by Labor. In Love Box Co. and Reynolds Metals Co., supra, Labor conceded that it had the burden of proving the feasibility of engineering or administrative controls.

n17 At the hearing, Labor objected to the introduction of evidence concerning the cost of compliance beyond the cited machines at three California plants. It is true that the only issue, at the time of the hearing, was whether Continental violated the standard with respect to the cited machines at the three California plants. But Continental was then, and is now, required to comply with the standard at all of its plants throughout the country. Thus, assuming feasible engineering controls exist to reduce noise in Continental's plants, it is clearly proper for Continental to attack the problem by first trying to quiet the machines which generate the greatest exposure of its employees throughout the country.

Furthermore, as Judge Burchmore observed, Continental's approach permits economies of scale to be given appropriate consideration. In approaching the problem on a company-wide basis, Continental is able to spread its research and development expenditures over a large number of plants.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

On this record, we cannot conclude that Labor has proven the feasibility of engineering controls by a preponderance of the evidence. Olin Construction Co. v. OSHRC, 525 F.2d 464 (2nd Cir. 1975); Armor Elevator Co., 5 OSAHRC 260, BNA 1 OSHC 1409, CCH OSHD para. 16,958 (1973). Even though Continental is a very large company and a number of plants would be affected, the cost to implement engineering controls is substantial. Without knowing the number of employees now exposed to excessive noise in these plants we cannot gauge the benefits which might be achieved. Additionally, the only expert testimony on the point indicates that engineering controls are not economically feasible.

Even if we only consider whether it is feasible to enclose only the cited machines at the three California plants, we reach the same conclusion. The cost to develop and install enclosures for these machines is $494,000. The exposure of approximately 125 employees would be reduced to within G-16 limits, and some lesser reduction would be experienced by an unspecified number of other employees. These figures, standing [*29] alone, are not sufficient to establish by a preponderance of the evidence that engineering controls are economically feasible.

It should be stressed that nothing in this decision relieves Continental from continuing to utilize means currently in effect to protect its employees from excessive noise. We hold only that, on this record, Labor has failed to sustain its burden of proving the economic feasibility of engineering controls. Nothing which has been said limits Continental's duty to comply with the standard by implementing whatever engineering or administrative controls may be feasible to reduce the noise levels in its plants.

Accordingly, the citations alleging violations of 29 C.F.R. 1910.95(b)(1) are vacated. It is so ORDERED.



MORAN, Commissioner, Concurring:

I agree with the foregoing opinion only insofar as it affirms the disposition ordered by Judge Burchmore in these cases and concludes that complainant failed to carry his burden of establishing that engineering controls are economically feasible. I cannot join in the remainder of that opinion, however, because it improperly interprets the facts and is permeated with unnecessary and misleading dicta. n18 [*30] The factual analysis stated in Judge Burchmore's decision is correct. I therefore attach that decision hereto as Appendix A and incorporate it herein by reference so that the facts of this case will be stated correctly in this decision. Furthermore, I specifically agree with and adopt Judge Burchmore's reasons for holding that 29 C.F.R. 1910.95(b)(1) cannot be validly enforced against this respondent on the basis of the record in this case because cost considerations in combination with the other factors discussed in his opinion make it infeasible for respondent to utilize engineering controls.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n18 For example, I disagree with the conclusion that 29 C.F.R. 1910.95(b)(1) requires the use of administrative or engineering controls even though complainant is unable to establish that such controls will reduce the noise in respondent's plants to the permissible levels specified in Table G-16.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -




The majority errs when it construes the term "feasible" in 29 CFR 1910.95(b)(1) [hereinafter 1910.95(b)(1)] to include considerations of economic as well as technological feasibility. My colleagues also err when they place upon the Secretary the burden of proving the economic feasibility of engineering controls in every case in which he seeks to enforce the noise standard. Therefore, I dissent from the majority's disposition of these cases. n19 In my opinion, economic factors are properly considered only in fashioning an appropriate abatement order.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n19 In his concurring opinion, Commissioner Moran rejects the Chairman's factual analysis as improper. Nevertheless, he concurs in the holding that the Secretary failed to carry his burden of establishing the economic feasibility of engineering controls. This concurrence is based upon his specific adoption of Judge Burchmore's reasons for vacating the citation.

Apparently, Commissioner Moran interprets the Judge's analysis as incorporating a finding that it was economically infeasible for Continental to institute technologically feasible controls here. This is a mistaken interpretation of the Judge's decision. As the Chairman observes, the Judge disposed of the case by ruling that the standard itself was economically infeasible and, therefore, unenforceable.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

I agree with the lead opinion insofar as it holds that 1910.95(b)(1), by its terms, "requires that engineering (and administrative) controls be utilized as the primary means of protecting employees against excessive noise." In addition, I concur in the Chairman's holding that 1910.95(b)(1) requires the implementation of feasible engineering or administrative controls even though such measures by themselves may not bring sound levels within the acceptable limits set out in Table G-16. One need go no further than the wording of the standard itself in order to reach these holdings.

Reading 1910.95(b)(1) in this manner, a comprehensive approach to the attenuation of excessive noise levels in work environments is apparent. An employer has a continuing duty to develop and implement technologically feasible engineering or administrative controls to reduce excessive sound levels in the workplace. This ongoing duty attaches regardless of the fact that at any given time the controls implemented would not necessarily be effective in bringing noise levels below G-16 limits. Personal protective equipment [*33] is to be used only as a form of supplemental protection during that limited period of time during which those engineering or administrative controls which have been implemented cannot by themselves reduce sound levels to within the limits of Table G-16. As such, 1910.95(b)(1) requires the development of new technologies and the adaptation of existing technologies so as to reduce sound levels by the use of engineering or administrative controls.

Although Chairman Barnako pays lip service to this reading of the standard, the practical effect of his holding in the instant case is the negation of the entire thrust of 1910.95(b)(1). Specifically, by reading the term "feasible" as it appears in the standard to include economic as well as technological feasibility in determining whether a violation of the Act has occurred, the Chairman has, in effect, relieved employers of the continuing duty to develop and implement engineering or administrative control measures. n20 This emasculation of 1910.95(b)(1) is compounded by placing upon the Secretary the burden of proving economic feasibility and by announcing a test that requires the weighing of "all the relevant cost and benefit [*34] factors" in order to determine whether an employer can be required to implement control measures as the primary method of compliance with the noise standard.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n20 I submit that Chairman Barnako recognizes the practical effect of his holding. In closing he asserts that:

Nothing which has been said limits Continental's duty to comply with the standard by implementing whatever engineering or administrative controls may be feasible to reduce the noise levels in its plants.

Notwithstanding this bare assertion, for the reasons stated in my opinion it is clear that he has in fact severely limited, if not totally negated, Continental's ongoing duty to implement technologically feasible engineering or administrative control measures.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Economic considerations are not implicit in the term "feasible" as it appears in 1910.95(b)(1). Under the noise standard, a duty to develop and implement engineering controls arises upon a showing that such measures are technologically feasible. In other words, if an employer is [*35] faced with excessive noise levels in the workplace and there are technologies available to reduce such noise, he must commence a program aimed at reducing noise levels by engineering controls. Failure to embark on such a program in the face of existing technologies constitutes a violation of 1910.95(b)(1). The cost of abatement is not relevant in the determination of whether an employer is required to abate excessive noise levels. Cost considerations are relevant only in determining the pace at which an abatement program should proceed. As such, cost factors may play a role in fashioning an appropriate abatement period. Accordingly, I dissociate myself from the Chairman's ultimate interpretation and application of 1910.95(b)(1).

A more complete understanding of standards that are structurally similar to 1910.95(b)(1) can be grasped from the court's analysis of the standard at issue in Society of Plastics Indus., Inc. v. O.S.H.A., 509 F.2d 1301 (2d Cir. 1975), cert. denied, 421 U.S. 992 (1975). In Society of Plastics the court entertained n21 and rejected the petitioners' challenges to a standard promulgated by the Secretary of Labor dealing with occupational [*36] exposure to vinyl chloride. Among other things, petitioners contended that the Secretary had exceeded his statutory authority n22 by requiring employers to reduce exposure levels to vinyl chloride to 1 ppm through engineering means. They asserted that the attainment of this level of exposure through engineering means was impossible.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n21 Jurisdiction to entertain such challenges to the validity of the Secretary's rulemaking actions is conferred upon the courts of appeals by section 6(f) of the Occupational Safety and Health Act of 1970, 29 U.S.C. 651 et seq. [hereinafter cited as "the Act"].

n22 Section 6(b)(5) of the Act.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The Second Circuit disagreed with the petitioners' assertions of impossibility. Rather, the court noted that the record revealed vast rapid technological improvements in engineering and work practice controls which in turn suggested that the petitioners merely lacked faith in their own technological potentialties. Society of Plastics Indus., Inc. v. O.S.H.A., supra at 1309. [*37] More important to the instant case, however, the court went on to note that:

In the area of safety, we wish to emphasize that the Secretary is not restricted by the status quo. He may raise standards which require improvements in existing technologies or which require the development of new technology, and he is not limited to issuing standards based solely on devices already fully developed. Cf. Chrysler Corp. v. Dept. of Transportation, 472 F.2d 659, 673 (6th Cir. 1972); Natural Resources Defense Council, Inc. v. E.P.A., 489 F.2d 390, 401 (5th Cir. 1974).

Society of Plastics Indus., Inc. v. O.S.H.A., supra at 1309 (emphasis added.) n23 Thus, the court recognized that the Secretary could enforce standards that would force the development and implementation of technological improvements.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n23 The Second Circuit's observations were echoed by the court in AFL-CIO v. Brennan, 530 F.2d 109 (3d Cir. 1975). In its decision the court noted that:

Although we hold that the Secretary may, consistent with the statute, consider the technological feasibility of a proposed occupational health and safety standard promulgated pursuant to 6(a), we agree with the Second Circuit in Society of Plastics, Industry, Inc. v. OSHA, supra, that, at least to a limited extent, OSHA is to be viewed as a technology-forcing piece of legislation. Thus the Secretary would not be justified in dismissing an alternative to a proposed health and safety standard as infeasible when the necessary technology looms on today's horizon.

AFL-CIO v. Brennan, supra at 121 (footnote omitted).


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The relevance of the court's analysis in Society of Plastics to the operation of 1910.95(b)(1) can be seen by reference to the compliance requirements set out in the vinyl chloride standard. At 29 CFR 1910.1017(f) n24 the approach to compliance is set out in detail. It is as follows:

1910.1017 Vinyl Chloride.

* * *

(f) Methods of compliance.

Employee exposures to vinyl chloride shall be controlled to at or below the permissible exposure limit provided in paragraph (c) of this section by engineering, work practice, and personal protective controls as follows.

(1) Feasible engineering and work practice controls shall immediately be used to reduce exposures to at or below the permissible exposure limit.

(2) Wherever feasible engineering and work practice controls which can be instituted immediately are not sufficient to reduce exposures to at or below the permissible exposure limit, they shall nonetheless be used to reduce exposures to the lowest practicable level, and shall be supplemented by respiratory protection in accordance with paragraph (g) of this section. A program shall [*39] be established and implemented to reduce exposures to at or below the permissible exposure limit, or to the greatest extent feasible, solely by means of engineering and work practice controls, as soon as feasible.

Although the provisions are more detailed than those of 1910.95(b)(1), it is clear that the abatement approach under the vinyl chloride standard is the same as that implicit in the noise standard. It is apparent, therefore, that 1910.95(b)(1), like the vinyl chloride standard, should be viewed as technology-forcing in its approach to noise abatement.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n24 At the time of the court's review of the vinyl chloride standard, it was codified at 29 CFR 1910.93q. It has since been recodified at 29 CFR 1910.1017 without any change in substance. 40 Fed. Reg. 23072 (1975).

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Thus, any interpretation of 1910.95(b)(1) must be made in view of its technology-forcing characteristics. This, of course, would include the interpretation of the term "feasible" as it appears in 1910.95(b)(1). Although the Chairman [*40] purports to pay heed to the abatement approach implicit in 1910.95(b)(1), his interpretation and application of the word "feasible" in economic terms negates the very approach he presumably endorses. Under the Chairman's approach, present economic difficulties can vitiate the technology-forcing process implicit in the noise standard.

Several flaws are apparent in the method by which Chairman Barnako reaches the conclusion that economic factors must be considered in determining whether a violation of the Act for failing to comply with the noise standard has occurred.

Reliance is placed on decisions of two courts of appeals in reaching the conclusion that "[a]s used in the Act, 'feasibility' contemplates economic as well as technological feasibility." It is important to note, however, that both decisions relied upon concerned challenges to the Secretary of Labor's rulemaking action under section 6 of the Act. This Commission, however, is not an appellate court reviewing the Secretary's section 6 action. Rather, we are interpreting a term as it appears in a technology-forcing standard regulating excessive noise that was properly adopted under section 6(a) of the Act. n25 Consequently, [*41] we must determine what effect, if any, economic considerations should have in enforcing 1910.95(b)(1). Any such determination must be made in light of the standard's ultimate objective of requiring employers to implement engineering or administrative controls that will reduce sound levels to within acceptable limits.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n25 Turner Co., 4 BNA OSHC    , 1975-76 CCH OSHD para.    , (No. 3635, Aug. 24, 1976).

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The Chairman relies heavily upon Idustrial Union Dep't, AFL-CIO v. Hodgson, 499 F.2d 467, 477 (D.C. Cir. 1974). He notes the observation of the court that:

This is not, however, the same thing as saying that Congress intended to require immediate implementation of all protective measures technologically achievable without regard for their economic impact (emphasis added).

A close reading of the emphasized portion of the above quotation reveals the fallacy of the Chairman's application of the court's observations regarding the asbestos standard to reach his interpretation of 1910.95(b)(1). The asbestos [*42] standard at issue in the Industrial Union Dep't case is not, strictly speaking, a technology-forcing standard inasmuch as the standard itself sets forth methods of engineering and work practice controls, and requires their immediate implementation. See 29 CFR 1910.1001(c) and (d). The noise standard at issue in this case bears neither of these features. It does not set forth specific engineering or administrative control measures. Nor does it require the immediate implementation of engineering or administrative controls so as to reduce sound levels below those limits set forth in Table G-16 without the use of personal protective equipment. n26 Thus, any reliance upon the court's observations regarding the asbestos standard is misplaced here.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n26 The asbestos standard at 29 CFR 1910.1001(d)(1) permits the use of personal protective equipment as an interim abatement measure. It provides as follows:

(d) Personal protective equipment.

(1) Compliance with the exposure limits prescribed by paragraph (b) of this section may not be achieved by the use of respirators or shift rotation of employees, except:

(1) During the time period necessary to install the engineering controls and to institute the work practices required by paragraph (c) of this section.

(ii) In work situations in which the methods prescribed in paragraph (c) of this section are either technically not feasible or feasible to an extent insufficient to reduce the airborne concentrations of asbestos fibers below the limits prescribed by paragraph (b) of this section; or

(iii) In emergencies.

(iv) Where both respirators and personnel rotation are allowed by subdivisions (i), (ii), or (iii) of this subparagraph, and both are practicable, personnel rotation shall be preferred and used (emphasis added).

These narrow exceptions to the requirement of immediate implementation of completely effective engineering controls or work practices clearly cannot be read as coextensive with 1910.95(b)(1) insofar as it permits the use of supplemental personal protective equipment during development and implementation of engineering or administrative control measures. In other words, 29 CFR 1910.1001(d)(1) does not lessen the duty to achieve full compliance immediately through the use of engineering controls or work practices.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

It is noted in the lead opinion that the history of 1910.95(b)(1) reveals that no attention was necessarily paid to its economic feasibility when it was first adopted as a Walsh-Healey standard. The Chairman purports to fill the gap during enforcement proceedings by noting that:

Thus, were we to hold that economic feasibility is irrelevant in enforcing the standard, the consequence would be that the standard would be applied without any consideration ever having been given to its economic consequences, either on a national basis or on an industry-by-industry basis. Such a result would be inconsistent with the intent of Congress as expressed by the court in Industrial Union Department, AFL-CIO v. Hodgson, supra. n27

I find this "gap-filling" approach disturbing. The Chairman implies that economic considerations applied during rulemaking should now be applied during enforcement proceedings. Surely, Congress never intended to introduce such a plainly unworkable element into cases in which the Secretary seeks to enforce an occupational safety and health standard which was derived from an [*44] established federal standard.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n27 As noted above, the Industrial Union Dep't case involved a different type of standard than that at issue in this case. Moreover, the court did not say that the Secretary must consider economic feasibility. Rather, it merely held that the Secretary could consider economic feasibility. Industrial Union Dep't, AFL-CIO v. Hodgson, supra at 123.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

I do not suggest that economic considerations are irrelevant during enforcement proceedings. On the contrary, if an employer affirmatively establishes economic hardship, this Commission may adjust the abatement period under its power to fashion "other appropriate relief" under section 10(c) of the Act. Such considerations should not, however, enter into a determination as to whether 1910.95(b)(1) has been violated and a duty to develop and implement technologically feasible engineering controls exists.

The Chairman rejects the Secretary's assertion that economic factors should only be considered if the cost of implementing controls [*45] would seriously jeopardize an employer's financial health. While I agree with his rejection of this assertion for somewhat different reasons, it is important to analyze the way in which Chairman Barnako rejects the Secretary's assertion.

It is stated that the Secretary's approach is unreasonable inasmuch as "[i]t ignores the nature of the hazard because it does not distinguish between hazards." The Chairman would have the Secretary weigh the relative severity of health hazards and choose with which applicable standard an employer should be required literally to comply.

I question the viability of balancing deleterious effects on affected employees. In his discussion the Chairman weighs serious hearing losses against life-threatening hazards such as those posed by carcinogens. While this approach seems logical, I submit that a better deaf-than-dead concept is too Draconian to be consistent with the statutory purpose of providing a safe and healthful workplace.

The Secretary should not be forced to choose between hazards. Rather, he should be free to require literal compliance with both standards. Any balancing that takes place should be in the context of determining [*46] the relative time needed to comply literally with both standards, i.e. in framing appropriate abatement orders. n28 An employer should be required to comply literally with both applicable standards, the relativity of the hazards being important only insofar as the fashioning of an appropriate abatement order for each hazard is concerned.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n28 Suppose for example that more than one life-threatening hazard exists in the same workplace. The Chairman's approach would not offer a conceptional solution. Also, the fact that the noise standard came from regulations under the Walsh-Healey Act is significant. See Turner Co., supra, note 25. The type of cost balancing contemplated by the Chairman would seem to make uniform bidding on Government supply contracts impossible.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

This approach does not lose force even if the Chairman's concept of "finite resources available for use to abate health hazards" is introduced. n29 The allocation of finite resources to abate hazards is but another factor in fashioning an appropriate [*47] abatement time for hazards. Abatement times can reflect the priorities in allocation of resources among several hazards. In other words, an employer faced with abating several hazards should be required to begin abating each hazard at the same time. The pace at which abatement should proceed can be adjusted to reflect the individual employer's financial condition.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n29 I have difficulty with this concept of finite resources allocable to safety and health matters. It would seem to presuppose the continued existence of employers in all situations regardless of the severity of hazards existing in the workplace. At least two courts of appeals have suggested to the contrary. In Industrial Union Dep't, AFL-CIO v. Hodgson, supra at 478, the court noted that:

Standards may be economically feasible even though, from the standpoint of employers, they are financially burdensome and affect profit margins adversely. Nor does the concept of economic feasibility necessarily guarantee the continued existence of individual employers (emphasis added).

Accord, AFL-CIO v. Brennan, supra at 122-3.

In any event, as noted by the Chairman, "Continental does not claim that the cost of the engineering controls involved in this case would jeopardize its financial health."


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

If, as in the instant case, a crude cost-benefit approach is employed to vacate a noise violation even though engineering controls are technologically feasible, the practical effect is to relieve the employer of the continuing duty to develop and implement engineering or administrative controls. This is particularly true when dealing with a standard like 1910.95(b)(1) which, by its very terms, envisions abatement over a period of time.

Perhaps the most disturbing aspect of the Chairman's holding is his placing of the burden of proving the economic feasibility of engineering controls upon the Secretary. This imposition is contrary to settled principles of law. More importantly, however, it places upon the Secretary a burden that is so onerous that it may be impossible to satisfy. As a necessary consequence, therefore, it will be virtually impossible for the Secretary to enforce the noise standard so as to assure the continuing development and implementation of technologically feasible engineering controls.

At this point it is necessary to review exactly what Chairman Barnako seems to be requiring [*49] the Secretary to prove in order to establish economic feasibility. First, he would require the Secretary to identify and establish the finite amount of resources an employer can allocate for the abatement of all safety and health hazards in his workplace. Next, the Secretary would have to identify all hazards existing in the work environment and show what it would cost to abate each hazard. In the event the finite resources allocable to the abatement of hazards are insufficient to cover the cost of abating all the hazards identified, the Secretary would then have to establish a priority schedule for abatement based upon the relative severity of the hazards and the benefits that abatement of these hazards will achieve.

Clearly, proving such matters would in most cases pose extremely difficult if not insurmountable problems for the Secretary. He would be required to master virtually every aspect of an employer's financial condition and physical operation before being able to require literal compliance with 1910.95(b)(1). Such efforts would be time-consuming and would, of necessity, severely strain the Secretary's enforcement resources. I believe that the imposition of this [*50] burden on the Secretary is contrary to Congress' goal of an effective enforcement program. n30

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n30 Section 2(b)(9) of the Act.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

Even assuming that the Secretary could embark on such an exhaustive investigation of an employer in order to require compliance with the noise standard, a quasi-judicial agency should carefully assess its own resources before committing itself to "cost-benefit" decision-making in every noise standard litigation. This Commission surely lacks the expertise and resources to enter into the "cost-benefit" analyses suggested by the lead opinion. The complexities of such analyses are manifest and extend beyond the situation of any single employer. Requisite societal "cost-benefit" analyses would require both intra-industry and inter-industry considerations. See Industrial Union Dep't, AFL-CIO v. Hodgson, supra at 479-81; AFL-CIO v. Brennan, supra at 120-23. Such analyses should only be made by the Secretary during rulemaking proceedings. They should not be attempted by the [*51] Commission during enforcement proceedings against an individual employer.

In my opinion the above policy considerations dictate that the term "feasible", as it appears in 1910.95(b)(1), should be interpreted so as to exclude economic considerations. Thus, the Secretary carries his burden under the noise standard by establishing the existence of technologically feasible engineering methods and a technologically feasible abatement time. Economic factors should be considered only if raised by an employer seeking adjustment of the technologically feasible abatement period.

I, therefore, disagree with the majority's action in vacating the citations at issue in this case. Continental conceded that engineering controls were technologically feasible. Consequently, the citations should be affirmed. These cases should, however, be remanded for the limited purpose of fashioning an appropriate abatement order.



Altero D'Agostini and Norman S. Nayfach for the Secretary of Labor

Charles M. Chadd and Thomas D. Nyhan for the respondent


By citations issued July 25, 1973 in No. 3973, and subsequently in the other cases, as amended, the Secretary [*52] charges that, on the pertinent inspection dates, respondent committed non-serious violations of section 5(a)(2) of the Occupational Safety and Health Act of 1970, 29 U.S.C. 651 et seq (the Act), in that respondent failed to comply with the occupational safety and health standard provided in respect of noise exposure, as contained in 29 C.F.R. 1910.93(b)(1). The regulation provides:

When employees are subjected to sound exceeding those listed in Table G-16, n1 feasible administrative or engineering controls shall be utilized. If such controls fail to reduce sound levels within the levels of Table G-16, personal protective equipment shall be provided and used to reduce sound levels within the levels of the table.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n1 Table G-16 is set forth in Appendix A hereto.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The citations pertained to the following plants of the respondent, and the penalties proposed in most instances were zero and in some cases nominal amounts of less than$ 100: Los Angeles, Calif. (No. 3973); Pittsburg, Calif. (No. 4397); San Jose, [*53] Calif. (No. 4501); Worthington, Ohio (No. 5327); Pascagoula, Miss. (No. 4853); Bedford Heights, Ohio (No. 7122); Cincinnati, Ohio (No. 7910); and Columbus, Ohio (No. 7920). Abatement of the alleged violations was proposed in several steps as illustrated by the title case: (1) research and study of engineering and administrative controls with monthly reports of progress, culminating (2) in a final study report on control implementation to be submitted no later than January 4, 1974; (3) installation of controls in accordance with final study report, and monthly reports of progress during such installation; (4) all feasible controls to be installed within 18 months following final report in (2).

Timely notices of contest were filed in each case and all cases have been assigned to the undersigned judge. A consolidated hearing was held in the California cases, Nos. 3973, 4397, and 4501, at San Francisco on January 24-6, and February 27 - March 1, 1974. thereafter, pursuant to stipulations between the parties, the remaining cases were consolidated with the title case and it was agreed that no further hearing would be required. The stipulations provide, in effect, that the decision [*54] in the California cases shall govern the decision in the other cases. The cases have been extensively briefed by the parties, the last reply briefs having been filed August 22, 1974. A motion of respondent for leave to file a short reply statement, being unopposed, is hereby granted.


Many of the basic facts of this case are not in dispute, even though the hearing developed a transcript of more than a thousand pages; as to several issues the parties reached agreement after the evidence was in. For example, complainant introduced extensive testimony as to the noise levels and employee time in various areas of each California plant; yet after the hearing it was stipulated that noise levels in all plants exceed the G-16 levels for the specified durations. Accordingly, the following statement of facts is largely undisputed; where there is a factual issue, the evidence on both sides is summarized and findings are made in the discussion which follows:

(1) Respondent, Continental Can Company, Inc., a corporation, at all times herein relevant maintained its principal place of business at 633 Third Avenue, New York City, New York 10016;

(2) Respondent, at all times erein relevant, [*55] maintained workplaces at Los Angeles, Pittsburg and San Jose, California; Worthington, Ohio; Pascagoula, Mississippi; Bedford Heights, Cincinnati and Columbus, Ohio; at which locations respondent's employees, were engaged in the manufacture of metal cans and/or metal can elements.

(3) Respondent, at all times herein relevant, had employees and was an employer engaged in a business affecting commerce within the meaning of Section 3(5) of the Act, in that its employees are engaged in the manufacture, sale, and distribution of metal cans and/or metal can elements to various locations outside the aforementioned states from points within those states.

(4) The noise levels in the plants in question in areas where employees were working exceeded the sound levels listed in Table G-16 of 29 C.F.R. 1910.95.

(5) The ambient noise levels in the cited areas of the three California plants were in the following ranges:

Los Angeles

97-100 dba


94-100 dba

San Jose

96-102 dba

(6) Personal ear protective devices are available that will reduce noise levels of the magnitude found in each of the plants below 90 dba and will effectively protect employees from loss of hearing due [*56] to harmful noise. For example, ear plugs properly seated will attenuate noise 22 to 27 dba; ear muffs properly seated will attenuate noise up to 35 dba. Proper seating is a very simple process.

(7) During the inspection of each of the Los Angeles, Pittsburg, and San Jose plants, all employees observed by each compliance officer were wearing ear protection.

(8) There is no issue and the Secretary is not alleging in the consolidated cases that Continental Can (a) has failed to provide its employees "with personal protective equipment" which "reduce sound levels within the levels of the Table (G-16)" or (b) has failed "to administer a continuing effective hearing conservation program."

The regulatory standard, as interpreted by the Secretary, rejects the use of personal protective equipment as a permanent alternative to administrative or engineering controls. The Secretary cites the following record references as showing a rational basis for adopting that position as a safety and health measure:

(9) Even a well-enforced, conscientiously-maintained mandatory ear protection program will present the chance of accidental or incidental exposure to noise in the case of those employees [*57] who may be poorly trained in the use of or improperly fitted with personal hearing protective equipment; similarly subject to exposure are workers who may, out of ignorance or obstinacy simply choose, from time to time, not to wear such equipment. (Tr 714, 718, 733, 738-740, 756, 1005, 1042). To the degree that feasible engineering controls can reduce the possibility of such exposure by shifting primary responsibility for the worker's protection to the employer, there is additional safety and health protection provided to the employee.

(10) Ear plugs and ear muffs may be uncomfortable to wear, particularly over extended periods of time (Tr. 346, 756, 1009-1010, 1018, 1048-1050) Ear protection devices create the possibility that an employee may not hear certain danger sounds (such as that of an approaching forklift truck) when the employee goes from a noisy area to a relatively quiet area of the workplace without removing the ear device (Tr. 755). In addition, ear plugs and ear muffs may produce an allergy or dermatitis (Tr. 756) and dirty ear plugs can generate infections (Tr. 758).

The Secretary concedes that his catalogue of drawbacks does not prevent interim reliance upon [*58] protective equipment, as the standard makes clear. But he insists that they form a rational basis for the decision of the Secretary to compel employers to get rid of the noise rather than try to protect the employee from it. As to the reasons given by the Secretary for rejecting permanent use of ear protection, the respondent presented medical testimony at the trial to show:

(11) a. There are no ear protectors so uncomfortable that they cannot be properly used (Tr. 715). Wearing ear protectors can be compared to wearing glasses as far as discomfort is concerned (Tr. 739).

b. Ear protectors will not cause headaches (Tr. 715, 735-36).

c. The proper hygenic maintenance of ear plugs merely consists of washing them with soap and water at the end of the day (Tr. 759).

d. Ear plugs, even very dirty ones, will not cause infection in a healthy ear (Tr. 758). A dirty ear plug may cause infection in a previously infected or scratched ear, but this can be avoided by dipping the plug in an antiseptic solution (Tr. 758).

e. Ear protectors do not generate psychological problems akin to claustrophobia (Tr. 716, 742-743).

f. In any environment where the ambient noise level is 70 dba [*59] or above, a person can hear and communicate better with personal ear protection than he can without such protection (Tr. 750-753, 721-722).

g. Every employee can be fitted with adequate personal ear protection either of the plug or muff type (Tr. 716-717). Proper fitting of an insert ear plug can be made by a nurse by measuring the ear with a measuring device (Tr. 719).

Respondent concedes that there are employees who dislike ear plugs or muffs and asserts that this creates a necessity for supervision of any hearing protection program.

(12) Continental Can Co., has a mandatory hearing protection program in all locations and no employee was found without hearing protection in the three California plants.

(13) An effective mandatory personal ear protection program is as effective a means for reducing employee exposure to harmful noise as is reducing the ambient noise levels by engineering controls.

(14) An effective mandatory personal ear protection program is more effective in protecting employees from harmful noise than is reducing the ambient noise levels by engineering controls to 90 dba in the respect that ear protection can protect people who need to be protected below [*60] 90 dba; it is more certain protection against noise induced hearing loss especially for those people with tender ears.

(15) Administrative controls are not feasible. However, it is technically possible to develop partial or total enclosures to reduce noise to G-16 limits on Continental Can machines.

The Secretary introduced the testimony of an expert, Dr. Salmon, to demonstrate the feasibility of engineering controls and to support the proposed abatement program.

(16) Dr. Salmon has never been involved in the implementation of a noise reduction program in a can manufacturing plant. He has never seen engineering controls applied to reduce noise in a can plant.

(17) With respect to double die and single die punch presses, Dr. Salmon suggested a partial enclosure consisting of individual enclosures over the major noise making portions of the machine. Dr. Salmon thought that it would take a month to design and two to four weeks to construct and install an enclosure for each machine of the same type.

(18) With respect to the body maker, Dr. Salmon suggested a complete enclosure. He thought it would take one month for design and installation of a prototype including consideration [*61] of visual and manual access to the machine and two weeks to produce and install subsequent enclosures.

(19) Dr. Salmon was engaged by a canner to make recommendations for noise reduction in a single canning plant. He started his study in February 1973 and by the end of January 1974, had not produced a report.

(20) Dr. Salmon admitted that the real problem in reducing noise by enclosures is not development of basic principles, but rather execution of those principles in given situations.

(21) With respect to technical feasibility, Dr. Salmon did not purport to do any engineering but merely testified as to general considerations that he would use in developing enclosures for specific machines.

(22) None of the suggestions for enclosures made by Dr. Salmon guaranteed reduction of the noise at the machine in question to 90 dba. Dr. Salmon's suggested solution to reduce noise on the double die press was represented by him to give only a 5-10 dba reduction; however, the machine's noise level measured by the compliance officer ranged from 97-100 dba. His solutions for the scroll shears ranged from 4 to 15 dba; while the noise level was 98-99 dba. Salmon's suggestion for the single [*62] die press supposedly could reduce the noise 5-10 dba; however, the levels measured by the compliance officer ranged from 96-97 dba. Salmon said his solution for the gang die press would reduce noise 5-10 dba; however, the levels measured by the compliance officer ranged from 94-100 dba. Dr. Salmon testified that an enclosure for the body maker constructed according to his suggestions would effect a 10-15 dba noise reduction; the compliance officer measured the noise in the range 96-102 dba.

Respondent introduced testimony as to its continuing program for the development of engineering controls, and the time and expenditures required to quiet the machines to G-16 levels.

(23) Continental Can has spent almost $400,000 on research and development of engineering controls to reduce noise.

(24) The most prevalent noise-making machines in respondent's Metal Division can be grouped into nineteen families of machines with a total of about 6300 machines. It would cost about an additional $1,230,000 to complete the research and development for prototype enclosures for each family of machines. (There are an additional 280 to 285 specialized machines in the Metal Division which would not [*63] fall within the nineteen families.)

(25) Fabrication and installation of an enclosure for each of the 6300 machines in the nineteen families would cost about $32,000,000.

(26) Yearly maintenance of the enclosures on the 6300 machines would cost about $175,000.

(27) In comparison to the cost of about $33,500,000 for developing, fabricating and installing enclosures and $175,000 a year for maintaining them, a complete self-evaluating hearing protection program using personal protective devices for respondent's entire Metal Division would cost about $100,000 per year.

(28) Continental Can has spent over three years on research and development in noise reduction and has yet to develop an enclosure that would both be successful in reducing ambient noise to 90 dba and be acceptable to operating personnel.

(29) To the best of the knowledge of all expert technical witnesses testifying in this case, no other company in the can industry has succeeded in successfully developing an enclosure for any of its can-making equipment.

(30) Because noise is a problem throughout Continental Can's Metal Division, research and development is being done on a centralized basis in the Company's Technical [*64] Center. The order of selection of machines for research was by a priority list which gave highest priority to the machines that emit the greatest noise and affect the greatest number of workers in order to relieve the greatest number of people.

(31) the priority list consisted of nineteen families of machines with each family including several machines similar in function. The priority list was as follows: Minster (Tab) Press; Bodymaker; Punch Press; Double Seamer; Conveyor; End Liner; Scroll Shear; Lacquer Spray Machine; Edge Cleaner; Process Air; Decorating Press; Decorating Coater; Coil Shear; Air Tester; Welder; Composite Winder; Palletizer; Flanger; and Necking-In Machine.

(32) Continental Can has been pursuing this program for over three years and to date has concentrated on Minster Press, Bodymaker, Punch Press (i.e., single die, double die, gang die) and Double Seamer.

(33) Although over three years have been spent in this research program no solutions have been developed which would both reduce noise to the point where the ambient noise level would necessarily be below 90 dba and provide the necessary access, visibility, safety and convenience during overhaul to be accepted [*65] by operating personnel.

(34) With respect to the Bodymaker (one of the cited pieces of equipment), it was decided, as a result of the initial study, that a total enclosure was required. A prototype was designed, built and tested for some months in the field; it reduced noise to an acceptable level, but needed many changes to permit proper access, adequate visibility and reasonable convenience during overhaul. Therefore, the enclosure must be redesigned. It took from 8 to 12 months to research, develop and install the prototype and 4 to 6 months to test it in the field.

(35) With respect to the double die press (one of the cited pieces of equipment), it was decided to design a partial enclosure. This enclosure reduced noise to 85 dba in the technical center, but when installed in a plant, the dba levels were higher, so that the additive effect would result in levels in excess of G-16. Furthermore, with time the partial enclosure continues to lose effectiveness; therefore, a total enclosure will have to be designed. The prototype that failed took 8 months to develop, 2 weeks to install, and from 4 to 6 months for field evaluation.

(36) The recommendation of Dr. Salmon with [*66] respect to the 314 punch press was virtually identical to the partial enclosure work performed by Continental can starting in 1971 or 1972 and that approach did not work.

(37) Following the present priority list in the manner adopted by respondent, it would take from seven to nine years from the present to complete the installation of enclosures on each of the cited machines. This approach would follow an orderly fabricating and installing process on the machines which are widely used throughout the Company. It would be planned on a master schedule which would involve the least amount of down time, the smallest amount of laid-off personnel, the least amount of transferring of cans from other locations to fill customer requirements while installation is being made.


As interpreted by the Secretary, the cited regulation prohibits the permanent use of personal protective equipment as an alternative to engineering controls, no matter how effective such protective equipment may be to prevent hearing loss and no matter how well administered is the program requiring its use. Further, according to the Secretary, the regulation requires the use of such feasible engineering controls [*67] as already exist even though the noise levels may still exceed the G-16 levels so that ear protection must still be worn. Finally, the Secretary contends that the economic feasibility of engineering controls is not a relevant consideration unless there is such severe economic distress as would involve the closing of the workplace.

Respondent contends that the standard, as interpreted by the Secretary, is invalid, that there are no feasible engineering controls to reduce sound levels to G-16 levels and that a reasonable time for development and implementation of adequate engineering controls would be seven to nine years.

The questions presented are: (1) Whether this Commission has the power to determine questions of the validity of the standard adopted by the Secretary; (2) If the Commission has that power, is the noise standard invalid (a) because it excludes the use of personal protective equipment as a permanent alternative to engineering controls or (b) because it requires the use of controls even though they do not achieve G-16 noise levels or (c) because it excludes considerations of economic feasibility; and (3) If the standard is valid, then what controls are feasible and [*68] what is a reasonable period of abatement.


The power of the Commission. No court has yet decided the question whether this Commission has the power to pass upon the validity of a regulation promulgated by the Secretary. The Commission itself has in fact exercised such power to invalidate a standard on the grounds that it was unenforceably vague; in that case one Commissioner dissented, urging that the Commission lacked such power. Secretary v. Santa Fe Transportation Company, No. 331 (December 20, 1973). It must therefore be concluded that the majority of the Commission is of the opinion that the Commission has the requisite power and authority.

Examination of the Act, and reflection upon the statutory scheme of regulation, supports the view that the Commission has the power and duty to pass upon the validity of the standards which are cited in proceedings brought before the Commission. For the Act sets forth a clear separation of power between the Secretary who was given the power to establish standards and prosecute enforcement actions, and the Commission which was authorized to carry out the adjudicatory functions of the Act. Section 2(b)(3). [*69] The power to adjudicate contested enforcement proceedings, which is accorded to the Commission under section 10(c), necessarily involves the power and duty to determine the validity of the cited standard, for an invalid standard would be a nullity which could not form a lawful basis for affirming a citation.

The scheme of regulation adopted in this Act is fundamentally different on this point from the familiar scheme of other statutes underlying such agencies as the Federal Trade Commission, or the Interstate Commerce Commission. For in such cases the agency possesses both rule making and quasi-judicial powers, and it would be a manifest absurdity for the agency to pass upon the validity of its own regulations. This Commission, on the other hand, is similar to the Tax Court in that it performs only an adjudicatory function in the exercise of which it is inherently sound to consider the lawfulness of the standard promulgated by another, wholly independent agency. n2 This is not to say, of course, that the Commission's determination is final and unreviewable, for section 11 of the Act clearly provides for judicial review of the Commission's decision. However, there is just as much [*70] reason for the Commission to make a reviewable determination of the validity of a standard as there is for the Commission to arrive at a reviewable decision on any other necessary element of an enforcement proceeding.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n2 Emerson Electric Manufacturing Company, 28 T.C. 1090 (1957) is a case in which the Tax Court passed upon the validity of an Internal Revenue Service regulation. And the jurisdiction of the Tax Court to pass upon constitutional questions has been established. See J. Mertens, Jr., THE LAW OF FEDERAL INCOME TAXATION, volume 9, page 57, 50.19.

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

The Act itself supports the view that this Commission is expected to make reviewable determinations of the validity of regulations, for it provides in section 11(a) that no objection that has not been urged before the Commission shall be considered by the court. Accordingly, if the regulation were not challenged before the Commission, it could not be challenged in court.

The view that the Commission can pass upon the validity of regulations is not contrary [*71] to the principle, urged by the Secretary, that at administrative agency lacks the power to pass upon the constitutionality of the very legislation that created it. It is the validity of a standard promulgated under the Act, and not the Act itself, which is brought in issue herein. Moreover, it is unquestionably true, as the Secretary further argues, that this Commission has only the power conferred upon it by statute, that the authority of the Secretary to promulgate standards is exclusive and that the Commission has no rule-making power. In passing upon the validity of the Secretary's standards, this Commission clearly has no authority to consider the wisdom or fairness of the standard, or whether it comports well or ill with the purposes of the Act; the Commission cannot lawfully "second guess" the Secretary. It's only proper function, so far as the standard is concerned, is to determine whether it is within the power of the Secretary to promulgate according to the provisions of the Act and of the Constitution. In performing that function, the Commission does not exercise a rule-making power; rather it exercises a purely adjudicatory power.

The Secretary points out that the [*72] Act provides in section 6(f) a specific method for obtaining review of a challenged standard. It is there provided that any person adversely affected by a standard may, within sixty days after the standard is promulgated, petition a court of appeals for judicial review. It is the view of the Secretary that this method of pre-enforcement review is exclusive. That view, however, is contrary to the legislative history on the matter, for the Senate Report on the bill explicitly recited, emphasis supplied:

Section 6(f) . . . provides that any person who may be adversely affected by a standard may, within 60 days of its issuance seek judicial review in the appropriate United States Court of Appeals. While this would be the exclusive method for obtaining pre-enforcement judicial review of a standard, the provision does not foreclose an employer from challenging the validity of a standard during an enforcement proceeding. n3

Finally, as to legislative history, the Secretary points to a bill that was debated and rejected (H.R. 16785) and which was supposed to have expressly provided for the Commission to review the validity of standards, citing a Commission judge's decision in [*73] Secretary v. Boise Cascade Co., No. 2944 (March 14, 1974). Examination of that decision reveals that H.R. 16785 actually provided for such review by the Secretary and not by the Commission, so the case does not stand for the proposition that Congress rejected the idea of having the Commission pass upon the validity of the Secretary's standards.

- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n3 LEGISLATIVE HISTORY OF THE OCCUPATIONAL SAFETY AND HEALTH ACT, by the Subcommittee on Labor, p. 148. See Senate Report No. 91-1282, p. 8 (91st Cong., 2nd Sess.)

- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -

I conclude that, as a matter of law, this Commission has the power and the duty to determine the validity under the Act and under the Constitution of any standard cited by the Secretary if it be properly challenged in an enforcement proceeding.

The test of validity. It is well settled under the constitution that due process demands that a law shall not be unreasonable, arbitrary or capricious, and that the means selected shall have a real and substantial relation to the object sought to be attained. [*74] Nebbia v. New York, 291 U.S. 502, 525 (1934). At the same time, as the Secretary contends, it is incumbent upon the opponents of a law to demonstrate that it is clearly unreasonable, Arizona Copper Co. v. Hammer, 250 U.S. 400, 426 (1918) Where the legislative judgment is called in question the inquiry is whether any state of facts either known or which could reasonably be assumed affords support for it; the courts will not pass judgment on the wisdom of the law. Railway Express Agency v. New York, 336 U.S. 106, 109 (1954).

A regulation adopted and enforced by an agency pursuant to an act of Congress must not only satisfy substantive due process, it must also conform to the requirements of the statutory grant of authority. H. & H. Tire Co. v. Department of Transportation, 471 F. 2d 350 (7th Cir. 1972). National Tire Dealers and Retreaders Assoc., Inc. v. Brinegar, 491 F. 2d 31 (D.C. Cir. 1974). Further, a regulation is not rationally related to the regulatory purpose just because an agency says so. Some factual basis must appear. Pan American Petroleum Corp. v. F.P.C., 352 F. 2d 241 (10th Cir. 1965). The courts require a showing by the agency that [*75] there is a rational basis for its regulations Coakley v. Postmaster of Boston, Mass., 374 F. 2d 209, 210 (1st Cir. 1967).

In this case the Secretary argues that the noise standard, was an established Federal standard at the time the Act was passed, and that it was therefore legislatively mandated by section 6(a) of the Act, which provides that the Secretary shall within two years after the effective date of the Act promulgate as an occupational safety and health standard any national consensus standard and any established Federal standard. According to the Secretary, the noise standard is therefore virtually unchallengeable because a holding that it was invalid would mean that the statute itself was invalid. I reject this argument because it is plain from a reading of the Act that Congress did not intend indiscriminately to adopt every national consensus standard and every established Federal standard, whether or not they conformed to the purpose and provisions of the Act and of the Constitution. Section 6(a) is itself conditional in that the mandate to the Secretary contains the qualifying phrase, "unless he determines that the promulgation of such a standard would not result [*76] in improved safety or health for specifically designated employees." Moreover, the entire section is governed by the definition of an occupational safety and health standard as contained in section 3(8), which provides, emphasis supplied:

(8) The term "occupational safety and health standard" means a standard which requires conditions, or the adoption or use of one or more practices, means, methods, operations, or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment.

When these provisions are considered in the light of the whole purpose of the Act, which is expressed in section 2(b) as the assuring so far as possible every working man and woman "safe and healthful working conditions", it becomes inescapably clear that any standard adopted by the Secretary must in order to be valid have some relationship to safety and health and must be reasonably necessary and appropriate therefore. I conclude that the test of a valid standard under this Act and under the Constitution is whether it bears a rational relationship to safety and health and whether it is reasonable and not arbitrary or capricious.

The noise standard. [*77] There is no dispute as to the validity of the G-16 standard of permissible noise levels. Also the evidence is uncontroverted that personal protective equipment protects employees from loss of hearing due to the excessive noise levels here involved. The question presented is whether there is a rational relationship between the statutory goal of safe and healthful working conditions and the Secretary's determination to get rid of the noise rather than to protect the employee against it. What rational safety or health basis exists, if any, for compelling the employer to quiet the machines, at great expense, when the employee can be protected by personal devices at relatively little cost?

In answer to interrogatories the Secretary gave purported reasons for rejecting personal protective devices; they included supposed unhealthy or unsafe side effects caused by ear protection such as headaches, claustrophobia, extreme discomfort, causes infection, and causes communication difficulties and difficulties in fitting and maintenance. On brief, the Secretary cited record references to testimony in support of those reasons, and those references are herein before set forth. Examination of [*78] the transcript reveals that the testimony in support of the Secretary's position is insubstantial and unconvincing. On the other hand, there is positive medical testimony that is virtually uncontroverted on the basis of which I find that ear protection does not cause headaches or sensations akin to claustrophobia. It causes no greater discomfort than wearing glasses. Ear protection will not cause infection in a healthy ear and infection can be prevented in a previously injured ear. In any environment over 70 dba, a person wearing ear protection can hear conversation better than without it. Anyone can be fitted with ear protection and the devices are easy to maintain. The alleged drawbacks to personal ear protection have been shown by respondent to be non-existent or lacking in substance.

Aside from the supposed drawbacks of personal protective devices, the Secretary points out that it is difficult to assure worker compliance in a mandatory ear protection program because some workers will choose not to wear the devices, out of ignorance or obstinacy. This point is emphasized by the final reason given by the Secretary for his rejection of personal protective equipment, i.e., [*79] that anything less than getting rid of the noise will "operate improperly to transfer responsibility for employee protection against excessive exposure from employer to the individual worker." This revealing statement indicates that the real reason for requiring engineering controls instead of personal protection is not so much a matter of safety or health as a decision to place the entire burden of protection upon the employer. Such a decision lacks any explicit foundation in the Act and seems contrary to the provision of section 2(b)(2) of the Act to the effect that "employers and employees have separate but dependent responsibilities and rights with respect to achieving safe and healthful working conditions", and the further provision of section 5(b) that "Each employee shall comply with occupational safety and health standards and all rules, regulations and orders issued pursuant to this Act which are applicable to his own actions and conduct." In any case, the evidence here reveals no instance of one of respondent's employees failing to wear protective equipment as required.

When all of the evidence is fairly weighed it is abundantly clear that the relationship between safety [*80] or health and the choice between engineering controls and the use of personal protective devices is insubstantial at most; the use of ear protection undoubtedly affects comfort to some degree, and people probably do prefer to work in a quiet environment rather than wear ear protection, but comfort and preference do not constitute safety and health.

I conclude that the rejection of personal protective equipment as a permanent alternative to engineering controls has not been shown to be rationally related to the purpose of the Act.

Even if engineering controls could be found to afford some advantage over ear protection that was rationally related to safety or health, to be valid, a regulation requiring such controls must be reasonable and not arbitrary or capricious. But the regulation here involved requires engineering controls regardless of cost and regardless of whether they result in G-16 noise levels. Such a regulation is in my opinion arbitrary and unreasonable.

As herein above stated, the evidence shows that ear plugs will attenuate the noise level by 22 to 27 dbA and ear muffs up to 35 dbA. Respondent's employees are therefore protected against noise levels even lower [*81] than the work level of 90 dbA. If engineering controls are required, and if they actually reduce the noise levels to G-16 levels, then the employees will not be required to wear ear protection and their hearing will be protected against any noise level over 90 dbA. The protection thus afforded would be less than at present, and all that would be gained is freedom from wearing ear protection. Yet the cost of so doing is shown to amount to $33,500,000 for design, fabrication and installation, plus about $175,000 per year for maintenance, as compared with about $100,000 per year total cost of a company-wide mandatory program of personal ear protection. Moreover, the evidence shows that this costly engineering program is not certain to produce G-16 levels in the plants; indeed the proof establishes that G-16 levels will not result in some plants, even though they may be attained in some locations. Where G-16 levels do not result, ear protection must still be worn and nothing has been gained except a partial and inadequate lowering of the noise levels. To require as the Secretary does that such inconsequential controls be undertaken without regard to cost is arbitrary to the point [*82] of capriciousness.

I conclude that the regulation, as interpreted by the Secretary, is invalid because it exceeds his statutory authority to adopt safety and health standards that are reasonable and appropriate to provide safe or healthful employment, and because it denies to the employer substantive due process of law as guaranteed by the Constitution.

The citations do not allege that respondent has in any way failed in the requirement for personal ear protection. The charge is that respondent has failed to utilize feasible engineering controls. Since the requirement for controls unlawfully rejects the alternative of effective ear protection and capriciously requires the application of inadequate controls without regard to cost, there is no valid basis for the citations and they must be vacated.

The matter of abatement. If despite all the above considerations it should be held on review that the regulation is valid and that respondent must design and install engineering controls, then the law requires (section 9(a)) that a reasonable time be fixed for compliance, i.e., abatement of the "violation". I find that a reasonable period for the design, development, fabrication [*83] and installation of engineering controls that will result in G-16 sound levels in respondent's plants is seven years.

The testimony of Dr. Salmon does not support the shorter, two year period proposed in the citation because Dr. Salmon only proposed elementary enclosure devices which would not result in G-16 sound levels. Dr. Salmon made no overall estimate of the period of time required to complete the program to the point where G-16 levels would be assured. On the other hand, respondent's witnesses described a complete, and orderly program that would meet the G-16 standard, and there is no evidence to controvert the resulting testimate of seven to nine years for completion. In this connection, the Secretary points on brief to purported settlements reached with other can companies which provide for a shorter period for abatement. Such settlements are not of record and have no proper bearing on this case. Aside from the proprieties, however, it may reasonably be observed that such settlements are subject to later petition for modification of abatement; it is wholly possible that the difficulties which this respondent foresees may actually be encountered by others as well and [*84] that their agreed target dates may turn out to have been unduly (if not intentionally) optimistic and unobtainable. In all fairness, if this respondent is going to be required to invest millions in engineering controls it should at least be given the time which the evidence shows is reasonably necessary to do the job.

In prescribing an abatement period, it should also be noted that there is no basis in the law for requiring monthly or any other reports of progress to be made by respondent to the Secretary. The law requires only that respondent comply with the regulation, and neither the law nor the regulation says anything about such reporting. The requirement for such reports amounts to an unwarranted interference by government in the performance by respondent of its statutory duty. There is no basis for the Secretary to look over respondent's shoulder as it complies with the law; the Secretary introduced no evidence to show that such supervision would in any way promote or insure compliance. On the other hand the submitting of one more report cannot but increase the expense of doing what respondent has to do on its own anyway. If respondent is going to be required to install [*85] engineering controls it should be left free to do so in its own way and without interference, so long as it makes every effort to do so in good faith. I find on this record that respondent has thus far conducted its engineering program in complete good faith and that there has been no showing that it has failed to implement adequate engineering controls as soon as they could be developed.

It is ORDERED that the citations in these proceedings, to the extent that they charge failure to utilize feasible administrative or engineering controls, be and the same are hereby vacated and that these proceedings be and the same are hereby terminated.

Robert N. Burchmore, Judge OSAHRC

October 1, 1974




Sound level

dBA slow

Duration per day, hours












1 1/2






1/4 or less


- - - - - - - - - - - - - - - - - -Footnotes- - - - - - - - - - - - - - - - - -

n1 When the daily noise exposure is composed of two or more periods of noise exposure of different levels, their combined effect should be considered, rather than the individual effect of each. If the sum of the following fractions: C 1/ T 1+ C 2/ T 2 Cn/Tn exceeds unity, then, the mixed exposure should be considered to exceed the limit value. Cn indicates the total time of exposure at a specified noise level, and Tn indicates the total time of exposure permitted at that level.

Exposure to impulsive or impact noise should not exceed 140 dB peak sound pressure level.


- - - - - - - - - - - - - - - - -End Footnotes- - - - - - - - - - - - - - - - -