September 18, 2001

Mr. Stephen E. Sandherr
Chief Executive Officer
The Associated General Contractors of America
333 John Carlyle Street
Suite 200
Alexandria, VA 22314

Dear Mr. Sandherr:

Thank you for your letter of August 27, 2001. Let me first thank AGC for continuing to work with OSHA to advance safety and health at the nation's construction sites through partnering programs.

I believe that several of the issues you raise arise largely from a misunderstanding of what was stated both in OSHA's proposed and final rules for steel erection, which were the product of the negotiated rulemaking process.

Fall protection for connectors and deckers

(1) You state that, "[i]n section 1926.760, OSHA gives an exception to connectors that allows them not to use their fall protection equipment while working at heights between 15 and 30 feet. In addition, OSHA allows workers engaged in decking in a controlled decking zone to work without conventional fall protection at heights between 15 and 30 feet. In sharp contrast, the proposed rule required all employees to tie off at 15 feet across the board . . ." [Emphasis added].

The proposed rule did not require that deckers in a controlled decking zone and connectors use fall protection equipment between 15 and 30 feet. In the proposed rule, §1926.760(b)(3) stated that connectors shall "[b]e provided, at heights over 15 and up to 30 feet above a lower level, with a personal fall arrest or fall restraint (positioning device) system and wear the equipment necessary to be able to be tied off; or be provided with other means of protection from fall hazards in accordance with paragraph (a)(2) of this section." Note that this provision does not require the worker to tie-off. There was a similar exception for deckers in §1926.760(c). See volume 63 of the Federal Register, page 43508. With slight wording changes, these are the same requirements found in the final rule.

6 feet vs. 15/30 feet

(2) You note that the fall protection threshold requirements in the final rule are higher than those in Subpart M, and that AGC believes that all workers should be tied-off at 6 feet.

This was a very controversial issue in the negotiations. Only after protracted discussions and numerous meetings of the negotiated rulemaking advisory committee (SENRAC) did all the committee members, including the member representing the AGC, agree to what is in the proposed rule. A 6-foot threshold was specifically discussed, and ultimately rejected. The reasons for adopting the threshold heights used in the proposed rule and the final rule are discussed in the preamble to the final rule. See 66 FR 5243 - 5249.

In deciding against adopting a 6-foot fall protection threshold, we stated:

SENRAC evaluated whether the trigger height in steel erection should be different than that in subpart M and concluded that it needed to be higher . . . The special circumstances of steel erection can make conventional fall protection very difficult to deploy below 15 feet. For many steel erectors, especially connectors, the work starts at the top level of the structure. This means that anchor points above foot level are often limited or unavailable. Because of the nature of the structure, the available fall arrest distance is usually about 15 feet.

66 FR 5244 - 5245.

Shear Connectors

(3) You assert that, in highway bridge work, it is safer to install shear connectors in the shop than in the field: "Unlike building construction, the fall distances from bridges are not mitigated by floors below. In addition, bridge work is often done over operating roadways, further increasing the hazards to workers."

First, the Agency did not consider the presence of a floor that is more than 15 feet below a worker to be a primary means of fall protection. Second, throughout the rulemaking, SENRAC members stated that the tripping hazard posed by pre-installed shear connectors was a greater hazard than that posed by exposure to falls while installing them from a deck. In fact, the final standard prohibits installation of the connectors until "after the metal decking has been installed, using the metal decking as a working platform." It also prohibits installation while in a controlled decking zone (CDZ), since the decking is not yet completely installed in a CDZ.

Multi-employer worksites

(4) The final rule contains several specific duties that have been placed on the controlling contractor. We stated in the preamble to the final rule that these duties are in addition to those placed on controlling contractors by the agency's multi-employer policy. You assert that "in promulgating the revised standard the agency has attempted to bootstrap its informal multi-employer citation policy into a set of principles having the binding effect of a standard . . . The [multi-employer] policy generally places primary responsibility for compliance on the superior entity, such as the owner, construction manager or general contractor..." [Emphasis added].

OSHA does not have an "informal" multi-employer policy. The policy is explained in both Compliance Directive CPL 2-0.124 and in the Agency's Field Inspection Reference Manual. Referring to the policy in the preamble of a standard does not convert it to a regulation. The purpose of the discussion in the preamble was simply to point out that controlling contractors have both the duties specified in the standard and those that arise under the multi-employer policy.

The Agency's multi-employer policy does not place "primary" responsibility for the safety of subcontractors' employees on the controlling contractor. To the contrary, the current policy makes clear that the primary responsibility rests with an employee's employer. The policy specifically states that:

The extent of the measures that a controlling employer must implement to satisfy this duty of reasonable care is less than what is required of an employer with respect to protecting its own employees. This means that the controlling employer is not normally required to inspect for hazards as frequently or to have the same level of knowledge of the applicable standards or of trade expertise as the employer it has hired.

Inconsistencies

You state that several provisions are "either inconsistent within Subpart R or with other OSHA standards. AGC believes that OSHA needs to address these inconsistencies, so that the final rule will be consistent with its other standards."

I will address each of your points in turn:

(1) "The definition of 'opening' should either be removed or conform with the definition in Subpart M.... Any other 'gap'...should be identified as a hole."

This point was raised by several commenters during the rulemaking. In the preamble to the final rule, OSHA acknowledged the difference in the use of the terms "hole" and "opening" in Subpart R as opposed to Subpart M, and explained that the use of these terms in the steel erection industry is different than in the rest of the construction industry. We decided that more confusion would result if we were to try to harmonize the terms than to stay with those terms as they are currently used in the industry. See 66 FR 5219.

(2) In this next point, you state that "section 1926.752(c)(2) prohibits the steel erection contractor from swinging a load over other workers." You note that this cannot be achieved in a steel erection worksite, and that the provision is inconsistent with §1926.753(b), which you say allows employees to work under suspended loads.

It is unclear which provision you are referring to; §1926.752(c)(2) in the final rule does not address overhead loads (it deals with providing a firm properly drained storage space for materials and equipment). I assume you are addressing §1926.752(d), which provides that "all hoisting operations in steel erection shall be pre-planned to ensure that the requirements of §1926.753(d) are met."

Section 1926.753(d) ("working under loads") requires that "routes for suspended load shall be pre-planned to ensure that no employee is required to work directly below a suspended load except for: (i) employees engaged in initial connection of the steel; or (ii) employees necessary for the hooking or unhooking of the load." This was carried over from an identical provision in the proposed rule.

The extent to which avoiding overhead exposure is infeasible was discussed in the rulemaking. In the preamble to the final rule, we explained that, in SENRAC's view, the two exceptions that were built into the standard should take care of those situations where overhead exposure is unavoidable. See 66 FR 5209-5210.

(3) "In §1926.753(a)(1)(v)(2) there is no definition for 'qualified rigger.' Do you take the term 'qualified person' and apply it to the term 'rigger' whatever that may be?"

We are not sure which provision you are referring to, since the cited provision does not exist in the final rule. In §1926.753(c)(2), the final rule defines a qualified rigger as "a rigger who is also a qualified person." In §1926.751, a qualified person is defined as "one who, by possession of a recognized degree, certificate, or professional standing, or who by extensive knowledge, training, and experience, has successfully demonstrated the ability to solve or resolve problems relating to the subject matter, the work, or the project." So, according to the final rule, a qualified rigger is a rigger who meets the qualified person criteria.

(4) You object to §1926.753(a)(1)(i), stating that "critical lifts are not isolated to steel erection and should be addressed in Subpart N. [This section] is not specific to steel erection and is redundant to subpart N . . .Any inspection requirements for cranes should not be addressed but should remain consistent in [Subpart N]."

The provision you refer to is a provision in the proposed rule regarding pre-shift visual inspection of cranes. The equivalent provision in the final rule is §1926.753(c)(1)(i). The provision applies to all cranes used in steel erection, not just those used for a "critical lift."

OSHA's crane standard, Subpart N, has not been changed since 1971. In SENRAC's view, additional protections are necessary, especially in light of new crane technology and the standard's allowance of routine use of personnel platforms. See 66 FR 5208. Since SENRAC was authorized to deal only with the steel erection rule, it could not recommend changes to Subpart N. Updating Subpart N would have required a separate rulemaking. The Agency is currently evaluating whether to undertake such a project.

You also object to the provision because "many times hoisting equipment is not owned or controlled by the steel erector. This can create questions about inspection authority on multi-employer construction sites. Leaving inspection requirements in Subpart [N]...assures that the responsibility and authority remains with the entity that controls the hoisting equipment."

Outside of the steel erection context, the "responsibility and authority" to comply with the inspection requirements in Subpart N is not limited to who controls the equipment. All employers with responsibilities pursuant to the multi-employer policy have obligations (to varying degrees, as explained in the policy).

That is unchanged under the steel erection standard. The fact that employers other than the one who controls the equipment have responsibilities does not relieve the employer who controls the equipment of its responsibilities. Note that an employer who controls a crane used in steel erection would be "engaged in steel erection" and therefore required to comply with Subpart R.

(5) You note that the steel standard, unlike Subpart N, allows the routine use of personnel platforms. During the rulemaking, commenters objected to this aspect of the steel standard. The Agency relied on SENRAC's reasoning and expertise in adopting this provision, which is explained in detail at 66 FR 5209.

(6) "Section 1926.75(f)(1) [sic] should be renamed 'perimeter cable guardrails' to be consistent with Subpart M §1926.502(b). Cable guardrails should meet the requirements of Subpart M...§1926.502(b)(9). This provision should also be moved to §1926.760."

You are apparently referring to the provision in the proposed rule on perimeter guardrails. In the final rule, we did require that perimeter cables comply with Subpart M, [§1926.502]. See §1926.760(d)(3). We also moved the requirement for perimeter cables to §1926.760. See §1926.760(a)(2). These changes were discussed at 66 FR 5250.

(7) "The wording of §1926.757(a)(10), 'unless written direction to do so is obtained from a qualified person' should be changed to 'without approval of the structural engineer,' to remain consistent with all of the other requirements."

This is a reference to a provision in the proposed rule that provided: "Steel joists and steel joist girders shall not be used as anchorage points for a fall arrest system unless written direction to do so is obtained from a qualified person." That was carried over into the final rule in §1926.757(a)(9). As explained in 66 FR 5236-5237, OSHA specifically considered whether to require approval from the structural engineer of record. We decided not to do that because "the SER [structural engineer of record] may not have the knowledge of steel joist erection necessary to approve tie-offs to joists. The qualified person, however, as defined is the appropriate entity to make the determination." 66 FR 5237.

(8) "Section 1926.759(b) is redundant to §1926.752(c). As already noted, the hoisting of materials on a multi-function construction site may require that loads be moved over areas where workers may be located. The movement of loads over personnel is a general site condition and not specific to steel erection. The section should be modified."

You are apparently referring to provisions in the proposed rule, rather than the final rule. Section 1926.759(b) in the proposed rule was titled "Overhead protection," and stated that the controlling contractor was to ensure that no other construction processes take place below steel erection unless adequate overhead protection was provided. Section 1926.752(c) in the proposed rule was also titled "Overhead protection," and stated that the controlling contractor was to pre-plan hoisting operations to comply with §§1926.753(b) (working under loads) and 1926.759 (falling object protection).

Your point is similar to comments received during the rulemaking, in which it became apparent that there was confusion about the purpose of §1926.759(b). As explained at 66 FR 5243, the purpose of §1926.759(b) was to ensure that workers who are below steel erection activities are protected from falling objects other than hoisted materials. Consequently, in the final rule, the title of §1926.759(b) was changed to "Protection from falling objects other than materials being hoisted." Also, §1926.752(c) in the proposal, which you refer to above, was moved to §1926.752(d) in the final rule, and its reference to §1926.759 was deleted.

(9) "Section 1926.760(c)(3) should conform to Subpart M §1926.502(g)(3), and not just...suggest compliance with a non-mandatory appendix."

Several commenters argued in the rulemaking that control lines should be required to meet the Subpart M requirements; the reasons for the agency's decision are discussed in the preamble at 66 FR 5247:

In the final rule, OSHA has made the provision more consistent with subpart M where possible. A new paragraph was added to subpart R's appendix D regarding flagging or marking of the control line with highly visible material. The only remaining difference in the control line requirements is the allowable distance from the leading edge. A control line for a controlled decking zone is to be erected not more than 90 feet from the leading edge, while the maximum distance permitted in Subpart M is 25 feet. The longer maximum distance in Subpart R is needed because of the size of the bays that are decked.

(10) "In §1926.760(e) OSHA should not establish contractual terms."

As explained in the preamble to the final rule, the controlling contractor is not obligated to accept responsibility for fall protection equipment. It may choose not to accept that responsibility; the standard simply requires that if responsibility is not accepted, the controlling contractor must make sure that the fall protection equipment is removed. See 66 FR 5250-5251.

At times there may be unusual instances where a limited reopening of a negotiated rule is necessary. However, in general I believe that the Agency should avoid revisiting the issues that were addressed in a negotiated rulemaking after the negotiating committee members have debated them, agreed on a proposed rule, and the rulemaking process has been completed.

I hope the explanations above clarify what the final rule requires and the extent to which these points were discussed and addressed in the rulemaking. All interested parties considered the data and issues associated with the points that you have raised. It does not appear that there is any new information involved. In these circumstances, I do not think that it would be appropriate to reopen the rulemaking. I would, of course, be happy to meet with you regarding this matter.

Sincerely,



John L. Henshaw
Assistant Secretary