About | Contact | JOIN

NRCA's comments regarding OSHA's Advanced Notice of Proposed Rulemaking for fall protection in the construction industry, January 2000

On Jan. 24, NRCA submitted comments to the Occupational Safety and Health Administration (OSHA) opposing the agency's Advanced Notice of Proposed Rulemaking (ANPRM) for fall protection in the construction industry. NRCA's comments follow.
January 24, 2000

OSHA Docket Office
Docket S206C, Room N2625
U.S. Department of Labor
Washington, D.C. 20210

Attention Docket Clerk:

The National Roofing Contractors Association (NRCA) is pleased to submit the following comments regarding the Occupational Safety and Health Administration's (OSHA's) Advanced Notice of Proposed Rulemaking (ANPRM) for Fall Protection in the Construction Industry. The roofing industry has numerous concerns about the far-reaching effects of potential new, onerous regulations that will neither enhance overall worker safety nor provide a practical solution to the complex issue of fall protection in the roofing industry. NRCA stands ready to discuss the issue further with OSHA and all interested parties.

About NRCA

NRCA was formed in 1886, and its primary mission is to represent and serve the interests of professional roofing contractors, as well as promote professionalism in the roofing industry. Professional roofing contractors strive to meet the highest standards for installing high-quality roof systems, using equipment and practices that protect workers, building occupants and the public from significant hazards. Professional roofing contractors recognize that worker health and safety are of paramount concern to their companies and employees.

NRCA is one of the oldest construction trade associations in the United States, celebrating its 114th year. It represents more than 4,600 members in 51 countries throughout the world. Its 3,500 U.S. contractor members represent approximately 65 percent of all roof system installations in the United States. More than 90 percent of roofing contracting firms are considered small businesses; many are family-run operations where family members work side by side to make their companies successful. Roofing contractors are keenly aware of the need for a safe workplace not only because safety makes good business sense but also because they care about their workers. Most roofing contractors have worked on roofs and fully understand the hazard. Often their sons, daughters, nieces or nephews work on roofs, too.

NRCA member concerns

NRCA's members who perform residential (steep-slope) roof system installations strongly and repeatedly voiced their dissatisfaction and frustration with OSHA's 1994 Subpart M when it was published and, more recently, reaffirmed their concerns about OSHA's ANPRM, particularly the suggestion in the ANPRM that OSHA may reinstate, perhaps without any rulemaking proceedings, the requirement for nearly universal use of what OSHA calls "conventional" fall protection systems in residential roofing work. The issue of fall protection is extremely important to our members and one that NRCA is uniquely qualified to address. NRCA is providing the following comments in an effort to assist OSHA promulgate a fall-protection rule for roofing work that is practical, feasible and effective.


    1.  Development of the 1994 OSHA Rule

On the very important subject of fall protection, NRCA and OSHA have had a long and constructive relationship. Both have recognized that roofing work can be dangerous and safety is of the utmost importance. In fact, when OSHA promulgated the first Subpart M, "Floor and Wall Openings," in 1971, it interpreted the standard as applying to low-pitched roof perimeters. However, the original standard created confusion among OSHA compliance officers and roofing contractors regarding the definition of a floor. In 1975, it was determined that a floor is not a roof and the standard could not be appropriately applied to roofing work. During the next five years OSHA and NRCA worked together to develop a roofing-specific standard. In November 1980, OSHA published a final rule, "Guarding of Low-Pitched-Roof Perimeters During the Performance of Built-Up Roofing Work." It introduced a warning-line system and safety-monitoring system as practical and effective fall-protection methods.

In November 1986, OSHA reopened Subpart M in order to update it. Eight years later, based on a rulemaking record that had largely been assembled during the mid-1980s, the final rule was published. OSHA's August 1994 Fall-Protection Standard included a dramatic change in the height at which the standard's requirements were triggered—from 16 feet to 6. In addition, steep-slope roofing fall-protection options were restricted to what OSHA called "conventional" fall-protection systems—guardrails with toeboards, safety nets and personal fall-arrest systems with snap hooks at each end. The standard introduced a specific section for residential construction that included the option for contractors to develop site-specific fall-protection plans when other "conventional" methods were shown to be infeasible or create a greater hazard. Roofing work, however, was not given this option.

OSHA's 1994 rule engendered controversy immediately. NRCA was inundated with phone calls from residential roofing contractor members expressing their serious concerns with the feasibility and practicability of the new rule. Specifically, the new 6-foot rule required fall protection for virtually every roofing job in the United States and OSHA's fall-protection options could not reasonably be implemented. Roofing contractors voiced serious concerns that the conventional fall-protection measures mandated by the rule would introduce new and serious safety hazards to the roofing workplace while doing little to reduce existing fall hazards. In addition, because of the highly competitive nature of the roofing contracting market and the heavy costs of compliance with the new rule, the rule was likely to increase the market share of "rogue" contractors who tend to disregard both legal and industry quality and safety standards and thus are often able to significantly underbid professional roofing contractors. As a result, the new standard not only would have forced at least some professional contractors out of the residential (steep-slope) roofing business, it also would likely have significantly increased safety hazards across the industry because of the increased penetration of rogue contractors using unsafe and substandard roofing practices. In short, implementation of the new standard would have done a great deal more harm than good.

As discussed below, although OSHA initially denied a petition submitted by NRCA to reopen or reconsider the 1994 Standard, the Agency ultimately did recognize that the rule had serious shortcomings and issued this ANPRM. OSHA initially attempted to address these problems by immediately issuing a series of no less than six "interpretations" and "clarifications" of the standard. In the end, however, after roofing industry concerns about the rule continued to be raised both to the Agency and in Congressional oversight hearings, OSHA agreed to expeditiously reopen the rule and, to provide interim relief, agreed further to issue a compliance directive (Instruction STD 3.1), which permitted roofing contractors to use alternative fall-protection measures, specifically slide guards and safety monitors, on many residential (steep-slope) roofing jobs.

    2.  Immediate Efforts to Address Industry Concerns

After denying NRCA's petition, OSHA offered to meet with NRCA to discuss the areas of concern and possible measures to address them. As a result of subsequent correspondence, OSHA issued a clarification stating that, among other things, the residential construction portion of the rule could be applied to residential (steep-slope) roofing work. As a result, roofing contractors were allowed to use site-specific alternative fall-protection plans on jobs where a specific demonstration of the infeasibility of "conventional" fall-protection systems was made.

Later that month, NRCA invited Dale Cavanaugh from OSHA's compliance office to speak at NRCA's annual convention to address questions regarding the new standard. During the program, he stated that a safety monitor can have other duties while acting as safety monitor as long as the monitoring function is not compromised. NRCA had requested that this and additional answers be documented as formal OSHA policy so that contractors could rely on them at job sites when a compliance officer questioned the modification. As a result of that request, Mr. Cavanaugh asked NRCA to submit the questions NRCA raised to OSHA to obtain official interpretations.

NRCA submitted questions and, as a result, OSHA issued four additional interpretations that addressed concerns the roofing industry raised with respect to the 1994 Standard. They are: Interpretation M-1 addressing whether Subpart M would apply to the delivery of product to a construction site; Interpretation M-2 addressing the Subpart M obligations placed on a supplier of roofing materials; Interpretation M-4 addressing which Subpart M provisions apply to work performed on a home; and Interpretation M-6 addressing whether Subpart M requires the hiring of an individual to be act solely as a safety monitor.

In April 1995, an NRCA contingent met with a delegation of OSHA representatives, including Mr. R. Bruce Swanson, Mr. Roy Gurnham, Ms. Barbara Bielaski, Mr. Ted Twardowski, Mr. Dale Cavanaugh, Mr. Tom Seymore and Mr. Stephen Jones, in a continuing effort to obtain more reasonable fall-protection solutions, specifically to ease the burden of the regulation on the residential (steep-slope) roofing industry and help ensure that the new regulatory regime would in fact provide significant worker safety benefits. OSHA offered to participate in a joint task force to develop generic fall-protection plans to address the unique fall hazards presented during residential (steep-slope) roofing work. It was OSHA's intent to include these generic residential fall-protection plans as additional appendices to Subpart M, similar to Appendix E.

In July 1995, OSHA issued another clarification explaining that fall-protection plans do not have to be site-specific. NRCA submitted its final fall-protection plan to OSHA that included support and comments from various groups, including the National Association of Homebuilders (NAHB).

    3.   Issuance of the 1995 Interim Fall-Protection Guidelines

Despite these efforts to address the deficiencies in the 1994 Standard, the public outcry over the Standard continued to mount and soon drew the attention of Congress. The House Small Business Subcommittee on Regulation and Paperwork held a hearing on June 15, 19951, and the House of Representatives passed an appropriations rider proposing to prevent OSHA from enforcing the 1994 Standard 2. When the issue came to the Senate for consideration, an agreement was reached between OSHA and interested parties that forestalled the possible legislation. As set forth in a November 2, 1995 letter to OSHA Administrator Joseph A. Dear from Senators Harry Reid, Jon Kyl and Don Nickles, the agreement in principal called upon OSHA, among other things, to:

  1. Publish an interim compliance directive allowing the use of alternative fall-protection measures—specifically, the use of slide guards on roofs with slopes of up to and including 8-in-12.

  2. Promptly reopen the rulemaking proceedings on the 1994 Standard; and

  3. Incorporate the roofing appendix that had been developed jointly with NRCA, together with the provisions of the compliance directive, into the body of the 1994 Standard 3.

Accordingly, on December 8, 1995, OSHA issued Instruction STD 3.1: "Interim Fall Protection Compliance Guidelines for Residential Construction," signed by OSHA Administrator Dear. As it applied specifically to the roofing industry, OSHA adopted the input NRCA provided regarding the use and installation of roofing slide guards and safety monitors on roofs with slopes of 8-in-12 or less in pitch and 25 feet or less in eave height. STD 3.1 provided a definition of residential construction that included structures where residential-type products and installation methods are used. It also provided exceptions for tile and metal roofs4.

Describing the reasons for issuing this Directive, OSHA, "acknowledge[d]" that the concerns that had been raised by NRCA about the benefits and impacts of the 1994 Standard "merit further consideration by the Agency." (STD 3.1, § G, emphasis added.) The Agency went on to state:

"Therefore, OSHA has determined that it is appropriate to initiate further proceedings regarding Subpart M so that the Agency can re-evaluate what constitutes adequate fall protection for various construction operations. OSHA is moving expeditiously to develop an NPRM. In particular, further proceedings will provide an opportunity to update OSHA's information regarding the fall hazards to which residential construction workers are exposed. This, in turn, will enable the Agency to evaluate the alternative methods and procedures suggested by roofing contractors for residential roofing work." (STD 3.1, § G, emphasis added.)
After the Directive was issued in 1995, NRCA worked hard to assist its members to comply with the 1994 Standard as modified by the Directive and continued NRCA's traditional role in developing and disseminating safety information, programs and materials for the roofing industry. In addition, NRCA has worked closely with OSHA and the United Union of Roofers, Waterproofers and Allied Workers (the Roofers Union) toward the common goal of improving worker safety and developed the Roofing Industry Partnership for Safety and Health (RIPPSH). In fact, RIPPSH is the first partnership that includes labor, government and management in the construction industry working jointly to promote and reward safety. NRCA's contractor members support and encourage the development of these safety-related programs. Roofing workers are vital to the industry's success and, unfortunately, there already is a serious industry labor shortage. The industry cannot spare even one worker and takes the well-being of all workers seriously.

It is with this long history of dedication to the improvement of worker safety on the roof and working with OSHA that NRCA provides the following comments and information in response to the issues and questions raised in the July 1999 ANPRM.

General comments about the 1994 fall protection standard

NRCA welcomes the reopening of the fall-protection rulemaking that OSHA agreed to undertake over four years ago. Based on the substantial input we have received from our members since the 1994 rule was published, and particularly in response to the ANPRM, it is clear that more work needs to be done to develop a practicable standard that adequately protects residential roofing workers from fall hazards without introducing significant new hazards to roofing work. NRCA is prepared to continue working with OSHA, the Roofers Union and other interested parties to develop the credible data needed to identify and support the best solutions.

NRCA sees the ANPRM as the first step in a process that can lead to that outcome. But to reach that goal, the formal rulemaking proceedings must be broad enough in scope to allow OSHA and the stakeholders in the process to develop and address the issues and evidence that must be resolved to make the best choice for worker safety, while maintaining the climate of cooperation that, we believe, currently exists. As developed below, NRCA believes that there are three principal ingredients to a successful rulemaking on the important issue of fall protection: (1) conduct a detailed evaluation of the effectiveness of the fall-protection measures permitted in the 1995 Compliance Directive; (2) develop an adequate assessment of the net benefits of the "conventional" fall-protection measures contained in the 1994 Standard as applied in residential (steep-slope) roofing work; and (3) maintain the status quo—keep the 1995 Directive in effect—until the rulemaking has been completed.

    1.  The Need to Evaluate the Effectiveness of the 1995 Compliance Directive

In reviewing the significant input NRCA has received from its members, particularly since the publication of the ANPRM, one point stands out more that any other. There is a broad consensus that the measures required by the 1995 Compliance Directive—especially the slide guard provision—are not only workable and practical for the great majority of residential roofing work, but also have yielded noticeable safety benefits. NRCA intends to investigate this issue among its membership more systematically in the near future. (In fact, NRCA is in the final contract negotiation stage with the National Safety Council to conduct a comprehensive prospective roofing falls study.) However, these assessments by NRCA's members are by themselves an important indication that the scope of the forthcoming reconsideration proceedings regarding OSHA's fall-protection standard must include a detailed and comprehensive assessment of the adequacy of the controls allowed in the 1995 Directive.

As noted above, the Agency made this point when it issued the 1995 Directive. And the agreement OSHA reached with interested parties, which led to the issuance of the Directive, also contemplated at least a genuine consideration of the merits of the fall-protection measures permitted by the Directive. In fact, it would appear that the agreement reflected a consensus that, in the absence of persuasive new evidence to the contrary, the provisions of the 1995 Directive should be incorporated into the fall-protection standard.

In any case, we now have more than four full years of practical experience industry-wide from the use of these "interim" measures. Given their apparent success, at least as indicated by the evaluations of some NRCA member contractors, a systematic evaluation of the interim measures' effectiveness certainly would make good sense and is in any case necessary to achieve the worker safety criteria of the Occupational Safety and Health Act.

    2.  The Need for an Adequate Assessment of the Net Safety Benefits and Feasibility of the 1994 Standard.

      a.  Net Safety Benefits

There continues to be serious concern within the roofing industry that requiring what OSHA calls "conventional" fall-protection measures for essentially all residential (steep-slope) roofing work will do a great deal more harm than good. Stated broadly, the concern is that the rule could well have such an untoward effect both by effectively increasing the safety hazards of roofing work and forcing some roofing contractors—specifically, professional roofing contractors who are committed to doing quality roofing work safely—out of the residential roofing market. Devising a well-conceived rule that provides the fall protection for residential roofing work will require additional analysis of three major sets of issues.

First, as discussed below, the experience of NRCA members using conventional fall-protection systems reveals that the use of these systems introduces new hazards to the roofing workplace and has resulted in injuries to roofing workers. NRCA believes that this important issue has not been adequately addressed and should be one of the major issues in a reopened rulemaking.

Second, NRCA's members have expressed strong reservations about the net benefits of the 1994 rule because it ignores the competitive structure and practical realities of the residential roofing industry. Based again on experience with the use of conventional fall-protection systems, contractors have indicated that mandating of the use of these systems for virtually all residential (steep-slope) roofing work will effect a substantial competitive realignment within the industry, with professional roofing contractors losing significant market share to rogue contractors and others who, thanks to the heavy compliance costs of OSHA's rule, will be better able to underbid professional contractors who endeavor to comply.

The impacts of such a realignment go well beyond the bankruptcies and layoffs that would directly result—although NRCA would hope that these effects alone would be sufficient to cause OSHA to seriously question the merits of the rule. Rogue contractors are likely to disregard not only the fall-protection standard but all or most industry and regulatory worker-safety standards, as well as disregard good roofing practices. As a consequence, increased penetration of rogue contractors in the market would almost certainly lead to an across-the-board increase in worker safety hazards and an increased incidence of poorly constructed roof systems. And a roof system that fails can present a risk of severe injury to building occupants and heavy damage to building contents.

Third, NRCA believes that additional data and analyses are needed to support a well-informed judgment of whether the conventional fall-protection systems the rule requires will achieve the significant safety benefits OSHA has projected. As discussed in NRCA's comments during the 1994 rulemaking, the data OSHA relied on at that time—which remain the only possible basis for a rule—show only that there are too many injuries due to falls from roofs, a proposition that no one, including NRCA and the roofing industry, seriously disputes. What the data do not reveal, however, are the underlying reasons for the falls that occur in roofing work. NRCA members continue to explain in no uncertain terms that the overwhelming majority of fall incidents are the result of failures to adhere to existing safety requirements or use simple common sense. There certainly is no convincing evidence that the use of conventional fall-protection systems would substantially reduce the current rate of fall incidents in the roofing industry.

      b.  Feasibility

Quite apart from the indications that the 1994 Standard would have yielded significantly negative net safety benefits, there remain serious concerns that the Standard would not be feasible in the residential roofing industry. As indicated above, the Standard would likely have driven a significant number of professional roofing contractors out of the residential roofing market, resulting in bankruptcies and employment dislocation. More broadly, it would have led to a wholesale change in the competitive structure of the industry, benefiting rogue contractors, who would prosper while professional contractors attempting to comply are forced out. This kind of economic restructuring is certainly a sufficient ground by itself to conclude that the 1994 Standard was economically infeasible.

A related but legally distinct deficiency with the 1994 standard was impossibility of compliance. Roofing work requires largely manual labor, which means of course that worker adherence to required safety measures is necessary if contractors are to hope to comply. The input we have received from our members since the 1994 Standard was published confirms the industry predictions that were made five years ago. Specifically, there is powerful worker resistance to the use of "conventional" fall-protection systems in residential roofing work. This observation comes even from NRCA members who have used conventional systems in recent years and thus have seen the results first-hand.

There are several reasons for the high degree of resistance seen in many roofing workers.

First, a great deal of residential roofing work is paid at a "piece" rate, i.e., by the job, not by the hour. Anything that significantly increases the amount of time needed to complete a job that is paid at piece rates takes money out of workers' pockets.

Second, actual experience in the field with personal fall-arrest systems confirms that use of the equipment involved, particularly body harnesses, is extremely uncomfortable, particularly in hot weather (when most roofing work is done), and can also be painful. It is no answer to point out that alternative "conventional" fall-protection measures are allowed under the 1994 Standard, because guardrails and safety net systems are simply not a practicable option for most residential structures, including nearby landscaping, that are involved in this rulemaking.

Third, roofing workers resist personal fall-arrest systems because they are unnecessary. This response is not, as some seem to believe, born of an unjustified "bravado." Roofing workers learn from their training and first days on the job that anything that unnecessarily prolongs the amount of time they spend on a roof or creates an obstacle to moving about or carrying out their tasks, creates a hazard that should be avoided if possible. The hazards created by these circumstances include not only falls but a host of other hazards, including trips and burns. Thus, workers know that the risk of a fall on all but steeply sloped roofs can be minimized by using proper footwear and common sense, and their resistance to the additional time demands and obstacles introduced by personal fall-arrest systems is certainly grounded in reason, experience and common sense.

Whatever the underlying causes for worker resistance to these requirements of the 1994 Standard, that resistance makes compliance with the Standard impossible in many instances, because contractors have little or no leverage over employees who refuse to comply in many markets. The most severe sanction available to the contractor is terminating the worker, but in far too many instances the worker can immediately find a job with another contractor who does not enforce compliance with equipment requirements the worker considers to be unnecessary at best, and dangerous at worst. As a consequence, disciplining or even firing workers who resist personal fall-arrest systems only hastens the day when many professional roofing contractors will be forced out of the residential roofing market.

Lastly, NRCA urges OSHA to revisit the important question of the costs of the 1994 Standard in connection with the forthcoming reconsideration proceedings. Industry concerns that OSHA seriously understated the costs of complying with the Standard were raised in the immediate aftermath of the rule's publication; they remain a substantial concern today. In addition to the specific cost issues raised at that time, we believe that OSHA's assessment should include costs such as: (1) lost earnings to roofing workers paid at piece rates; (2) increased costs to contractors and, ultimately, homeowners and building owners, arising from increased labor costs applying to work requiring the use of personal fall-arrest systems; and (3) increased insurance and other costs of worker injuries resulting from the use of these systems. It is incumbent upon OSHA to fulfill its requirements under the Regulatory Flexibility Act to do an accurate analysis of the economic impact of the 1994 Standard on the roofing industry as the basis for future modifications. Obviously, it will also be necessary for OSHA to comply with all other requirements under the Small Business Regulatory Enforcement Fairness Act.

      c.  Building the Necessary Evidentiary Basis for a Sound Decision

Under the Occupational Safety and Health Act, it is, of course, OSHA's burden to justify its standards by "substantial evidence," and it is therefore necessary for the Agency to develop or compile additional data along the lines indicated above. NRCA recognizes that this is no small task. We also understand, however, the importance of completing the work that was started in 1995 with the issuance of the interim guidelines and recognition by all concerned, OSHA included, that the 1994 Standard must be reopened for reconsideration. As indicated above, NRCA is committed to continue to work with OSHA and the Roofers Union, as well as other interested parties, in developing the data and information necessary to resolve these critical issues. In that regard, we are currently investigating possible avenues for obtaining the additional data needed, and once a feasible approach has been identified, we will contact OSHA and the Roofers Union to discuss cooperative implementation of the investigation.

    3.  The Need to Maintain the Regulatory Status Quo Pending the Completion of the Reconsideration Proceedings.

In the meantime, it is critical that the regulatory status quo be maintained. Our members were extremely concerned about the indication in the ANPRM that OSHA may summarily rescind Instruction STD 3-0.1A before reopening the 1994 rulemaking, i.e., without following the standard-setting procedures required by the Occupational Safety and Health Act. In NRCA's view, such an action would not only be manifestly unlawful, but also be a breach of good faith with the industry great enough to threaten, and perhaps undermine completely, the potential for a cooperative Government-Labor-Industry effort to resolve the difficult and important issues raised in the ANPRM and these comments.

As indicated above, OSHA has expressly acknowledged that the 1994 Fall Protection Standard needs to be re-evaluated and agreed further that until that re-evaluation has occurred, roofing contractors should be allowed to implement alternative means of fall protection—specifically safety monitors and slide guards—for residential roofing work. Although OSHA committed in 1995 to reopen the rulemaking "expeditiously," it still has not done so. In the five year period since OSHA made these agreements, the Directive has remained in effect, and as a practical matter has supplanted the standard itself insofar as residential roofing work is concerned.

NRCA believes that the courts would have little difficulty striking down a rescission of the Directive without following the standard-setting process required by the Occupational Safety and Health Act. But perhaps more important, as the background of this proceeding makes clear, the Directive was issued as part of an acknowledgement by everyone concerned, including OSHA, that the 1994 Standard needs to be re-evaluated. The expectation going forward, of course, was that OSHA and the affected parties would proceed to address the underlying deficiencies in the Standard. And that remains the hope and expectation of the roofing industry today.

What is more, the roofing industry has over the past four years made a sizeable investment in the implementation of the 1995 Guidelines, in the form of education, training, compliance assistance and promoting diligent on-site supervision. Attaining routine compliance with the Guidelines was a task far more difficult than anyone imagined, primarily due to the need to change the way roofing work has been done for a great many years. And as already noted, NRCA has received some input from its members suggesting that implementation of the Guidelines has resulted in tangible worker safety benefits.

In short, for OSHA now to summarily rescind the 1995 Guidelines and reinstate the admittedly flawed 1994 Standard, would represent an unjustified and unfair renunciation of the findings and commitments the Agency made in 1995, and surely would re-ignite the controversy the Standard originally created. In addition, such an action would seriously undermine OSHA's credibility in the roofing industry, an industry where the Agency acknowledges that it is heavily dependent on voluntary compliance if its worker safety objectives are to be achieved. The immediate impact of such an arbitrary decision, we fear, would be an increase in fall-related safety hazards in residential roofing work. For these reasons, NRCA strongly encourages OSHA to keep Directive STD 3-0.1A in effect until the completion of rulemaking proceedings to re-evaluate the 1994 Standard.

Questions about roofing work

In response to the questions posed by OSHA in the ANPRM, NRCA offers the following responses.

"Are monitors an effective means of preventing falls?"
OSHA has long considered the use of safety monitors to be the least effective method of fall protection. This is based on the premise that 100 percent positive protection is best. On a roof, it is nearly impossible and virtually impractical to attain 100 percent positive fall protection. Installation of a residential-type (steep-slope) roof system happens quickly. NRCA members strongly feel that monitors are an effective means of preventing falls on residential roofs with slopes of 4-in-12 or less.

A safety monitor cannot be compared to the conventional fall-protection options (safety nets, guardrails or personal fall arrest systems). The conventional methods stop or arrest a fall. A safety monitor is more closely aligned with a restraint system, because when employed properly a potential fall victim is prevented from falling by a verbal command. OSHA must begin to recognize that fall protection cannot be limited to only conventional methods of protection. Fall protection is situational—the method must meet the situation and not the other way around. And in roofing work, the choice of the most appropriate method must always be based on an accurate appreciation of the underlying causes of the hazard and of the new hazards that may be introduced by the method selected.

In the case of low-slope roofs, fall exposures can occur when workers are too close to an edge or opening in the roof, for example. The use of safety monitors is a viable fall deterrent. Although the use of conventional fall-protection systems might also work in theory, the experience of NRCA's members reveals that they can also introduce new hazards in a variety of ways. These include creating a false sense of security, unnecessarily adding additional tripping hazards to the workplace, needlessly prolonging the time needed to complete a job and creating greater fall hazards for those who install conventional fall-protection systems, such as catch platforms and guardrails. These situations can expose workers to trip hazards leading them to take shortcuts or prolong their fall exposures and, in turn, compromise safety.

"Are there slide guard brackets/devices that can be used on [metal] roofs?"
NRCA is unaware of any slide guard bracket/device that does not require roof deck penetration. This precludes the use of any such system because penetration of a metal roof compromises the water-shedding ability of the roof system. There are stages during installation where tying off is possible, but once the metal panels are in place, using brackets is no longer possible without penetrating the metal roof.

"Are these [slide guard] specifications appropriate?"
NRCA's members agree that the current specifications are adequate and that installation of the roof jacks should be as per manufacturers' recommendations.

"Are slide guards effective as replacements for personal fall protection?"
Again, OSHA must consider the situation. In its own safety program guidelines, OSHA recommends that hazards be identified and the appropriate abatement strategy employed. Under the unique residential roofing situation, slide guards work. They are easy to install and use, cost is not a significant factor and they assist workers by providing a foothold on steeper roofs and a ledge for properly stacking materials.

"Since roofing work is done after the roof is fully sheathed, are there technical or other reasons why roof anchors could not be used for roofing work?"
There are no known technical reasons why roofing contractors couldn't use another's roof anchor. However, so many things could happen to that anchor from the time it was originally installed by a framer and eventually used by a roofing worker that the anchor's integrity could be compromised.

Roofing contractors, as well as other residential contractors, do not want and should not be made to be held responsible for others' work. An employer generally should take responsibility for his workers and their actions. However, relying on another contractor's worker to properly install anchors is a recipe for a lawsuit—and, more importantly, potential cause of severe injury to the contractor's employees. In addition, once a framing contractor, for example, leaves the job with his anchors in place, he will be held responsible for the integrity of the anchors. Again, this is not a fair position to for the framing contractor. Further, approximately 80 percent of steep-slope roofing involves repair and replacement work; therefore, using a left-over anchor that is many years old, possibly rusted or damaged, is unwise at best and deadly at worst.

General liability and workers' compensation insurance costs are already a major expense for roofing contractors and not necessarily easy to obtain. Adding liability coverage for roof anchors installed by another contractor will only widen that gap of availability and force more contractors "underground" or out of the residential roofing market altogether. As discussed above, this will only lead to greater safety hazards for roofing workers, as rogue contractors who are likely to disregard safety standards across the board fill the vacuum created by the departure of professional roofing contractors from the market.

"Why is it infeasible to remove a roof anchor?"
OSHA is missing the liability concern when it asks this question. The issue of liability does not involve the removal of the anchor, but the use of it by workers other than that of the installing contractor. See previous answer. It is for this reason that the overwhelming majority of roofing contractors will, in fact, remove any and all anchors they install.

"Are there other roof anchors that are designed to be readily removed?"
Please see above.

Questions about the definition of residential construction

"Is this an appropriate definition of residential construction for the purposes of the rule?"
It is unlikely that there will ever be consensus on this definition. Yet the current definition goes a long way toward reaching agreement from the standpoint that the focus should be on materials and work practices and not on the end use of the building. And in that sense, NRCA agrees with the definition. Again, OSHA must allow employers to consider their situations and apply the appropriate safety method(s) to the hazard(s).

"Does this definition adequately distinguish between projects where conventional fall protection is feasible and those where, for some operations, it is not?"
See above.

"Is this a workable definition--can employers readily use it to determine whether their project is considered residential construction?"
NRCA is concerned with some of OSHA's rewordings that occurred in the rewrite of STD 3.1 to STD 3.01A. As stated above the initial definition went a long way toward a focus on materials and work practices. The current definition appears to narrow the intent of STD 3.1 and, therefore, has created confusion. Not all buildings are made of stick frame, some are constructed with metal studs.

Therefore, NRCA suggests the following definition: An employer is engaged in residential construction where materials, methods and procedures being used are essentially the same as those used in building a typical single-family home.

"Does the STD 3—0.1A definition adequately deal with these types [single story stick framed commercial malls; and single story stick-framed retail structures] of projects?"
NRCA believes the definition is clear except for the point made above.

"Should OSHA define residential construction in terms of the end use of the structure?"
As stated above, NRCA strongly feels that the emphasis should be placed on the best way to protect workers, not on the building's use. For example, workers reroofing a doctor's office that used to be a house must be protected from fall hazards using measures that are not appropriate based on the type of structure, not its use.

"Should the economic scale of the project be a factor in determining the fall protection options available to the builder?"
NRCA fails to see any justification in using the economic scale of a project to determine fall protection.

"Would it be appropriate for OSHA to allow the use of alternative fall protection procedures on portions of a commercial structure that meet the definition of residential construction?"
Absolutely. Again OSHA needs to recognize that the construction process, whether commercial or residential, is not necessarily different just because the end use is different. The focus must be on the hazard and the best way to address it. The overriding issue must always be the structure itself.

Questions about fall protection for vendor employees

"Is there a reason why conventional fall protection for vendors is infeasible?"
Conventional fall protection for vendors who deliver materials directly to the roof is not necessarily infeasible, but OSHA should approach this issue differently. OSHA must ask what are the hazards presented in the work process and what are the appropriate hazard abatement methods. Sometimes, technically feasible abatement methods create other hazards and are not appropriate for hazard reduction. That is typically the case with vendor-loaded roofs.

Usually, the work of loading a roof is of short duration, fast moving and performed with equipment, such as any number of mechanical lifts, where safety can be compromised if the supplier introduces conventional fall-protection devices, such as a personal fall-arrest system. As stated above, new hazards are created by spending more time on the job to install or dismantle a personal fall-arrest system, tripping over safety lines and creating a false sense of security. In addition, it would not be at all surprising to find worker resistance to conventional fall protection systems to be as strong among loading workers as it is among roofing workers. OSHA should certainly study that issue very carefully because, as developed above, such resistance could well mean that a decision by OSHA to require conventional systems will reduce existing levels of fall hazard protection for these workers.

"Why would tripping and being in an arrested fall be a greater hazard than the risk of falling, unprotected, to the ground?"
Again OSHA should approach this issue differently. What is causing the tripping in the question asked? The material loading process is methodical and fast-paced. Introducing lifelines creates tripping hazards that otherwise wouldn't exist.

"Are there reasons why these types of systems [retractable lanyards] cannot be used to protect vendor employees?"
Please see above.

"Is there a reason why vendors delivering supplies to a roof could not install this type of anchor and use it for fall protection for their employees?"
Please see above.

"Is there a reason why these [strap anchor] systems are infeasible or would pose a greater hazard?"
Please see above.

Questions about restraint system criteria

"Specifically, what are the maximum loads expected to be imposed on a system designed to restrain an employee from stepping past an edge?"
NRCA cannot even begin to guess. Again, OSHA should approach this issue differently. The issue with restraint is preventing the wearer from reaching a fall. Unlike fall arrest devices, a fall restraint device only needs to stop a person who is doing nothing more than moving toward an otherwise potential fall hazard. It will not, and does not, need to be designed to arrest a fall. A more appropriate question would be, "What is the minimum strength requirement of the restraint system to stop a worker before reaching a fall hazard?" The answer to that question is best left to engineers, but NRCA imagines that a typical lifeline and safety belt would suffice.

"What are the appropriate strength requirements for restraint systems anchors and other components?"
Please see above.

"Is there a need for the requirements in Subpart M for snap hooks and other connecting hardware also to apply to restraint systems?"
To avoid confusion, NRCA recommends using the same requirements.

"Alternatively, should components of a restraint system meet the same strength and other criteria as those for personal fall arrest systems?"
No; please see above.

"Is there a significant likelihood that restrain system components would get mixed up with personal fall arrest system components?"
Not if the requirements were the same as above using safety belts or other similarly constructed belts were used instead of harnesses.

Other issues

Although OSHA has not asked for comment in the following areas, NRCA would like to take the opportunity to raise three other issues. First, regarding the 6-foot height trigger for Subpart M, NRCA is requesting that OSHA expand this rulemaking process to include a discussion on the 6-foot height trigger issue. NRCA feels that a 10-foot trigger may be appropriate given OSHA's scaffolding regulation requirement.

Though NRCA recognizes the need for special requirements for industry-specific hazards, the height trigger issue for falls from scaffolds and roofs is no different except the origination point. If OSHA believes that 10 feet is appropriate for scaffold-related work, then why the difference for roofing work? In fact, the difference that currently exists creates confusion on a roofing job where one side of a building is scaffolded (for any number of reasons) and the other is not. In this situation, on one side of the roof fall protection is required at 10 feet but on the other side it is required at 6 feet.

NRCA would welcome the opportunity to present further discussion on the matter and urges OSHA to consider the matter.

The second issue involves the exceptions to the OSHA definition of mechanical equipment. The original definition of mechanical equipment excepted wheelbarrows and mop carts because of the nature of built-up roofing work. In fact, as stated above, the first OSHA fall-protection standard that applied to the roofing industry was known as "Guarding of Low-Pitched-Roof Perimeters During the Performance of Built-Up Roofing Work". It was more commonly referred to as the "built up roofing standard".

Wheelbarrows and mopcarts were identified as not falling under the mechanical equipment definition because they are manually moved and because it was necessary for these pieces of equipment to be close to the edge when performing certain roofing duties, such as dumping debris or installing roof membranes. In 1975, when the discussion first began about the need for an industry-specific standard, built up roofing was the predominant type of roof system installed.

In the 25 years since, many more roof systems have become popular, each having their own unique installation methods and procedures. The current Subpart M does not account for these new systems and the unique pieces of equipment required to install them. Roofing contractors who install an EPDM roof system (i.e., a "rubber" roof system) can use a seaming device that is "motorized" but only to the extent it is pushed by hand to seam the rubber membranes together. Roofing contractors have received OSHA citations for using these pieces of equipment outside of the warning line, where the current standard prohibits "mechanical equipment".

There are numerous examples of these pieces of mechanical equipment that NRCA has already submitted to OSHA for consideration. NRCA asks that OSHA reopen the rule to consider these pieces of mechanical equipment that provide parallel service to these different roof system applications.

Lastly, NRCA requests the re-opened rulemaking should include a review of how the warning-line and safety-monitoring systems are properly used when installing roof systems other than built-up systems. Again, the current Subpart M allows these fall-protection methods to be used only as they apply to built-up roofing work. However, when installing an EPDM system, for example, the membrane, which can be 25 feet wide, must be rolled out. During the process of rolling out the membrane, the warning lines must be temporarily removed or they will be buried under the EPDM membrane. During this rolling out, roofing contractors have received citations for not having the warning line erected, even though a safety monitor was present.

Technically a violation may have occurred, but realistically the standard does not fit the situation. And as a result, an otherwise compliant roofing contractor is exposed unfairly to a citation.


To be both practicable and effective, a fall-protection standard for the residential roofing industry must be based on a firm appreciation of the unique characteristics of the industry. It must be compatible with the specific competitive environment in which roofing contractors must operate. It must accommodate the capabilities and limitations of the firms competing in the market, which are overwhelmingly small businesses that are typically family-owned, thinly capitalized and only marginally profitable. It must be designed to work in light of the characteristics of the work force and the nature of the work itself—particularly its variability from job to job—its heavy emphasis on manual labor and hand tools, and the many unique safety hazards posed. These characteristics usually call for education, training and the use of commonsense work practices as the best and only practical solution to the constantly changing safety challenges of roofing work.

Although OSHA's 1994 Fall Protection Standard came nowhere close to meeting these criteria, OSHA took a major step forward in 1995 when it acknowledged the need to re-evaluate the Standard and, in cooperation with the industry, developed and issued the 1995 Interim Guidelines. From the limited evidence that is available at the present time, it appears that the Guidelines have been successful in at least some important respects, including the achievement of a substantial degree of acceptance within the industry, and some indication, albeit far from conclusive, that positive safety benefits have been realized. The job for OSHA and the stakeholders now is to build on what appears to have been a positive start, working together to fine tune the Guidelines and formally incorporate them into the Standard.

As discussed above, work remains to be done to accomplish this mission in a way that serves the objectives of the Occupational Safety and Health Act. The effectiveness of the 1995 Guidelines must be systematically studied. The 1994 assessments of net benefits, costs and feasibility of extending "conventional" fall-protection measures throughout the residential roofing industry need to be updated and improved to incorporate all the relevant factors. And a much more refined understanding of the factors principally responsible for the current rate of fall-related injuries and deaths in the industry must be obtained through more careful study. In addition to these investigations, consideration must be given to adjusting key elements of the Guidelines, including the 6-foot height trigger.

While it is OSHA's burden to justify its rules with persuasive evidence, NRCA for many years has worked together with the Agency and Roofers Union to obtain the necessary data and remains willing to continue those efforts today. NRCA welcomes OSHA's decision to reopen the 1994 Standard and hopes to work with the Agency in formulating and documenting a proposal that meets the criteria set forth above for devising a workable and effective fall-protection standard for both the residential (steep-slope) and commercial (low-slope) roofing industry.

William A. Good, CAE
Executive Vice President

Roofing industry news

[ More news ]


Find a contractor

Roof type

ZIP Code

Find roofing contractors by state

Sponsored link