Is the safety of U.S. electrical products in danger?
View the full story, including all images and figures, in our monthly digital edition
To sell many types of products in North America, manufacturers must have their products certified by an independent certification organization. That certification process verifies that products comply with relevant safety requirements prior to entering the marketplace. For many products in Europe, however, manufacturers can self-declare that their products comply with requirements. In many cases, they are not obligated to draw on independent certification organizations.
In addition, under the North American system, those independent certification organizations are accredited by both government agencies and recognized private-sector accreditors as being technically competent to make a compliance judgment and as having a certification process that can verify a product's compliance over the life of its manufacture.
In an SDoC environment, by contrast, a similar accreditation (competency) structure is not employed. Manufacturer labs might be accredited (but not required) as competent in administering the tests. That accreditation, however, is insufficient in ensuring the technical competency to interpret the data and to render a compliance decision. And under the SDoC system in Europe, all compliance checks occur after a product enters the market and typically when an incident is reported. Post-market monitoring in Europe rests in the hands of the regulators, who admit insufficient resources for effective oversight.
Yet many European regulators and a number of companies are in favor of SDoC. They believe that by using self-declaration, companies can lower compliance costs and move their products to the market more quickly than before. These companies want to shift the North American system away from pre-market to post-market verification and change the regulatory infrastructure to do so.
The current debate has its roots in a set of sectoral United States—European Union (EU) mutual recognition agreements negotiated in the late 1990s. It resurfaced in 2005, with an industry-driven request to evaluate the merits of SDoC. OSHA decided at that time that the data did not support a change to its current structure. In the final years of the Bush administration, the United States and Europe agreed to initiate a framework dialogue that included conformity assessment matters. And in 2008, Europe identified the OSHA system as its No. 1 priority issue.
Through a request for information (RFI) issued in late 2008, OSHA sought to secure data on many aspects of SDoC: wanting to know how the declaration was equivalent or superior to a third-party system, understanding product risk and specifications of SDoC, determining the type of infrastructure required to administer SDoC, figuring the costs of SDoC, acquiring data on the declaration's effectiveness, and examining what affect SDoC would play on international trade. (See “Call to Action,” sidebar, page 20.)
In response to this, three experts joined Consulting-Specifying Engineer editor-in-chief Michael Ivanovich for a Webcast on June 25 titled, “ Product Certification and Safety Standards: What You Need to Know .” The experts were: Ann Weeks, vice president of government affairs for Underwriters Laboratories (UL); Dr. John R. Hall Jr., National Fire Protection Assn . (NFPA) division director of fire analysis and research; and Al Scolnik, vice president of the National Electrical Manufacturers Assn . (NEMA) Technical Services Dept. Each discussed issues surrounding SDoC, from the basics of what it is and how it works, to the arguments for and against it, to recently created advocacy activity in favor of traditional third-party certification. The Webcast was produced and sponsored by ASCO , a company of Emerson Network Power.
Weeks' presentation, The Who, What, Where, When, and Why of SDoC, provided background for the current debate, including how the World Trade Organization's (WTO) Technical Barriers to Trade (TBT) Agreement treats SDoC and third-party certification as viable options for government regulators and how the OSHA-based system aligns with WTO obligations.
Weeks defined SDoC, noting that a product within this system is declared compliant by only its manufacturer. Any independent verification of compliance is conducted by the government and after the product enters the market and, in Europe, typically is driven by a reported incident. The SDoC method is supported by a number of companies within the IT community, including its industry association the Information Technology Industry Council , and many officials within the EU.
Understanding how the WTO addresses SDoC is critical because SDoC proponents, especially EU regulators, assert that the U.S. system is a trade barrier. Simply put, the WTO treats SDoC as one of a number of compliance mechanisms from which a regulator can choose. Regulators typically choose the compliance mechanism based on such factors as product risk. The TBT Agreement also states that members must ensure that conformity assessment measures do not create unnecessary barriers to trade.
When OSHA developed its Nationally Recognized Testing Laboratory (NRTL) program, it identified products in the workplace that, if noncompliant, could pose substantial risks to workers. OSHA then chose third-party certification as the compliance mechanism and identified the acceptable standards to which a corresponding product must be certified. OSHA also established a recognition program for independent certification organizations—NRTL program—to assess their competency in evaluating products against the standards' requirements and making a determination of compliance. Moreover, OSHA requires any NRTL process to include a mechanism for monitoring ongoing compliance of the product.
“In the case of the OSHA program, one of the things that SDoC supporters allege is that the U.S. system is a trade barrier,” said Weeks. “But when you look at the OSHA program, it provides for 'national treatment,' which complies with WTO obligations for the U.S. It complies with trade requirements in the sense that both U.S. and non-U.S. certifiers can apply to become a NRTL, and it applies equally to both imported and domestically produced products. So it's not disadvantaging imported products over those produced domestically.”
In most circumstances, SDoC doesn't work
In his presentation, Product Certification, Hall said that in most settings, SDoC has been found not to work. Market surveillance authorities have found a high level of noncompliance of SDoC-certified products, and surveys of customers and other knowledgeable parties reveal a high level of skepticism that SDoC-certified products comply with standards.
Hall went on to address other frequently offered arguments regarding SDoC. For example, SDoC supporters claim the generally lower fire loss rates in Europe are evidence that SDoC-certified products are actually safer than North American third-party-certified products. SDoC supporters do not offer a detailed argument that fire loss rates are significantly driven by product safety differences attributable to differences in certification and, in fact, no such case can be made from the available evidence. Too many other differences such as infrastructure and voltages are critical. Many of these differences provide important reasons to use third-party certification, which incorporates national conditions and requirements explicitly.
The argument on the basis of cost is also fallible. The conventional thinking is that using SDoC lowers compliance costs. But the counter-argument states that spending is needed no matter what type of compliance mechanism is used, and in-house spending can exceed outsourcing. Someone must conduct the tests. Someone must interpret the data. Someone must make the compliance decision. Someone must verify ongoing compliance. And the competencies of those making the decisions and auditing the process must be validated.
Regardless of whether this is done by the manufacturer, the government, or a third-party certifier, someone must maintain the process, the equipment, the competencies, etc. If a company finds it too expensive and difficult to have its products tested by an independent third party, then they might just as easily consider it too expensive or too difficult to make the products safe and effective in the first place.
SDoC supporters suggest that other methods of managing product safety, such as risk assessment and litigation, make third-party certification unnecessary. In fact, risk assessment cannot be validly performed without the same kind of detailed, credible test data produced by third-party certification. Litigation is very uncertain in its application and its outcomes, and even when it is successful in removing unsafe products from the marketplace, it does so only after considerable loss has already occurred.
The ideal conditions for SDoC are where the supplier and the buyer are part of the same company. In that case, both supplier and buyer will share safety goals, expectations, and assumptions about conditions of use. Under those narrow conditions, SDoC can provide “adequate confidence” to a buyer that the supplier's product is acceptable.
In the end, according to Hall, SDoC does not work as well as third-party certification and cannot be made to work acceptably well without costly, technically complex added features that remove most if not all of the reason for interest in SDoC in the first place.
Stakeholder coalition formed
Following Hall's argument was Scolnik, who discussed the goals and actions of a coalition of supporters of retaining third-party certification for products used in the electrical infrastructure. He explained that the coalition includes 18 organizations representing broad interests such as labor, electrical contractors, code officials and electrical inspectors, consumer groups, product suppliers, testing organizations, and others, and was formed in 2008 in response to serious concerns about OSHA, which was, for the second time in three years, considering a move away from a system that has a long history of supporting electrical safety in this country.
“These groups were concerned that the SDoC proposal would eliminate the current requirement of third-party testing and certification, which assures the safety of products intended for use in U.S. workplaces. Instead, a manufacturer could self-declare that its products have been tested and found compliant with safety requirements with no independent verification. Our workplaces would move from one where safety is addressed prior to installation to a post-market system where problem equipment could be addressed only after an accident has already occurred.”
He asked, “In this day and age of counterfeit and substandard products, of report after report of unhealthy and unsafe product imports, does it really make sense to move away from a tried and proven safety system to one that is well known for its shortcomings?”
He noted that NEMA believes that the marketplace should determine the requirements for selection and use of products, including the electrical safety requirements and practices. “While SDoC has appropriate applications in some markets,” he added, “the U.S. marketplace has relied on third-party certification of electrical products used in the electrical infrastructure. Those who now wish to institutionalize SDoC must accept that the checks and balances of our current system have played a fundamental role in driving U.S. electrical products to be the safest in the world.”
Scolnik then encouraged Webcast viewers to become involved by contacting OSHA and key congressional leaders who will have a voice in deciding the matter. As SDoC gains popularity in Europe, he believes that Americans need to do what they can to ensure that products stay safe in this country. He said, “One way to do this is to encourage your elected representatives to stand up for workplace safety.”
While the main debate dealt with whether SDoC should be the international mechanism for evaluating products, another discussion arose during the Webcast's question-and-answer portion about which types of products would be affected by SDoC. Weeks explained that the European proposal seems to focus on products that use fewer than 1,000 V.
The problem with that, she said, is that products are composed of a wide variety of materials and parts. Some of these materials and parts that make up the larger product are considered “safety critical.”
The question then becomes, according to Weeks, “If you were to move the end product to an SDoC system, then how do you treat the components or separate out the treatment of components in that same system?”
Call to Action
The request for information (RFI) closed on Jan. 20, 2009, after more than 60 submissions, the majority of which supported the existing NRTL program. Four European submitters asserted that OSHA's 90-day RFI period failed to provide them with adequate time to compile data supporting their cause. As a result, OSHA briefly considered reopening the RFI period, but later decided to address these concerns when it takes up the issue later this year. As a result, OSHA may now consider reopening the RFI at some point this fall. Regulators still have the right to choose the conformity mechanism that meets their confidence needs, either a third-party system or SDoC. But that choice may not be available for long.
“[SDoC] remains a question of safety,” said NEMA's Al Scolnik. “I think there is every good reason to retain the OSHA NRTL program. I think the concern that the coalition has had is that it may get caught up in some trade issues, and that's where we've been talking to people who are involved in workplace safety both on [Capitol] Hill and in OSHA to make the case that you really shouldn't be jeopardizing safety in the workplace for the sake of some trade agreement. Electrical safety is too important to be used as a negotiating tool.”
About the Webcast
The live Webcast, “Product Certification and Safety Standards: What You Need to Know,” was held on June 25, 2009. An archive, which contains the complete event including Q&A and a copy of all slides, is available at www.csemag.com/webcasts . Registration is required, but access is free.
Ann M. Weeks
Vice President, Government AffairsUnderwriters LaboratoriesWashington, D.C.
John R. Hall Jr. , PhD
Division Director - Fire Analysis & ResearchNational Fire Protection Assn.Quincy, Mass.
Al B. Scolnik
VP Technical Services, NEMA Washington, D.C.