July 10, 2015

 “When as kids we came to an orchard wall that seemed too high to climb, we took off our caps and tossed them over the wall, and then we had no choice but to follow them.” – Frank O’Connor, An Only Child


Today, the Commission voted to provide “only limited”[1] relief for the enormous burdens placed on the economy by our requirements under the Consumer Product Safety Improvement Act (CPSIA)[2] that children’s products be tested by third-party labs.[3] That limited relief would come from the Commission’s determination that unfinished and untreated wood does not contain any of the seven heavy elements[4] subject to limits by the voluntary toy standard that was made mandatory by the CPSIA,[5] so long as the wood comes exclusively from a tree’s trunk.[6]


I wanted to provide more robust relief. I lost.


In concept, third-party testing has real value to consumers. Having a neutral lab make sure products meet the safety standards the Commission has established substantially increases the likelihood that the products consumers give to their children will meet those standards.


The safety benefits of testing do come with high costs, as a full suite of testing can, depending on the product, cost hundreds or thousands of dollars. Multiplied by multiple units and across dozens or hundreds of product lines, testing can easily be an investment of millions every year for a single company.


Where a test genuinely does add safety value, those costs can be worthwhile. Testing for testing’s sake, however, is a waste of resources that could go to genuine safety improvements. Part of our obligation as public servants is to ensure that we maximize the gains consumers see not just from our own spending, but also from the spending we demand of the companies we regulate. Today, we did not fulfill that obligation.


I offered an amendment that would have implemented three substantial burden relief ideas that Chairman Kaye and I discussed in our letter to Senate Commerce Committee Chairman Thune in September of 2014.[7] I believe these three areas are well-supported by the evidence in hand and are well within our statutory authority to implement immediately through interpretive rules or even guidance policy. While there are other avenues we should explore, these three are, I think, entirely reasonable, but my amendment was not adopted.


The limited relief we have instead provided – exempting “unfinished and untreated wood . . . harvested from the trunks of trees”[8] from testing for heavy elements – is hardly an early first step. Congress reminded us of our regulatory duty in 2011,[9] expressly directing us to find ways to reduce testing burdens (without compromising safety). We have precious little to show for four years of efforts.


To be sure, we have made efforts. We have spent hundreds of staff hours and at least a million dollars examining the question of testing burden relief. The problem lies in how we have framed the task before us, how we have instructed our talented, dedicated staff to pursue this mission.


Congress directed us to find changes that would “reduce third-party testing costs consistent with assuring compliance with” our statutes and rules.[10] Congress did not ask us to – nor should we – compromise our safety goals in the name of cutting costs. However, Congress also did not ask for absolute guarantees, perhaps because of an understanding that there are no absolute guarantees.


Sound regulation, like sound science, trades in probabilities and risks. And CPSC’s statutes and rules – including on the subject of testing – reflect that reality. We do not seek to eliminate all injuries from consumer products, only unreasonable risks.[11] We do not subject every unit of every product to testing, only representative samples of those products.[12]


We have even been willing to introduce some possible risk of non-compliance into the testing regime. In 2011, we issued a rule that permits manufacturers and importers to rely on testing of the various component parts of a product rather than waiting until the end to test the finished product.[13]The rule does require manufacturers and importers who choose to rely on their suppliers’ test results to do so with eyes wide open, exercising “due care” in evaluating the proficiency of the supplier and the reliability of the testing.[14]


Even with a due care requirement, introducing more links in the chain unquestionably creates some added possibility of a non-compliant product or component, some potential for slippage. However, we made the wise decision to weigh that minimal added risk against the significant potential cost and efficiency gains of allowing, for example, a paint manufacturer to test in huge batches and toy manufacturers to rely on that testing.


For whatever reason, we have lost – or abandoned – the notion of making decisions based on risk and probabilities. If we allow manufacturers to skip third-party testing for materials that make up less than 10 mg of a product – the de minimis exemption that was one of my proposals for this vote – do we marginally increase the odds of a non-compliant product?[15] Yes. Do we actually increase the odds of a harmful product? I do not believe we do, because a tiny amount of an element is still a tiny amount, even if it makes up a non-compliant concentration of a part of a product. Nonetheless, we are unwilling to embrace this approach.


If a test does not have the potential to make consumers safer, then it is an unnecessary test and precisely the sort of burden we should look to eliminate. Such a practical, risk-based approach would be welcome in all of our efforts, but it is particularly relevant in the burden-relief context. Looking for ways to make testing cheaper is looking for ways to make complying with the testing requirement easier. If complying is easier, the odds of compliance go up, and the excuses for non-compliance go down.


Beyond just regulatory wisdom, however, cutting unnecessary testing costs was Congress’ direction to us. The proviso that we do so “while assuring compliance” certainly puts a wall between the universe of theoretical relief opportunities and the world of practicable solutions. That wall, however, is only insurmountable if we over-build it with unreasonable expectations that bear little relationship to the risks consumers actually face. Each needless restriction we place on the ingenuity of our staff is just another brick in the wall.


If we have failed to climb the wall because we think it can’t be done, it’s not time to throw our hands up. It’s time to throw our caps over.

[1] U.S. Consumer Product Safety Commission, Briefing Package: Recommendation for Determinations on the ASTM Elements in Unfinished and Untreated Woods and Other Natural Materials, 13 (2015).

[2] Pub. L. No. 110-314, 122 Stat. 3016 (2008).

[3] CPSIA, 122 Stat. at 3022, § 102.

[4] Those elements are “antimony, arsenic, barium, cadmium, chromium, lead, mercury, or selenium,” with lead regulated separately under the CPSIA. ASTM Int’l, Designation: F963-11, Standard Consumer Safety Specification for Toy Safety, § (2011).

[5] CPSIA, 122 Stat. at 3033, § 106.

[6] U.S. Consumer Product Safety Commission, draft Federal Register notice, Toys; Determinations Regarding Heavy Elements Limits for Unfinished and Untreated Wood, 22 (2015).

[7] Those proposals would extend the “lead-free” determinations we made for a lengthy list of materials to the heavy elements outlined in the ASTM F963 toy standard that was the subject of this vote, codify a de minimis exemption from chemical testing for materials that contribute less than 10 mg to the mass of a product, and recognize that two highly regarded international toy standards are the functional equivalents of F963 and successful testing to either of those is sufficient to conclude a product meets the requirements of F963. Elliott F. Kaye and Joseph P. Mohorovic, Letter to The Honorable John Thune, Ranking Member, Committee on Commerce, Science, and Transportation, United States Senate (Sept. 26, 2014).

[8] U.S. Consumer Product Safety Commission, draft Federal Register notice, Toys; Determinations Regarding Heavy Elements Limits for Unfinished and Untreated Wood, 22 (2015).

[9] Pub. L. 112-28, 125 Stat. 276, § 2 (codified at 15 U.S.C. § 2063(a)(3)).

[10] Id.

[11] 15 U.S.C. § 2051(b)(1).

[12] Admittedly, the fact that much of the testing that we require results in the destruction of the product would make an every-unit testing scheme impossible. We could, however, require that each unit be subject to each non-destructive test. Thankfully, we have made the wise judgment – informed by our statutory mandate – that testing fairly chosen samples is sufficient.

[13] 16 C.F.R. part 1109.

[14] 16 C.F.R. § 1109.5. We define due care as “the degree of care that a prudent and competent person engaged in the same line of business or endeavor would exercise under similar circumstances” and state plainly that “[d]ue care does not permit willful ignorance.” 16 C.F.R. § 1109.4(g).

[15] Even with a de minimis exemption for testing, these materials would still have to comply with each underlying rule. So, while a manufacturer would not face penalties for not sending 9 mg of paint to a third-party lab, that manufacturer could still face product seizure and penalties if that 9 mg of paint raised the product’s concentration of lead, for example, to 101 parts-per-million.