Recently, the Commission raised eyebrows around the product safety community by effecting what many see as a broad policy shift through two privately negotiated settlement agreements. Of course, I am referring to the issue of requiring all-inclusive compulsory compliance programs as a condition to settle alleged failure to report violations, using as a justification for this action the existence of past voluntary recalls. This blog is to bring you up to date on the state of play.

First, I had an exchange with my colleagues about our settlement with Kolcraft, where my colleagues insisted on adding the compulsory compliance program language at the last minute after an agreement had been reached between counsel for the company and the agency. I expressed my disagreement with the decision, while my colleagues jointly endorsed it. Second, mere weeks later, we approved a settlement with Williams-Sonoma that included the identical requirement. Though I voted to approve—on the basis that the company was well-represented and agreed of its own will—I still felt uneasy about using the enforcement vehicle for this apparent policy shift, undermining both the efficacy of the settlement and the legitimacy of the policy, which deserves public input. This began an “unusual” back-and-forth with one of my colleagues, with his statement, my supplemental statement, his supplemental statement, my further supplemental statement and so on, with the latest ping-pong volley occurring today.

These settlements, the identical language they share, and the emerging policy they represent all demonstrate that this is an important issue. So important, in fact, that it’s a conversation we should be having more publicly and with more participants.

As I have stated many times, I fully support corporate compliance programs. They are one way a company can build a commitment to safety into all of its products. And there may be a role for CPSC in getting more companies to institute compliance programs, but, whatever that role is, it’s one we should play in public. If we are going to require them, we should not do that through privately negotiated settlement agreements but instead should engage the public in the conversation so that we fully understand the implications of this policy.

The whole purpose of a multi-member commission is to make sure various views are heard and rules are the product of engagement and discussion. So, let’s engage and discuss.

That being said, I supported the broad outlines of this package. One key reason I voted to move ahead? I believe it’s high time we asked the public what to do about certificates. So now it’s your turn to let us know how we could improve this rule. Talk to me here, but more importantly, talk to the all of us at the Commission by submitting comments here.

Electronic Records Access

The proposal requires electronic certificates to be in an accessible location without password protection and identified on the product, its packaging, or its invoice, so our staff, our partners at CBP, distributors, and retailers can have ready access. In principle, that requirement seems reasonable. My concern, however, is that this makes them available to everyone (something the statute does not require), even people who might use them for illegitimate purposes. We have included a request for comment on how to ensure easy access only for the people who need the certificates.

Direct-to-consumer International Sales

I recently purchased from a British company whose shoes are unavailable in the US. Who should certify them? The company, with no U.S. expertise (and, arguably, no legal title to them after they accepted my payment)? The shipper, with no direct knowledge? Me? And what value would that certificate have?

I wanted to consider exceptions for single transactions between entirely foreign companies and US consumers. I did not expect agreement on the idea, but I also did not expect vehement opposition to even thinking about it.

First, my colleagues worried reasonably that we lack the authority to make an exception. But this was merely a question, and we plainly have statutory authority for that. A lack of authority might be a hurdle to a solution, but it should not prevent us from exploring a problem. Second, they raised policy disagreements. Those are fine reason to vote against the policy but insufficient justification for squelching debate.

Redundant Certificates

I wanted to request comments about potential redundancy. For example, we have a rule governing small parts in toys for children 3 or under and a broader toy standard that incorporates that rule. As it stands, a certificate will have to show compliance with the overarching standard—which incorporates small-part compliance—and compliance with the small part rule. This redundancy adds to the proposal’s nearly half-billion dollar annual price-tag, so it we should at least inquire into its value.

My colleagues rejected the notion, calling my question “murky” and expressing preference for the status quo. With respect, dissatisfaction with language calls for improving it, not dismissing it wholesale, and defaulting to the status quo is the hallmark of unresponsive government. An idea that has outlived or exceeded its usefulness should not survive by inertia.

One colleague added that he was initially inclined to include my question, but reversed after our staff suggested redundancy eases enforcement. This is the kind of discussion we could and should have had with the public, and it demonstrates we lack information public comment could provide. Perhaps consensus would have formed that the enforcement benefits outweigh any waste, but we should solicit public input, not presume it.

Even though my colleagues were unwilling to ask these questions, nothing prevents you from answering them anyway. The comment period will run 75 days after the proposal hits the Federal Register, and we will all be better off the more we hear from the all corners of the public about what we got right and what we could do better.

While I have focused on issues raised in the public discussion of this proposed rule, there are other equally (if not more) important issues that need to be addressed.  For example, requiring electronic certificates is an important change that merits public input, as is the proposed mandate for common carriers, as “importers of record,” to be liable for certificates.  The new content requirements are also important changes.

I do hope that the agency will take the time and make the effort to craft a certification rule that improves safety without imposing undue burdens.  But if any hope of doing that demands public input. So, you need to provide that input, and we need to listen to it.

The 1110 rule requires manufacturers and importers to include certificates with their products attesting to the products’ tested compliance with all applicable CPSC safety standards. These can be Children’s Product Certificates or General Certificates of Conformity, depending on the product. Most CPC requirements are already covered in Parts 1107 and 1109, leaving GCCs to 1110. As proposed, this revision contains a one-size-fits-all requirement to keep GCCs for five years. On its own, this might not be a problem; we have to set a mark, and five years might be as good as three or seven. But we cannot look at this rule on its own, as my colleagues are unfortunately doing, because doing so creates unnecessary confusion.

This rule rests on the testing rules, and those rest on safety standards. These other rules frequently contain their own recordkeeping provisions. Our rule on flammability standards for mattresses, for example, requires manufacturers to test prototypes and then keep the records for as long as the product is on the market, plus three years. Those test records, then, could be discarded after four years or hang around for decades. Even within one product, the requirements that already exist vary with circumstance, and that variation exists throughout our regulatory arena.

Recognizing the differences in the standards’ requirements and my colleagues’ preferences for keeping records for longer, I proposed a compromise tying the retention requirement for GCCs to their underlying standards, defaulting to five years for any certificate based on a standard with no recordkeeping mandate. My colleagues did not agree, and insisted on a universal five-year mandate. Since CPCs all have a five-year minimum, they argued, imposing the same requirement for GCCs would “harmonize” our certificate rules.

“Harmonizing” requires that the notes be in the same key. Imposing one regulatory scheme’s parameters on another simply for the virtue of nominal similarity while ignoring their underlying differences is as unsound as “harmonizing” your children’s closets by giving them all the same size pants, no matter their ages. Yes, they’re in harmony, but someone’s going to wind up with a bad fit.

In this instance, the more valuable harmonization would have been matching the certificate’s retention time to that of the rule that creates the standard and the test on which that certificate is based. I was pleased my colleagues were at least willing to include a request for the public to comment on the recordkeeping retirement, and I hope they read those comments with open minds.

Our staff originally proposed what I thought was an acceptable approach: If your product is subject to multiple rules and exempt from testing for only some of them, then you have to certify to the ones in force and claim your testing exemption(s) for the rest. But if your product is exempt from testing under any applicable standard—whether your product has one or more testing exceptions—you don’t need a certificate just to say that. To me, this seemed not only a reasonable opportunity to minimize unnecessary burdens but also more consistent with the law, which bases certificates on testing.  Requiring a certificate with no information other than an exemption is wasteful and contrary to the purpose of the testing regime.

My colleagues were uninterested in these benefits. Arguing that having more pieces of paper to shuffle would expedite work at the ports, they amended the proposed rule to require companies to create, provide, and maintain certificates that say nothing more than, “I’m exempt from testing to the standard.” Although I do not think such a certificate is necessary, I thought public input on the question could be helpful, so I proposed returning to the staff’s original language and asking for comment on the safety, efficiency, and cost implications of my colleagues’ approach. My colleagues were not interested in asking a question, and decided to plow ahead. (My colleagues did less-than-helpfully note that the public could still comment on the approach.)

The rule they insisted on might turn out to be the efficient one. We might hear from commenters that consistency in certificates is more useful than skipping hollow ones. What baffles me is my colleagues’ refusal to even solicit public input on the point, particularly when they are claiming benefits that, if real, the regulated community would likely endorse. Dogged refusal to invite any other perspectives is not the hallmark of reasoned decision-making.

Tomorrow, we’ll continue this discussion of the areas where the 1110 rule could use improvement before it’s final.

Not a lawn dart.

Yesterday, I discussed how the Commission fell short in considering and presenting the costs of our 1110 rule (on certificates of compliance). Today, I examine how we (mis)handled the tricky question of how the certificate rule fits with banned products.

The Consumer Product Safety Improvement Act of 2008 handed us a seemingly strange mandate. We were to require any manufacturer or importer of a consumer product to issue a certificate specifying any “rule, ban, standard, or regulation applicable to the product” and certifying conformity thereto. For rules, standards, and regulations, that makes plain enough sense, but certifying that your product is not banned might seem a little odd. Read too broadly, requiring something to certify that it is not banned could lead to absurd results. Think of requiring baby blankets to certify that they are not lawn darts. Or a hammer needing to certify that it is not general apparel containing asbestos.

Congress left it to us to implement this requirement reasonably. Our proposal almost does so, but, as with the costs, we have buried the lede. Portions of the rule discuss how individual bans interact with the certificate rule, but the rule does not provide any general guidelines to help a company determine if a ban relevant to its products requires certifying, “this is not banned.”

Our staff strived mightily to give some reasonable content to the requirement. They looked through the CPSC’s bans and saw that some appear to ban whole categories of products, while others ban only part of a product category. In particular, if a product was subject to a specific test, it appears easier to identify the products to be tested, and those that could pass the test would have to certify to the ban. (Those that failed would, obviously, be banned.) Some of staff’s helpful language appears in the rule’s preamble, but preambles do not appear in the Code of Federal Regulations, where companies and their counsel will look in years to come. Following staff’s efforts, I proposed amending the draft NPR to include this principle in the rule.

Certificates are required for products which are subject to a ban when the banned characteristics defined by the language of the ban do not define the whole product category within which the banned products fall and the products are not specifically excluded from the ban.

I also wanted to ask, in our request for comment, if our ban certificate language was clear. My colleagues rejected both these ideas, though they did at least agree to ask the public about staff’s assessment as expressed in the preamble.  Where failing to include a required certificate could trigger civil or even criminal penalties, I think we owe a cogent explanation of the rule, and I hope you will tell us what you think of the principle, its application as staff indicated, and whether the principle belongs in the actual rule.

More of my thoughts on the weaknesses of our 1110 proposal tomorrow.

One of my biggest frustrations was that my colleagues were unwilling not just to do more to reduce the annual costs of this rule but even to be upfront in the way we presented its costs to the public. I wanted to include a chart that would lay out what we expect this change to cost in concert with the rule it replaces and the rules it supports (parts 1107 and 1109, the testing and certification rules for children’s products and their component parts). My colleagues insisted that we only needed to talk about the costs presented in the analysis prepared for this specific set of amendments. None of our rules operates in a vacuum, and the original 1110 rule was short on economic analysis because of the tiny 90-day window we had to pass it, so I thought it more transparent to give the public a chance to see and comment on the total tab. After all, companies don’t have the choice of conforming just with the new language—they have to follow all of our rules—so it is inconsistent with reality to tell people about only the new costs.

Since my colleagues refused to include this vital information, I have posted it here. The chart below breaks down how each component of the certificate rule adds to the prices consumers pay. Remember these are annual costs—and will be incurred year after year. Note also that this does not account for the actual set-up costs already expended to create and maintain the certificates, nor does it necessarily account for the costs of changes needed to comply with this proposal. And remember that these numbers apply only to paperwork—the costs of these sister rules are surely much, much larger.

I’ll talk about more of my concerns with the rule change we just proposed—and the stilted process that led to it—the rest of the week, but I wanted to put the bottom line up front, just as we should have done in the rule.

While I like spirited debate, I firmly believe that this debate should be limited to the issues and not devolve into personal attacks. Yet, in one colleague’s statement, she resorts to just that. I do believe that there must be room on the Commission for differing points of view and regulatory philosophies.  That, of course, is the point of a Commission. So, no matter how loud or petulant the protestations to the contrary, I will continue to fulfill my duty to evaluate our regulatory landscape, form my own opinions, and engage in the debate. After all, that’s what I was hired and sworn to do.

Our testing and certification rule places enormous burdens on companies with too little benefit to consumers. In 2011, Congress and the President tried to focus the agency on the issue through Public Law 112-28, telling us to ask the public to help us find savings, fix what we could without weakening compliance, and ask for more authority if we needed it. We have been dragging our feet on that work, and the latest chapter—our Fiscal Year 2014 budget request—makes clear that we won’t pick up the pace anytime soon.

In this budget, the extent of our burden reduction effort is to acknowledge that P.L. 112-28 exists. I tried to get agreement on an amendment that would have added a statement that we “may undertake activity to reduce the burdens identified” and that our staff would, as appropriate, prepare briefing packages on specific proposals. Of course, I would have preferred stronger language, but I wanted my colleagues’ agreement to this small commitment to action and so I offered this as a compromise. My colleagues found that too bold, explaining instead that we had already fulfilled our obligations under the law, voluntarily followed up on some of the comments we received, and might do more in the future.

I do not concur with my colleagues’ cramped and nonsensical view that all the law requires is that we seek comments on how to reduce burdens. (Would Congress really have asked us to get public comments and not intend us to review, analyze, and act on them?) Once presented with real options for reducing burdens, we have an obligation to take some action. Since my colleagues were not willing to make even this small commitment I could not in good conscience support a budget that asks for more resources but ignores basic regulatory obligations, especially as other agencies expect cuts to their resources. (My official statement on the budget can be found here.)

In 2012, our staff suggested 16 (non-exclusive) ways to reduce testing burdens and in the FY13 operating plan, the Commission whittled its to-do list down to sending out further requests for more information on just four ideas. We’ve asked for comments upon comments. Information is good (and people should again respond to our request), but Congress wanted us to do something about costs, not just consider doing something at some future time.

In response to my objections, I’ve heard the “door is not closed” on reducing burdens. The tone underlying that statement is that we’ve already done what we need to do, but we might do more. As discussed, I don’t think we have done much at all, but let’s take the statement at face value. Is there any reason to believe the door isn’t closed? Agencies only do the work they budget for, and not designating any resources for testing burden reduction is a sign that we won’t be doing that work.

I’m also told the budget is not really the appropriate place for burden reduction, that our operating plan would be the better vehicle. If it’s like the FY13 operating plan, the next version won’t even be written until halfway through FY14, when most of our resources are already committed. That’s the regulatory equivalent of “when we get around to it.” It’s not consistent with either the law or our obligation as public servants to regulate with no heavier a hand than necessary to reduce unreasonable risks to consumers.

Even liberal Slate blogger Matt Yglesias agreed that health and safety agencies tend to under-estimate the costs and over-imagine the benefits of their work, leaving consumers with net losses. Matt was particularly concerned about pure compliance costs, the paperwork costs of figuring out the regulation and documenting compliance that do virtually nothing to make consumers safer but cost big companies money, small companies jobs or existence, and consumers choice and affordability.

The worst part may be that these costs actually lessen companies’ ability to make consumers safer, because every dollar they spend identifying the hoops and deciphering just how to jump through them is a dollar they can’t spend on innovation, including safety innovation. I have been dismayed by the repeated refusals by some at CPSC to consider the full effects of our rules before imposing them, and this habit is starting to get noticed outside the building. I hope these growing voices lead us to make our work better before Congress has to step in and force us to improve.