HOW MUCH TESTING IS TOO MUCH TESTING? Or, Button, Button, Who’s Got the Button? (and who tested the button? certified the button? materially changed the button? periodically tested the button? randomly sampled the button?)

Today the Commission is putting out for comment two proposed rules for doing the testing required by the CPSIA.   It is important that the basic rule be in place before the testing and certification stay of enforcement is lifted in February 2011.  Given the importance of this rule, we need to hear from the public about whether this proposal gets close to being right. 

This proposed rule, in an unprecedented way, puts federal regulators onto the factory floor.  The proposed rule is over 170 pages long and is by its nature very complex.  I am not confident we have written a rule that works for all the products under our jurisdiction.

One important aspect of the proposal deals with component testing and certification.  If suppliers of the paint or the resin or the buttons and zippers that go into a product are willing to test and certify for compliance, it is wasteful to also require the manufacturer who is using those components to repeat the testing.  This only makes sense.  However, if the components are going into children’s products, they need to be tested by a third party laboratory approved by the CPSC.  Further, the component maker is also responsible for doing periodic testing and for testing when a material change takes place, as is also required by the statute. 

We are hoping the component testing rule will result in the development of a market for third party tested components to especially help small businesses.  During my recent trip to China, I specifically asked various manufacturers whether component testing would address the concerns I was hearing about the immense costs of testing.  Several manufacturers told me they were already seeing tested components being advertised.  Others told me that for commodity type products, such as wire, it is unlikely such a market will develop.  I would be interested in your thoughts on how component testing will help you and what more we can do to develop this process.

In a statement I filed today, I lay out some of the serious concerns I have with the proposed rule.  This proposal will be open for comment for 75 days.  The text of the rule tells you the process for submitting comments.  I urge you to weigh in.  We need to hear from you before you’re told what testing you’ll have to do year after year, product after product.


4 Responses to “HOW MUCH TESTING IS TOO MUCH TESTING? Or, Button, Button, Who’s Got the Button? (and who tested the button? certified the button? materially changed the button? periodically tested the button? randomly sampled the button?)”

  1. 1 Steph H June 8, 2010 at 4:07 pm

    The magnitude of this law blows my mind. It is utterly absurd and wasteful to require test, after test, after test on the exact same component as it goes to each step of the supply chain. Does lead magically appear if a component arrives at my door from a supplier that has tested said component and has documentation to prove it complies? So the manufacturer has to get it tested, then the supplier has to get it tested, and then I have to get it tested? What a way to completely smother home businesses.

  2. 2 Ed Loewenton May 11, 2010 at 4:43 pm

    Dear Ms. Nord,
    Although I used to devote a fair amount of time researching and writing about safety issues regarding toys and other consumer products ( – see also the pages linked from this URL), I have had little time to keep up with progress on CPSIA. I am a member of the Handmade Toy Alliance but have not been active, despite intentions.
    I am not sure what the proposed rule-making has evolved to, but one post – just where, I forget, probably on the Handmade Toy Alliance website – especially caught my eye. The raw number of tests that would be required would be staggering. Testing of components for each product would require, eventually, millions of tests annually, since testing, to be meaningful, would have to be repeated periodically.
    And yet periodic testing, as opposed to batch testing, is meaningless. In fact, no testing regime guarantees strict product compliance. Any test is just a snapshot in time. Every time a component of an assembly is depleted and restocked, there is the opportunity for substitution by a non-compliant product. This is most likely to be inadvertent, when the assembler/manufacturer has no reason to suspect the new part, paint, etc., is any different from the last batch. In fact, I actually saw that happen to a manufacturer, who discovered to their astonishment that a purchased plastic part was pigmented with lead, and had to recall a season’s output of otherwise safe product.
    So no product safety paradigm or system is any better than its weakest link. The present attempt to tighten the entire system sufficiently to avoid any but the least probable mistakes appears to be leading to an absurd result. Full component testing, or full finished product testing (each of the manufacturer’s SKU’s) on an every-batch basis of the finished products will lead to a volume of testing too large to be supported by lab capacity even several times greater than now exists. The expansion of bureaucracy needed to monitor and enforce such a program would be extremely untimely in face of the cutbacks in spending which our deficit economy really requires. The testing program will exist on paper but not in fact; and there will be a disruption in shipped product, as manufacturers wait for testing. (This actually happened in China in fall of 2007, when the Chinese government decreed that products had to be tested before they could be loaded. I knew one toy importer/”manufacturer” who did not receive product for 4th quarter.) The only real beneficiaries would be the labs.
    You are suggesting a model which is already largely in place: certification by component vendors of their products. Material Safety Data Sheets (MSDS) are a pretty good prototype of the kind of document that could satisfy the new regulations.
    I would add to the MSDS several requirements:
    1. The entire formula of the upstream components must be specified. In the case of paint, the paint manufacturer must certify everything in the mix, protestations of proprietary information notwithstanding.
    2. In the case of a component supplier purchasing sub-components purchase elsewhere, these components would have to travel with a current-batch pedigree of their own.
    2. The specified blends, subcomponents, etc., would require lab testing for banned substances periodically. Any user could elect either their own testing, or to maintain evidence thereof from an upstream blender or assembler. The documentation would have to be more recent than a specified date. A final assembler (e.g., the consumer-level manufacturer or owner of the brand name of the final product) would be required to maintain a current certification of content for the entire bill of materials.
    3. It would be a civil and criminal offense to substitute any non-documented materials in product for sale to the public in the categories covered by CPSIA. At the level of the small, home-based craftsperson, this means that if you run out of red paint for your wooden toy and need only a few ounces to finish a batch, you cannot run down to the hardware store for whatever they are selling, unless it happens also to come with the sort of lab-certified documentation outlined here. You would have to buy a standard quantity from your regular supplier.
    4. Such a model as this already exists in employment regulation. Employers are required to maintain I-9’s (immigration status) W-4’s, and raw payroll data such as hours, but are not required to send these anywhere except in certain circumstances. But when there is a problem and a visit from an agency, they had better have their documents in order!
    5. This model reduces the volume of testing to the minimum, while ensuring that consumer products will be tested thoroughly at some point (or various points) in their assembly.
    6. This model moves responsibility for testing as much as possible upstream in the generation of each product. This makes logistic and economic sense. The upstream supplier tends to be larger, with greater financial and physical resources, makes larger batches, and has fewer individually distinguishable products as a proportion of revenues. Certainly, the number of industrial-level manufacturers are far fewer in number than their universe of customers, and their variety of products far smaller than the variety of final, consumer-level products that might be made with those industrial supplies.
    In the case of custom-designed or custom-blended components, it is assumed that the buyer would specify certification, and the costs of this added testing would be part of the price structure, or otherwise negotiated.
    This model thus reduces substantially the volume of testing, the total non-productive dollar cost to the economy, and the burden on the small producer.
    7. The final or small producer does have the responsibility for maintaining an up-to-date paper trail. This is the sort of thing we have computers for! If you have chosen to sell products that pose a potential hazard by virtue of the special vulnerability of the expected end user (in this case, children), you have the responsibility to demonstrate to the buying public the safety of your products. A small annual revenue is in itself no indicator of product quality or safety; it is just as likely an indicator of the opposite.
    8. Apparently, some suppliers are moving toward the sort of certification discussed in this proposal. It is likely that formalizing this model as regulation would make it economically attractive and thus common.
    10. In regard to exempt materials: There are materials that either have not been shown ever to contain the banned substances, or by their nature cannot contain these substances in a form which can present a hazard.
    *The most obvious are minimally processed naturally-grown materials: *unfinished wood; natural undyed fiber such as wool, cotton, hemp, silk; Paper from natural sources.
    Other materials are equivocal:
    *Clear wood finish will not contain lead, but may contain metal-based dryers. Safe for toys, but not for teething.
    * Dyes (as opposed to opaque finishes) will not contain lead or phthalates, but may offer other chemical hazards if chewed or sucked on.
    *glass- I was impressed and convinced by a study I read of the refractory nature of leaded glass to everything but hydrofluoric acid – a child with hydrofluoric acid in his saliva needs no one’s protection!
    *structural metals: the cheapest steel an aluminum parts may contain very small amounts of alloyed lead, but structural steel and aluminum are generally used in applications such as vehicles and sports gear not accessible to children who are still mouthing and chewing (under 3 years). Brass always contains some lead; it makes it machineable. Almost all the environmental exposure comes from faucets! Brass plating does not require the addition of lead, and I would doubt if any were to be found in plating.
    * Plastics such as polyether sulfone (PES), Medical grade silicone, or polypropylene and polyethylene do not require plasticizers (phthalates) or metal stabilizers (lead). These are found almost exclusively in PVC (vinyl).
    These materials should be either outside the scope of the ACT when not combined with regulated materials, or be subject to a modified set of regulations.
    Reading the posts from various groups asking for special deals, however, I do note some of the worst safety offenders fabricating reasons for their own exemption. I STRONGLY URGE THAT EXEMPTIONS BE KEPT TO A MINIMAL CLASS OF MATERIALS DEMONSTRATED TO BE NON-HAZARDOUS BY VERY STRICT STANDARDS. Otherwise, the inevitable political backlash, whenever it finally comes, will have you going through all this fuss all over again.

    Thank you for your time and attention. I would actually really like a response or acknowledgment:

  3. 3 Wacky Hermit May 7, 2010 at 8:09 am

    I know you’ve encouraged us to comment before on CPSC proposed rulemakings etc., but it would be really helpful for me (and probably for others) if you could include a link to the appropriate docket on Most of the time when I look at the CPSC documents (I subscribe to the email feed) it says [insert docket number here] in the documents, and I can’t comment without the docket number, so that means I have to spend some time trying to figure out the right search string on that will bring up the document I want to comment on. If you already know the docket number and you’d like to invite me to comment, it would be helpful if you either gave the docket number so that I can search it up quickly, or provided a link to the docket on Thanks!

    BTW I know you posted earlier that our comments are read by the CPSC. But that’s a wholly different thing than our comments mattering. I can comment all I want to CPSC, but even if CPSC totally agrees with me, their hands are tied by the stupid laws Congress keeps force-feeding all of us. It’s not a good investment of my time to flail away at something that neither the CPSC nor I can change, unless the sheer volume of comments has some effect on Congress– and judging from their reaction to the sheer volume of comments they got on all sorts of topics last year, I think their minds are already made up.

  4. 4 Mary B May 5, 2010 at 5:34 pm

    I think the testing should take place before the component (buttons, zippers, ribbon, paint, etc.) ever reaches the consumer level. Small businesses and crafters should be able to purchase a button shaped like a bunny and use it on a baby sweater without having to be concerned about whether it is safe or not. And what about a gramma who makes a sweater and gives it to her grandbaby and uses an untested button – is that sweater any “safer” because it wasn’t offered for sale?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Enter your email address to subscribe to my blog and receive notifications of new posts by email.

Join 986 other followers


RSS CPSC Breaking News & Recent Recalls

  • An error has occurred; the feed is probably down. Try again later.

Nancy's Photos

  • 80,875 visits

%d bloggers like this: