Rick Woldenberg has been looking into it:
It’s certainly true that children have been checking books out of libraries for many, many years with only paper cuts to show for their reckless behavior. And thrift stores have sold children’s clothing and strollers for years without incident, but apparently no one knew the “dangers” that lurked within. Indeed, all of these items have been considered safe until February 10, when magically they will become unsafe unless proven otherwise.
According to reports from the field, zippers and snaps on kids’ garments are among the components most likely to flunk the new CPSIA standards. In all probability, millions of existing, already sold garments would flunk for similar reasons. Yet have CPSIA advocates pointed to even a single instance in which an American child has been poisoned by garment zippers or snaps? Could this be because kids do not as a rule detach and eat zippers and snaps? “Why aren’t we seeing many claims of injury from all the dangerous children’s products still legally available in the market?” Woldenberg asks. “Shouldn’t we be seeing outrageous injury statistics right now, currently”?
To put it differently, advocates seem to have taken a few genuine instances of injury from distinctive, atypical products (lead jewelry, powerful ball magnets) and used them to manufacture an imaginary crisis in the safety of children’s products generally. But there was never any general crisis of children’s product safety.
Following up, as part of a more general critique of CPSIA advocates’ misrepresentations — which should be read in its entirety, as it makes many other valuable points — Woldenberg analyzes the purportedly alarming data on product recalls involving children (Excel spreadsheet). Separately, Wacky Hermit has been looking at the recall figures as well. She concludes:
Of the 63 recalls that would have been prevented by CPSIA, only 1 resulted in an injury (a child ingested lead paint from a crib and had elevated blood levels of lead). This means that had CPSIA been in place for 2008, one child would have been helped.*
If we’re going to extrapolate one or a handful of injuries into a supposed national crisis, we might as well deduce a “children’s bathing crisis” from a bathtub drowning, a “children’s kitchen crisis” from a stovetop scalding or a “children’s transportation crisis” from a highway smashup. Common Room has a wrap-up which also should be read in full:
the majority of recalls (by an astronomical number) are not because a item has actually harmed anybody, but because the CPSC or the company determine that perhaps a particular item might possibly harm a child- and, while it flies against our intuitive, emotional reaction to the news that an item has lead, not all lead products are the same. The lead in a kid’s mini-bike tire valve is not as dangerous as the lead in paint on a toddler’s block. … The CPSIA treats real, imagined, and nonexistent threats exactly the same, and that is not sound policy.
*More: Jennifer Taggart writes to say that looking at past recalls does not make it possible to assess the law’s full effect since most items covered by CPSIA had not been subject to federal regulation (except under general catchall provisions). So the generalization quoted above should probably have included some limiting language to that effect.