Crash Test Dummies
Crash Test Dummies
Op-Ed Contributor
Washington
IN early January, parents were shocked to read about findings by Consumer Reports that
10 of 12 infant car seats had failed — most “disastrously” — in simulated 38 miles per
hour side-impact crash tests. Many parents were infuriated. These seats can be expensive,
and parents rely on them to protect their most precious cargo.
How the testing mistake was made is instructive not only for Consumer Reports but for
everyone who cares about public safety.
As it often does, Consumer Reports was attempting to fill a void in the government’s
work. In 2000, Congress instructed the National Highway Traffic Safety Administration
to issue within two years a new standard to improve the safety of child restraints. One
requirement was to minimize head injuries from side-impact collisions.
In May 2002, the safety administration announced it needed more research and testing
information before it could design a test for the new standard. While some research is
being conducted, the agency has not taken any further steps on side-impact protection for
infants and children. As a result, there is no government-approved side-impact crash test
for infant car seats as there is for frontal-crash testing.
So Consumer Reports had to decide how to conduct its test. The magazine tried to
replicate on a test sled the safety administration’s car-to-car side-impact test, which is
conducted at 38 m.p.h. However, a car crashing at this speed transmits a force to the
stationary car of about 20 m.p.h. because, unlike the sled, both vehicles move after they
hit.
Because Consumer Reports does not conduct crash tests, save for low-speed bumper-
impact tests, it has limited expertise in designing such tests. The magazine publishes
government and insurance industry-financed crash test information and contracts with
outside labs for child restraint testing. Its communications with its outside lab, Calspan,
concerning this side-impact vehicle test at 38 m.p.h., were apparently unclear. The lab
conducted the sled test at 38 m.p.h. instead of 20 m.p.h, thus doubling the crash forces of
the government standard.
None of this is to excuse the magazine. It should have considered the alarming testing
results a warning sign. Rather than rushing the results to print, the organization should
have consulted outside engineers, including one of its own board members, who is an
auto safety engineer. And while it apologized last week, and announced the creation of a
panel to study the error, it should have explained what went wrong.
But none of this should excuse the highway safety administration, either. For five years, it
has not completed the work mandated by Congress — though it took less than 10 days to
evaluate Consumer Reports’ testing and find the error. Further, the safety administration
must address deficiencies in its consumer information testing program, which is woefully
inadequate, out-of-date and underfinanced.
Several key areas come to mind: it still does not test to evaluate survivability in rollovers;
it doesn’t measure the effects of size differential when a passenger car collides with, say,
a light truck; it doesn’t test vehicle structure in frontal, off-center crashes; it doesn’t test
fuel tank vulnerability in rear-end collisions; finally, it fails to test how badly pedestrians
are injured when hit by vehicles. Most of these consumer information tests are
performed, or are being developed, in Europe, Australia and Japan.
What’s more, the tests that are done are simply too weak. Cars are not tested at high
enough speeds; sport utility vehicles are tested against insubstantial barriers. Take all this
together, and consumers are ill served. Last year, Congress finally added more money for
crash testing and required that test results be posted on the vehicle price sticker in the
dealer showroom, but the safety administration should have done this on its own.
Consumer Reports will get past this mistake. Those of us who value its legacy expect a
full public explanation of how its testing went wrong and how it will ensure that similar
mistakes are avoided in the future. At the same time, the safety administration should
devote its resources and energies not just to pointing out the grave mistakes of others, but
to fulfilling its own mandate.
Joan Claybrook, president of Public Citizen, was the administrator of the National
Highway Traffic Safety Administration from 1977 to 1981. She served on the board of
Consumers Union, which publishes Consumer Reports, from 1982 to late 2006.