http://www.cariboo.bc.ca/dsd/cced/faculty/dcharbon/forum/usa.htm
Canadian Forum
LYING WITH STATISTICS
What's the real U.S. unemployment rate?
by ED FINN
"There are three kinds of lies: lies, damned lies, and
statistics." - Benjamin Disraeli
Even in Disraeli's time, it seems, statistics were often
distorted to give a misleading impression. Today, the misuse
of data to "prove" a theory, support an opinion or excuse a
policy has been elevated to a fine art. Forty-four
years ago, the various ways that figures, charts, graphs,
tables and averages can be manipulated were described by
Darrell Huff in his book How to Lie with Statistics. From
the sample with the built-in bias to the unqualified
average, from the calculated omission to the selective
comparison, Huff demonstrated how easily statistics can be
used to deceive rather than inform.
Huff's book has long been out of print, but I suspect
that dog-eared copies are still carefully hidden in the desk
drawers of many advertising copywriters, public relations
consultants, politicians and even some economists. Or maybe
they learned how to lie with statistics without Huff's help.
Examples of statistical fabrication abound: the UN's bogus
ranking of Canada as No. 1 on its Human Development Index;
the Gross Domestic Product, which rates all economic growth
as good, even crime and pollution; and Canada's official
unemployment rate, which omits discouraged and involuntary
part-time workers.
If I were to pick the most dishonest case of statistical
skullduggery, it would probably be the official unemployment
rate in the United States. As in Canada, this rate -now
claimed to be down to five per cent - completely disregards
the millions of people who have given up looking for work,
as well as those who are working fewer than 20 hours a week
but would prefer full-time jobs. The calculation of the U.S.
unemployment rate, however, is done much more deceitfully,
and with some of the most blatant statistical perversions
ever devised.
For that country's business and political leaders, it
is important that the national jobless rate reflect the
merits of their policies. Mass layoffs, part-time work, job
insecurity, big corporate tax breaks, cuts in welfare and
UI benefits are not conducive to a lower rate of
unemployment. In fact, they invariably have the very
opposite effect. But the big corporations and their
political flunkeys want to convince the American public
that their free market approach benefits workers as much as
shareholders. And how better to peddle that lie as the
truth than with the crafty misuse of statistics.
According to the Council on International and Public
Affairs (CIPA), the real U.S. rate of unemployment, if
properly calculated, would be 11.4 per cent - more than
double the official rate. The CIPA listed seven major
changes in the definitions of "employed" and "unemployed"
that were made in the U.S. methodology that have had the
combined effect of substantially reducing the number of
Americans officially listed as being jobless.
Among the categories dropped from the labour force
survey, in addition to the discouraged, were the under-16
group, those on strike or locked out and those who weren't
actively looking for work in the four weeks prior to the
survey. But by far the largest group omitted from the list
of jobless in the U.S. are the working-age men who are out
of work because they are in prison or on parole.
The 1.5 million American men in jail and the 8.1 million
on parole make up nearly 10 per cent of that country's male
workforce. By not including them in its labour force survey,
the U.S. is able to reduce its official unemployment rate by
more than five per cent.
Just as the omission of a large group of unemployed can
drastically skew the statistics, so can the inclusion of a
group whose members are virtually 100 per cent employed -
such as the members of the U.S. armed forces. By lumping
these 1.5 million army, navy, air force and marine personnel
in with the civilian workforce, the official unemployment
rate is reduced by nearly another one per cent.
Given these egregious survey distortions, even the
CIPA's estimate of an 11.4 per cent rate of unemployment in
the U.S. could be far too modest. An actual rate of 15 per
cent would probably not be an exaggeration.
The CIPA points out that the ridiculously low official
rate of five per cent also conceals the real jobs crisis in
the U.S. by ignoring what it calls "the pauperization of
work" - the replacement of higher-paid jobs by those at or
close to the minimum wage, usually temporary or part-time,
and often below the poverty line.
Between 1979 and 1995, the greatest loss of jobs in the
U.S. - 22 million -- was in the durable goods sector of
manufacturing, where the highest-paying blue-collar jobs are
located. In sharp contrast, most of the newly created jobs
over the same period - 29 million - were in the service
sectors, and three-quarters of those were in the
lowest-paying categories: retail trade, health care, and
business services.
Over the past four years, according to the U.S. Bureau
of Labour Statistics, 8.4 million American workers were laid
off. Of the 3.8 million who had held their jobs for three or
more years, 64.8 per cent either did not find new jobs,
found only part-time jobs or found jobs that paid less than
their previous earnings.
On top of this shift to insecure and low-paying jobs,
the real wages of American workers as a whole have declined
by 19.5 per cent below their level of 25 years ago. Not
surprisingly according to the CIPA, poverty in the U.S. has
soared to unprecedented levels, with one out of four
Americans - about 64 million - now living in a state of
destitution.
Canadians should keep these grim realities in mind
whenever they hear our CEOs, politicians and academics
citing the artificially low U.S. unemployment stats as
"proof' that their market-driven policies create more jobs
than they destroy. The need-cons go on to argue that the only
way to bring down our high unemployment rate is to make
Canadian society a carbon copy of the U.S. and intensify the
scorched-earth tactics of public sector cutbacks,
privatization and deregulation.
That is their only recourse. They can't Americanize our
method of calculating the rate of unemployment, as much as
they might wish to do so, because Canada has much smaller
numbers of both convicted criminals and soldiers. Excluding
the former and including the latter wouldn't bring our rate
down even a fraction of one per cent. So all our need-cons
can do is (mis)use the phoney U.S. figures to claim, in
effect, that unemployment can be reduced by throwing more
people out of work.
To the many Canadians who have been fooled by this and
other exercises in statistical deception, I would implore
you to beware of data that don't gibe with your perception
of reality.
Disraeli, if he were still alive, would say it's time -
way past time - for a reality check.
Ed Finn is a research associate with the Canadian Centre for
Policy Alternatives.
