Unemployment is even worse than the numbers say

Business columnist Don Pittis on how jobless numbers can get so wonky.

Just now, there is almost nothing Barack Obama would not do to bring the unemployment rate down.

And that is even though high levels of unemployment are among the most powerful evidence there is against government conspiracy theories.

If governments were truly omniscient and omnipotent, as the conspiracists among us like to say, this is not a time they would sit on their hands.

A job centre in central London, in the U.K. Unemployment appears to be on the rise throughout the industrialized world. (Alastair Grant/Associated Press)

The latest U.S. figures show unemployment rose to 9.6 per cent, but the statistics say the economy did not lose as many jobs as feared. Canada's report comes next week, the result of a slightly different statistical method.

But as jobless numbers stay high, governments stay nervous. U.S. mid-term elections are only two months away and a Canadian vote of confidence is possible anytime after Parliament resumes on Sept. 20.

If our leaders could sit down with a cabal of Illuminati or Freemasons and make unemployment go away, I tell you, they would.

It is an incontrovertible fact that both the U.S. president and our own prime minister, Stephen Harper, know that unemployed voters do not make for happy ones.

Be worried

Still, no matter how worrying our jobless numbers are just now, there are many voices telling us that we should be even more concerned.

In this context, I think it is important to note that neither the U.S. Bureau of Labour Statistics, nor our own Statistics Canada, is trying to pull the wool over our eyes. This is not a conspiracy.

As labour market expert Martin Dooley of McMaster University told me in an email, although labour statistics may be imperfect, the most important thing is to have a standard method of counting so you know if things are getting better or worse.

"The best use of the unemployment rate is to measure changes in labour market conditions," Dooley wrote. Nobody wants to mess with that equation.

However, I have spoken to and read many critics who say that jobless statistics underestimate the number of people out of work.

And, from my research over the last couple of days, it seems pretty clear that, although the method may be standard, that method also appears to minimize the real pain of the unemployment effect during a downturn.

The atrophying effect

One of those critics is Charles Beach, an economist at Queen's University in Kingston. He says that, at a time like this, the figures hide useful information and disguise how much current unemployment truly affects the economy.

For one thing, he says, the statistics don't distinguish between plain old unemployment and long-term unemployment.

Long-term unemployment, says Beach, makes a much bigger hole in the economy. When people stay jobless for a long time, they end up "selling their assets and getting rid of RSPs. Human skills atrophy."

These people are among the so-called "discouraged workers," those who have been out of work so long they are no longer looking. And if they aren't looking, officially, they aren't unemployed. So the numbers skew further.

But there is also other suffering missed in the counting.

According to the U.S. definition, an employed person is someone who "did any work at all (at least one hour) as paid employees, worked in their own business … or worked 15 hours or more as unpaid workers" in a family business.

Canada has provisions that amount to the same thing.

But that means that if you are so desperate for work that you pick up an hour's fill-in as a crossing guard, or you shovel some driveways, or you work for free at your uncle's corner store, you are in the statistics as "employed."

And that kind of desperation is certainly more common when jobs are scarce.

Riding the rails

Another way statistics underestimate the jobless is by taking non-working people out of the official workforce.

That happens because statisticians characterize unemployment as the number of people looking for work compared to the total workforce.

But if you don't count the discouraged workers who aren't looking you don't get the full picture.

In the 1930s, when hobos road the rails searching for work or food, unemployment was estimated at around 25 per cent. But our modern-day hobos, the growing number of homeless and transient, aren't counted.

Today's data is collected by household. But, ah, the homeless aren't in households.

Aboriginal Canadians can also help the workforce shrink. Measured if they live in the city, they suddenly disappear from the numbers if hard times force them back to family on reserves.

A similar kind of sleight of hand applies to migrant workers who tend to drift in and out of the number counters' sight.   Interestingly, though, the same statistical quirks can make the unemployment rate stay stubbornly high even after a recovery begins. We might call this "the revenge of the discouraged worker."

According to StatsCan's Vincent Ferrao, when the labour market picks up, discouraged workers cheer up. "Suddenly," he says, "you feel this is the moment to look for a job."

That means that just as some people find work and leave the ranks of the unemployed, the no-longer-discouraged join the statistical workforce. As unemployed.

The other factor at play here is "the revenge of the under-employed."

Finding a full-time job after a year working a few hours a week would be a thrill.

But, statistically, moving from even one hour a week to 40 hours a week with benefits doesn't reduce the unemployment rate by one iota.

With all this in mind, I have just thought of a perfectly cynical solution to rescue our troubled political leaders from the political brutality of high unemployment rates: All they have to do is borrow a little more cash and hire every unemployed person for one hour a week.

Street sweeping. Bring your own broom. Problem solved. Statistics don't lie. Now can I join the Illuminati?