Updated jobs prediction

Back in early February, when economists were talking about 8% unemployment by year end (we hit that in spring!) and shuddering at 8% unemployment and 325,000 job losses by year end, I made this prediction that seemed somewhat alarmist:

For 2009, I would look for an average of 50-60K in job losses (600K-720K annualized). This is optimistic, and it could be worse. I project an increase to 9% or 10% unemployment by the end of 2009.

I also predicted

it (unemployment) will be at 8% in April. By mid year, at the absolute latest. 9% by year end is looking conservative. Remember in the last 2 recessions, unemployment went to 12%.

In this job loss hurricane for the next few months, 50 to 100K should be the range.

Let's revisit this:

When did it hit 8%? April. good prediction

9% by year end? That is now the consensus (doesn't mean it will happen, but no longer outlandish). 10% I address in my prediction below.

In the job loss hurricane (which ended in March, as I acknowledged on May 11th when the April numbers were out) 50 to 100K was about right.

So far in 2009, job losses YTD are 331K. That projects to roughly 662K by year end, again roughly in line with 600K-720K. Too early to say...

Notes on the July numbers:

Every month, we get tons of new self-employment jobs (75K and fluctuating wildly since October). That to me is complete BS. That number is way too high and volatile to be believable. Since my predictions thus far have been quite good on the jobs front, I am now going to make predictions excluding this number, as it seems silly. I will predict the numbers ex self-employment.

The July losses excluding self-employment were 79K. That is astronomical for an economy supposedly exiting recession. Even when you add the 35K of self-employment, the number becomes 45K. Now, some of the July losses were probably due to the fact that April and June were quite good. Perhaps August will be better. Also, interesting that Quebec finally had a bad month (-37K; more than 50% of the job losses since October). I live in Quebec and I am shocked and stunned by the complacency here. Quebec is on a different planet. Perhaps this report will take Quebec from a 2007 mentality to something closer to 2009.

Now for my new predictions:

1) Unemployment in the low 9% range by year end. 10% will only happen (I don't want it to happen) if we get a repeat of last fall's market carnage.

2) Lowering the job losses excluding self-employment to the lower end of the range (600K). We are close to 400K right now, so this is a deceleration from the first half. The 720K only happens if we get a repeat of last fall's market carnage.

3) The jobs report be relatively good for August & September as companies become more optimistic that the economy has bottomed and therefore, they don't lay off as many. October to December may be very bad though as I believe more people will realize that this optimism was misplaced, that this was not a regular recession and the recovery will be anemic. Again, this assumes no market carnage.