Go figure!

The wild monthly gyrations in Canada's employment data continue.  StatsCan reported today that the economy added 74,000 jobs in September, almost all of them full time.  This was a much stronger result than most experts had predicted.  However, the erratic nature of the employment data over the last several months, together with growing fears over the outlook for the global economy and commodity prices, warrants caution.

The most bizarre aspect of the previous month's employment report was StatsCan's claim that about 98,000 private sector jobs had been lost, the most for any single month in the history of the data, with an offsetting and equally unprecedented leap in self-employment offsetting most of the loss.  Guess what? In September, StatsCan estimates that 130,000 private sector positions were created.  Neither the August nor the September data seems to bear any resemblance to what can be observed in the economy -- there have been no major layoffs, but equally no big plant openings either.

It still looks as though there may be something amiss in StatsCan's data collection, or perhaps its seasonal adjustment techniques.  Q3 is normally the time of the auto makers' model year switchover, which sometimes leads to workers being furloughed, but that wouldn't account for the reported jump in self-employment in August, which was largely reversed in September.

Unreliable data make it difficult for even the most plugged-in experts to forecast the future, and one such has today announced that it's throwing in the towel:  the Bank of Canada!  Governor Stephen Poloz has announced  that the Bank will be dropping its "general market guidance" -- the use of terms like "easing bias" or "tightening bias" -- in the belief that doing so will help to reduce market volatility.  Poloz's predecessor, Mark Carney, was big on guidance, but his attempts to introduce it at the Bank of England have backfired quite badly.

From now on, the BoC intends to lay out its analysis of the economy and let market participants draw their own conclusions. In principle this sounds like a good thing -- economists and analysts will have to examine the economic data, rather than carrying out an exegesis on Gov Poloz's pronouncements. It's a bit unfortunate, however, that the Bank is unveiling this new approach at a time when faith in StatsCan's data is as low as it's ever been.