Was last November a turning point?

Statistics Canada's March GDP release was the fourth consecutive month in which output failed to go above what it had been in November 2007. So are we going into a recession as in 1981-82 or 1990-91? Or will the Canadian economy bounce back with stronger growth as in 2001-2002?

Before answering, it might be a good idea to look at what happened towards the end of previous expansions. Since Canada doesn't have an equivalent of the NBER's business cycle dating panel, I'm going to use monthly GDP to identify the months that marked the end of the last expansions: April 1981, March 1990, December 2000, and (for the purposes of the graphs that follow) November 2007.

I'm going to use the same graphical device I used here: all variables are logged and re-scaled to zero at the peaks of the expansions (1981:04, 1990:03, 2000:12 and 2007:11). They are then multiplied by 100, so that they can be interpreted as the per cent deviation from their values at the peak.

Let's look at GDP:

Looking at this graph, it'd be hard to dismiss out of hand the possibility that we're already in a recession: the fall in GDP is almost exactly what we saw going into the 1990-1991 recession. But then again, we're not far from what we saw early in 2001, either.

But GDP isn't the only thing we look at. Employment tells a very different story:

In previous recessions, employment continued to rise for a month or two after the peak in GDP. But employment in 2008 is not doing what we'd expect five months after a turning point. Moreover, this increase in employment is being accompanied by strong real wage growth. Real average hourly wages are 3% higher than they were a year ago, and this real wage growth is sustaining retail sales. As Statistics Canada's Philip Cross notes, "A recession does not have record-high auto sales and a very tight labour market."

Finally, here's StatsCan's composite leading indicator:

Again, the leading indicator seems to be more consistent with the 2001-2002 experience than in the two previous recessions.

I suppose I could go through more series (and I may do so in the future as more data become available), but right now, I'd assign a probability between 0.3 and 0.4 that the answer to the question in the title is "yes". Although the GDP data are worrisome (if they weren't, my assessment would be much lower), when set in context of the other data - especially data that are available in a more timely fashion than are GDP data, with its two-month lag - the most plausible scenario is still one in which the declines in GDP are due to transitory events (as discussed in the March GDP release) that do not - yet! - point to a continued decline in output.

Comments

[Ben]

I would say that the employment figures are skewed because much of the employment growth has been public sector.