Canadian Recession? Not So Fast

• Recession headlines are premature…

• ...as they mix GDP concepts in an apples and oranges sense…

• …that doesn’t consider uncertainty around trade and inventory effects

• Distortions should merit caution when throwing around dangerous headlines…

• …as wildfires knocked ½% off Q2 GDP growth…

• …and then strikes knocked another ½% off Q3 GDP growth…

• …and the absence of both would have kept growth in the black in both quarters

• This changes nothing for BoC observers given inflation’s many drivers

• Canadian GDP m/m % change SA, August:
  
• Actual: 0.0
• Scotia: 0.1
• Consensus: 0.1
• Prior: 0.0
• September guidance: “essentially unchanged”

My oh my, some folks are quick with their recession headlines. Too quick in my opinion.

Not only is their math debateable, but the added issue is also what constitutes a real recession in order to merit the potentially very damaging and confidence-sapping label.

I think markets saw through all of this better than some of the media coverage. Canadian two-year yields are unchanged compared to just before GDP estimates were released. They are outperforming the US front-end where the two-year yield is higher because of strong US data such as an upside surprise to the Q3 Employment Cost Index (1.1% q/q SA versus 1% consensus), the biggest repeat-sale house price gain for a single month (+1% m/m SA) since May that extends the string of gains to six months, and better than expected US consumer confidence figures that included slightly higher one-year ahead inflation expectations. The USD rallied on a DXY basis after all the US data and hence the main reason why the C$ weakened after combined US and Canadian data.

CANADA IS NOT IN RECESSION USING MONTHLY GDP ACCOUNTS...

First the data. GDP was flat in August and hence a tick softer than Statcan’s preliminary ‘flash’ guidance that said it grew by 0.1% m/m SA. Statcan said the weakness in August was due to “factors such as higher interest rates, inflation, forecast fires and drought conditions” while leaving out strikes which was a curious omission for an agency that grappled with them earlier this year and when strikes are playing a major dampening role on Q3 GDP growth (more on that later).

Statcan’s preliminary guidance for September GDP is “essentially unchanged” which basically means it could be a tick or so on either side of 0% given the uncertainties at this point. Let’s go with zero for the rest of the math, but it could be slightly higher or slightly lower. Chart 1 shows that GDP has basically been trending flat for a while which doesn’t really look like a recession to me.

What that translates into is tracking -0.08% q/q at a seasonally adjusted and annualized rate (SAAR) in Q3. This is using the monthly, production-side GDP accounts. Using the same set of accounts showed GDP growth of +0.1% q/q SAAR in Q2. The light blue bars in chart 2 provide this measure of growth tracking plus the red one for Q3.

By this definition using the monthly GDP figures we cannot say the economy has entered technical recession. All we can say at best is that it has stalled and is bouncing on bottom.

...AND IT’S ALSO A QUESTIONABLE CALL USING QUARTERLY GDP ACCOUNTS

Using quarterly expenditure-based GDP accounts involves a tonne of hand waving but as yet we need to be careful with the tracking as we continue to incorporate new information and data.

BoC’s Fresh Q3 GDP Forecast Wasn’t Shocked Enough to Dismiss It

One quick point is that the BoC just printed a forecast of 0.8% q/q SAAR for Q3 GDP growth in last week’s Monetary Policy Report. Today’s monthly GDP data is probably not enough of a shock to drop that figure negative on its own and so we might infer that what they are seeing is a touch more growth using the more complete set of quarterly GDP accounts than what is being tracked using the incomplete monthly GDP accounts.

You Can’t Bolt Monthly GDP Tracking Onto Quarterly GDP Tracking

Why ‘incomplete’? A main difference between quarterly expenditure-based GDP accounts and monthly production-based GDP accounts is that the former asks how higher or lower output of goods and services was achieved by considering inventory changes and contributions from swings in exports and imports. The latter does not. That’s why we need to be careful in applying Q3 tracking using monthly GDP as an extension to what quarterly GDP accounts showed for Q2.

I think net exports may be tracking a significant positive contribution to growth based upon changes in volumes shown in chart 3. Export volumes are looking up and import volumes are tracking lower. Lower imports serve as an addition to GDP growth given the accounting because they serve as less of a leakage of activity out of the country. Net trade (exports minus imports) is therefore tracking an improvement but we need to be careful in drawing overly strong conclusions because the weakness in imports might also reflect a weaker pull effect from the domestic economy for which we have limited tracking estimates.

The tricky part, however, is how this is reflected in inventory swings. On that count, the data is highly incomplete. We really only have monthly manufacturing and wholesale inventories and only up to August. Retail inventories are only available on a quarterly basis and up to Q2 and data on farm/resource inventories are incomplete. Using only the inventory data at hand offers tracking of a negative contribution to GDP growth as inventories fell (chart 4) but I wouldn’t overstate the reliability of that figure given the caveats that have been offered. Still, that would be consistent with tracking a decline in import volumes given that Canada imports much of what it consumes.

RECESSION SEMANTICS?

All of which makes for very unappealing uncertainty in the eyes of the press that was very quick to latch on to recession headlines as one story dominoed throughout the highly concentrated industry’s production chain.

But let’s step back from the math and use some common sense. Is the economy really that bad to merit labelling it as in recession? I don’t think so. Not yet anyway.

For one thing, some of the weakness has been temporary as anyone who spent two seconds in this country over the summertime can attest. There are at least two effects that require consideration.

One is that wildfires enveloped much of the country and the BoC now figures they subtracted around ½% from Q2 GDP. Add that back as a transitory shock that had nothing to do with interest rates and Q2 GDP growth would have been slightly positive. Not great, but not a negative.

Another is that there were strikes aplenty over Q3. Witness chart 5 that shows hours lost due to strikes and how they soared in Q3. We’ve seasonally adjusted data from Statcan that tracks hours lost due to strikes and lockouts and then added that figure back to total hours worked from the Labour Force Survey in order to compare what happened to hours as reported versus hours after adding back the ones that were lost due to strikes. This is shown in chart 6 as the largest hit to hours worked that we’ve dealt with in a long time and equal to roughly ½% of total hours worked. The gap between the bars for 2023Q3 shows the impact of lost hours worked due to strikes and how big it was compared to earlier quarters.

Now, since GDP is hours worked in aggregate times labour productivity, the roughly ½% q/q SAAR hit to hours worked due to strikes would, by corollary, translate into a ½% hit to q/q SAAR GDP growth in Q3 due to strikes assuming that the production lost by the striking workers is reflective of the overall average output per hour worked of the typical employee in Canada.

The bottom line is that growth has been weak, but not weak enough to merit the recession label in any genuine sense. Controlling for just two of the transitory shocks (wildfires and strikes) would suggest the economy has yet to post a quarterly contraction. Then we would need to control for the effects of droughts in parts of the country and extreme wetness in other parts.

It Should All be a Wash to BoC Observers

Does this change a whole lot? Well, yes and no. It doesn’t help confidence if the media is running around screaming recession if it isn’t fully justified to do so. It took about a nanosecond for the headlines to scream recession in big red bold highlighted lettering. Nonsense.

In a grander sense, however, the connections to monetary policy have not really changed one bit. The economy needs a protracted period of weakness in order to open up a material amount of disinflationary slack.

Then there are all of the other drivers of inflation beyond just slack arguments. Like ripping wages in nominal and real terms, tumbling productivity, excessive immigration, tight inventories of things like housing and autos, inadequate infrastructure, an undervalued and tumbling C$, high oil prices, still overly stimulative fiscal policy, labour hording arguments, strong corporate balance sheets, and exaggerated drag effects from mortgage resets. Painful, but exaggerated, and only one part of the Canadian macro picture.