The US and Canada updated a suite of releases covering US consumers, US industry and Canadian housing. Here are what I think are some worthwhile takeaways accompanied by additional charts at the end.

TAKEAWAY #1

US industrial capacity use remains very high at 2018 levels with no signs of slack emerging (chart 1). Demand and damaged capacity likely combine to explain this. For as long as capacity use remains high, the broad sector will have pricing power. Industrial production fell 0.2% m/m but manufacturing was a touch firmer at +0.1% m/m. 

Chart 1: US Capacity Utilization Rate

TAKEAWAY #2

The US job market remains strong. Weekly initial jobless claims fell again and have been declining for several weeks (chart 2). I’m a little leery toward the quality of the data given uncertainty around whether seasonal adjustment factors were reliable on the prior upswing and in the more recent downswing period. Still, smoothing through it all would still leave behind little evidence that the job market is weakening.

Chart 2: US Initial Jobless Claims

TAKEAWAY #3

US retailers are at best treading water into Q3. That may be partly because consumption patterns continue to shift toward services not well captured by retail sales and partly due to ongoing product shortages in some sectors. In any event, sales were up 0.3% m/m in August (-0.1% consensus) in nominal terms but were revised down to -0.4% (from 0%) the prior month for an overall draw relative to expectations and in terms of the two-month trend. Breadth was soft. The surprise is that autos and parts increased by 2.8% m/m despite lower new vehicle sales volumes. Recall that these are nominal numbers needing adjustment for prices. In volume terms, overall retail sales fell again last month given CPI was strong in case anyone needs a reminder of the market effects. US nominal retail sales are tracking a Q3 q/q SAAR gain of just 2.4% after 10.9% in Q2 (chart 3). In real terms there is no growth in Q3 as higher prices are hitting demand at retailers.

Chart 3: Retail Sales Growth Tracking

TAKEAWAY #4

US supply chains are putting in a mixed performance at the manufacturing level into this month. The Empire manufacturing gauge jumped higher which is a positive signal around the NY/NJ region, but the Philly Fed’s gauge barfed on all that (charts 4, 5). Global supply chains are healing, but be careful with respect to how far one goes with that argument as some of the metrics (Baltic Dry, semiconductor prices, order backlogs etc) may be falling partly because of improved supply chains but partly on increased uncertainty toward the economic outlook.

Chart 4: Empire State Manufacturing Survey; Chart 5: Philadelphia Fed Business Outlook Survey

TAKEAWAY #5

Canada’s housing market continues to weaken, but for how long and how much is a big source of uncertainty. I have no faith in the ability of economists to forecast percentage price changes in house prices as splashy attention-seeking forecasts always miss the mark on the upswing and downswing and by a band roughly equivalent to the width of Canada. Home sales fell for a sixth month and were down 1% last month. New listings fell 5% m/m and fell just about everywhere across the country in a nationwide sellers’ strike. That kept the market pretty tight with months’ supply at just 3.5 months which is below the long-term average of 5 months and well below peaks but ~double where it was at the start of the year, while sales to new listings increased to 54.5% which remains in mild sellers’ territory (chart 6). And so prices controlled for compositional shifts (splashy forecasts often choose average prices instead…) were down 1.6% m/m SA, 5% over 3 months, 7.4% over 6 months but are still 7.6% higher y/y and 44% higher than 3 years ago. There is a tonne of padding in home equity. 

Chart 6: Rising but Still Low Supply of Canadian Homes

I’ll conclude by repeating points I’ve made with clients on Canadian housing. It’s a different housing market in Canada than elsewhere. Generalist articles on global housing markets fail to take into account important distinctions as follows:

  • OSFI's B20 already prepped the market for much of the rate shock facing originations, unlike elsewhere. The rise of the US 30 year mortgage rate has been a 300bps shock to originations this year whereas the way to look at the rate shock in Canada is relative to how the market had adjusted to B20’s qualifying criteria at 5¼%. Relative to that de facto qualifying rate before the run up in mortgage rates, Canada is getting about half of the incremental rate shock to originations through the dominant 5-year product than the US is through 30s.
  • Canada is the only G7 country with accelerating population growth. Italy is still shrinking. The Euro-area is flat. The US has practically stalled out and not just because of complications around the Mexico border. Canada is getting 1% population growth just through higher immigration targets alone as Canada is now importing the equivalent of a city of Edmonton or Ottawa roughly every couple of years and with a tighter focus upon the economic class (destined to labour, housing, consumer markets, versus family class).
  • Canada is getting a terms of trade lift to incomes that every other G7 country is wishing they had.
  • Canada is run by serial stimulators across all levels of government. Enough said!
  • Then we have all the usual arguments on differences in the Canadian housing finance model, like well capitalized banks subject to regulatory stress testing exercises, little product tossed into off-balance sheet leveraged conduits, gov't guarantees provided to insured mortgage borrowers, attractive pick-up on mortgage bonds, the fact that it’s hard to do a 'big short' and that most hedge fund types who have tried over the years have repeatedly gotten their heads handed to them, and tighter macroprudential rules that started in 2008.

Instead, I'm more focused on the likelihood that housing corrects as the economy and job market weaken but doesn't stay down which has been a repeated pattern. There is still a wall of cash on household balance sheets. There remains investor appetite in tight rental markets from people who balance the cost of carry against the income stream relative to uncertainty facing financial assets. There remain a lot of sidelined first-time buyers who will find more attractive entry points to pounce upon and each time that happens housing charges higher. Canada needs to expand the housing stock by over 1 million every 3–5 years and won't and so demand leaks into resales.

If I’m right on the medium-term drivers, then the BoC needs to be very careful if and when we get to the point of risking a premature declaration of victory over inflation as it could well boomerang the other way again. Decisions made into the environment of peak house price weakness may not be the best ones.

Please see additional charts for US industrial output and retail sales on the next page.

Chart 7: August Changes in US Industrial Production by Industry; Chart 8: August Weighted Contributions to US Industrial Production; Chart 9: August Changes in US Nominal Retail Sales; Chart 10: August Weighted Contributions to US Nominal Retail Sales; Chart 11: US Retail Sales Recovery to Pre-Pandemic Levels by Sector; Chart 12: US Nominal Retail Sales Recovery