• Canada added 14k jobs with mixed details
  • Sickies and weather remained as drag factors which may point to later gains
  • Wages exploded but with two-sided interpretations
  • BoC watchers should continue to fade backward data
 
  • Canadian jobs m/m 00s / UR %, SA, March:
  • Actual: 14.1 / 6.7
  • Scotia: 30 / 6.6
  • Consensus: 15.0 / 6.7
  • Prior: -83.9 / 6.7

Markets largely ignored Canadian jobs numbers as they should. They’re backward looking and peering into the abyss where surging commodity prices in a commodity producing country dominate attention alongside the outlook for the US economy and trade policy.

Jobs were up by 14.1k. Details were mixed as I’ll explain but for now check out charts 1 and 2 for summary measures. Canada’s 2-year bond yield fell by 1–2 bps post data while the US 2-year yield rose by 1–2bps post US CPI. CAD strengthened by under a quarter cent to the dollar.

Chart 1: Canadian Jobs Break Down
Chart 2: March Changes in Canadian Employment Levels by Sector

Details

After losing 84k jobs, a gain of 14k isn’t great. Details were mixed to boot. Before turning to that, however, I think that among the factors that kept job growth back in February were some that repeated in March.

How so? You don’t generally lose a job if you’re sick or it’s bad weather, but both can prevent the interviewing and hiring process. Charts 3 and 4 continue to show unusually high numbers of hours lost due to sickness and weather. This was an awful flu season in the US and Canada and there is no need to tell Canadians about the worse than usual winter (US eastern seaboard too).

Chart 3: Hours Lost Due to Illness/Disability for the Month of March; Chart 4: Canada Hours Lost Due to Weather for the Month of March

If that’s a contributing factor to a soft pattern for jobs of late then maybe chirpin’ birds and blossoming buds will bring something better. We’ll see.

The gain was driven by part-time jobs (+15.2k) as full-time jobs were flat (-1.1k).

The gain was mostly in private payrolls (_15.4k) as public payrolls were little changed (+4.6k) and so was the number of self-employed (-5.8k).

BC’s drop of 19k was the biggest drag among provinces (chart 5).

Chart 5: Job Growth in Canadian Provinces

By sector, goods added 12.5k, services were flat at 1.7k. Within goods, resources were up 10.3k with little change elsewhere. Within services, it was mixed with ups and downs across several subsectors (chart 2 again).

Wages are exploding with a caveat. Wages were up by 11.2% m/m SAAR in March. I expected a drop after the 16.4% m/m SAAR gain in February (chart 6). Statcan does note that wages were distorted by compositional shifts. Wages were up by 4.7% y/y but controlling the compositional shift yields wages up 3.6% y/y which has been stable over the past three months. Does this matter? Yes and no. Yes in that you want to control for shifts in terms of wage pressures at the margin. No in that whether people are getting paid more because of accelerating wages or because they're changing jobs/sectors doesn't much matter in a pay is pay and macro sense.

Chart 6: Canadian Hourly Wages

Ignore Canada's unemployment rate. It's fiction. It held at 6.7% in March which is marginally higher than the OECD’s estimate for the NAIRU equilibrium rate. Chart 7. The Labour Force Survey lags the decline in population in Statcan's quarterly numbers because it applies a 12-month smoothed moving average to the temps category of immigration and is therefore slow to turn lower as that's the category being curtailed by immigration policy. Chart 8. The point here being that the UR may fall/tighten when LFS catches up to falling population and LF counts.

Chart 7: Canada's Unemployment Rate; Chart 8: Canada's Population Estimate

Hours worked were up 0.25% m/m SA which is good for monthly GDP, not so good for quarterly (chart 9). 

Chart 9: Canada Total Hours Worked

The gain in hours worked feeds into February GDP tracking 0.3% m/m SA versus Statcan's earlier 0.2 guidance. March is too early (we only have 0.2% m/m for hours, no other data) but this may indicate 0.1–0.2 for March GDP. That would give 3/4% q/q SAAR Q1 GDP using monthly figures, onto which we may add the reversal of the large 4%+ inventory drag on Q4 GDP. That could still leave within reach a 1–2% q/q SAAR GDP expansion, but a) final domestic demand may be tracking decently again like it did in Q4, so it's a textbook case of how GDP isn't always the bees knees especially given violent trade/inventory swings caused by Trump, and b) the BoC is focused on everything that the commodity surge changed, not backward data.