From @JustinTrudeau this morning: EI claims reach 500k this week. By far the most ever recorded. #cdnecon 
That 500k is ~2.6% of employment, which is the first confirmation we've had that March 2020 will be the largest employment drop in recorded Canadian history. Previous record was -2.6% for the whole of July 1932. We've reached that drop now in a single week.
Given the interest in this tweet, I should give a citation: historic employment data is available through the Canada Year Books: www66.statcan.gc.ca/acyb_000-eng.h… It's labour intensive to collect, but it's there. The -2.6% is the change in my own estimate of the seasonally adjusted data.
Here's how it looks (from page 829 of the 1934/35 yearbook). www66.statcan.gc.ca/eng/1934-35/19… The data are reported monthly in tables (but you have to go through various yearbooks one by one. sometimes you luck out and the provide two years of monthly data in one place.)
You can see the seasonality in the data. So, using the ggseas package in R I seasonally adjust the raw data. This allows for comparisons to how employment changes are reported today (it's the s.a. data we typically look to). The result.
And it's the historic data where you need to look for the massive monthly drops. Over the past half century, one can access www150.statcan.gc.ca/t1/tbl1/en/tv.… easily. No month has *ever* seen a drop larger than 0.74% (which we saw in June 1982).
Of course, there's uncertainty over what March 2020 data will show. The full effect might not be known until the April LFS numbers (released early May). But it seems clear to me that this is the sharpest negative shock we've ever seen. Despite the depth, hopefully it'll be short.
