Been away for the past week or so fighting the flu. Hoping to come back to regular blogging soon. In the meantime, two interesting stories from the Globe:
Government study reveals significant errors in voluntary census
In another example from the report, the real 2006 long-form census found that visible minorities as a share of the population increased by 2.77 percentage points between 2006 and 2001. The simulated voluntary approach would have reported an increase of only 0.74 percentage points.
EI harder to get in Ontario, Western Canada during last recession
In 2008-2009, only 46 per cent of unemployed Canadians received EI benefits, compared with 71 per cent and 76 per cent in the recessions of 1981-82 and 1990-91…
The current EI program is designed so that benefits become easier to access in each region of the country as unemployment increases. However, the report finds that partly because of regional programs in high unemployment areas – such as those that assist seasonally employed workers in Atlantic Canada – the program never fully adjusts to serve unemployed Canadians in Ontario and parts of Western Canada.
Worthwhile Canadian Initiative 
That was some pretty bad reporting in that article, about the findings, but here’s what I’ve reconstructed from it:
– StatCan examined the 2006 census results, relative to the 2001 ones, and as a logical proxy for a voluntary survey, they compared what the projected totals for 2006 would be if they didn’t do any follow-up past the initial 74% responses to the long form (which of course is a cheat w/r/t undercounting the small populations they enumerate & didn’t send a long form to)…. to what they actually ended up getting with the 96% or so after all the interventions.
And the results showed big discrepancies in the two areas this article covers: 1) the percentage of households that rent (vs. own); and 2) the visible minority population.
1) I’ve now gathered that the actual final 2006 results reported that 3,841,815 — or 31.5% — of Canada’s 12,210,320 non-collective, private households were rentals. The article notes that was a 3.08 point drop in rental housing units relative to the 2001 census, which’d mean that in 2001, 34.5% of households were rentals.
So, if that proportion were maintained, we’d predict 4,217,893 rental households in 2006 — 376,078 more than there actually were — which represents a potentially alarming 9.8% loss of rental stock (possibly due to condo conversions, demolitions, etc), altho’ it could be a good sign, if it meant hundreds of thousands more were in a position to buy.
But if they’d ceased the long form data collection at the 74% response mark, they’d have found that only 26.5% of the households (the reported 8.07 pt. drop from the 2001 percentage) were rentals: or 3,232,520 altogether, if they projected it right. That would be 985,373 fewer than expected from the 2001 proportions, which’d be a 23.4% drop from prev. level, relative to total no. of private households: which might lead to a huge hue and cry for building a massive amount of rental and social housing stock, costing billions, potentially unnecessarily.
2) The actual 2006 census reported a visible minority population of 5,068,095, which was 16.2% of the 31,241,030
total. The article notes that was a 2.77 pt. real increase as a share of the population relative to 2001 (which means that in 2001 they were just 13.5% of the pop.). But the initial, non-nagged, 74% response found only a 0.74 pt. increase as a share of the population — so, 14.2%
of the population, which would be projected to 4,433,902, and undercount the true(r) visible minority population as of 2006 by 634,193, or -12.5%.
I’m an american college student interested in comprehensive economic policies. I was reading up on germany’s strategy of government sponsored enterprises that are pseudo-private banks that lend to small and medium manufacturing businesses. What are your thoughts on GSEs?
The EI piece is also junk – they compared the total number of unemployed with a measure of newly unemployed (EI recipients, i.e.). The reason a smaller percentage of unemployed in Ontario and the west drew EI during the last recession is because a greater percentage of their unemployed during that period were the chronically unemployed who never held a job prior to the recession (apparently, about 60%). The Quebec and Atlantic economies, being more fragile, suffered more from the recession itself and consequently only about 30% of their unemployed during that period were of the “chronic” variety.
If I had to guess (and it would need to be a guess, since the authors of this “study” didn’t trouble to do the research), I would say that it is quite likely that a higher percentage of unemployed in Ontario and the west were covered by social assistance benefits of some variety during the period of the recession than in Quebec and Atlantic Canada.
In short: Quebec and Atlantic Canada lost more jobs during the recession itself, so their EI claims were higher. Ontario and the West lost fewer jobs, so their unemployed were mostly covered by social assistance. Is this supposed to be a reason to feel bad for Ontario and the west?