Quantifying the effect of MB Infrastructure spending

So anyone who follows me on twitter has seen that I have been debating back and forth with my NDP MLA about the effect of the MB government infrastructure spending on the provincial unemployment rate and how it compares to other provinces.

My position is that unemployment is driven more by the nature of a province’s economy and global factors as opposed to direct government investments, policy or the political party in power. To support my argument I have provided historical unemployment rate data ( below) that shows Manitoba’s comparative position hasn’t really shifted too much over the last 25 years. It usually falls between 2nd and 4th in Canada. I’ve also pointed to more detailed Stats Can data beyond government news releases and press clippings which shows that our current position is more of a result of other provinces falling off than Manitoba doing better.

Historical Unemployment Rates by Province:

unemply history

Note that MB position since 1976 is between 2nd and 4th

My MLA’s argument is that the PST infrastructure spending is driving the increase in MB’s comparative unemployment position. To support his argument he states that the gov’t is spending lots of money on infrastructure.

But what if we could quantify the effect of the provincial government infrastructure spending? What if we were able to see exactly what the effect of this infrastructure spending on unemployment is?

I think we can.

Using some documents I’ve pulled from the provincial government website and stats can data, we can see the effect of the provincial infrastructure spending on Manitoba’s unemployment rate.

The provincial government has a couple of Conference Board of Canada reports on their website. Here’s the initial one that purports to show the effect of the PST infrastructure spending. Now, I’m going to assume their methodology in calculating the effect on an annual basis is correct, but that’s a huge stretch since the application of the methodology is so deeply flawed. I’d be embarrassed to work for the Conference Board putting out material like this. The report assumes each year on a stand alone and then adds them. But that’s not correct. It’s essentially quintupled the effect, and also neglected to factor in prior spending.

So this graphic?

58900 jobs

It’s not 58,900 jobs, it’s actually less than 4,000. But more on that later.

But lets just assume their behind the report calculations are correct, even if the application is so deeply flawed.Let’s look at the Conference Board’s report on 2013-2014 infrastructure spending.

Specifically, lets look at table 2 from page 6:

Table 2
Now, note the 1st line, we can link that to another publication in a few minutes. Lets just look at the 1st line and the employment line. This indicates that for $844M in spending, 8,200 jobs are created. Or $102,926.83 per job created. Let’s just save that for later.
Now lets look at another document from the same site. Lets look at the 2014/2015 annual report on infrastructure spending put out by the provincial gov’t.
In this case, lets look at the 5 year spending plan from page 7:
5 yr plan
First, note the planned investment of $844M, that links to our earlier Conference Board of Canada document. So we’re looking at comparable data here. But we have to do some major restatement of the data as listed to correct for flaws in the data:

The base funding level for each year seems to be the 2012/13 base. But that doesn’t factor in inflation or any annual increase to the baseline that would have occurred without PST spending. So we’ll go and adjust the base by 2.5% per year. and remember that’s an incremental amount, not just increase each year by 2.5%. Our adjusted numbers now look like this:

Adjust 1

Quite a change, isn’t it? But we want to look at incremental PST funding. That’s PST spending above what would otherwise be spent. Put another way, Planned investment less the restated base 2012-2013. Those numbers look like this:

adjust 2

The highlighted numbers are the incremental PST spending per year based on a reasonable 2.5% increase to base funding each year. Now those numbers are on a stand alone year by year basis. We also need to look at the year over year change. This is important – if you’re hired to work on a job over 5 years are you one job or 5? The above numbers reflect each year as being a separate job. So let’s look at the incremental PST spending change from year to year:

adjust 4

So now we know an adjusted amount and how much that increases year over year. The increases from year to year are what we need to calculate additional jobs. Now remember back to the 1st document  we looked at? Where we calculated this: “Or $102,926.83 per job created. ” We can apply that rate to the above numbers to come up with the following:

adjust 5

So 2,550 jobs created to-date due to MB Gov’t PST spending. Okay, we’re close to the end of this now!

Let’s go back to that 1st table though:

Table 2

 It also states that for $844 million in spending, the labour force increases by 3,800. That 3,800 is 46% of 8,200 listed employment. Those are jobs taken by people from outside the province and are not jobs for unemployed Manitobans.
Now that we have our total on jobs created to-date, let’s look at the Stats Can unemployment data for October 2015:
Okay, now that we have all that data, lets put it into this and see what happens. We’ll use October. Note that the above numbers are in thousands.
We have 2,550 jobs from PST spending and 46.3% of those or 1,173 are people moving to Manitoba to work. That’s a net 1.327 jobs to date from MB PST infrastructure spending.
So if we back out those numbers we see what would be without MB infrastructure spending:

Labour Force decreases from 678,000 to 676,827

Employment decreases from 641,800 to 639,250

Unemployment therefore is 676,827- 639,250 = 37,577

Unemployment rate is 37,577 / 676,827 = 5.5%

So there you have it. The effect of MB PST infrastructure spending on unemployment:  0.2%

And since Saskatchewan’s unemployment rate in October is 5.6%, the effect of MB Gov’t PST infrastructure spending on unemployment position relative to rest of Canada? Nothing. Still #1 regardless.

So much for the argument that infrastructure spending is driving Manitoba’s #1 unemployment position.