Crawl Across the Ocean

Tuesday, March 29, 2005

Canadian Government and the Three Bears

So I was thinking to myself that one of the most controversial aspects of government in Canada (and in most places), is whether government here is too big (papa bear gov't), too small (mama bear gov't) or just right (baby bear gov't) - but I don't really have a good sense of just how big the government of Canada is - especially in comparison with other countries.

Luckily, google was able to bring the relevant information to my fingertips within seconds via this UN site which is a collection of statistics compiled by people trying to compare the size of government across countries. Seriously, how did we ever find out anything 10 years ago?

Anyway, there's a lot of interesting data there and perhaps later in the week, if I get time, I'll try putting some of it in chart form, but for now, I'll pick out some of the highlights, from a Canadian perspective.

There are a few things we need to be careful of (actually there's a ton of things as you'll see if you read this summary of the UN's findings but I'm going to hit the high points):

1. Comparisons which look at central government spending will understate the size of Canadian government because we have a relatively large non-central (Provincial + Local) government segment (more on this later).

2. There are different measures used. Government consumption typically doesn't include cash transfers whereas Government Expenditure does. Government Expenditure is roughly equivalent to Total Tax Revenue, except total Tax Revenue tends to be lower since governments have other sources of income besides taxes (resource royalties for example). Revenue will also be lower than expenditure if the government is running a deficit (I know I'm stating the obvious here, but it's relevant later on).

3. When comparing rich and poor countries it is advisable to adjust for the fact that services tend to be much more expensive relative to goods in rich countries. Or to put it another way, labour is cheap in poor countries. Because government expenditures tend to be service-oriented/labour intensive, government spending - all else being equal - will appear lower in poor countries (because they're not paying their teachers and doctors and civil servants and policeman and soldiers very much).


OK, enough caveats, what do the numbers tell us?

Looking at consumption (which excludes government transfers) as at 2002, Canada had the 11th highest government spending out of a list of 24 'developed' countries. i.e. we were right in the middle, along with the U.S.

The lowest country was Ireland at 13.3% of GDP devoted to government consumption (perhaps reflecting the efficiency of having just one level of government?) followed by Switzerland at 15.2% and Greece at 15.8%

At the high end are Sweden (28%), Denmark (26.1%) and Iceland (25.1%).

Canada came in at 19%, while the U.S. was at 18.9%.

Interestingly, the UN researchers built a statistical regression model to look at the data they collected and determine what factors influence the size of government. What they found was that government consumption tended to be lower (as a % of gdp) in countries with large populations and higher in countries with large geographical areas. This makes sense if you figure there are economies of scale in many government services and that many services (such as road building for example) are more expensive per capita in sparsely populated countries.

Note that both of these factors work in favour of a relatively large government sector in Canada. As the report says:
"Thus, from the raw statistics, Canada appears, relative to its population, to have larger government than the United States. Taking into account the diseconomies of serving a small population dispersed over a wide area, Canada actually has a rather small government compared to that of the United States!"

note: they must have thought this was noteworthy since it merited one of only 2 exclamation points used in their entire 17 page report

Interestingly the results were different when they looked at total government expenditure (which includes transfer payments - i.e. pension plans, child benefits etc.). Here they found that the strongest predictor of government expenditure was how 'open' the economy was, where 'openness' indicated a 'globalized' country which was heavily integrated onto the world economy (low trade barriers, large number of exports/imports as a % of gdp etc.). As the report says,
"on average, governments of open economies spend a significantly larger portion of GDP and collect the additional taxes needed for this task."

The figures for government expenditure were only given for central governments (since they are not collected for many local governments) so I'll look at total tax revenue instead. I found two sources for this data (the UN, and the OECD), and while they were quite similar, they didn't line up exactly. I'll go with the OECD figures since they have cover a greater range of dates.

The Canadian government collected 31.9% of GDP in taxes in 1975. Looking at the data every five years, this figure climbed to a peak of 35.9% in 1990 before levelling off and gradually starting to decline to 33.9% as of 2002.

By way of comparison, the lowest figures in the OECD (2002) are Mexico (18.1%), S. Korea (24.4%), the United States (25.4%) and Japan (25.8%).

At the high end are Sweden (50.2%), Denmark (48.9%) and Belgium (46.4%).

Canada is similar to other countries with British roots with Australia at 31.5%, New Zealand at 34.9% and the U.K. at 35.8%.

Of course one thing we need to consider is the different deficits in different countries. For example Canada's revenue (33.9%) is 8.1% higher than Japan's (25.8). But as we can see here (scroll down), Canada was running a 0.6% of GDP surplus in 2002 while Japan was running nearly an 8%(!) of GDP deficit. So the truth is we were spending around the same, just that we were collecting the tax to cover it and they weren't. Similarly, when you factor in the deficits that most other OECD countries are running, you can see that Canada's spending is lower than the revenue figures make it look (of course factoring in resource royalties may have the opposite effect but I couldn't really find any figures on how big this effect is).

This also means that if Canada's tax take is lower as a % of GDP now than it was in 1990, plus we have turned a deficit of 3.4% (1989) into a surplus, then government expenditure has come down fairly significantly as a % of GDP in the last 25 years.

Now I know what you're thinking, how much of this reduction was reduced program spending and how much was reduced interest payments on our debt? According to the Fraser Institute (source OECD), government debt interest payments were 5.3% of GDP in 1990, and were down to 4.9% in 2000. They've probably fallen more between 2000 and 2002 (maybe to 4.5%?) but it still suggests that most of the change has come from reduced spending as a % of GDP.

The one final number I wanted to look at was the ratio of non-central government spending to central government spending. i.e. How big is the local/state/provincial government sector relative to the national one. As you can see from this chart, Canada leads the world in this category as one of 3 countries (China and the Netherlands Antilles being the other 2) where local/regional government actually outspends the central government. I guess it doesn't tell us much that we didn't already know - Canada is a very decentralized country. But at any rate, it's something to keep in mind next time you hear about the 'fiscal imbalance'.

But wait, you say, the provinces may spend money but the federal government collects the taxes and transfers them to the provinces so even though the provinces spend a lot they don't have control over their spending. So we really need to look at tax revenue collected by local/provincial governments vs. the central government. OK, here's the figures based on tax revenue. It's true that the ratio of local/provincial taxes to federal ones is lower than the ratio for expenditures but (with the exception of China) this is true everywhere - and even if you look at tax revenues Canada remains a world leader (now #2 behind China) in decentralization.


So, tons of numbers (which I really should put into some charts) but what does it all mean? Looking at the historical record, government spending is lower as a % of GDP now than it was 25 years ago. Also, just looking at the raw numbers, the size of Canadian government doesn't look out of place in any international comparisons. When you factor in Canada's dispersed population, its open economy and the fact that it's running a surplus instead of a deficit (not to mention the public pension system being adequately funded) Canada's government appears to be one of the smaller ones among developed countries. Given that it also has one of the world's most decentralized systems of government, our federal government can be seen to be one of the smallest in the developed world.

Or to put it another way, Andrew Coyne ended this lament on our high spending (written in 1994) by saying, "Switzerland and Japan can run a welfare state on a third of GDP or less. Why can't we?". So he should be happy to know that as of 2002 there is very little difference in total government expenditure between Canada and Japan and Switzerland and that government expenditure as a % of GDP in Canada is running right around 1/3 of GDP.

Or to put it another way, most signs seems to suggest that, in contrast to those who would suggest that a) Canada's government is an ever growing monster that is threatening to swallow us all, or b) Canada's government is getting ever-stingier and is callously dooming poor people to a life of poverty, the size of our government seems both stable and reasonable.

Or to put it another way, Goldilocks went to Sweden but the taxes were too high. 'This government is too big', she whined. So Goldilocks went to Mexico but her car broke down on the bumpy roads, her tires were stolen and the air was thick with smog. 'This government is too small', she whined. So Goldilocks went to Canada. "Ah, this government is just right", she sighed.

So how does the story end? I'm guessing some people will end it with:

Goldilocks fell asleep in the land of good government and while she was sleeping the left wing Papa Bear returned and ate her, dooming her to a life of ever larger government, stifling her incentives, curtailing innovation, trampling on her freedoms, spreading inefficiency and turning the country into Sweden. The End.

While others might end it with:

Goldilocks fell asleep in the land of good government and while she was sleeping the right wing Mama Bear returned and ate her, dooming her to a life of tax cuts for the wealthy, corporate oligarchy, environmental degradation and disregard for the suffering of the poor and unfortunate, turning her country into Mexico (with a stop in the U.S. along the way). The End.

For me, I agree that we can't afford to fall asleep, but I think we need to make friends with the baby bear. After all a mama or papa bear wouldn't eat one of its kid's friends would it?

Labels: , , , ,


  • You should post at Blogs Canada e-group.

    By Anonymous koby, at 10:00 PM  

  • Good post Declan.

    My head is spinning now though - I'm going to have to re-read it all to try to take it all in.


    By Blogger Andrew, at 7:00 AM  

  • Excellent post.

    I recall that comparisons in government spending between Canada and the US have often had us neck and neck.

    However, America spends huge amounts on its military compared to us, so some people point to that as evidence that we overspend in other areas.

    Then again, without such a huge military, America would have to find jobs for a lot of people, pay for better college access, or make due with a less-educated workforce.

    Also, America's deficit is huge - around the same size as the military budget. Hmmmm....

    By Blogger Mark Francis, at 7:35 AM  

  • koby - Yeah, I'm never sure whether to cross-post stuff at the e-group or not - stuff like this just seems to long and involved to post there, but maybe I'm wrong.

    Andrew - thanks - yes my head was spinning a bit too. It's certainly possible that I have misinterpreted stuff in the piles of charts but I tried to keep it straight.

    mark - You make a good point about military spending as it is a factor which I didn't consider but which no doubt makes somewhat of a difference (about 3% of gdp I figure, Canada vs. U.S.). It's probably not much of factor for other developed countries since they most likely all spend around 0-2% of gdp on the military.

    By Blogger Declan, at 11:16 AM  

  • Social entitlemenets equal social engineering (health care, official multi-culturalism, gun control,bilingualism,CBC) equal spending your tax dollars to keep the ruling party Liberal Party in power. Less government ,less social engineereing . Government spending as part of GDP ? The government does not make anything ,civil servents shuffle paper , that does not create any value to the economy.LTP

    By Anonymous South of Bloor st,Leon., at 8:55 AM  

  • "Social entitlemenets equal social engineering"

    Not true, the vast majority of social 'entitlements' are transfers of money to poor people - primarily children and the elderly.

    "(health care, official multi-culturalism, gun control,bilingualism,CBC) equal spending your tax dollars to keep the ruling party Liberal Party in power."

    Health care is simply a service provided by the government, there's no 'engineering' there. As for the rest a) they amount of money spent of htem combined is trivial in terms of our GDP and b) they generally seem like good ideas to me.

    "The government does not make anything ,civil servents shuffle paper , that does not create any value to the economy.LTP"

    The government build roads, sewage treatment plants, schools, embassies, the list is endless. Not all of what government does is tangible, and a lot of it involves paper, but last time I checked nobody was arguing that all banks or accounting firms or law firms should be shut down even though all they do is 'shuffle' paper.

    OK, maybe some people would argue this, but it's not really a point of view I can take seriously.

    By Blogger Declan, at 11:44 PM  

Post a Comment

<< Home