Playing With Numbers & Assumptions

Paul Wells suggested I should read this column by Andrew Coyne, and it looked promising at first, since I'm always partial to anyone pointing out that the Premiers are a bunch of whiners who should stop blaming Ottawa for their troubles and focus on managing their own affairs properly.

But I have to take issue with a couple of implicit assumptions in Coyne's piece. Says Coyne:

"That Ontario, on the other hand, still manages to show a deficit is not because it is short of revenues. No government in the history of Ontario has had as much revenue at its disposal as the current government of Ontario. Total revenues from all sources this year will exceed $6,600 per capita. By way of comparison, the Harris government took in about $5,700 a year, on average, in constant 2005 dollars. (Because it slashed taxes? No: the previous NDP government under Bob Rae, which raised taxes, got by on just $4,900 and change from each of its citizens.)"

The thing is that, while per capita revenue has gone up (in constant dollars), so has per capita income. Coyne seems to be suggesting that government spending as a percentage of GDP should always fall as incomes rise. But what does the Provincial government spend money on? Largely on health, education and infrastructure (not to mention interest on debt - another reason revenues have had to rise). Can we assume that Ontarians want to spend a smaller percentage of their income on health, education and infrastructure than they did a decade ago? Aren't we facing a 'crisis' because people want to spend a *higher* percentage of their income on health than they used to?

Coyne may have a point about provinces spending irresponsibly, or he may not, but we'd need to see the revenue and spending numbers as a percentage of GDP to really know one way or the other.

The second assumption Coyne makes is that,

"a cut in marginal income tax rates isn't a one-time thing: it's a permanent contribution to raising productivity across the province's economy -- an investment, if you like, that will pay dividends every year in perpetuity."

Here's the thing, if the money is collected in tax it will be spent by the government. If it isn't collected in tax, it will be spent by taxpayers instead. I'm not sure it is so obvious that taxpayers will spend it more productively. Now if government spending was, say 80% of GDP, it would be pretty clear that society would be more productive with taxes lower. Similarly, if government spending was 2% of GDP, it would be pretty clear that society would be more productive with taxes higher. So somewhere in between is a tipping point. Of course it all depends on how the money collected is spent.

So I see no basis for this assumption that a reduction in marginal income tax rates will boost productivity. In fact I'm sure I could easily round up example of countries with higher marginal income tax rates than ours which also have higher productivity and higher productivity growth, along with examples of countries with lower marginal tax rates and lower productivity and/or productivity growth.