It is widely believed that the March 29 federal budget will make cuts in program spending, and the Conservative government seems to have done a very good job of preparing public opinion for them. I've written a few Economy Lab posts (, , ) on the topic without actually coming down on any side of the should-they-or-shouldn't-they question, but I guess it's time to get off the fence.
That graph should also make it clear that Canada is in a different situation than the UK and the US. The UK started cutting well before employment had recovered, and the US is still nowhere near that point. But we shouldn't be drawing general lessons about how contractionary fiscal policy is always and everywhere a terrible idea.
Of course, a severe contraction on the order of magnitude of the Chrétien/Martin years would be a very bad idea: the Bank of Canada has very few bullets in reserve, and we can't count on strong US growth to compensate for a severe fiscal contraction (although developments on that front are more promising than they have been for years). But that's not going to happen, because there's no need for it. The PBO's estimate for the structural deficit for 2012-13 is $14b, less than one per cent of GDP.
The Conservatives aren't going to raise taxes to reduce the deficit. Neither did the Chrétien Liberals - the last Prime Minister who raised taxes in an attempt to balance the budget was Brian Mulroney. I don't have anything against cutting programs per se - there are doubtlessly some programs that wouldn't pass a cost-benefit analysis and should be cut. And it would be a good idea to prune back some of those tax expenditures.
But I don't like Procrustean budget cuts: chopping programs simply to reach a spending target. Too often, programs that are cut for budget reasons end up being reinstated anyway, and laid-off public servants being re-hired as consultants at a higher cost. But that's what we're going to get.
I keep referring to Table 5.9 of last year's budget, because it's the only available indicator of where the government want to go, and past experience suggests that when this government says it will do something, it generally does it - for good or for ill.
So here is Table 5.9:
The strategy is to hold direct program spending constant - which means a reduction in terms of share of GDP - and letting transfer programs grow with GDP. Revenue growth - personal income tax revenue growth in particular - is supposed to close the gap between spending and revenues (Table 5.8 of the budget).
I have my doubts about whether that plan will last four years: that's a long time to sustain a regime of continual program erosion ("cuts" doesn't seem to be the right word). Stuff happens.
But if the plan is implemented, I have a hard time seeing how it could be all that big of a deal as far business cycle analysis goes. Spending cuts on the order of $4b-$8b are just under half of a per cent of GDP, roughly equivalent to a one percentage point increase in the GST. That's not going to put us back into recession on its own.
I suppose I have to pause here and comment on this report, in which the Canadian Association of Professional Employees (CAPE) claims that budget cuts of $8b would produce job losses of some 116,000, of which 61,000 would be in the private sector. This didn't pass my smell test (there's no way an equivalent tax increase could produce a mini-recession), so I decided to take a closer look. Here is the press release, which in turn links to explanatory notes here and here. It turns out that the analysis is based on passing the budget cuts through StatsCan's input-output tables and adding up all the changes.
As macroeconometric policy analyses go, this is a remarkably amateurish attempt: using the input-output table coefficients to predict behavioural responses to a change in fiscal policy is simply wrong. It's a static model with no markets, no prices, no budget constraints, only y = A∙x → ∆y = A∙∆x. The CAPE analysis adopts the same methodology as all those dumb "economic impact" studies that promoters use to show how a government subsidy for their pet project will generate enormous spillovers, and deserves the same weight we give those exercises. That is to say, none at all.
Last year, the arguments for delaying a fiscal contraction were that the recovery was still not sufficiently advanced, and there was the risk of something bad happening in the US and/or Europe. This year, the recovery is one year older, the US is finally showing signs of a sustained recovery, but there's still the risk of something bad happening in Europe.
We already know that waiting until the moment is exactly right before dealing with the deficit is in itself a risky strategy: there is always going to be a risk that something will go wrong. If things do go pear-shaped in the next few months, the policy stance can be adjusted accordingly.
Although I disagree with the way that the Conservatives have chosen to bring it about (my preference would be an increase in the GST), I think a mild fiscal contraction is the right choice for this year's budget.