On the history of the federal government's structural balance

In the report it released last month, the Parliamentary Budget Office provided estimates and projections for the federal government's structural balance - that is, with the effects of the business cycle stripped out. This is always a tricky business, but their historical numbers resemble those produced by the Department of Finance (Figure 4-1), and the methodology used to generate them looks reasonable enough given the time and resources the PBO had available. So they are certainly worth taking seriously.

They very kindly provided me with the numbers used to produce Figure 5-2. Here it is, without the numbers for debt service charges:

And here is the structural balance, along with the structural primary balance (that is, not counting debt charges):

Some comments:

• Brian Mulroney's government doesn't get near enough credit for the progress it made in dealing with the federal deficit. Eliminating the structural deficit required an 8-point swing in the primary balance, and half of it occurred under his government.
• The Mulroney government split its 4 points equally between tax increases and spending cuts. The Chrétien government's contribution consisted entirely of cuts to spending.
• Debt charges as a share of potential GDP have fallen by more than 2% since 2000, so the deterioration of the total balance hasn't been as severe as that of the primary balance.
• Since 2000, revenues have fallen at a rate roughly twice that of the increase in spending.
• Stephen Harper's government has followed the trends set out by those of Paul Martin and Jean Chrétien.

Comments

[Frances Woolley]

And the point is?
- taxes have been cut so much over the last few years there's lots of room to raise them again?
- the only way to really reduce spending a lot and fast is with a Paul Martin (the Finance Minister version) program review?
- part of the healthy budget balance we've had in recent years has been due to lack of debt service charges?
- the federal government is much less important than it once was in terms of program spending, and - though you don't show it - the provinces are more so?

Just wondering what the worthwhilers think...

[Stephen Gordon]

I really wasn't going anywhere here, just playing with the PBO numbers.

If spending is cut, it's hard to see how it could be done without cutting the transfers to the provinces - which is a lot of what Martin did. And the 2:1 tax cuts:spending ratio is useful for context on just how the surplus turned into a deficit. We're hearing a lot about the Conservatives spending spree; this puts that into perspective.

[Andrew F]

Did you see the blog that Coyne wrote about his suggested plan to return to balance? When weighing that versus a 2% hike in GST, I can't believe he thinks the >1/6th program spending cut is more appealing to politicians or the general public. This is not to say that there is no room for re-evaluating spending priorities, but given the collective shrug over the reduction in the GST, I think it would be straightforward (if not easy) to spin a yarn about needing to restore the GST to return to balance.

[DR]

Why were revenues so random during Trudeau's final term? They covered the whole range in 2-3 years.

[Patrick]

DR: side effect of the CB engineering a severe recession to squeeze out inflation?

[Jim Rootham]

The chief statistician at StatsCan wrote a paper identifying the high interest rate regime as the primary cause of the deficit and got mugged by the Finance department for his troubles.

[Geoff NoNick]

I'm always leery of data interpretations like this - while it's useful to display the data this way for the purposes of historical analysis, it's a dangerous way to make policy because it inherently buys into a certain social-democratic mindset: namely, that program spending does or should move in relation to total GDP.

While revenue is obviously directly correlated to GDP (the tax base is drawn from the wealth of the nation), I can see no reason why program spending decisions should bear any relation to GDP. Isn't it conceivable that GDP could rise markedly without a commensurate increase in program spending? Certain social/political elements would like us to believe not.

Not saying you're of that bent - just an observation.

[Stephen Gordon]

Well, I did once write a post with the title "Why share-of-GDP is the correct measure for the size of government". The basic argument revolves around Baumol's Disease as applied to the public sector.

[Jim Sentance]

On the Trudeau question, haven't looked at the numbers but I am reading Just Watch Me and my quick guess would be oil revenues - big jump with the NEP, then a big fall as oil prices tanked.

[Neil]

So it all comes back to the ever unpopular GST. It's a really effective tax, but because it's so politically unpalatable, and basically destroyed the PCs, nobody is willing to bring it back to where it needs to be.

[Geoff NoNick]

The Baumol's Disease argument is interesting (being a recent follower of Worthwhile, I hadn't seen it), and I particularly respect your identifying the central debate to be one of politics (what is the role of government?) rather than economics (how much should government cost?). I also agree that given the traditional Canadian answer to those questions, share of GDP is a reasonable yardstick for government spending.

But I remain leery of the assumption of the share-of-GDP by, say, "progressive think-tanks" (which may be using it as a means of advancing less economically sound spending policies), or certain sectors of the public service, who should recognize that use of that model implies an answer to a political question that has yet to be decided. In fact, it could be argued that the present government has decided its stance on that question, and that use of the share-of-GDP model advocates against the government on a matter of policy.

[Style]

An important point is that the Chretien government shifted many responsibilities onto the provinces already, so Coyne's suggestion that Harper abrogate the current agreements in order to balance the federal budget is pretty simple-minded. Also, the Chretien government cut federal spending out of other programs, like EI, in a way that shifted burdens onto rate-payers in wealthier provinces.

On the GST, the Paul Martin approach would be to maintain the lower tax rate but cut the GST rebates by a sufficient amount to rebalance the books. Anyone want to advocate that?

[Frances Woolley]

Style - by GST rebates do you mean the refundable GST credit to lower income Canadians, lower input tax credits, or the scope of zero rated and tax exempt commodities? I don't think the first is big enough to give you much by way of budget savings and would be a bit harsh, the second would compromise the integrity of the GST, but the third I could see some scope for.

On the Trudeau era, this extends Steve's analysis back a few years... http://dsp-psd.pwgsc.gc.ca/Collection-R/LoPBdP/CIR/903-e.htm
No smoking gun for 1980, except (my guess) the big recession that year and possibly expansions in RRSPs and RHOSPs.

[Style]

I mean the refundable GST credit (for which I cannot find numbers right now). The other two seem more like tax increases.

[Just visiting from macleans]

I once intervened against a provincial crown corporation (electric utility)and a private gas utility that were both seeking rate increases from their captive customers.

The Chair of the Commission was an economist, and the formuala that they were using for Operating Maintenance and Admin expenses will look familiar to some: OMA(this year) = OMA (last year) x (1 + growth in customers) x (1 + inflation)

Here was the problem: The gas company had just amalgamated with another a few years earlier, and had not reduced their fixed costs (duplicate management and admin). So, they had yet to enjoy economies of scale.

The electric utility had invested $$$ in IT systems (based upon a business case of increases in productivity as is often the case - do more with less), and in addition, had reached maturity in its operations. So, while in the past, it may have cost a lot of money to add a new customer, once the infrastructure was built, and computer systems installed, it didn't now, or at least shouldn't. (IT business case justifications have a history of being forgotten when making the budget for the following yr. Savings in one area are consumed in another.)

So, in both cases the formula utilized by the commission didn't apply. Additional revenue so granted just ended up being consumed by the utilities in inefficiencies - ie their budgets grew based upon the (1+g)(1+i) formulas.

My point of intervention: The companies will NEVER meet economies of scale if you are starting at a point where there are large redundancies or inefficiencies built in, and perpetuated by how you set budgets through formiula.

As a marketing prof once pointed out: All fixed costs are variable; All variable costs are fixed. Perhaps a microcosm of what happens on a larger scale in Fed budget making.

The general shape of the declining red line over time, flattening out at a lower % of GDP would be what I would generally expect - notwithstanding Baumol's Disease. Similar to Coyne's arguments made elsewhere. In fact, doesn't Canada's recent experience over the past 35 yrs as shown largely refute Baumol's?

[westslope]

Style: If I understand correctly, these EI reforms lead to raising more Employment Insurance revenues in rich, low unemployment areas. That means the net subsidy to unemployment workers--including seasonal resource workers and self-employed fishing category workers--increased in areas experiencing high unemployment rates.

Translation: The Chrétien Liberals increased the net subsidy to overharvesting quasi-renewable resources. If fish populations are going to collapse, why not help pay for the fish stock collapses?

Chanted to the tune of the beer advert:
My name is Self Loathing and I AM Canadian.