How the federal government went from persistent surpluses to persistent deficits

Update: See this follow-up post, which corrects some mistakes made below. I've added comments pointing out which parts needed revision.

In the early years of the millenium, the federal government ran what appeared to be an indestructible series of budgetary surpluses. No matter how many tax cuts were implemented, it ended each year with a sum of money so large that its biggest fiscal policy challenge was to find legal ways to spend it.

Those days are over, of course. Dealing with the deficit will - or should - be a theme of the upcoming federal budget. So it's worth spending some time thinking about just how the policy decisions of the current and previous governments managed the trick of transforming that indestructible surplus into what looks to be a stubbornly persistent defict.

Many or most of the numbers you'll see on public finances use annual data, but I've found it useful to keep track of the monthly numbers published in the Fiscal Monitor. The most recent annual data are for the 2009-10 fiscal year that ended last March, the latest Fiscal Monitor release is for December. The downside with these numbers is that there are strong seasonal cycles in many of the components of revenues and expenditures, so I'll be working with 12-month moving sums (when expressed as shares of GDP, they will be 4-quarter moving sums). With these numbers, it's easier to pick up turning points and inflection points in revenues and spending.

Here are the basic numbers; debt service charges are included in the expenditures series:

I should add that I've removed the one-time transfer of $7.2b made to the provinces in March 2005. This was an extraordinary event that would otherwise obscure the trends in the moving sums: a sharp jump followed by a sharp reduction a year later.

Of course, the proper scale for the size of government is as a share of GDP, so here are the same numbers as a per cent of GDP:

The numbers after 2009 are driven by the recession, but my point is that the seeds of future deficits were sown by policy decisions made before then. Let's zoom in on 2003-2008:

During the three years before 2005, the federal government ran surpluses on the order of 1% of GDP, based on expenditures and revenues that were fairly stable. Then something - many things, actually - happened. Here are the questions for which I've been trying to find answers:

1. Why did revenues rise in 2005?
2. Why did revenues level off over 2006-7?
3. Why did expenditures rise in 2005?
4. Why did expenditures level off over 2006-8?
5. Why did revenues fall in 2008?

1. Why did revenues rise in 2005? It looks like the answer is simply strong economic growth, led by personal income tax revenues:

 

It should be noted that the 2005-06 jump in PIT revenues is quite impressive when put in context:

The sharp reductions after 1999 are due to the Chrétien-era cuts to personal income tax rates.

I haven't gone that deeply into the PIT numbers, but it seems to me as though the story behind the 2005-06 surge involves a working-age population that was still growing, an employment rate that was hitting an all-time high and strong wage growth. If I've missed something, please let me know in the comments. [eta: The answer is that I missed a data redefinition; see the follow-up post]

2. Why did revenues level off over 2006-07? It's clear from the second-last graph that the answer is not that all sources of tax revenues leveled off. PIT revenues continued to increase in 2006 and then fell in 2007, but other sources had the reverse pattern: the result was a wash.

Here's what was happeniing in those other sources:

The newly-elected Conservative government cut the GST by one percentage point in July 2006, and this brought about a drop in non-PIT revenues. But this was a one-time effect; by 2007, the growth in corporate tax revenues was enough to bring non-PIT revenues up.

Once again, it's useful to put those GST and CIT revenue swings in perspective:

So the apparent stability of revenues over 2006-07 is a bit misleading; there were many important shifts in PIT, GST and CIT revenues that just happened to cancel each other out.

3. Why did expenditures rise in 2005? Here is a thing that I had heard, but which I really hadn't appreciated until I started looking more closely at the numbers: for the most part, the federal government is in the business of writing cheques and running the armed forces. In the fiscal year 2009-10, these activities cost the federal goveernment $216.6b. The cost of operating all of the other ministries and agencies was $58.2b - and in 2009-10, that latter number was inflated by the government's stimulus program. Any discussion of government expenditures that focuses on operating expenses of the non-defence ministries is going to miss the point by a wide margin.

So it's probably not surprising that the surge in expenditures came from transfer payments:

Interestingly enough, the increase in transfers was fairly evenly distributed between transfers to persons and transfers to provinces:

I don't have a short answer for just where those increases came from. A longer answer would include the Canada Health Transfer and the Universal Child Care Benefit, but a complete answer would be beyond the scope of this post. What does matter for the purposes of this post is that these increases were the result of policy. The major transfer that is driven by the business cycle - Employment Insurance payments - were declining over this period. [eta: The increase in transfers to persons is part of the data redefinition problem; see the follow-up post]

The last graph also answers the next question:

4. Why did expenditures level off over 2006-8? The growth rates of transfer payments to persons and the provinces were matched by the growth rate of GDP.

5. Why did revenues fall in 2008? It's important to note that revenues were falling before the recession started:

Personal income tax revenues were stable in 2008; the decline was due to the fall in GST revenues following the reduction of the GST rate from 6% to 5% at the beginning of 2008, as well as a return of CIT revenues back to its previous trend:

 

Here are the policy decisions made between 2005 and 2008 that generated that swing in the pre-recession budget balance:

• Transfer payments increased by roughly 1% of GDP. [eta: Make that 0.5%]
• The GST cut reduced revenues by roughly 0.75% of GDP.

Put the two together, and we get a trend towards a deficit even before the recession hit.

In a subsequent post, I'll go into the reasons why exiting the recession won't be enough to restore the federal budget balance.

title('Nominal GDP and CPI: UK and Canada')

Comments

[Shiner]

Fantastic post, thanks for this!

[dj]

They could start by cutting "Indexing" of the Pension plans

[Robert McClelland]

They could start by cutting "Indexing" of the Pension plans

Why is every rightwing solution to screw over Canadians who aren't in the upper income bracket?

[Joseph]

Is there some reason why returning the GST to 7% is an issue? I was never a huge fan of paying for the GST when I was in Canada but I recognize that one has to finance government in some manner and it seemed to be a pretty reasonable way to do it. Is the effect of going from 5 to 7% really so profound? Can't we mitigate the impact on the truely impoverished?

But with the aging population in Canada (and here in the US) one would tend to think that this is not really the time to undermine government finances. Curiously enough, that is precisely what is happening in both countries. It's a strange world . . .

[Mark]

"Is there some reason why returning the GST to 7% is an issue?"

If you believed a media, raising a VAT is equivalent to committing seppuku.

If you talked to individual citizens about factors influencing their household, I think it would be greatly different.

[Georgian]

Excellent.

[Bob Smith]

"If you talked to individual citizens about factors influencing their household, I think it would be greatly different".

Ask Gordon Campbell if raising the VAT is a media myth.

On an individual level you're probably right, but tell that to the governments of BC and Ontario, who are getting kicked in the teeth because they REPLACED one tax (and a badly designed one at that) with the HST. That change (and the revenue growth that came with it - who are we kidding, it wasn't revenue neutral) will probably be permanent (though we'll see, voters might kill it in BC), but not before it costs the politicians who implemented it their jobs.

[Andrew F]

Well, increasing the GST would be an embarrassing reversal for this government, so they would never implement it. It's not obvious to me which source they could tap if they chose to increase revenues. Similar to the GST, CIT won't be increased as it would be a reversal of their policy to decrease rates there. A carbon tax is out of bounds because of their wrong-headed campaign against that policy. What remains? Personal income tax? I can't see them raising these rates either.

So, any improvement in budget balance by this government will likely occur in the form of spending cuts. They like to claim now that they won't touch transfers to people or provinces, or the military. So they intend to squeeze ~$15 billion in savings from ~$30 billion in spending. Good luck. My bet is that provinces face a cut in transfers.

Either that, or this government is replaced by one that is not so constrained by their previous positions.

[Tuzanor]

It will be interesting to see what the Conservatives inevitably do, as their rhetoric up until now has constrained them. Even excluding the stimulus, they've been increasing most of the structural spending by a rather large amount on Conservative friendly programs (the military, elderly, etc) and leaving Liberal friendly ones alone for now (probably due to the minority situation). There was a symbolic $1B cut early in the administration, but the spending it seems was mostly redirected.

It seems to me that the government inherited such a great fiscal position within what seemed like a healthy economy that they thought they could cut taxes and increase spending and continue to pay down debt. Remember how smug Flaherty was in his first 2 years? These graphs show great speed in their changes.

If the current government survives, especially if they get a majority, I'm guessing we're going to see a repeat of the 1990s and a squeeze to the provinces or maybe some thinning of the civil service. Stephen, do you think transfers can be cut enough to return to balance, ignoring what it will do to provincial budgets? In the 1990s the spending cuts were rather across the board, from the military down to the civil service and privatization of crown assets and the brand new GST. The structural deficit isn't the same caliber as it was in 1993, but neither are the options. Could they cut spending elsewhere and make it look like ending the stimulus? Defer infrastructure funding other than arenas for for-profit organizations?

[Bill M]

I don't think any politician is constrained by previous positions.

[Jim Rootham]

You say the government is writing cheques and the armed forces. Where does the RCMP fit into that?
Of course, a large chunk of RCMP spending is financed by the provinces and municipalities that hire them.

[Stephen Gordon]

The RCMP is a biggish - $2b - item in the Dept of Public Safety and Emergency Preparedness. So it's part of the non-defense, non-transfer pie.

[Bob Smith]

$2 billion is only "biggish" for us mortals, for the feds, that's a rounding error.

Tazanor said: "'m guessing we're going to see a repeat of the 1990s and a squeeze to the provinces or maybe some thinning of the civil service. Stephen, do you think transfers can be cut enough to return to balance, ignoring what it will do to provincial budgets?"

I think that's exactly what's going to happen (we'll certainly see cuts, in real terms, to the CST - the federal government should never have agreed to lock in a 6% increase every year). And truth be told, that's what should happen. Setting aside equalization (for which, for obvious reasons, you need federal involvement), it makes no sense to have the federal government raise revenue to give to the provinces to fund their programs, when the provincial governments have the self-same instruments of public finance as the federal government(at least the major ones, i.e., the ability to impose value added taxes and income taxes). I suspect that the provinces will be told that, if they need to fund their spending commitments, it's high time they explained to their voters that those commitments have costs. (In fairness, there will be a quid pro quo here, I suspect the provinces will be encouraged to make more use of the HST system and will also be given greater flexibility in terms of how they deliver health care).

That would certainly explain the apparent reluctance of the Tories to go to the polls now, instead of in late 2012. With a spring election, the subsequent election would be in spring 2015, while if the election is deferred until fall 2012, that gives the Tories until fall 2016. Fighting a knock-drown, drag-out battle with the provinces in late 2013-early 2014 (when the current deal with the provinces expires) will be a hell of a lot easier in latter scenario than the former. Perhaps the Tory reluctance to go to the polls now (despite positive polling numbers) isn't so feigned.

That said, politically, it might not be as suicidal as some might think. Certainly in Ontario, BC and Alberta, federal transfer to the provinces should be a source of some frustration, because, inherently, these sorts of roughly per-capita transfer involve further transfers from taxpayers in those (relatively rich) provinces to those in poorer provinces (over and above equalization). God alone knows why the premiers of Ontario are always yapping about increased federal transfers (other, I suppose, is that they don't mind the federal government taking more from their citizens than they get back, provided that they don't bear the political cost). A federal Conservative government with sympathetic allies in those provinces (entirely possible, given the way things are going in all three provinces) might be able to have that argument made for them in those provinces without having to do it themselves.

[Andrew F]

As a resident of one of those three provinces, I would largely be satisfied with a large transfer of tax room with concomitant reduction in transfers. Ontario is really going to need to do something drastic--the deficits here seem less under control even than Ottawa. If Ottawa uses this method to eliminate its structural deficit of about $15 billion, that implies a worsening of Ontario's fiscal balance by another $5 billion.

[Dieter H]

Is no one going to mention the forthcoming $16 billion untendered contract for fighter planes that we could be sparing ourselves? Planes that have no plausible function in the post-cold war era....

More people are going to suffer through underfunded public services than will ever suffer from our lack of supersonic hardware. You can take that to the bank.

[Bob Smith]

Andrew F: " I would largely be satisfied with a large transfer of tax room with concomitant reduction in transfers."

Except that I doubt very much we'll see the feds say "we'll reduce our taxes, so you can reduce yours", i.e., the traditional way of transfering tax room. I think the more likely scenario is either "we already reduced the GST, so if you need more money, you can increase the HST" or, "if you need more money, get it yourself, health care/education/etc. is your responsibility, not ours" (I could especially see them making that last argument to Quebec and daring the government of Quebec, by then likely a PQ government, to assert that, no the federal government plays an important role in providing those provincial services, and that Quebec simply can't do it without more money from the feds).

Now that I think about it, though, the upcoming negotiations with the provinces might explain a lot of what is going on in Ottawa. For example, why aren't the feds blowing their brains out trying to reduce the deficit (they didn't push very hard in negotiations with civil servants, for example)? Answer: They don't want to be going into negotiations with deficit ridden provinces while running a surplus. The federal case is a much easier to make if (like in 1995) the federal government is running a structural deficit (although, since you don't want to end up like the old PCs, not one that's so massive, as to potentially snowball). Of course, that might explain some of the dynamic in Ontario as well. Why hasn't the Mcguinty government fought it out with civil servants over wages? Sure it's an election year, and they don't want to jeopardize potential voters (although, not doing so just gives a card to Hudak to play to his base), but more to the point, the provincial position is a lot more compelling if you're running a long-term deficit roughly equal in magnitude to what the feds are running.

[Bob Smith]

Sorry that should read: "we'll reduce our taxes, so you can INCREASE yours". Obviously, what I wrote was wrong.

[Andrew F]

I hope you're wrong, Bob. I'm not much impressed with my provincial and federal governments playing a game fiscal-ruin chicken.

[Kosta]

Bob Smith wrote: "Answer: They [the Feds] don't want to be going into negotiations with deficit ridden provinces while running a surplus. The federal case is a much easier to make if (like in 1995) the federal government is running a structural deficit."

Nice observation Bob, and great post Stephen!

[Bob Sears]

Not increase the GST?
Why does no one mention that to solve the mess the UK is in they increased the VAT?
A "temporary" increase, until deficit spending is increased with a commitment to no new programs. Introduce improved child care then reduce spending in an existing program to pay for it.

[Bob Sears]

CORRECTED POST
Not increase the GST?
Why does no one mention that to solve the mess the UK is in they increased the VAT?
A "temporary" increase, until deficit spending is eliminated with a commitment to no new programs. Introduce improved child care then reduce spending in an existing program to pay for it.

[Tuzanor]

"Not increase the GST? Why does no one mention that to solve the mess the UK is in they increased the VAT?"

Because the conservatives campaigned hard on that tax cut, it arguably won them the election, and they can't be seen raising taxes or everything they've presented themselves on will fall apart. They got a huge blowback on the income trust about face, how will it look on a tax that actually has an effect on their base (this is ignoring the economic advantages of consumption taxes over other forms)? It'd be an embarrassment. Also, the UK is perusing across the board spending cuts on top of the VAT increase, it's government is a much larger percentage of the economy, and its deficit is an order of magnitude larger in both real and GDP terms. It's not really fair to compare.

[Tim]

I guess for those outside of Canada there is a certain degree of legal and historical history that makes the GST a politically difficult issue in Canada especially in places such as BC. Under Canada's constitution provinces are only allowed to collect "direct" taxes whereas the Federal government can collect taxes by any "means or method." However, in the early 1960s Canada's federal government essentially promised to refrain from imposing "direct" sales taxes at the retail level and would only impose so called wholesale excise taxes such as the old Manufacturers Sales Tax. This is essentially how for example cigarrette taxes still work in Canada. Provinces impose a tobacco tax collected at the retail point of sale while the federal government imposes a "hidden" excise tax on tobacco manufacturers and importers.

Essentially when the GST was introduced in 1991 replacing the hidden Manufacturers Sales Tax the feds were seen by many including Bill Vander Zalm when he was premier of BC as essentially breaking the promises previous federal govts have made regarding "direct" sales taxes and thus the tax was opposed by almost every province other than Quebec. Several provinces including BC lead by Vander Zalm challenged its constiutionality in court. Now twenty year later tensions are not what the used to be on the issue but they still exist especially in places like BC.

[Bob Sears]

The hardest thing that any government can do is admit it was wrong. Decreasing the GST bad idea, good politics.
The Conservatives in a former life aka Reform thought the route to go was to decrease personal income tax in favour of a GST.
Increase the GST but make real cuts to the low and middle income tax levels and while you're at it increase the floor for the top bracket to $250,000.

[Bob Smith]

"while you're at it increase the floor for the top bracket to $250,000"

That would entail a massive selling job, and wouldn't be much of a vote getter.

I mean, let's face it that would translate into a $4,000 tax break for the richest 1-2% of the population (I seem to recall that the top 1% have incomes starting at $169k, so it would be more than 1%). A back of the envelope calculation suggests that pushing up the threshold for the top marginal rate would result in an approximately $1 billion tax cut for the top 1% (approximately 246,000 people), the total cost would probably be closer to $1.5-$2 billion. Good luck selling a $2 billion tax cut to the richest 1-2% of the population (especially if its financed with a GST increase).

As for increasing the GST, but providing tax cuts for the low and middle income levels, I've always wondered about the economics of that idea. In theory, I suppose, it's an improvement, since you're replacing an income tax with a consumption tax. In practice, though, because Canada's tax system provides for tax sheltered savings, I would have thought that Canada's "income" tax is already, to a large degree, a consumption tax for people in lower income brackets since any savings they have are likely to be tax sheltered (either through RRSPs/RPPs or TFSAs - TFSAs in particular are an important development since they are a more flexible savings vehicle than the existing registered plans). In that case, increasing the GST just to cut the "income" tax is a case of replacing a defacto consumption tax with a de jure one and of debatable policy merit (I suppose there may be some redistributive effects, though, which may or may not be desirable).

Obviously, for people in higher tax brackets whose savings are sufficiently high to cap out their RRSP/TFSA limits, the income tax is a tax on income, but for those people cutting the lower tax rates is akin to a lump sum rebate (since it involves a cut to their infra-marginal tax rate), and so of debatable policy merit.

[Determinant]

Increasing the GST is a "courageous decision" in the Yes, Prime Minister meaning of the phrase, a decision that could cost you the election, for the reasons Tuzanor stated.

[dj]

Wow..I'm new here...don't know this blog was about left or right...indexing is the great unknown ,just ask AIG. But running are health care out of a Tim Hortons seems like a good plan.

[Simon van Norden]

Great post, Stephen.
Do you know whether these figures get revised much (if at all) ex post? Or is this all on a cash-flow basis?

[Stephen Gordon]

Some of the numbers do get revised. Each Fiscal Monitor provides numbers from the same month in the previous year, and so I've been using those - and hoping they don't get revised later.

It also turns out that surge in PIT revenues was due to a redefinition, so I'm going to have to revisit that.

[Georgian]

You have in the past claimed that a cut in CIT of say $6 billion is lower due to dynamic effects.

Can one not say the same for GST cuts? ie the additional $12 billion or whatever loss in revenue is less when dynamic effects are taken into consideration?

[Stephen Gordon]

No, that's part of the big difference between the CIT and the GST. The supply of savings is much more elastic than the demand for consumer goods and services, so behavioural responses to the CIT are very important. The behavioural response to the GST cut is much more modest, and zero isn't a bad approxiation for its size.

[Georgian]

So, are you suggesting the $12 billion or whatever in GST cuts ends up in savings (ie bank or investment) rather than consumption?

If so, how would one confirm this is what happened?

[Jacques René Giguère]

As Krugman said, a modern western government is an insurance company with an army and a navy.

Equalization?
Yes provinces should run and finance their own things. But equalization is about something else. It is not subsidizing the lazy or paying ransom to Quebec ( in that case a simple solution : let my people gooo (low bass voice))
We are in a common currency area and some provinces ( let's say Alberta) wreck the other's exchange rate. As long as AB doesn't endure the costs of a soaring exchange rate and leave Central and Eastern Canada stew in the Dutch disease cooker you must have equalization. As i say to my students in the Québec, Canada, régions course. what we need monetrily is not two countries with the same currency but one central bank overseeing five or sic regional currencies. The second best solution is fiscal federalism as the Europeans are now discoveriing. After 1 1/4 century we have still not understood. Europe will a have a tough slog learning.

[anonymous]

Why are the graphs plotting comparable values on incomparable axes? If both red and blue lines are percent GDP, plot them on the same axis with the same scale. The graphs are misleading and confusing the way they're being presented.