Scott Sumner  

Canadian success

Zeke Emmanuel on Optimal Life ... Social Desirability Bias and A...

David Henderson has a very good post on the Canadian economy.  David argues that austerity can be expansionary, and points to the example of Canada in the 1990s. Josh Barro counters that Canada was aided by an expansionary monetary policy:

Canada's 1990s austerity was able to be expansionary because of an export boom that coincided with government spending cuts. The boom owed something to the North American Free Trade Agreement, which fits the conservative narrative. But a weak currency also favored exporters. In January 1995, the Canadian dollar stood at $0.71, down from $0.89 in November 1991, which prompted The Journal to mock the currency as the "northern peso." But even as Canada's broad economic metrics improved through the 1990s, the currency kept falling, hitting $0.65 in August 1998. Conservatives urging the United States to follow Canada's fiscal example usually leave out the "weaken your currency" part of the plan.
That's a good criticism of some conservatives. But Paul Krugman praises Barro's argument, even though it indirectly discredits Krugman's claims. Recall that Krugman argued that expansionary austerity won't work at the zero bound, because monetary policy cannot offset fiscal austerity. And that means, ipso facto, that monetary policy cannot depreciate currencies at the zero bound. But in the time since Krugman made that argument both Switzerland and Japan have done exactly that, using expansionary monetary policy to depreciate their currencies at the zero bound. This shows that Krugman is wrong about monetary policy, and hence wrong about expansionary austerity. The Canadian experience of the 1990s is very relevant to the problems faced by Europe, Britain and Japan at the zero bound.

Barro also points out that both Canada and Australia have benefited from strong commodity markets in recent years. That's true, but it's a two-edged sword. Because commodity markets are far more volatile than other sectors, countries that rely on commodities will tend to have a more unstable business cycle than those that don't. And yet both Australia and Canada have recently had a more stable business cycle than the US. Indeed Australia hasn't had a recession since 1991, even avoiding recession in years when commodity prices were low. That's because they've had better monetary policy.

Paul Krugman takes David Henderson to task for making false claims about government spending in Canada:

When dealing with right-wing claims about economic data, you should never forget Moore's Law: not only shouldn't you accept their assertions, you should assume that what they say is probably wrong. Barro:
Squeezed by high interest rates, a left-of-center government instituted big spending cuts in the 1990s; as a result, Canada's level of public expenditure as a share of its economy has fallen to match America's.

From the IMF database:

And indeed the IMF data shows that government spending in Canada is more than 6 percentage points higher (as a share of GDP) than in the US. This surprised me, as I recall the two being quite similar. And other data sources like Wikipedia show them to be similar. Then I checked an earlier link to the IMF, from their October 2012 data set, just 18 months before the one cited by Krugman. Sure enough, the Government spending/GDP ratios for the US and Canada were only 1.2% apart in 2011. So why has the gap suddenly widened? It hasn't, the IMF data set cited by Krugman shows the gap to be 5.75% back in 2011. Indeed the figures also differ sharply for 2010, and earlier years. It seems the IMF has great trouble determining the actual ratio of government spending to GDP. It would be nice if there were an international organization that could produce reliable figures; does anyone know of one?

Comments and Sharing

COMMENTS (15 to date)
MichaelT writes:

Bob Murphy has a post on this as well, but argues that monetary policy was not as "loose" as you and Barro claim:


Scott Sumner writes:

Michael, Bank of Canada asset growth is not a good indicator of the stance of monetary policy.

I'm not sure how loose the monetary policy was, I was just reacting to Barro's claim.

RPLong writes:

I am unclear on how this relates to Krugman's claim about the zero lower-bound. Canada in the 1990s was not anywhere near the zero lower-bound.

From what I can tell, Krugman's claim is that Canada succeeded with austerity because it had a weaker currency than its best trading partner. That is difficult to dispute; it certainly couldn't have hurt.

Another important dimension here is the enormous rise in oil prices over the 1990s. That finally enabled Canada to tap oil that was previously too expensive to extract and sell. For my money, this was a much, much more important development than the weak Canadian exchange rate.

And remember, during the 90s, Canadian oil extraction companies used an income trust corporate finance arrangement, which had the power to attract a great deal more investment capital than traditional corporations could. It's important to remember that it was the Conservative government that changed corporate tax laws to the point that the 90s-era income trust approach was no longer viable.

J Mann writes:

It's hard to unwind, but I think that Krugman is calling *Josh Barro* an unreliable liar, not David Henderson. Krugman is quoting Barro, and Barro doesn't source that claim to anyone.

Presumably, since cockroaches are involved, he feels no need to be civil to Barro, either.

David R. Henderson writes:

Another important dimension here is the enormous rise in oil prices over the 1990s. That finally enabled Canada to tap oil that was previously too expensive to extract and sell. For my money, this was a much, much more important development than the weak Canadian exchange rate.
I considered this explanation when I wrote my Mercatus piece. I rejected it for a very simply reason: oil prices were low throughout the 1990s. See here.
I probably should have had a paragraph on that because the issue has come up a number of times.

Kenneth Duda writes:

Scott, I love your writing and your ideas.

But once again I'm going to complain about your characterization of Krugman, because it annoys me. Krugman does not assert monetary policy impotence at the ZLB. He asserts *conventional* monetary policy impotence at the ZLB, where by "conventional" he means interest rate targeting, which is in fact (obviously/tautologically) impotent at the ZLB. He seems to be quite careful about this, leaving the door open for "unconventional" monetary policy (QE, NGDPLT?) to be effective without pointing out that it is very likely that NGDPLT would be really effective and a really big improvement over the fiscal stimulus alternative.

Now, it really annoys me that Krugman expresses things that way. He should be arguing for NGDPLT, rather than coyly dodging it. His posts pointing out the lack of a political base for NGDPLT are annoying and irrelevant, because as you've pointed out, it's professional economic consensus that drives fed policy. And he annoyingly pronounced a test of MM, only to never follow up when MM passed the test.

But I think you should be more accurate in your criticism. Nothing he's said about monetary policy is wrong. He's just leaving out some really important things, like (as you kindly pointed out to me in comments once) the Fed has the wrong target.


Kenneth Duda
Menlo Park, CA

[Visible email address removed. We do not publish email addresses in public, not even by request. --Econlib Ed.]

RPLong writes:

Prof. Henderson - Thanks, I stand corrected! That's all the more surprising to me because what I wrote in my previous comment seemed to be part of the conventional wisdom when I lived in Alberta. Very glad to have learned something today!

Andrew_FL writes:

The whole idea of "monetary offset" strikes me as tantamount to an admission that reducing government spending is in and of itself a bad thing, which needs to be mitigated.

This makes it very tempting to reject the notion on it's face.

honeyoak writes:

Scott, the oecd keeps track of these statics here

if the link does not work try:
OECD iLibrary / Home / Statistics / General government / General government spending

Michael Byrnes writes:

Andrew wrote:

"The whole idea of "monetary offset" strikes me as tantamount to an admission that reducing government spending is in and of itself a bad thing, which needs to be mitigated."

I think it is more an argument that government spending to stimulate the economy is a bad thing. There are obviously other reasons why the government would spend money.

And it is difficult enough for government to spend money wisely even if they aren't allowed to play the "stimulus" card, which could be used to justify just about anything.

Andrew_FL writes:

@Michael Byrnes-Huh? Monetary Offset seems to be the doctrine that what would otherwise be negative effects of cutting government spending can be mitigated by loosening Monetary policy at the same time. In other words, sure austerity is bad, but it doesn't have to be!

It doesn't seem, to me, to say anything about stimulus spending whatsoever. But it does seem, to me, very far from a correct view of the impact of actually reducing government spending.

Scott Sumner writes:

RPLong, You said;

"Another important dimension here is the enormous rise in oil prices over the 1990s."

I don't think so.

Regarding exchange rates, yes Krugman focused on the weaker currency. But as I said, countries at the zero bound can weaken their currencies. That's his mistake--he assumes they can't.

J Mann, I think you misunderstood him.

Kenneth, Yes, I know exactly what Krugman has been saying---no one has followed him any more closely than I have. He says many things, but his critique of expansionary austerity only makes sense if all forms of monetary stimulus are ineffective at the zero bound. Which is false. I hope you read the post of his I linked to, it's a real eye-opener. He's flat out wrong about Japan and Switzerland.

The distinction between conventional and unconventional monetary policy that he tries to make is actually pretty meaningless, which is why even he ignores the distinction when he criticizes the expansionary austerity claim. He usually says expansionary austerity doesn't work, not that it doesn't work unless the central bank uses unconventional tools.

An open market purchase of bonds is 100% "conventional" no matter how he tries to twist the English language. Look at almost any money and banking textbook and they will describe normal expansionary monetary policy as open market purchases of bonds.

Andrew, That's not at all a correct interpretation of what I am saying. If there is a stable monetary policy (either inflation or NGDP targeting) then the fiscal multiplier is zero. The central bank doesn't have to "do anything." Just keep a steady policy.

Honeyoak, Thanks, so how do their US/Canada estimates compare to the IMF estimates?

Mr. Econotarian writes:

Can someone show me where greatly expanded government spending has ever actually been linked with increases in private economic growth?

J Mann writes:


Thanks for replying. It's a minor point, so feel free to ignore but . . .

I can't see any way that the "Moore's law" bit of Krugman's article is a critique of Henderson. Krugman is literally quoting an assertion made by Barro in Barro's article. I think it's fair to assume that Krugman believes that Barro was inaccurate because he was fooled by some right winger, but I'm not sure why that would be Henderson. Barro's article mentions several writers, but doesn't mention David at all.

I looked through David's 2011 article, and I can't find any place where David argues that "Canada's level of public expenditure as a share of its economy has fallen to match America's" Am I missing some source that indicates that Barro got that statement by reading Henderson?




Pithlord writes:

In addition to whatever concrete changes in the inflation target, there was a huge change in the *ethos* of the Bank of Canada in the mid-1990s. The Governor between 1987 and 1994 was John Crow, who was by all accounts an extreme inflation hawk, and wanted a zero inflation target. Once he left, the Bank adopted a more reasonable target. What was particularly controversial in Canada at the time was that the Bank also decided to ignore the exchange rate if it did not affect core inflation. That was the basis for the massive drop (which corresponded to low commodity prices).

In Canada, it is important not to just look at the federal level for fiscal policy, since Canadian provinces do not have the kind of fiscal restrictions American states have. The biggest political change in the decade was not at the federal level, but the transfer of power from the NDP to the market-oriented Harris Conservatives in Ontario. The NDP had tried to offset the Bank's restrictive Crow-era monetary policy with Keynesian fiscal stimulus, but failed. In contrast, in the second half of the decade, federal and provincial fiscal retrenchment corresponded with looser monetary policy, and the economy recovered.

I think this experience is why a lot of Canadians are receptive to Scott's ideas.

Comments for this entry have been closed
Return to top