David R. Henderson  

Response to Krugman on My Canada Study

PRINT
There is nothing tautological ... Uber Wars Update...

UPDATE BELOW

I think this qualifies as a cockroach idea (zombies just keep shambling along, whereas sometimes you think you've gotten rid of cockroaches, but they keep coming back.) I thought we had disposed of all this four years ago. But nooooo.

This is from Paul Krugman's post today, "Conservative Canadian Cockroach." The link he provides is not to a post by him, but to a short critique of my study of the Canadian government's budget cuts from the mid-1990s to the mid-2000s. The critique is by Stephen Gordon, an economics professor at Universite Laval in Quebec.

The "this" that Krugman thinks qualifies as a cockroach idea is the idea that "austerity is expansionary." Krugman doesn't define austerity, but a random look at his past posts on austerity suggests that he doesn't carefully distinguish between austerity brought about through cuts in government spending and austerity brought about through increases in taxes. I think there's a huge difference. The Canadian case that I dealt with here is one in which the government cut spending by about $6 for every $1 of tax increases.

For those of you who don't read Krugman regularly, cockroach ideas are bad ideas that "you think you've gotten rid of" "but they keep coming back."

First to set the record straight, although I think austerity with government budget cuts can be expansionary, my goal with my study of Canada's budget was more modest: to show that one could cut government budgets substantially even during a period of slow growth of GDP without slowing the economy. Here's how I put it in the study:

The second big lesson is that the Keynesian argument that big cuts in government spending will slow an economy receives no support from Canada's experience.

In his critique, Professor Gordon makes two points. I'll respond to both.

Gordon writes:

Henderson suggests that "the Canadian experience does not support the Keynesian view that policymakers should not cut government spending during an economic slowdown." In point of fact - and it remains an open question in my mind whether this was by luck or by design - Paul Martin's famous austerity budget of 1995 was brought down at a point when private-sector employment had reached its pre-recession levels.

Notice that Professor Gordon doesn't take on my point that there was an economic slowdown. Rather, he shows, correctly,that "private-sector employment had reached its pre-recession levels." Well, as any good macroeconomist knows, private-sector employment can often reach its pre-recession level during a period of slow growth. As Gary Burtless of the Brookings Institution pointed out, in March of this year private-sector employment finally exceeded its pre-recession level. Does anyone care to disagree that this happened during a time of slow growth?

Indeed, the time that Professor Gordon refers to was a period of slow growth. Here are the World Bank data on Canada's growth rate of real GDP during the period at issue:
1994: 4.6
1995: 2.7
1996: 1.7
1997: 4.3
So 1994 was a year of high growth, but 1995, when the big budget cuts began, and 1996, were years of low growth.

Gordon writes:

Monetary policy could and did offset the fiscal contraction.

Henderson hardly mentions monetary policy, except to note that inflation was low and stable, and that by February 1995, the CAD had depreciated to USD 0.71. But there's much more to the story than that:
[He follows with a graph of interest rates and exchange rates.]
In the two years following the 1995 budget, the Bank of Canada reduced interest rates by more than 500 basis points. And by 1999, the Canadian dollar had depreciated by 10% in USD terms, fueling the export-led expansion.


He's right that I hardly mentioned monetary policy. I do believe, as he seems to also, that monetary policy can offset fiscal policy. Great! I have no problem with that. Co-blogger Scott Sumner has written extensively about how monetary policy can offset fiscal policy.

I will point out, though, that Gordon hardly mentions monetary policy either. At best, Gordon looks at the effects of monetary policy. That's at best. But his claim that "the Bank of Canada reduced interest rates by more than 500 basis points" is probably false. Central banks have only small effects on interest rates.

Final note: I do thank Professor Gordon for his tone (hey--he's a fellow Canadian--what do you expect?) and for his statement:

Much of the paper sets out the political context that made this possible, and is very useful in that respect.

UPDATE: David Beckworth argues that, indeed, Canada's central bank did offset fiscal policy with monetary policy.


Comments and Sharing






COMMENTS (14 to date)
Rick Hull writes:

I keep reading that CBs have little control over interest rates, but how does that square with the Fed's ability to drive down interest rates to near zero after the dotcom bust and in response to the subsequent housing bust and resulting financial crisis?

Was the Fed really just tagging along with market rates in an illusion of control? I find that very hard to believe.

I like the Hummel piece you linked but I haven't digested it. His conclusion is that the Fed has little control over real interest rates, which, even in the simplest scenarios, are subject to inflation expectations as a side effect of interest rate policy. This is true even if the Fed has full control over nominal (short term) rates.

Again, given the Fed's interest rate policies over the last 15 years, and the nominal interest rates in the market over the same period, how much of the conventional wisdom -- that the Fed controls short term nominal rates -- is actually wrong?

David R. Henderson writes:

@Rick Hull,
I keep reading that CBs have little control over interest rates, but how does that square with the Fed's ability to drive down interest rates to near zero after the dotcom bust and in response to the subsequent housing bust and resulting financial crisis?
Rick, you’ve begged the question. Of course, it doesn’t "square with the Fed's ability to drive down interest rates to near zero after the dotcom bust and in response to the subsequent housing bust and resulting financial crisis.” But what you haven’t established is that the Fed had that ability. All we observe is that interest rates fell to near zero. That doesn’t mean the Fed was responsible.
Was the Fed really just tagging along with market rates in an illusion of control?
Basically, yes. Imagine that 3-month Treasuries pay 3% per annum. Do you think the Fed could target, say, a 0.5% Fed funds rate and make it stick? I don’t.
His conclusion is that the Fed has little control over real interest rates, which, even in the simplest scenarios, are subject to inflation expectations as a side effect of interest rate policy.
So, back to the Canada example, do you think Canada’s central bank caused expected inflation to fall by 5 percentage points? It’s possible but I doubt it.

Rick Hull writes:

If I have begged the question, it's because I find it unfathomable that the market's response to crisis is lower interest rates, all the way down to zero. Low nominal rates imply low time preference and low default risk. The market's response to crisis should surely be higher time preference reflect higher default risk. The CB wants to counteract this tendency, and it seems they do have the ability.

This desire to counteract the market tendency is noble on its face, attempting to avoid a catastrophic spiral, where e.g. IBM can't make its payroll. This is analagous to schemes like FDIC to counteract the catastrophic spiral of bank runs.

Am I totally off base in my understanding of market responses to crisis?

David R. Henderson writes:

@Rick Hull,
If I have begged the question, it's because I find it unfathomable that the market's response to crisis is lower interest rates, all the way down to zero. Low nominal rates imply low time preference and low default risk. The market's response to crisis should surely be higher time preference reflect higher default risk.
I agree that it IS hard to fathom. But it doesn’t make me give up my understanding of monetary policy.
This desire to counteract the market tendency is noble on its face,
But then, as you probably already know, the Fed would have just pumped liquidity in. But as Jeff Hummel and Michael Bordo have pointed out, the Fed sterilized much of its increase in the money supply.
Am I totally off base in my understanding of market responses to crisis?
No. It IS a puzzle.

David Boaz writes:

Haven't both Alberto Alesina and Christina Romer done work on the results of tax cuts vs. spending increases? Which seems relevant, even if the specific question here is spending cuts and tax increases.

David R. Henderson writes:

@Rick Hull,
I was watching my 49ers going down in flames when I wrote my last response. So I wasn’t all there. Actually, Alan Greenspan, Ben Bernanke, Jeff Hummel, Martin Wolf, and I all pointed out back in 2009 that the low interest rates were lower real rates and that they were due to, choose your term, a savings glut or an investment dearth.

Rick Hull writes:
pointed out back in 2009 that the low interest rates were lower real rates and that they were due to, choose your term, a savings glut or an investment dearth.

Is it possible that this explanation begs the question? The fact of SG/ID -- is this measured or determined independently, or are we reasoning from interest rates?

The savings glut doesn't make much sense to me, at least in the US, where (seemingly) half the country had just taken a massive hit to net worth via the housing market crash. Also, hasn't McArdle pointed out how little savings American households actually have, living paycheck-to-paycheck now (and recently) moreso than ever?

Income goes to consumption or savings, right? And savings may go to investment. Where most income goes to consumption, as I baldly assert in the case of US households, savings gluts are precluded, unless I'm misunderstanding the term.

Investment dearth could mean one of two things: a seeming lack of (or reduction in) investment opportunities (lower supply), or investors choosing to hold cash to avoid risk (lower demand). Considering the former case, I don't believe that investment opportunities themselves were lacking. If it seems this way, it's rather that investors judged a larger portion of opportunities as too risky for their return. Considering the latter, the choice of cash over investment is clearly indicative of higher time preference and increased default risk.

Regardless, if we're talking about investors choosing cash over investments, I don't see how this pushes interest rates lower.

Of course, it's not just households that save, so maybe there really is a savings glut. However, since we only measure inflation at the household level, and for limited goods at that, perhaps unmeasured price inflation is the missing link.

How do the government statisticians construct the CPI? Month-by-month, the BLS tracks the purchases of 6,100 households across the country, which are taken to be “representative” of the approximately 320 million people living in the United States. The statisticians then construct a representative “basket” of goods reflecting the relative amounts of various consumer items these 6,100 households regularly purchase based on a survey of their buying patterns. They record changes in the prices of these goods in 24,000 retail outlets out of the estimated 3.6 million retail establishments across the whole country.

Emphasis mine, from Richard Ebeling.

Note, I'm not a fan of that article, but the fact I quoted struck me. I'm assuming the fact is true.

Perhaps there is a great deal unmeasured price inflation for consumption goods only of interest to firms. Likewise financial assets. Do we really want to ignore all of that when analyzing "real" returns? Could that be the confounding factor?

David R. Henderson writes:

@Rick Hull,
You don’t look at one country’s savings to look at what affects interest rates. It’s a global market. A huge amount of savings from Asia and the petroleum exporting countries is the key factor.

Rick Hull writes:

At the risk of going too far off track, I'd like to refine my concluding question. I believe the Fed now targets CPE moreso than CPI, and I'm not sure which is used to compute "real" returns. But CPE is consistently lower than CPI, perhaps due to chaining, which you and I have tangled over previously, much to my benefit -- though I'm still not sure how to pronounce Laspeyres.

The monetary authority always wants to understate inflation, so given a choice of two index calculations where one is always lower than the other, it's not hard to predict which will be preferred.

I'm considering whether there is a great deal more money sloshing around the system, bidding up various goods and assets, than we might expect from the standard inflation measurements. If so, might that explain lower market interest rates? Admittedly, this is pure speculation at the level of underpants gnomes theory.

Finally -- given that our inflation measure ignores final good consumption by firms, financial assets, commodities, land, land improvements, etc. -- doesn't that make the calculation of real rates via simple subtraction totally invalid? After all, isn't the market that determines nominal rates much larger than household investment? Aren't we trying to subtract apples from oranges? Doesn't all of this (needless) complication dramatically confound economic calculation?

Apologies for veering off into inflation, but I greatly value this discussion and your feedback. I'd love to continue via a different post or channel as desired.

Rick Hull writes:

Yes, of course, I totally overlooked global savings, which if increasing and increasingly allowed to cross national borders, explains lower domestic rates as market rates less subject to domestic policy. Though to the extent global savings result from helicopter money from various CBs, our "market" rates are influenced by entities not subject to market discipline.

Forex arbitrage, to the extent it is allowed, should dampen this effect. Likewise international government bond rate arbitrage, confounded by forex, should tend to reinforce CB activity as a whole. That is, Chinese monetary policy may "help" the Fed target interest rates in the US.

I'll bow out here, hoping I did not distract too much from the specifics of your post. Thanks again for the discussion.

Noah Carl writes:

Incidentally, the Alesina paper mentioned by David Boaz is called 'The Design of Fiscal Adjustments'.

ThomasH writes:

If monetary authorities have allowed a recession to develop (failed to maintain NGDP growth) and are not promptly fixing it (not returning NGDP to its pre-recession growth path) ,the world of 2008-10 in the US and 2008 and counting in the Euro Zone, then neither kind of austerity is expansionary. At full employment either kind can be increase/decrease real income depending on which taxes (on consumption or investment?) and which expenditures (in consumption or investment) are increased/decreased.

The zombie/cockroach idea is that either kind of austerity during a recession is expansionary.
***

"Do you think the Fed could target, say, a 0.5% Fed funds rate and make it stick? I don’t."
I do. The Fed just announces its policy and starts buying stuff until the Fed Fund rate falls to 0.5%. It's hard to imagine when that would be a good policy but the Fed could do it.

Bob Murphy writes:

David,

I'm a huge fan of your Canada fiscal work (as you know), so consider this question a minor quibble: When you said in the comments above that you didn't think the Fed could target a low Treasury rate and "make it stick," isn't that saying, "The Fed would have to continually expand its balance sheet in order to keep the interest rate lower than the 'correct' level." ? And isn't that exactly what the Fed has been doing since late 2008?

On a related note, the conventional story is that Volcker came into power and "jacked up interest rates to get inflation under control." Do you (and Hummel?) think that story is totally wrong?

David R. Henderson writes:

@Bob Murphy,
I'm a huge fan of your Canada fiscal work (as you know),
Thanks, Bob. I’m a huge fan of your work too, especially your recent analytic pieces on global warming.
so consider this question a minor quibble:
Glad to see “so” rather than “but.” :-)
When you said in the comments above that you didn't think the Fed could target a low Treasury rate and "make it stick," isn't that saying, "The Fed would have to continually expand its balance sheet in order to keep the interest rate lower than the 'correct' level." ? And isn't that exactly what the Fed has been doing since late 2008?
I don’t think so. Even continually expanding its balance sheet couldn’t have much effect on interest rates. For more, see Jeffrey Rogers Hummel, “The Myth of Federal Reserve Control Over Interest Rates."
On a related note, the conventional story is that Volcker came into power and "jacked up interest rates to get inflation under control." Do you (and Hummel?) think that story is totally wrong?
I can’t speak for Jeff but I think the story is wrong. What I think he did was start reducing the growth of the money supply. But I would be interested in what Jeff, who knows more about this than I do, thinks.

Comments for this entry have been closed
Return to top