David R. Henderson  

Take That, Keynes and Lerner

PRINT
Bumps on the Treadmill... What's the Worst War Since WWI...

Paul Martin, the Minister of Finance for Canada from 1993 to 2002 and Prime Minister from 2003 to 2006, was the person most responsible for bringing down government spending and government debt as a percent of GDP. Between FY 1993-94 and FY 2005-2006, federal spending on government programs fell from 16.8% of GDP to 12.8% of GDP and federal spending on interest on the federal debt fell from 5.5% of GDP to 2.5% of GDP, for an overall drop in federal spending from 22.3% of GDP to 15.3% of GDP. In his memoirs, Hell or High Water: My Life in and out of Politics, he writes:

It is important to understand that the no-deficit rule was a sharp break with tradition. In the postwar years, many economists argued that you did not need to be in the black every year, as long as budgets were balanced over the course of the economic cycle, so that deficits during slumps would be paid off with surpluses in good years. Whatever the economic rationale for that approach, it didn't work in the real world of politicians. Once you break the spell--once governments find that they can get away with borrowing instead of taxing to pay the bills--it is almost impossibly tempting for politicians to do it again and again until the debt is out of control.

Jim Buchanan and Dick Wagner couldn't have said it better. Take that, John Maynard Keynes and Abba P. Lerner.

The reason for the title of Martin's book is that when he pledged to hit his deficit targets in 1995, he said that he would do it "come hell or high water." He kept his pledge.


Comments and Sharing


CATEGORIES: Fiscal Policy



COMMENTS (11 to date)
Daniel Kuehn writes:

Could you provide some context in terms of output, the state of demand, etc. over this period? I'm having trouble understanding what we're supposed to make of this.

Unless the point is just "we didn't do what they said we should" - in which case the output and demand information would still be nice because "what they said we should" is of course contingent on that.

octavio writes:

It is also worth pointing out that since Martin was voted off the island and the reform party/canadian alliance party/conservative party took over the parliament, the budget has been "balanced". Canada is no longer on track to completely eliminating it's debt.

David R. Henderson writes:

@Daniel Kuehn,
Good question. I could provide such context but doing so is very time consuming. Let me give you one important time series to give context. He started the budget cuts with the 1994-95 budget. The government's fiscal year started April 1, 1994. I've had a heck of a time finding monthly unemployment rates--does anyone have a URL to recommend?--but here are average unemployment rates by (calendar, I think) year from 1994 to 1997, by which time most of the budget cuts had occurred:
1994: 10.4%
1995: 9.5%
1996: 9.6%
1997: 9.1%

@octavio,
Even worse, the budget is no longer balanced. Harper appears to be a Keynesian and in Martin's memoirs, he criticizes Harper for giving himself no fiscal room to maneuver.


Barnes writes:

I'd like to make a point that Keynesians didn't really have such a bad idea. I don't think that running a deficit to get you through a rough patch is such a bad thing. In fact a country could probably do quite well if they did.

But we seem to have gotten it backwards though. We take our deficits and try to pay them back in the good times to come. Of course this approach is doomed to failure.

Really the way to do it, is to run a surplus and save it. Hold onto it for that inevitable rough patch. You can spend money in a slush fund all day and it will never bankrupt you. You are still just paying for the deficit in a slump, with a surplus from a good year, the difference is that you prepay instead of running up a debt.

And if you saved for that "rainy day" and used that money to finance your deficit, politicians would never be tempted to let spending get out of hand. Of course this is all a hypothetical, I could be wrong, politicians might still spend way too much this way.At any rate, I just think that a country should have some sort of fiscal redundancy.

Chris Koresko writes:

The whole Keynesian approach of using government spending to smooth out fluctuations in aggregate demand seems intuitively plausible at first, but the more I think about it the more confusing it becomes.

When an individual saves money, he is foregoing consumption today in exchange for a promise that he will be able to consume more tomorrow. This is possible because not everyone is saving at once; there's always somebody willing to consume extra today and promise to consume less in the future in return.

But how can a government save? If it saves during a boom by pushing money into a bank (or the whole banking system) and withdraws that money during a recession to finance higher spending, then won't the saving accelerate the boom part of the cycle by making too much credit available, and exacerbate the recession by making credit more expensive? In other words, isn't this pro- rather than counter-cyclical?

Then there's that multiplier thing. I have not been able to figure out how to do the accounting, and I worry that the economists who have calculated it might have gotten it wrong.

But if it's really greater than 1.0, so that $1 of deficit spending creates $1.5 or whatever of extra economic activity, it seems to me that that doesn't imply that we're better off for it.

In equilibrium it ought to be true that the amount of economic activity is such that the marginal wealth produced by a dollar of increased activity is zero. That should be more or less true during a recession as well, but with the equilibrium shifted to a lower level of activity. That would imply that stimulating more activity would destroy wealth rather than create it.

Finally, when the Government spends that $1 it probably does that less efficiently than a tax-payer would have, since it can't know the needs of the recipient as well as a person knows himself, and because it's subject to political economic issues and administrative costs. So when that $1 is spent, some fraction of it is effectively wasted. I don't have any reliable way to estimate what that fraction is, but my gut tells me it's at least 0.5.

I've heard that governments in the US spend around $75K/year per poor person on anti-poverty programs. If the poverty level is $22.5K/year for a family of 4, or $5.6K/year/person, then the efficiency for that spending should be around 5.6K/75K, or 0.075. So my estimate of 0.5 may be very optimistic.

Does any of this make sense?

Bob Murphy writes:

Incidentally, the Keynesians don't even bother talking about running surpluses in good times anymore. I'm not going to bother looking up a link, but lately when Krugman talks about the deficit problem, he says that all we need to do is get the deficit as a share of GDP back to the growth rate of the economy. That way, the total debt-to-GDP ratio will be stabilized.

So with this as the goal, the absolute level of the federal budget deficit grows exponentially. He doesn't even pretend that the government will ever run surpluses again; that's not even the goal.

Bob Murphy writes:

Chris wrote:

When an individual saves money, he is foregoing consumption today in exchange for a promise that he will be able to consume more tomorrow. This is possible because not everyone is saving at once; there's always somebody willing to consume extra today and promise to consume less in the future in return.

Actually Chris I think you're missing the role of savings and investment in a modern economy. What you are saying would only be true if the total amount of output every period were independent of previous savings.

But in reality, if everybody (in his role as an individual) saves a big chunk of his income, then everybody's income can grow over time. That's not a contradiction.

What happens is that by refraining from consumption in the near term, people free up real resources to get switched out of consumption sectors (like making TVs and restaurant meals) and into capital goods sectors (like making 18-wheelers and oil rigs). So that makes workers more productive in the future, because they have more tools and equipment augmenting their labor.

Chris Koresko writes:

@Bob Murphy: Yes, thanks for pointing that out. I was sloppy in not making the distinction between consumption and investment.

Snorri Godhi writes:

Barnes' view is pretty much the same as my own, but I would put it as a stringent condition:
it is not enough for a government to run a surplus during the good times; a deficit during a recession will help the economy only if the public debt is zero or, preferably, less than zero. That means that no country today should run a deficit during a recession.

Boonton writes:

I think my post got lost in the moderator's filter (no problem, it's a holiday weekend anyway)....but the gist of it was that I'm unimpressed by the PM's 'accomplishment'. Between 1993 and 2006 Canada had pretty good GDP growth.

http://www.bcstats.gov.bc.ca/data/bus_stat/bcea/tab1.asp shows that from 1993 to 2006 Canada's GDP growth has been pretty good, many years have been above 5% growth, 2000 was over 9%.

Not only is reducing spending as a % of GDP not that hard during a boom, it's not even anti-Keynesian! In fact its quite pro-Keynesian. The PM is patting himself on the back for having no major snow storms in July!

Boonton writes:

Chris

But how can a government save? If it saves during a boom by pushing money into a bank (or the whole banking system) and withdraws that money during a recession to finance higher spending, then won't the saving accelerate the boom part of the cycle by making too much credit available, and exacerbate the recession by making credit more expensive? In other words, isn't this pro- rather than counter-cyclical?

A few points of confusion:

1. In terms of classical Keynesian policies, gov't spending is not stimulus in itself. Its gov't purchases. This means Social Security, unemployment, Medicare and all other entitlements are part of personal consumption as they are transfer payments. Although this could be used in a stimulative manner (such as expanding benefits in a recession, contracting them in a boom). This area generates a lot of confusion, though, because entitlements are not really a Keynesian policy but one of income redistribution. These policies were not put in place to fight recessions but to redistribute income at all times, even if we had full employment from here to infinity and beyond!

2. In simple Keynesian models the magic question is does the aggregate spending in the economy sufficient to sustain full employment? The spending here is I (investment), G (gov't), C (consumption) Exports minus Imports. Gov't spending less reduces G. You're question is does this increase I because gov't pays off its bonds (or lowers the rate it issues new bonds). I'd say no because only a fraction of activity in the financial markets involve real investment (I). Most gov't 'savings' flows into the financial markets where paper is traded back and forth. Only a fraction of that paper is being sold to directly finance I.

In equilibrium it ought to be true that the amount of economic activity is such that the marginal wealth produced by a dollar of increased activity is zero. That should be more or less true during a recession as well, but with the equilibrium shifted to a lower level of activity. That would imply that stimulating more activity would destroy wealth rather than create it.

This is only true in a full employment equilibrium but Keynes's insight is that there's no reason to assume the economy's natural equilibrium is at full employment. This is where there's a real split among Keynesians, some who argue that deviations from full employment are temporary and others who argue that modern economies always generate a below full employment equilibrium requiring some among of gov't support.

I personally think both are true, there is a permanent equilibrium below full employment requiring some measure of gov't support. This is why most gov'ts never pay off their debt and always increase it (even if it may fall relative to GDP). There are also temporary shocks that require spouts of either stimulus or anti-stimulus (in the case of overheating).

Finally, when the Government spends that $1 it probably does that less efficiently than a tax-payer would have, since it can't know the needs of the recipient as well as a person knows himself, and because it's subject to political economic issues and administrative costs. So when that $1 is spent, some fraction of it is effectively wasted. I don't have any reliable way to estimate what that fraction is, but my gut tells me it's at least 0.5.

Again only at full employment. Here's the question, right now 10% of the labor force is doing nothing productive. Why is it impossible to find nothing that they can do that will generate some benefit?

The only argument that this is not so is some variation on Kling's 'recalculation' which argues that the market is putting 10% of the labor force 'on hold' as it figures out where to efficiently deploy them. This argument gets very strained as unemployment over prolonged periods of time is very destructive and very expensive.

Look at it from a personal level. Maybe it would make sense for you to quite your job and take a month off as you 'think' of something more productive to do with your life. If you take a month off, you loose 1/12th of your yearly income. If you come up with something that makes you a bit more, you'll quickly earn back that loos. Taking a year off, though is a full years income. Take two, three, four years off and its even more (esp. since your working life is only something like 40 years). If 'recalculationists' are right then all this wasted, unemployed resources needs to produce a huge boom pretty soon or else its not efficient at all. Pretty soon Arnold Kling will need the 'singularity' to arrive in order to pay off the losses incurred during his 'recalculation' period!

Comments for this entry have been closed
Return to top