But what would happen to the debt if we didn't run a deficit now?

If the government runs a deficit now, the debt-burden on future taxpayers will increase. But maybe the debt-burden would increase even more if we didn't run a deficit now?

The proper way to handle policy questions (in economics, or anywhere) is to compare what will happen with the proposed policy to what would happen under some alternative policy -- the counterfactual. Sometimes the counterfactual is simple: things will stay the same as they are now. But if we are talking about running a deficit to get the economy out of a liquidity trap, the counterfactual is not simple. About the only thing we can be certain of is that things will not stay the same as they are now.

In normal times, comparing a proposed deficit to the counterfactual is relatively simple. If the government runs a deficit, the central bank would tighten monetary policy to keep inflation on target. So inflation stays the same, real income stays about the same, the interest rate is higher, the exchange rate appreciates, the debt grows faster, etc.

But what happens in a liquidity trap, when the central bank wants to cut interest rates further but cannot?

I am reasonably confident that a large enough temporary deficit could stimulate aggregate demand sufficiently to get the economy out of a liquidity trap, and back to normal. I am much less confident in my (or anyone's) ability to predict what would happen in the counterfactual.

The most optimistic counterfactual is that the economy would immediately recover by itself. In this case, of course, the deficit merely causes the debt to be higher than it otherwise would be. The most pessimistic counterfactual is a deflationary spiral where the economy disappears down the drain. Any realistic counterfactual lies somewhere between those two extremes.

Here is my proposed counterfactual: after a deflationary recession lasting (say) 10 years, a deus ex machina rescues the economy from the liquidity trap and brings it back to normal. I know this is a totally unjustified assumption; that's my main point! Have you got a more reasonable one? The only better counterfactual I can think of is this: after enduring a worsening deflationary recession for a couple of years, an even bigger deficit is used to rescue the economy. In this case the effect on the debt is obvious: the sooner we run a deficit, the smaller the debt will be.

So, the proposed policy is a temporary deficit worth a cumulative (say) 10% of GDP which brings the economy back to normal immediately. The proposed counterfactual is a balanced budget and 10 year deflationary recession followed by a miraculous recovery in the 11th year. Let's compare the debt-burden in the 11th year under the two scenarios.

Inflation. Inflation will be different under the two scenarios, and it's the real debt, not the nominal debt, which creates a burden on taxpayers. Suppose there were 2%  inflation in normal times, and 5% deflation in a deflationary recession. A 10 year period of 7% inflation differential would mean that the price level would be doubled under the proposed policy compared to the counterfactual. If we assume that the debt is not indexed to inflation, the real debt would double under the counterfactual, while it would increase by only 10% under the proposed policy. So a deficit would actually reduce the real debt burden by about 90%.

Real Growth. Real income growth will be higher under the first than under the second scenario, and it's the debt/GDP ratio which matters to taxpayers, and which determines a country's solvency. If the growth rate were a mere 1% higher under the proposed policy than under the counterfactual, this would leave the debt/GDP ratio about the same, so average future tax rates would also be about the same. The growth rate differential would likely be higher than 1%, so the deficit would reduce future tax rates.

Money-finance. Part of a deficit will be financed by printing money. Assume nominal income grows 7% faster with the proposed policy than under the counterfactual (though a combination of higher inflation and higher real growth). Assume that the demand for money is proportional to nominal income. Then the nominal money stock in the 11th year will be double what it would be under the counterfactual. If the demand for money is 10% of nominal income, this doubling of the money stock would be enough to pay for the 10% deficit, so the nominal debt in public hands would be the same under the two scenarios. And it's the debt in public hands which creates the tax burden, since interest paid on debt held by the central bank is returned to the treasury. So a deficit will not increase the tax burden.

Interest rates. This effect goes the other way. Nominal interest rates will be positive if the deficit rescues the economy, and zero in the counterfactual. If some of the debt comes due during the 10 years, it will be rolled over at higher rates under the proposed policy than under the counterfactual, so the debt will accumulate faster than under the counterfactual (for given taxes and government spending). If all the debt gets rolled over every year, and if interest rates are 7% under the proposed policy, then the original debt would double, and so a 10% deficit would cause a 120% increase in the nominal debt in the 11th year.

The counterfactual matters a lot. My guess is that running a temporary deficit to escape a liquidity trap would cause a decreased debt-burden on future taxpayers.

Canada is not in a liquidity trap quite yet, but other countries are. And we had better get our arguments straight in advance.

Comments

[O.G.]

By what means should governments raise the money they claim that they require to run their foolish and corrupt redistributive spending programs ... that's a mug's question.

The real question is why is government so big and why does it spend so much money in the first place? Up until the point of complete collapse when they are strung up by the heels, is there any motivation for a politician or bureaucrat to manage other people's money wisely? That is, to manage other people's money according to the other people's best interests, and not to manage it in their own interests? I say "no". What's the point of anyone dedicating their life to controlling a monopoly if not to use it for their own interests?

"The most pessimistic counterfactual is a deflationary spiral where the economy disappears down the drain."

I don't think the real economy will spiral down the drain in such a case. Only the fake economy. Government spending would have to be severely curtailed, which would of course be a catastrophe (of sorts) for the people who live off government money. They will have to say "bye bye" to free government money and say "hello" to the real world where private property, productivity and profits actually matter. Other than this painful readjustment, deflation would greatly benefit everyone by giving them greater spending power and more bang for their investment buck, instead of being constantly screwed by inflation and taxes eroding away the fruits of their labours.

What you should really fear is government attempts to "ignite" the economy. That's like trying to "ignite" your house when all you need is a stove for cooking. Put in plain language, the only thing that governments know how to do is give money away to their friends and cronies and buy elections with money that they have stolen, borrowed (against future theft) or counterfeited. Period. That's your government in a nutshell. They are not a mechanism for improving society or for preserving its wealth. Therefore I don't think that it's productive to consider the ways in which government should put its hands on money, but only to consider the most efficacious ways of keeping as much money as possible out of their hands.

[Phillip Huggan]

O.G., Canada has the least corrupt government on Earth. Where this is an issue here is the optics of minor intransigence during an election, and how to export what we've done right elsewhere.

Where your rhetoric falls apart severely is in considering the children of poorer parents. With government spending they create wealth when they graduate university, without government spending they drain coffers in prison. Under your Republican/Libertopian model you get 50% of the researchers and entrepreneurs you get under a Socialist model, for much less than double spending.

Given Canada's environmental record is 56th of 57 accoring to GermanWatch (yay, beat Saudis), and we are last among 20 OECD countries in providing childcare (Chretein's legacy that Harper cancelled), taxing wealthy corporations and individuals or a carbon tax (still a swear word even though a 100x as much drop in oil hasn't been as stimulating as a tax was to hurt us according to Harper) are superior to deficit. This country is now about making rich people richer. Money has made AB evil. Ontario handled cash no problem.

[brendon]

Very interesting post - my train reading this morning was bond finance vs. tax financed government expenditures (from Michael Wickens new textbook). Not something i've delved into much so I might post something later this week if I get the chance, if only to learn something new.

I'm curious, given Canada's past history with national debt, what the impact on international credibility would be from large bond-financed deficit spending? Would it impact our credit rating and cost of capital?

[Phillip Huggan]

Typo above, double spending should be 1/2 spending.

Harper changed the federal debt formula so that interest savings on debt repayments immediately get eaten as a tax cut, instead of going to general revenues. Shouldn't it also have been changed so deficits immediately claw back these tax cuts in a symetrical fashion?

At our income/corporate tax schedules, it doesn't really matter if we run a deficit and f$%^ the poor/kids tomorrow, or maintain surplus and f$%^ the poor/kids today. Rich Canadians have the opportunity to turn poor kids away from prisons to the workforce, and the opportunity to prevent the USA from annexing our farmland and lakes two generations from now. Instead they are choosing urban sprawl. I turn my soles up at you. O.G. speaks for all Conservative MPs.

[Nick Rowe]

O.G.: Some people want a smaller government; other people want a bigger government. The political equilibrium reaches some sort of compromise. Whether it's a good compromise, I don't know (or at least, it's a different question). The question is whether, in times like these (the US and UK now, and Canada possibly quite soon) we want a bigger government than we normally have. And on that question, I answer "yes", even if I would prefer that government normally be smaller than it is.

The biggest and best argument against using fiscal policy for macroeconomic stabilisation is that arguments for TEMPORARY CHANGES (how do you do italics in comments?) in government spending and taxation get mixed up (both logically, and politically) with arguments over the NORMAL size of government spending and taxation.

On the one side of the spectrum, people who want a smaller normal size of government (like yourself) argue against a temporary increase for that reason. At the other side of the spectrum, people who want a larger normal size of government (check out the Progressive Economics Forum blog) argue for a temporary increase for that reason. (To paraphrase Mandy Rice-Davies "They would say that, wouldn't they?").

A functioning market economy needs a functioning financial system. One of the miracles of capitalism is that, most of the time, it has a financial system that can convert water into wine -- convert long, risky, illiquid investments into the short, safe, liquid assets that savers want to hold. But it has always been a confidence game, (and probably always will be) and sometimes that confidence game collapses. If it stays collapsed, the real economy will be badly damaged. It will revert to medieval finance, where the only capital is directly owned by the individual proprietor. (And that by the way, is something the left never seems to understand, when they rail against corporations: without corporations, what assets will you hold in your pension plan? who will own the capital in any large factory [ and no, "the workers" is not the answer])

brendon: thanks! Government spending can be tax-financed, bond-financed (borrow), and money-financed (print money). Bond-finance would hurt our credit-rating, and increase the government's interest rate, but not by much, unless the debt got very high. History is reassuring. We now have a debt/GDP ratio of about 30%. A couple of times in the past (after World Wars) it has been around 100%. UK debt/GDP ratios have been 200% a few times in history (basically, whenever it fought the French or the Germans).

Phillip: I can't see the typo. Mine? Yours? On Harper and the debt formula: this is getting into two questions: the normal size of government (which I don't want to go into here); and the optimal size of the normal national debt (on which economic theory says very little, we barely even have a theory of the optimal long-run debt/GDP ratio).

[Phillip Huggan]

To give one example of my disgust of Conservative greed: Canada healthcare costs are scheduled to increase by $80B/yr well within the next decade.
Conservative tax rollbacks from 2005 scheduled to 2011 are clawing back around maybe $20B/yr in general revenue.
Harper has frozen transfers to Provinces; he basically agrees with O.G. that public schools and hospitals are a waste of time. What if the US economy undegoes capital flight and this lasts ten years? There is a looming $100B *annual* structural deficit, or Cons won't let boomers die with dignity. Canada of my future will be spent putting out boomer fires that were so cheap to extinguish now by funding childcare and Dion's Green Shift. And for what, to overtax civic budget with sprawl?
I know the goal is to encourage a diversity of debate, but this is like debating whether to invade just Iraq or Iraq and Iran. That damn libertopian/PNAC media marketing wasn't even aimed at Canadians, WTF?? Words of warning to Canada's rich: build those fences high; the future won't be as forgiving as I am.

[ramster]

You consider the two scenarios in terms of inflation, real growth, money/finance and interest rates. How about the actual impact on people? The unemployment rate (or something similar) would be a good approximation. We don't have our system of political economy to control inflation or always have GDP growth. These are intermediate metrics; means to an end. The end is always and ultimately about people. That's particularly important when considering the political dimension of economic policy.

[Nick Rowe]

ramster: agreed, but it IS ALL about people. ALL economics is ALWAYS ALL about people (OK, sometimes a very small number of economists study animals, just for fun). The only reason that inflation, interest rates, money, real growth, taxes etc. matter is that they affect people (directly or indirectly). Unemployment is not there explicitly in my discussion, but it's very much there. "Real growth" is a proxy for changes in employment. "Recession" is a short hand for "unemployment is higher than normal". What this post is really about is whether there is a trade-off between two groups of people: today's unemployed, and future generations of taxpayers.

Phillip: I think you're going a bit overboard there. But anyway, this post is about TEMPORARY CHANGES in Government spending, not about the long-run LEVEL of government spending. It is very wise to keep those two debates distinct, otherwise there is a very big danger that good short-run fiscal policy gets de-railed by long-run debates. In fact, the sort of points you are making now are the strongest argument against my arguing for a temporary change in fiscal policy. If short-run emergency fiscal policy changes are going to get hijacked, perhaps it really would be better to have balanced budgets fixed in the constitution!

[Phillip Huggan]

Fine. I'm just trying to rant what I know about economics before Obama takes office. Then my economics blog posting volume goes to zero. Thought we had this debate in Canada this fall and the interventionists lost bad.

Obama's stimulus will be $700B-$1T, and ours, I've seen estimates of $4B-$40B. Given that this blog has provided a resonable estimate of USA GDP growth/contraction having an effect on ours of 1/3 as strong, their stimulus would affect us $233B-$333B (little less because much is public sector). Shouldn't the real question here be about crystal ball-ing Obama's stimulus and how to capture/compete-against it?

I'm all for mining. Given maybe Mountain Pine Beetle onslaught, plastic paper and IT, and USA housing deadening, not forestry. Not coal, oil sands. How much would it cost to retool HEAVILY subsidized car plants to make wind turbine components (Obama is green)? If it is anything less than building new wind turbine plants I say go for it. Or, Honda is making palliative car robots. I'm guessing since our big-3 auto industry is branch plants and not ownership, it is really a question of giving free money to dumb CEOs (some say Ford isn't so bad) under the guise of temporary stimulus. How about subsidizing Zenn and giving the assembly lines to it? Glen Murray says on CBC light rail transit between Red Deer-Cgy is the way to go. Basically the platforms of Canada's left wing parties, *temporarily*.
I'd rather see MDS get all the big auto money to grow tenfold in Ontario (Cgy has a spacehub too), make surgery robots and play off growing national space programmes against eachother for one of a kind expertise. I'd rather just give all stimulus dollars to Barrick to Gold-PEC monopoly the world's best hedge. Derivatives and Republicans aren't going away.
For the record, a $100B deficit bankrupts our country in a decade.

[ramster]

I inferred that you were using growth as a proxy for welfare and unemployment. The problem is that the experience of the last decade (if not longer) in the US is one where there was economic growth but stagnation in median income. So growth is only a proxy for widespread prosperity if it's well distributed. After all, people often applaud GDP growth without improvements in employment (I believe it's called productivity growth). As far as your statement:

"What this post is really about is whether there is a trade-off between two groups of people: today's unemployed, and future generations of taxpayers."

It seems from your first sentence that your interest is in the future debt burden in the two possible cases (run a deficit now to minimize the recession vs let the recession run its course). Fair enough, it's your blog and you can focus on whatever it is that interests/concerns you. As it happens, I wholeheartedly support increasing the current debt burden through immediate spending increases. I just think it's funny sometimes how economists frame things.

[Phillip Huggan]

(my post copied from redtory in the context of Ontario big-3 auto bailout discussion:)

Normally we can’t have this discussion because US debt and US bankers as ponzi sellers hasn’t been a problem . But I’d think NAFTA would be ignored and if we find a good way to sop up unemployment, USA would encourage it.
Once again, I’d love to see the big-3 estimates based on a North American car market contracting over next ten years vis-a-vis last ten years. I’m guessing at least half the 500000 jobs would be lost anyways.
List of Canada R+D top 100 spenders, 2/5 are HQed in Ontario: http://www.researchinfosource.com/2008-top100.pdf
I’d think an assembly line worker could easily transition to any job requiring community college skills, but not a university degree. I’d guess any company would accept free money provided it or a college uses it to retrain and hire a laid off auto/auto-supply-chain worker, and build operations (just like the big-3 were subsidized) in Southern Ontario near existing auto infrastructure.

Growing industries would be:
1) communications/telecom; Nortel, RIM, Alcatel, Ericsson..
2) biotech/pharma; GlaxoSmith, Pfizer, Bayer..
3) software/hardware; IBM, Cognos, Corel…
4) Aerospace; Pratt n Whitney, Bombardier, CAE…
5) Machinery seems a natural; Husky injection, Xerox…
6) Others include; MDS, NOVA chemicals, maybe all electronic parts makers, maybe any Magna/Linamar product destined for Asian car market (much of this is already in other countries).

Mulroney had no problem putting an aerospace contract into higher bid Quebec and it doomed lower bid Wpg’s Bristol Aerospace. Assembly line workers don’t have a double-digit wage future, IMO. This is a very rare opportunity where it is politically viable to get out of the NAFTA assembly line subsidy. If Obama brings in Universal Healthcare (not given) these plants are dead wood. Just a guess, I’d think this means assembly line workers mostly wind up watching gauges (process control) and watching robots instead of manual labour.

[Nick Rowe]

ramster: OK, I see where you are coming from. This is where I was coming from: there's been this debate about whether a country in a liquidity trap (interest rates at zero) should run a big deficit to reduce unemployment. Some people have said "OK, it might help the unemployed, but we need to worry about the increased future debt." And I'm saying "No, I don't think it will increase the future debt burden, if we think about it properly". I was setting aside one part of the debate (the impact on unemployment), not because I think it's unimportant, but because I don't have anything new to say. Instead, I want to focus on a part of the debate that I think everybody else (other than me!) is wrong about. But yes, economists do frame things in very weird ways at times.

The inequality and productivity questions are for another day's debate.

Phillip: you are now going waaaay off topic!

[ramster]

Clarification accepted. The recent discussions have largely been about running deficits to raise GDP and boost employment (e.g. by bailing out the auto sector, building infrastructure, etc.). Since private sector spending is collapsing, let's replace it with public sector spending. What about the alternative of getting it directly in the hands of people? I don't mean tax cuts. I know that I'd take every penny of a tax cut and use it to pay off my debts, as would many people (i.e. the multiplier is < 1). Since we're concerned about the welfare of those losing their jobs and we want any economic stimulus to actually be, well, stimulating, why not just boost EI payments? We could lift the maximum insurable amount (it's currently $41000/yr), the benefit rate (it's currently 55% of your salary) and make it last longer (it currently lasts 45 weeks). Anyone who loses their job is going to need to spend every penny they have so all the extra EI payments will flow directly into the larger economy while helping those who need help the most.

[ramster]

BTW, given that the BOC rate is 1.5%, how close are we to an actual liquidity trap? The deflation thing feels counterintuitive to me. I recently bought a house with only 5% down so I have a honking mortgage. People say that deflation is bad for me because the real value of my mortgage is rising while the real value of my house is dropping. But as of now, all I see is that my variable rate mortgage payments have been steadily dropping so my cash in hand is increasing. I realize that things get ugly if I have to sell (I'm probably underwater right now once I factor in transaction costs) but as long as I stay put (and employed) deflation doesn't feel that bad to me. What am I missing?

[Nick Rowe]

With a variable rate mortgage, if your interest rate rises and falls point-for-point with inflation/deflation (as it normally does, over longer periods), and if your wages also increase/decrease point for point with inflation/deflation, and you don't plan to sell your house, you are unaffected by inflation/deflation. So you are nearly immune (there's a "front-end-loading" problem when inflation gets high, and a "back-end-loading" problem when inflation gets negative, because the principal part stays the same, but that all washes out over the life of your mortgage). But your interest rate cannot go negative, so if deflation gets big enough, your interest rate won't match it point-for-point. That's the only problem.

We are close to a liquidity trap.