Economic growth is not contractionary, and other confusions about stimulus and spending

Let’s say that private gdp is 100 and government spending is 100.  Gdp then suddenly goes up to 200, so government spending as a percentage of gdp falls from 50% to 33.3%.  This is not a contractionary event.  It is fully possible to argue “government spending should go up too, to slot more public goods into the larger output,” but the initial change is expansionary, even though government spending as a percentage of gdp took a steep dive.

Let’s say that private gdp is 100 and government spending is 100.  Government spending stays the same in nominal terms but there is overall price inflation from a nominal change.  It is fully possible to argue “government spending should go up too, to restore the percentage of public goods in national output,” but the initial change is again expansionary in macroeconomic terms.  Nominal values are up.

In other words, when judging whether fiscal policy is contractionary or expansionary in macroeconomic terms, we do not automatically adjust for percentage of gdp and inflation.  Start instead by looking at nominal government spending, and then perhaps take a glance at nominal gdp or related measures.  The theory, after all, is about nominal values, most of all in the short run.

If you wish, I could construct a similar exercise for population or an increase in the labor force and come out with a similar result.

There is one case you could embrace (though I don’t see the critics doing it).  Say gdp is 100 and government spending is 100.  A negative real shock lowers real gdp and creates some price inflation.  Nominal government spending stays the same but in real terms it falls and that is quite possibly contractionary.  In that case the “adjustment for inflation” makes more sense, because the boost in prices is not producing some other positive, expansionary pressure.  That scenario is fully coherent, but of course suddenly the negative real shock is the major problem and the fall in government spending is a secondary and derivative problem, albeit a potentially important one.

Going back to my initial post on European fiscal policy, there are many upset commentators but in general they are not grasping these points much less responding to them or showing some level of understanding which is deeper than my own.

I am more than willing to admit that there are deeper understandings yet than what I offer in this post, but we are not at them yet, not in this discourse at least.

I don’t wish to respond point-by-point to some of the writings in the blogosphere, but given the above, Ryan Avent also is not looking deeply enough.  Both he and Brad Plumer did not see that the posts in question clearly distinguished between spending cuts and “austerity” (Brad did issue what is arguably a correction.)  I admire both bloggers and read them regularly, but these two posts both fail; here are some comments from Veronique.  I would say there is a dominant narrative, repeated many times in not always precise language, which people find it very hard to think outside of.

Most of the time “austerity” is a misleading word and more precise concepts — readily intelligible I might add — are available.  There really are some times when we should relabel austerity as “mostly tax increases,” but many people are reluctant to do so.


I heard on the radio (really) a commentator agreei with you. He said something to the effect that the word "austerity" is too charged and another word such as "retrenchment." I like "reversion."

A change in vocabullary may result in less hystrerical reactions.

In the run up to the current, global economic "wailing and gnashing of teeth", many nations experienced public sector growth that was significantly greater than private sector growth. That may be an indicator of whether reversion may be required. As in reversion to 2006 public sector spending levels and/or recoupling public sector growth to private sector growth.

This is Tyler at his best

Really? I think this might be the most confusing post of his I've ever read. I have at least three ways of interpreting the 2nd paragraph, and I can't tell which (if any) of these was intended.

That's because you, sir, are a retard.

I was baffled at this debate, since it's referring to a useless chart. Some of the public spending was to pay off debt, rather than the usual social welfare beneficiaries. Where is the breakout of that? It's possible for spending to go up but the size of gov't in the economy to go down in a stimulus sense. Paying down derivatives creates precious little employment. Austerity can still look like increased spending if you have a large debt to pay off or spend irregularly (the US DoD comes to mind).


Who is on your blog roll? That seems worthy of a what I have been reading post.

He has them listed on the bottom left.

I'm a bit surprised at the focus on this, since it doesn't really affect Veronique's graph much. After all, a lot of pixels were spilt railing against spending cuts that never happened.

Most of the time “austerity” is a misleading word and more precise concepts — readily intelligible I might add — are available.

Ahem solvency ahem.

The argument in the second paragraph seems misspecified, or at least incomplete. I take it that simultaneous with the inflation, the private sector grows in real terms enough to make up for the real decline in government? But that still doesn't make the event "expansionary"; the real output of goods and services has stayed the same. I don't think most people consider nominal values going up while real output holds steady to be what is traditionally meant by macroeconomic expansion.

So I'm still not sure why we wouldn't adjust for inflation, but point taken on adjusting for economic growth. However, shouldn't the baseline be the counterfactual rather than t=0? And unfortunately, I don't think Greece's GDP (or that of the various others) has doubled lately.

If GDP=100 and Gov spending =100 then Gov spending is 100% of GDP. If GDP rises to 200 (unlikely in that scenario!) & Gov spending stays at 100 then it is then at 50%.
Problem for many countries is GDP contraction made Gov spending a higher than affordable % of GDP.


GDP = *private* GDP (in Tyler's post). 100 private + 100 govt spending = 200, making govt spending 50%.
200 private GDP + 100 govt spending = 300, making govt spending 33.3%.

This comment is more illustrating of what is at stake than all other comments I have seen. Govt spending is extracted *at the threat of imprisonment* from private GDP, and those that object to de Rugy's post simply do not see it that way.

This reminds me of when someone claimed a borrowed dollar of gov't spending has a positive multiplier to GDP not just in the year spent, but for every following year, and was promptly asked how then gov't debt could ever be larger than GDP.

And then Krugman responded that those cases were explained by the government not borrowing enough.

Another embraceable case: justice and other public goods are assets & important factors of production that require maintenance. Required spending to maintain the "adequate policing" asset may well be a function of GDP, population, etc.

Sure. The socialists won in France promising to re-establish the retirement age to 60 from 62. Among other things.

IOW your point is irrelevant.

All that makes sense, but if unemployment increases by 30% (say, from 7% to 9.1%) and, as a result, (1) unemployment benefits go up, (2) requiring the government to layoff teachers and not open a new school as planned, people are going to look at that as austerity, even if it really isn't contractionary. People are going to be especially mad if, in addition to that, their taxes went up. They might be mad even if they are not paying more in taxes because, even though rates went up, they pay no more than they did before because they had to replace a higher paying job they lost with a lower paying job. We need a new framework for understanding and arguing about this stuff.

What I found most perplexing about the arguments that there have been "savage" cuts to government spending, only they are masked by booming automatic stabilizer transfers (unemployment benefits), is that these very transfers are trotted out as the ideal speedy, targeted stimulus. Why would people be so upset about this, unless they thought that were "structural" issues that were better treated by longer term spending.

Just to be clear, you are implying Greece's government/GDP drop is because the Greek economy is growing so strongly? (You may replace Greece in that sentence with France or Spain.) And since Avent didn't answer this hypothetical, his understanding is less than your own?

"Most of the time “austerity” is a misleading word and more precise concepts — readily intelligible I might add — are available."

This seems to be the main thing to me.

But the problem starts here:

"Start instead by looking at nominal government spending"

Austerity measures are a very particular kind of spending cut, from the perspective of those against austerity.
Nominal government spending erases distinctions in kinds of spending.

What does austerity consist in for those against austerity?

Political cuts to welfare and discretionary income.

What is that reflected in? Maintaining a certain *level* of public service provision and a certain *rate* of taxation among the poor.

Note in the case of public service provision that non-austerity policy could actually require an increase in public spending, and so a failure to increase public spending would be austere.

Now this conception of austerity may end up having little to do with government spending as a whole or fiscally contractionary policy in general. It may not mean the same thing as what economists are talking about when they are talking about austerity. But it nonetheless means what many people are talking about when they talk about austerity. If we want to talk about fiscally contractionary or fiscally expansionary policy, use those words. If we want to talk about policy aimed at solvency, talk about solvency, debt repayment, or deficit reduction. But when people are talking about their living conditions being made austere as a result of policy choices, maybe don't tell them the policy choices they observe aren't actually austere because on net government policy doesn't appear to be contractionary or deficit reducing or debt reducing or whatever..

To tell those people that austerity is not occurring because they're using the wrong word is just pointless and inane. In fact, their use of the world is quite close to the plain meaning of the phrase: 'austerity measures,' read as 'measures which make people's living conditions austere.'

On a blog named "Marginal Revolution", the reason so many people care about inflation should be obvious.

Also, given how much of the Obama stimulus consisted of tax cuts, are you unhappy with "stimulus" as well?

Great post.

On Monday I wrote: "Ultimately the debate over whether or not “The austerity experiment” has failed comes down to one’s definition of austerity. Moving beyond this debate requires policy makers and pundits to clarify their intended meaning of austerity before passing judgment on its effects."

The last paragraph sums it up and I think the idea that austerity in Europe has largely consisted of increasing taxes is too frequently missed.

I am sorry, when we talk about whether austerity happened or not in Europe, clearly we need to be talking about on a per capita basis, in euro terms. inflation adjusted. Instead, we saw a graph of nominal spending in dollars ( ! ) unadjusted for population. Heh? You do realize dollar depreciated 20% since 2004

Answer: Abstraction from particulars with no loss of substance. See Varian chapter one: "A model's power stems from the elimination of irrelevant detail, which allows the economist to focus on the essential features of the economic reality he or she is attempting to understand. ... In general we want to adopt the simplest model that is capable of describing the economic situation we are examining."

yeah, I get that. there was 17% total inflation between 2004 and 2012 and there was a depreciation of the dollar roughly equal to 20% and there was probably some population growth in Europe during those 8 years So here is the chart I want to see: the same chart adjusted for all of the above. That is the basis for comparison that makes any sort of sense. My guess is the charts will show a much steeper drop when charted this way.

Yes, you are really confused. The whole argument for stimulus from the Keynesian AD types is about NOMINAL spending and that's what Tyler is addressing.

Sorry, you are wrong. The conversation is about whether or not there is austerity in Europe. Austerity is felt in real terms per capita so posting a chart that is in nominal terms in the wrong currency and not per capita does not tell you whether or not austerity is happening.

When people talk about "austerity" in this context they are generally talking about gov't policy, not any circumstance that happens to reduce living standards.

You really missed the seven € symbols on the side an the part where it said "euros"?

Please take another look at the chart, especially the part that says euros(!).

If three economists cannot agree on terminology, it makes you wonder what else they are missing? Seems a lot of semantics and not much substance in this debate. Then again, that is true of a lot of economics, with simplistic linear models "proved" in exact closed form by authors, and that passes for scholarship. If you laid all the economists in the world end to end you would reach no conclusion, even if they are one armed economists...

I think the root of the problem is that the Veronique de Rugy graph in your "misunderstood" post is itself misleading.

It has nothing to do with nominal vs real or the size of government. It has to do with the fact that all of economics (including all of its hyper-partisan aspects on either side) amounts to worrying about tiny fluctuations on a massive base.

[graph of NGDP from 2002-2012, plotted from zero]

is basically a straight line. I could say: Look at this graph and tell me how bad is this recession, really? But our experience says it is bad. And that's because it is bad. We live in the crema of the espresso, not the coffee. Those tiny fluctuations in the curves are meaningful.

I submit this graph and ask, How generous has US stimulus been?
[graph of federal + state + local expenditures 2002-2012, plotted from zero]

(Looks like austerity to me ... which would explain the sluggish growth.)

Jason, the flip side is we live in the coffee and you worry about the "crema."

Only the marginal are unemployed, and we could put them to work tomorrow with my Guaranteed Income plan, but the idea of actually making its recipients be auctioned for private use weekly by their neighbors and friends, annoy liberals.

When you want to get serious about letting guys like me really unstick wages, and we fail, then you can call this a CRISIS.

But until I get out my Internet ax and start using the unemployed to reduce the rolls of public sector union workers... this is just a blip in your crema, when you are serious, I'll make your coffee.

Spain is richer than Honduras (by about a 100x in GDP), yet Spain is in economic crisis with 20% unemployment and Honduras is not in a recession and has like 5%.

Now tell me about the crema vs the coffee again?

I think there is a short way to make this point that might have helped Ryan Avent see what you mean. To paraphrase Scott Sumner:
"Never reason from a ratio change"

Tyler:"A negative real shock lowers real gdp and creates some price inflation. Nominal government spending stays the same but in real terms it falls and that is quite possibly contractionary. ... That scenario is fully coherent"

I am a 100% Sumnerian market monetarist, but I recognize that MMT is telling us an AS story where the government has got a temporary supply-side advantage in financial intermediation after the credit crisis, and any reduction of structural deficit is an AS shock. So basically Post-Keynesian crazies have got a coherent argument about austerity, but they are mistakenly expressing it in AD terms, and it is valid in AS terms only.

Where you say "this is not a contractionary event", I think it would be more accurate to say "this is not a spending cut". You have abstracted from any relationship between government spending and private output, and you have assumed no change in fiscal policy. The term “contractionary” normally refers to the effect of a policy change on output. You seem to be describing the effect of the shock, not the effect of keeping fiscal policy unchanged, on output when you say “the initial change is expansionary”.

I am most unclear about your position on inflation. If “ Government spending stays the same in nominal terms but there is overall price inflation” so that real government expenditure has fallen, what do you mean “the initial change is again expansionary in macroeconomic terms”? What change? The positive shock which raises output? But this says nothing about whether a reduction in real government spending is contractionary or not.

Then you say:

“A negative real shock lowers real gdp and creates some price inflation. Nominal government spending stays the same but in real terms it falls and that is quite possibly contractionary.”

What is it that you mean by contractionary then? If you are referring to the relationship between government spending and output, why would the nature of this relationship be different in the context of a positive versus a negative exogenous shock which has nothing to do with policy?

You don't think people are thinking deeply about this because of mood affiliation.

This is too much like the arguements two years ago about what is a liquidity trap. Everyone argues, someone is wrong, and then the definition changes so no one is wrong.

The European governments position was, and is, that if they balance their budgets market confidence would increase, borrowing rates would stay low, and recovery would spring forth. The only people screaming don't do it held the position that whatever they do to balance budgets will be counter productive. The pro austerity crowd was fine with the tax increases until they didn't work (or landed on the rich) and now the definition of austerity has to change. If Europe had gone pure spending cuts, and the deficit remained as tax revenues dropped, we'd be hearing they should have done the opposite.

It's the old saying about teaching economics vs engineering. Econ prof can use the same test every year because the answers always change.

This situation does not require "deep understanding." It's not even really an argument about economics. It's about agreeing on how the term "austerity" is used. Here's a rough-and-ready definition of "austerity" that is consistent with how the Krugmans and the Avents and the Camerons are using the term:

"Austerity: A set of changes to the laws controling government spending (and/or the tax rate or base) such that government expenditures (and/or tax revenues in a static analysis) during the specified time period are lower (or tax revenues higher) than would have occurred without the laws passing."

The word is descriptive. There is no consideration of what spending or taxes "should" be other than assuming a "current-law" baseline for the purposes of analyzing policy. It does not require spending cuts in nominal, inflation-adjusted, or inflation-and-population-adjusted terms. It simply requries an action to cause spending to be lower (or taxes higher) than would have been without the action.

While de Rugy's graph does not provide a baseline against which actual expenditures are to be compared, I think we can all nevertheless agree that the countries whose data is included in the graph have passed laws that caused government expenditures to be lower than would have been the case had those laws not passed. So no matter what de Rugy says, these countries have implemented "austerity" because that's just what the word means. This is a totally separate debate from whether "austerity" is wise, whether "keeping spending on autopilot" or "keeping tax revenues on autopilot" or "balancing the budget on the backs of group X" has higher moral status in the reader's mind.

I don't know where you got this definition but I think it proves a relevant point. Austerity means different things to diff people . There is your rather technical definition and there is a much more emotionally charged meaning which probably dominates media and the general public.

First, I think many people would agree that "austerity" is a misleading term. That is what set a lot of people off at the post with the title, "How savage has European austerity (spending cuts) been?" If austerity is a misleading term, why use it at all in economic discussions?

Second, in monetary economics, Milton Friedman dispensed with the notion that we could measure expansionary or contractionary monetary policy simply by looking at interest rate levels or changes in interest rates. What you need instead is something like the Taylor Rule to supply a benchmark of what interest rates should be. An interest rate higher than the Taylor Rule-dictated one is contractionary in relative if not absolute terms. And relative is what we should really care about. Without a "Taylor Rule" for fiscal policy, discussions of whether fiscal policy is relatively contractionary or expansionary won't get very far because there is no benchmark against which to judge policy.

One key point that goes missing is that whilst Austerians prefer to have the system "reset" at an equilibrium that is sustainable without much further debt or monetary stimulus in order to reposition itself and grow from there, is that due to various regulations, laws and outside competitive pressures (e.g.coming from Asian countries that do not stand still in the meantime) that rebalancing is made very hard, if not impossible. These nations are stuck between a rock and a hard place, stimulus is not an option and purging the debt to start afresh is also not really an option (politically at least). The crux of the matter is that these nations are not as rich as they thought they were. How many people would advice Asian or Latin American nations, if they found themselves in a similar situation, to engage in stimulus rather than get their political economy in order?

Tax revenues in the eurozone as a percentage of GDP are lower now than since sometime before 1995 (I don't have earlier figures).


"Second, I wish anti-austerity critics would start acknowledging that taxes have gone up too–in most cases more than the spending has been cut. "

is as touch misleading. Tax receipts are down - way down. Some countries which are in trouble - Greece, Ireland, Spain and recently Italy have raised taxes to compensate for the decline in receipts. They are all under severe pressure from the bond market, and have to show willing or get frozen out. Raising taxes is the fastest way to do it. There's no way politically to make rapid reforms. The governments are too weak (other than Ireland, which has cut).

The UK has cut some taxes and raised others in what is supposed to be a business friendly fashion. I don't how it nets out.

In addition, to mean anything, her charts should be real spending per capita net debt service. Otherwise that could be substantially lower and still show higher spending - a very peculiar view of austerity.

This is Spanish inflation

2000 3.43
2001 3.59
2002 3.06
2003 3.04
2004 3.04
2005 3.37
2006 3.52
2007 2.78
2008 4.09
2009 -0.28
2010 1.80
2011 3.20
2012 1.96

This comes to more than 40% or 2/3 of the increase shown in her chart.

She's pulled this kind of stuff before.

As Tyler noted, typically you look at these things in nominal dollars.

But real dollars doesn't change the picture much. Nearly all these countries are spending more than they were in 2007, which was not exactly a time of savage austerity.

Also, tax receipts have only fallen by about 1% of GDP, they're actually remarkably steady -- they've been in the range of 39-43% of GDP since 1995, and the largest drop was actually in the early 2000s.

As always the devil is in the details. For instance, since 2007 Greek tax revenue is down around 13 1/2 percent though it's almost unchanged as a percent of GDP, so spending as a percent of GDP is a bit misleading.

These sorts of calculations are hard to make meaningfully without a lot of careful thought, and you really need to consider more than one metric, because many countries are special cases. Spain, Greece, Ireland, Italy, France, Portugal, the UK and Germany all are for various reasons.

For instance, the UK has an enormous financial sector debt.

Spain has horrendous unemployment, out of control regional spending, and several large insolvent banks (It just committed 10 billion euros and counting to one this week). They're stuffed with sovereign debt and bad real estate loans. Spain has no idea how much it's going to spend or what it will receive in taxes. Still, most people would characterize 22.4% unemployment as reasonably austere.

De Rugy is just spinning a political agenda, not measuring anything meaningful. That would be hard work.

I read all of those critical blog pieces disagreeing with Tyler.

They are shrill and suffer from mood affiliation.

Good point! Everyone who disagrees with Tyler at any time on any subject is shrill and suffers from the fallacy of mood affiliation. This should be the blog heading.

I think Tyler and Rugy both have mischaraterized (misread, misheard) overlooked what the Keynsians (Stiglitz, DeLong, Krugman, etc. ) have been saying about European austerity. The Keynsians have criticized both the short term tax increases and the spending reductions for economies in a liquidity trap. They would repeal the huge increase the U.K. had in its VAT in 2011, as well as the immediate spending cuts and cuts in Government employment. When engaging in stimulus spending, they would try to spend on projects that are likely to improve Economic productivity long term and the would make permanent structural adjustments the major spending programs that would reduce the growth of those expenditures long term. If they have tax increases, they would include provisions for that they would only go into effect when economic growth, employment, and inflation reach certain trigger points. And, in addition to Fiscal expansion now, they would have the ECB act like the Fed and the Bank of England, as a lender of last resort to both banks and countries as long as the liquidity trap lasts. For an example read:

If Tyler and Rugy are upset that the Keynsians are not understanding them or mischaracterizing them, they should at least acknowledge that they have done the same to the Keynsians (who are very clear that they would cut spending and increase taxes if the economy was growing). Of course there is the ideological difference, Krugman and Stiglitz belive the New Deal and post WWII welfare state were good and should be preserved and expanded where appropriate to repair market failures and organize society along J.S. Mill's principles of the "greatest good for the greatest number " (utiliitarianism, not Marxism, by the way). Tyler and Rugy think it is a kind organized theft frorm the Nietzschian Superman and Woman who bears society on their shoulders and shoud be abolished.

Tyler and Rugy, as I understand it from reading their recent posts, therefore believe that "expansionary austerity" has not failed, but rather the particular kind that they and their co-believers would like to try (dissetablishment of the post WWII Welfare State with deep tax cuts on high incomes and capital) has not been tried. I think the examples they cite have that not faced the current conditions where world-wide demand is insufficient to fully employ the labor available to it and the Central Banks of the U.S., Japan, U.K, and Europe cannot engage in traditional monetary policy because interest rates are at the zero bound (although perhaps with a Republican President and the Ryan Budget in place, Conservatives would find less problems with QE). Again, I think that cutting benefits and income assistance programs at a time of mass unemploiyment is problematic, but if you follow the logic of Harry Lime that I quoted yesterday (from the movie "The Third Man"), losing a few million dots while obtaining trillions of Dollars, Euros, and Steriling in tax cuts is perfectly acceptable and in fact highly moral.

If I may put my market monetarist hat on, the election of the anti-Austerians is very contractionary, judging from the market reaction. The Dow, oil, and TIPS spreads have all fallen.

On the other hand, if Bernanke even so much as utters a sentence with the letters "Q" and "E" in it, the market rallies at least 1%.

Furthermore, whenever a EURO country commits to austerity, the market is happy. This may not mean that austerity itself is expansionary. But perhaps austerity keeps the EURO and LTRO-type programs alive, and that is expansionary compared to the alternative.

I'm going to try another example using, what I think, is pretty much the same logic:

Let’s say that private gdp is 200. The gov't collects 30 of that as taxes. Let's say that GDP falls by 100 (down to 100), and the gov't still collects 30 in taxes*. Gov't taxes as a percentage of GDP have risen from 15% to 30%.

Government revenue stays the same in nominal terms. It is fully possible to argue “government revenue should go down too, to restore the percentage of revenue in national output.” ...when judging whether fiscal policy is contractionary or expansionary in macroeconomic terms, we do not automatically adjust for percentage of gdp and inflation.
It really is the same logic as Tylers, and that logic shouldn't change just because you're switching from one side of the ledger to the other (ie, from expenditure to revenue). Call me cynical, but I think if Tyler tried to claim that a tax rate that doubles from 15% to 30% was NOT contractionary purely because the nominal amount stayed the same, people would have a field day with him.

*Essentially, I'm using a lump sum tax, just like in many economic models, and not really any different than the lump sum spending Tyler presents in his example. It's just revenue instead of expenditure.

Comments for this post are closed