The Bundesbank demonstrated the Sumner critique in 1991-92

I have recently written a number of posts (here and here) in which I have been critical about Arthur Laffer’s attempt to argue against fiscal stimulus. As I have stressed in these posts I do not disagree with his skepticism about fiscal stimulus, but with his arguments (and particularly his math). It is therefore only fair that I try to illustrate my view on fiscal stimulus and why fiscal stimulus (on it own) is unlikely to work.

My view of fiscal policy is similar to that view Scott Sumner as articulated in what has been called the “Sumner critique”. According to the Sumner critique if the central bank for example targets inflation or nominal GDP any action by the government to “stimulate” aggregate demand will only work if it does not go counter to the central bank’s nominal target.

Imagine that the central bank is targeting 2% inflation and inflation and expected inflation is at exactly 2%. Now the government in an attempt to spur growth increases the government spending by 10%. In a normal AS-AD model that would shift the AD-curve to the right from AD to AD’ as illustrated in the graph below.

The increase in government spending will initially increase real GDP (output) from Y to Y’, but also push up the price level from P to P’ and hence increase inflation.

However, as the central bank is a strict ECB type inflation targeter it will have to act to the increase in inflation by tightening monetary policy to push back the price level to P (yes, yes I am “confusing” the level of prices and growth in prices, but bare with me – I might just have written the whole thing in growth rates or argued that the central bank targets the price level).

Hence, once the government announces an increase in government spending the central bank would announce that it would reduce the money base (or the growth rate in the money base) to counteract any impact on inflation from the “fiscal stimulus”. The reduction in the money base would push the AD curve back to AD.

This is the Sumner critique – the government can not beat the central bank when it comes to aggregate demand. The central bank will ultimately determine aggregate demand and if the central bank targets for example inflation, the price level or nominal GDP then fiscal policy will have no impact on aggregate demand and note that this is even the case in a situation where unemployment is above the natural rate of unemployment. Hence, we have full crowding out even in a model with sticky prices and wages and underutilization of production factors (involuntary keynesian unemployment).

Furthermore, if the inflation target is credible then investors will realise that any fiscal expansion will be counteracted by a monetary contraction. Therefore, once the fiscal expansion is announce the markets would react by starting to price in a monetary contraction – leading to a strengthening of the country’s currency, falling stock markets and lower inflation expectations – this on its own would counteract the increase in aggregate demand. This is the Chuck Norris effect in fiscal policy.

Obviously in the real world neither monetary policy nor fiscal policy is ever 100% credible and there will always be some uncertainty about the scale and commitment to fiscal expansion and uncertainty about the central bank’s reaction to the fiscal stimulus. However, anybody who have follow developments in the euro zone over the past two years will realise that “promises” of fiscal austerity have been led to a rally in the stock markets (and fixed income markets in the PIIGS countries) as the markets have priced in the impact on aggregate demand of the expected monetary easing from the ECB. This is the reverse Sumner critique – fiscal tightening will not lead to a drop in aggregate demand if the markets expect the central bank to “cover” the short-fall in aggregate demand.

Hence, I think that the Sumner critique is highly relevant for the discussion of fiscal policy today both in Europe and the US. Below I will try to illustrate the Sumner critique with an episode from recent economic history – the German reunification.

The Bundesbank took all the fun out of German reunification 

After the fall of the Berlin wall in 1989 West Germany and East Germany was reunified. Due to the nature of the collapse of communism in East Germany the reunification of Germany happened extremely fast. Hence, most economic-political decisions were highly influenced by political expediency and geo-political and electoral concerns rather than by rational economic considerations.

One such decisions was the imitate political unification of the two Germanys. In fact East Germany was “absorbed” into West Germany. That for example mend that all social benefits and pensions etc., which were available to West German immediately (or more or less so) became available to East Germans and more or less from day one the benefit levels became the same in the entire unified Germany. This obviously led to a rather sharp increase in German government spending. The unification obviously also led to other forms of increases in public spending – for example the Capital was moved from Bonn to Berlin.

It is always hard to estimate how large a fiscal expansion is as the budget situation is not only influenced by discretionary changes in fiscal policy, but also by so-called automatic stabilizers. However, judging from calculation made by the Bundesbank (in the 1990s) the fiscal expansion due to reunification was substantial. In 1989 the cyclical adjusted budget surplus was around 1% of GDP. However, after unification the budget swung into a deficit. In 1990 the cyclical adjusted budget deficit was 2.5% GDP and in 19991 it had increased to 4.2% of GDP. Hence, the fiscal expansion from 1989 to 1991 amounted to more than 5% of GDP. This by any measure is a substantial fiscal easing.

It is very hard to assess what impact this strong fiscal easing had on the German economy – among other things because the Germany of 1989 was not the same country as the Germany of 1990 and 1991. Furthermore, this fiscal easing coincided with significant monetary easing as it controversially was decided to exchange one East Mark for one West Mark. That led to a rather substantial initial increase in the unified Germany’s money supply. However, while it can be hard to assess the direct impact on growth from the fiscal expansion it is much easier to assess the German Bundesbank’s reaction to it.

The Bundesbank was horrified by the scale of fiscal expansion and the potential inflationary consequences and the Bundesbank did not led anybody doubt that it would have to tighten monetary policy to counteract any inflationary consequences of the unification. Secondly, it also pushed strongly for the German government to fast tighten fiscal policy to reduce the budget deficit. Hence, market participants from an early stage would have had to expect that the Bundesbank would tighten monetary policy and that it would “force” the government to tighten fiscal policy. This in many ways is the exact same thing we see in the eurozone today, where the Bundesbank dominated ECB is telling policy makers if you don’t tighten fiscal policy then we will effectively allow monetary conditions to become tighter.

Already in 1991 the Bundesbank moved to counteract perceived inflationary risks and started tightening monetary policy. In a series of aggressive interest rate hikes the Bundesbank increased its key policy rate to nearly 10% in 1992. In that regard it should be noted that the Bundesbank hiked interest rates at a time when global growth was weak due among other things a spike in global oil prices in connection with the first Gulf war. Furthermore, the Bundesbank also put significant pressure on the German government to tighten fiscal policy, which it did in 1992.

There is no doubt that the Bundesbank wanted to demonstrate its independence to the government and probably for exactly that reason chose to be even more aggressive in its monetary tightening that was warranted even according to its own thinking. As a consequence of disagreement between the German government and the Bundesbank the governor of the Bundesbank at the time Karl Otto Pöhl resigned in October 1991 after having initiated monetary tightening.

The monetary tightening in 1991-92 not only sent Germany into a deep and prolonged recession it also was the direct cause of the so-called EMS crisis in 1992-93.

This particular episode in German (and European) monetary history is a powerful illustration of the Sumner critique. It is pretty clear that even substantial fiscal easing (around 5% of GDP) did not have long lasting impact on growth in Germany due to the Bundesbank’s counteractions to curb the perceived (!) inflationary risks.  I do not claim to have proven that the fiscal multiplier is zero, but I hope I have demonstrated that it is that it is unlikely to be positive if the central bank does not play along.

In the case of Germany in the early 1990s the fiscal multiplier was probably even negative as the Bundesbank decided to punish the German government for what it perceived as irresponsible policies. Anybody who is following the political struggle among European governments and European central bankers would have to acknowledge that it is very similar to the situation in Germany after the reunification.

Consequently I think it can be concluded that monetary policy will never be able to lift aggregate demand if the central bank refuse to do so – and that will often be the case if the central bank is worried about its credibility and independence.

I am no Calvinist and I tend to think that some of the calls from certain economists for austerity is rather hysterical given our problems particular in Europe primarily are monetary, however, I do think that the Sumner critique is highly relevant and we under normal circumstances (that is circumstances where the central bank for example pursues an inflation target) should expect the fiscal multiplier to be close to zero.

We all of course also know there are numerous other problems with fiscal easing – for example any temporary increase in public spending seem to become permanent and that is hard good for long-term growth in any economy, but that discussion is more or less irrelevant for the present crisis, which in my view mostly a result of misguided monetary policies rather than failed fiscal policies.


My discussion above was among other things inspired by Jürg Bibow paper “On the ‘burden’ of German unification” (2003) and a discussion with chief economist in the Danish think tank CEPOS Mads Lundby Hansen

Related posts:

“Meantime people wrangle about fiscal remedies”
Please keep “politics” out of the monetary reaction function
Is Matthew Yglesias now fully converted to Market Monetarism?
Mr. Hollande the fiscal multiplier is zero if Mario says so
Maybe Jens Weidmann and Francios Hollande should switch jobs
There is no such thing as fiscal policy

Leave a comment


  1. Lars, coincidentally today I clicked on this old post. The nature of the post is the same and the example is from a long time ago:

  2. Hi Lars,

    A convincing argument during “normal times”. How do you think the Sumner Critique applies when interest rates are zero, the central bank needs to ease further to boost AD but will not do so sufficiently (i.e. the situation in Europe, UK and arguably the US and Japan as well)?

    Do you think fiscal stimulus would be beneficial in just such a situation, where the offsetting monetary tightening is effectively already in place?

    • PoorRicardo,

      Thanks for the question. I fundamentally do not think the Sumner Critique is less valid in the present situation than in others. It seems like you are implyng that we are in a liquidity trap where monetary policy is inefficient. I do not think we are in a liquidity trap. Rather I think that monetary policy is extremely potent.

      The problem today is not that central banks can not ease monetary policy – they can easingly do that – but rather that they don’t want to ease monetary policy. ECB could easily double the level of nominal GDP or the price level – it is just a matter of printing money.

      I could, however, see a special case where you might be right. If the problem is that central banks don’t think that they can determine nominal GDP and the price level because they think that we are in liquidity trap. Then fiscal policy might actually work – because the central bank will not counteract when fiscal stimulus increases prices and nominal GDP. However, don’t think that that is the case in the real world. Central banks full well know that monetary policy “works” – they are just not willing to ease monetary policy for odd reasons.

      See this post on why I don’t think there is a liquidity trap:

      • Hi Lars,

        Thanks for responding so quickly. I’ve been an avid reader of your blog, and Scott Sumners, for some months and find market monetarism quite convincing. I therefore agree with your view that we are not in a ‘liquidity trap’ (in the sense that monetary policy is far from impotent) and the first best response to the current woes of the western world would be to give NGDP level targeting a shot.

        However, as you note in your response to Robert below, there isn’t a snowballs chance in hell that the ECB will do this, and it’s looking unlikely at the Fed and Bank of England too (for reasons unknown).

        In these circumstances, do you think use of fiscal policy, as a “second-best” policy response, is better than austerity and an underactive set of central banks?

        I understand your concerns about supposidly ‘temporary’ government spending easily becoming permanent, but tax cuts can be credibly temporary (e.g. the VAT cut in the UK in 2009, enacted and reversed exactly as promised by the then-Chancellor). Is that not a sensible course of action while we wait for the necessary intellectual revolution within central banks?

  3. Robert

     /  August 19, 2012

    It’s hard to tell what central banks actually think. One question that bugs me frequently is “why do they set a quantity of long term asset purchases instead of a target yield for long bonds?”. Is it because they don’t think they can actually affect yields, or because the don’t want to?

    • Robert,

      Yes, it remans a mystery why central banks have implemented QE in the way they have done. They have focused on the “input” rather than on the “output”. I however, would also suggested that it would be an bad idea to focus on getting bond yields down. In fact if QE is working and increasing expectations of higher future NGDP growth then we should expect bond yields to INCREASE and that would be good news.

      But you are right – central banks should focus on the the outcome. For example the ECB could announce that it will buy whatever amoung of assets needed to bring euro zone nominal GDP back to the pre-crisis trend level. That of course is never going to happen…

  4. “In that regard it should be noted that the Bundesbank hiked interest rates at a time when global growth was weak due among other things a spike in global oil prices in connection with the first Gulf war.”

    A lot of reasoning from price changes there – unfortunately we are used to it…

  1. The Bundesbank demonstrated the Sumner critique in 1991-92 « dajeeps
  2. Exchanging comments with Lars Christensen | Poor Ricardo
  3. The Bundesbank demonstrated the Sumner critique in 1991-92 « Economics Info
  4. Monetary disorder in Central Europe (and a some supply side problems) « The Market Monetarist
  5. The fiscal cliff and why fiscal conservatives should endorse NGDP targeting « The Market Monetarist
  6. In New Zealand the Sumner Critique is official policy « The Market Monetarist
  7. “Conditionality” is ECB’s term for the Sumner Critique « The Market Monetarist
  8. Skepticlawyer » They did it again
  9. Bernanke, Obama and the political business cycle – and some research ideas « The Market Monetarist
  10. Answering questions on “Quora” about Market Monetarism « The Market Monetarist

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: