The counterfactual US inflation history – the case of NGDP targeting

Opponents of NGDP level targeting often accuse Market Monetarists of being “inflationists” and of being in favour of reflating bubbles. Nothing could be further from the truth – in fact we are strong proponents of sound money and nominal stability. I will try to illustrate that with a simple thought experiment.

Imagine that that the Federal Reserve had a strict NGDP level targeting regime in place for the past 20 years with NGDP growing 5% year in and year out. What would inflation then have been?

This kind of counterfactual history excise is obviously not easy to conduct, but I will try nonetheless. Lets start out with a definition:


where NGDP is nominal GDP, RGDP is real GDP and P is the price level. It follows from (1) that:


In our counterfactual calculation we will assume the NGDP would have grown 5% year-in and year-out over the last 20 years. Instead of using actual RGDP growth we RGDP growth we will use data for potential RGDP as calculated Congressional Budget Office (CBO) – as the this is closer to the path RGDP growth would have followed under NGDP targeting than the actual growth of RGDP.

As potential RGDP has not been constant in the US over paste 20 years the counterfactual inflation rate would have varied inversely with potential RGDP growth under a 5% NGDP targeting rule. As potential RGDP growth accelerates – as during the tech revolution during the 1990s – inflation would ease. This is obviously contrary to inflation targeting – where the central bank would ease monetary policy in response to higher potential RGDP growth. This is exactly what happened in the US during the 1990s.

The graph below shows the “counterfactual inflation rate” (what inflation would have been under strict NGDP targeting) and the actual inflation rate (GDP deflator).

The graph fairly clearly shows that actual US inflation during the Great Moderation (from 1992 to 2007 in the graph) pretty much followed an NGDP targeting ideal. Hence, inflation declined during the 1990s during the tech driven boost to US productivity growth. From around 2000 to 2007 inflation inched up as productivity growth slowed.

Hence, during the Great Moderation monetary policy nearly followed an NGDP targeting rule – but not totally.

At two points in time actual inflation became significantly higher than it would have been under a strict NGDP targeting rule – in 1999-2001 and 2004-2007.

This of course coincides with the two “bubbles” in the US economy over the past 20 years – the tech bubble in the late 1990s and the property bubble in the years just prior to the onset of the Great Recession in 2008.

Market Monetarists disagree among each other about the extent of bubbles particularly in 2004-2007. Scott Sumner and Marcus Nunes have stressed that there was no economy wide bubble, while David Beckworth argues that too easy monetary policy created a bubble in the years just prior to 2008. My own position probably has been somewhere in-between these two views. However, my counterfactual inflation history indicates that the Beckworth view is the right one. This view also plays a central role in the new Market Monetarist book “Boom and Bust Banking: The Causes and Cures of the Great Recession”, which David has edited. Free Banking theorists like George Selgin, Larry White and Steve Horwitz have a similar view.

Hence, if anything monetary policy would have been tighter in the late 1990s and and from 2004-2008 than actually was the case if the fed had indeed had a strict NGDP targeting rule. This in my view is an illustration that NGDP seriously reduces the risk of bubbles.

The Great Recession – the fed’s failure to keep NGDP on track 

According to the CBO’s numbers potential RGDP growth started to slow in 2007 and had the fed had a strict NGDP targeting rule at the time then inflation should have been allowed to increase above 3.5%. Even though I am somewhat skeptical about CBO’s estimate for potential RGDP growth it is clear that the fed would have allowed inflation to increase in 2007-2008. Instead the fed effective gave up 20 years of quasi NGDP targeting and as a result the US economy entered the biggest crisis after the Great Depression. The graph clearly illustrates how tight monetary conditions became in 2008 compared to what would have been the case if the fed had not discontinued the defacto NGDP targeting regime.

So yes, Market Monetarists argue that monetary policy in the US became far too tight in 2008 and that significant monetary easing still is warranted (actual inflation is way below the counterfactual rate of inflation), but Market Monetarists – if we had been blogging during the two “bubble episodes” – would also have favoured tighter rather than easier monetary policy during these episodes.

So NGDP targeting is not a recipe for inflation, but rather an cure against bubbles. Therefore, NGDP targeting should be endorsed by anybody who favours sound money and nominal stability and despise monetary induced boom-bust cycles.

Related posts:

Boom, bust and bubbles
NGDP level targeting – the true Free Market alternative (we try again)
NGDP level targeting – the true Free Market alternative

Leave a comment


  1. “Hence, if anything monetary policy would have been tighter in the late 1990s and and from 2004-2008 than actually was the case if the fed had indeed had a strict NGDP targeting rule.”

    I’ve read the posts that Marcus Nunes has up that shows the NGDP trend through the 2000s. and particularly between 2003 to 2005 NGDP was below trend. It looks like the Fed took its sweet time returning it back to trend from the recession in 2001, getting it back there just before Greenspan left. And of course the Bernanke Fed began tightening almost immediately, so it wasn’t there for very long.

    So I am confused about your conclusion that the Beckworth view that money was too loose is the correct view. Perhaps it is just the difference between the “what if” and reality?

    • Dajeeps, I think the issue is really what base year is used and what trend growth is assumed. NGDP growth during the Great Moderation was actually slightly above 5% and it makes a major difference to these calculations whether you assume 5 or 5.5% NGDP growth.

      But again my view is probably somewhere between the Nunes-Sumner-Hetzel view and the Beckworth-Free Banking view in the sense that I do think monetary policy was too loose, but it might not in itself be enough to explain what happened in the US property markets. On the other hand it is very clear that monetary policy in the euro zone was far too easing in 2005-7 – and that probably is influencing how I think of that period.

    • Might the difference have to do with how long it took the Fed to return to trend NGDP from the 2001 recession that there was some wage and price adjustments going on? Or it could possibly be that because money might have been tighter than trend NGDP that would be why inflation in your counterfactual is a bit higher from 2003-06? Sorry if I am dragging you off topic, but the discrepancy between this and Marcus’ posts is perplexing, since below trend NGDP is a partial indicator of tight money.

      • One thing is NGDP relative to trend, but if we look at other indicators such as the US current account deficit, property prices, the level of risk taking and money supply growth and interest rates relative to different measures for the natural rate then I think the indication is that monetary policy became overly loose in 2005-6. That, however, was not the cause of the Great Recession, but it played a role initially.

        The Great Recession was rather caused by monetary policy becoming overly tight in 2008-9.

  2. Becky Hargrove

     /  October 6, 2012

    Even though my experience is only anecdotal, the Great Recession still feels like an inevitable result of the events of the late 90s. While the Great Recession of course played out more slowly, people were scrambling to rearrange their lives and plans after the tech “bubble”, and it became clear at that point for many that they would need to permanently adjust their life plans accordingly. While my anecdotal reasoning does not sit well with everyone, I am and will remain a strong proponent of NGDPLT because it shows what is POSSIBLE for monetary activity to achieve. My work then has to continue with what money cannot always achieve.

    • Becky, I think that a lot of people had the “feeling” of a bubble and that is probably one of the reasons policy makers are so reluctant to do the right thing today – “we just had a bubble and we don’t want another one…”

  3. Becky Hargrove

     /  October 6, 2012

    Lars, thanks for your response. I know that my stance confuses people because I am very much pro growth. It’s just that I believe the growth of the future has to happen on different terms. We need to find better ways for knowledge to utilize what basically amounts to single scale productivity, that is the technology that allows the individual to create product, but not in a way that the usual economies of scale pay off monetarily. While I have focused on the structured collaboration that could improve services in this regard, such collaboration is especially going to matter for the digital manufacture of the (near) future that will also mean some physical product benefits from economies of scale, but other product will be created which takes the opposite approach. Structured collaboration will also need to happen for local digital manufacture which does not rely on a Coasian framework.

  1. Bullard defends Price Level Targeting. Unfortunately it’s the wrong Price Level. Even the wrong target altogether | Historinhas
  2. Duncan Brown’s interesting NGDP wonkery | 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 )

Twitter picture

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

Facebook photo

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

Connecting to %s

%d bloggers like this: