Unintentionally, Kocherlakota makes a strong case for NGDP Level Targeting

In “Blackouts and the Burden of Uncertainty”, he writes:

From the mid-1980s through 2008, central banks had the tools, the will, and the knowledge to protect the economy from sharp swings in the demand for goods and services. They raised interest rates to head off surges, and lowered rates to prevent severe slumps. As a result, households and businesses could count on an economy in which aggregate demand grew relatively steadily. Nobody had to think about, or plan around, the possibility of persistent shortfalls in prices and employment.

That has changed. Since 2008, central banks haven’t been able or willing to defend against a sharp and highly persistent fall in aggregate demand. They have used much of their toolkit, and seem reluctant to employ the tools that remain. As a result, the flow of demand has become uncertain.  Market participants and others are focused on what could go wrong, and how central banks might — or might not — respond.

Before 2008, global aggregate demand was like electricity in the U.S. — just something in the background that everyone could count on. After 2008, it became like electricity in India — desperately needed, but subject to random and persistent shortages. Just as the uncertainty of electricity provision hobbles India’s economy, the uncertainty of aggregate demand impairs the global economy. To reduce uncertainty and promote higher growth, both systems need overhauls.

How should the world overhaul its system for providing aggregate demand? To me, this is the key question facing macroeconomists today. Answering it will require a big change in the discipline. Before 2008, most macroeconomists studying the U.S. and Europe largely ignored the possibility of long-lasting shortfalls in demand. This may (at least arguably) have been appropriate for most questions of interest before 2008. Now, however, they need different models and approaches to understand the effects of aggregate-demand uncertainty, and figure out how best to eliminate it.

The chart gives a visual of Kocherlakota´s ‘allegations’.

Note that during 1985 – 2007, NGDP (Aggregate Demand) growth was very stable (comparatively). In other words, “Before 2008, global aggregate demand was like electricity in the U.S. — just something in the background that everyone could count on.”


But note, after 2007 NGDP growth was initially quite unstable, “running off at high speed” through the Southwest corner of the “stability compound”. Contrast that with NGDP “running off at high speed through the Northeast corner of the “stability compound” in 1970 -84. While the 1970s defined the “Great Inflation”, the 2008-09, by symmetry, characterized a strong disinflation period. In 1985 – 07, aggregate demand growth is contained wholly within the “stability circle, and we had the “Great Moderation”.

Again note that contrary to Kocherlakota´s musings, after the “recovery” from the Great Recession was established in early 2010, what we observe is a very stable aggregate demand growth and not random and persistent shortages”.

However, given the low level of NGDP and it´s rather low average growth, the post 2010 period could be called “Depressed Great Moderation”!

From that perspective, again contrary to Kocherlakota, there´s no need to “overhaul the system for providing aggregate demand”, and also no “big change in the discipline” is required.

It is clear that the Fed can target NGDP growth at a stable rate. After all that´s what it did from 1985 to 2007 and from 2010 to 2015. What´s missing now is the definition of an adequate NGDP level and the most promising growth rate along that level path.

If that´s done the economy will, once again, prosper in a state of nominal stability.

9 thoughts on “Unintentionally, Kocherlakota makes a strong case for NGDP Level Targeting

  1. Great graphs.
    Central bankers are not targeting the obvious—growth in GDP, also called nominal GDP.

  2. “What is missing now is the definition of an adequate NGDP level and the most promising growth rate along that level path”.

    Scott Sumner has convinced me over the years that NGDPLT is the way to go as far as monetary policy. But I am very shaky on how the right level of NGDP growth should be determined for the central bank to target. Right now my understanding is that you would estimate population growth, add in some estimate of productivity growth, add in some estimate of potential real growth and come up with a percentage rate of growth that NGDP should be targeted at. Sumner generally says that this would be between 4 and 6 percent for the U.S. Is my understanding at all accurate? I have a feeling I am missing something. Is it some average optimal inflation rate target? How would you determine the optimal NGDPL target for a country?

    • Jerry, After the 1981-82 recession, the Fed “experimented” to find the appropriate level from which to start growing at around 5.5%. Likely, the previous level is not attainable anymore (in part due to hysteresis effects from the prolonged slump), but we also know (given the low inflation and growth) that the “adequate level is higher than the one we are on…
      I don´t believe in finding it through “estimates” of potential output and the like!

      • Thanks for the reply! The idea that it worked in the past so lets try that now is a very good argument to me. But I am afraid that it is far to simple an idea for many economists to grasp or support. You should dress it up with some delta symbols and logarithmic regressions or something like that. What’s the point of having fancy maths if you can ‘t use them. Your 1985-2007 graph would leave far to many economists with nothing to do.

  3. I learned from you guys that what matters is stable NGDP growth. Money is neutral, but not superneutral, therefore what matters is that the growth rate of NGDP is stable, or in other words, that the second derivative of NGDP is zero, and the variety of shocks that hit the economy don’t end up change the NGDP growth path. The actual NGDP growth target rate, under this view, is unimportant. I still don’t get why larger rates are better than lower ones.

      • Yes, Marcus, but your words: “What´s missing now is the definition of an adequate NGDP level and the most promising growth rate along that level path”. Implicit in this phrase is that you favor a higher LEVEL, and an implied higher growth rate. That is what I don’t understand, because your theory, and your own empirical tests don’t support the claim that a higher LEVEL (and corresponding growth rates) are better for the general economy, since money is neutral in the long run. To me, longer run money neutrality implies that any growth rates is acceptable, as long as they remain reasonably constant.

  4. JRR, A higher level is clearly required. Just look at inflation and inflation expectations. AD is clearly too low. To get to a higher level, a higher NGDP growth is temporarily required. The “steady state” growth rate from then on is a matter of discussion!

Leave a Reply

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

WordPress.com Logo

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

Google photo

You are commenting using your Google 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.