Credit bubbles and GDP targeting

In 2010 Scott Sumner first proposed that the Fed use GDP targeting rather than targeting inflation, which is prone to measurement error. Since then support for this approach has grown, with Lars Christensen, an economist with the Danish central bank, coining the term Market Monetarism.

Sumner holds that inflation is “measured inaccurately and does not discriminate between demand versus supply shocks” and that “Inflation often changes with a lag… but nominal GDP growth falls very quickly, so it’ll give you a more timely signal….” [Bloomberg]

The ratio of US credit to GDP highlights credit bubbles in the economy. The ratio rises when credit is growing faster than GDP and falls when credit bubbles burst. The graph below compares credit growth/GDP to actual GDP growth (on the right-hand scale). The red line illustrates a proposed GDP target at 5.0% growth.

US Credit Growth & GDP Targeting

What this shows is that the Fed would have adopted tighter monetary policies through most of the 1990s in order to keep GDP growth at the 5% target. That would have avoided the credit spike ahead of the Dotcom crash. More importantly, tighter monetary policy from 2003 to 2006 would have cut the last credit bubble off at the knees — avoiding the debacle we now face, with a massive spike in credit and declining GDP growth.

While poor monetary policy may have caused the problem, correcting those policies is unlikely to rectify it. The genie has escaped from the bottle. The only viable solution now seems to be fiscal policy, with massive infrastructure investment to restore GDP growth. That may seem counter-intuitive as it means fighting fire with fire, increasing public debt in order to remedy ballooning private debt.

Rising public debt is only sustainable if invested in productive infrastructure that yields market-related returns. Not in sports stadiums and public libraries. Difficult as this may be to achieve — with politicians poor history of selecting projects based on their ability to garner votes rather than economic criteria — it is our best bet. What is required is bi-partisan selection of projects and of private partners to construct and maintain the infrastructure. And private partners with enough skin in the game to enforce market discipline. I have discussed this at length in earlier posts.

4 thoughts on “Credit bubbles and GDP targeting

  1. Everett Goodwin says:

    The only fallacy is that the government’s bastardization of how GDP is measured makes the GDP number less relevant than ever. For example, GDP was redefined during the Obama presidency to include Research & Development (R&D). On the surface that might seem to make sense, but in reality it defies even the basic definition of “product”. U.S. Government R&D strictly forbids the use of R&D funds to create a product, yet the billions of dollars that go into R&D are now included in the GDP number! Even the effort to write poetry and music is now included. The Feds own GDPNow tool will start each month around 3.8 and then drop as each day goes by until by the end of the month it struggles to stay above 1.0. I suspect that the real issue with GDP is that the U.S. economy is less than 20% production oriented vs the 80% that it has been in the past. The Services sector now 80% of the U.S. economy which is also home to the majority of minimum wage jobs. As an aside, this is why there is such a push for doubling the minimum wage in the U.S. because the largest job growth falls into this region. This would also explain why the economy continues to limp along just above the recession level as it would be very difficult to grow any economy that is so inwardly focused. It would appear that we once again have folks not liking the answer (no inflation still!) so let’s use an answer we like better (redefined GDP). Once again looking at the color of band-aid to apply rather than noticing the sucking chest wound beneath it.

    • ColinTwiggs says:

      But GDP has always included services produced where there is no physical product.

      Definition from The Economist: Gross domestic product, a measure of economic activity in a country. It is calculated by adding the total value of a country’s annual output of goods and services…. It is usually valued at market prices; by subtracting indirect tax and adding any government subsidy, however, GDP can be calculated at factor cost. This measure more accurately reveals the income paid to factors of production.

      • evgoodwin says:

        To some extent, you are correct. The services portion that was previously included were mostly related to production services, but production represented 80% of the economy then. The double whammy is that those same services are marginalized because production now represents less than 20% of the economy, and the consumer is not likely to be able to be the spark the economy with the major job growth being at the low end of the wage scale. It is not so much the physical product as it is the intent of the development related to the product. Billions of dollars are spent on spaghetti code that is written for R&D projects because it is only used to validate concepts, not ever to be used in a product. It has some value, yes, but not something you would use to measure output of the economy. Again, if it were like the production related services that are now just a round-off error in the total picture, it wouldn’t matter much, but we are talking game changing numbers in the billions.

      • ColinTwiggs says:

        “…but we are talking game changing numbers in the billions.”

        What are the actual numbers? In a $13 trillion economy, billions are simply lost in the rounding.

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 )

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