Robin Harding Nominal consumption, not nominal GDP?

Nathan Sheets, head of the Fed’s international division until August, clearly plans to enjoy his freedom as head of international economics at Citi. In a new note he proposes a fairly dramatic communications option for the Fed – setting a target for the level of nominal consumption – which is definitely not the kind of thing you’re allowed to say in public when you work at the Federal Reserve Board.

As Mr Sheets notes, the Fed has ruled out any dramatic changes to its framework for the time being, but “our view is that in the event of a sizable financial shock from Europe—or evidence that the economy was slipping into recession—the Fed would be looking for a ‘bazooka,’ and such a regime would again be considered”.

The concept is similar to a nominal GDP target – you combine the Fed’s goals on inflation and activity into one number and you also target a level, rather than a rate, so if inflation and activity are too low now you promise that you will keep policy loose even if they go “to high” in the future to catch up.

“Similar to nominal GDP targeting, the FOMC would announce that it seeks to move the value of household purchases back toward the trend that it was on before the financial crisis. The Fed could broadly promise to use ‘all available tools’ to achieve this objective or specifically commit to purchase some quantity of assets with an eye toward lifting nominal consumption toward its pre-crisis trend. The basic mechanics of the regime would be broadly similar to those we discussed for nominal GDP targeting. In particular, the goal would be to strike a stance of policy that would plausibly be expected to bring nominal consumption back to its trend level over some medium-term horizon.”

Mr Sheets reckons that nominal consumption has several advantages over nominal GDP as a target:

  • It keeps the focus on consumer prices as the measure of inflation rather than the GDP deflator. That is closer to what consumers actually experience and to what the Fed does now thus reducing the communications challenge.
  • A stable path for consumption, rather than a stable path for GDP, is what consumers want and thus is closer to maximising their welfare.
  • Consumption is less volatile than GDP making it easier to target. Implied but not stated is that this would lead to less volatility in monetary policy.
  • Targeting consumption excludes components of GDP over which the central bank has no influence, notably government spending.
  • It might be easier to rally political support for stimulative policies in a crisis: “If politicians see that consumption is well below a sustainable level—meaning that their constituents are consuming less than they could be—this might help mobilize political support for stimulative monetary policies and perhaps also incentivize supportive fiscal policies.”

It’s an interesting idea but I can see some serious issues:

  • As Mr Sheets notes, setting a target for nominal consumption means that you have to make a decision about the “correct” level of savings. Mr Sheets adopts the post-recession level of 4-5 per cent but this is fairly arbitrary.
  • I disagree that this would be easy to communicate to the public or that it would attract political support – I think the opposite. Might opponents depict it as “compulsory consumption” ordered by the Fed? What are manufacturers going to say if a strong dollar kills exports but the Fed keeps policy tight because nominal consumption is strong? By considering only consumers you would alienate businesses and savers from Fed policy.
  • It does nothing to deal with the really serious objection to nominal GDP targets: you are taking a huge punt that inflation expectations will remain stable even when inflation goes above 2 per cent and you tell everyone that it was intentional. The cost of blowing up inflation expectations would be huge, so you have to believe the risks are low, and the rewards of targeting a nominal level very high to consider it.