Monday, October 23, 2017

Cowen and Sumner on Voters' Hatred of Inflation

A recent Scott Sumner piece has the declarative title, "Voters don't hate inflation." Sumner is responding to a piece by Tyler Cowen in Bloomberg, where Cowen writes:
Congress insists that the Fed is “independent"...But if voters hated what the Fed was doing, Congress could rather rapidly hold hearings and exert a good deal of influence. Over time there is a delicate balancing act, where the Fed is reluctant to show it is kowtowing to Congress, so it very subtlety monitors its popularity so it doesn’t have to explicitly do so. 
If we imposed a monetary rule on the Fed, even a theoretically optimal rule, it would stop the Fed from playing this political game. Many monetary rules call for higher rates of price inflation if the economy starts to enter a downturn. That’s often the right economic prescription, but voters hate high inflation. 
Emphasis added, and the emphasized bit is quoted by Scott Sumner, who argues that voters don't hate inflation per se, but hate falling standards of living. He adds, "How people feel about a price change depends entirely on whether it's caused by an aggregate supply shift or a demand shift."

The whole exchange made my head hurt a bit because it turns the usual premise behind macroeconomic policy design--and specifically, central bank independence and monetary policy rules--on its head. The textbook reasoning goes something like this. Policymakers facing re-election have an incentive to pursue expansionary macroeconomic policies (positive aggregate demand shocks). This boosts their popularity, because people enjoy the lower unemployment and don't really notice or worry about the inflationary consequences.

Even an independent central bank operating under discretion faces the classic "dynamic inconsistency" problem if it tries to commit to low inflation, resulting in suboptimally high (expected and actual) inflation. So monetary policy rules (the topic of Cowen's piece) are, in theory, a way for the central bank to "bind its hands" and help it achieve lower (expected and actual) inflation. An alternative that is sometimes suggested is to appoint a central banker who is more inflation averse than the public. If the problem is that the public hates inflation, how is this a solution?

Cowen seems to argue that a monetary rule would be unpopular, and hence not fully credible, exactly when it calls for policy to be expansionary. But such a rule, in theory, would have been put into place to prevent policy from being too expansionary. Without such a rule, policy would presumably be more expansionary, so if voters hate high inflation, they would really hate removing the rule.

One issue that came up frequently at the Rethinking Macroeconomic Policy IV conference was the notion that inflationary bias, and the implications for central banking that come with it, might be a thing of the past. There is certainly something to that story in the recent low inflation environment. But I can still hardly imagine circumstances in which expansionary policy in a downturn would be the unpopular choice among voters themselves. It may be unpopular among members of Congress for other reasons-- because it is unpopular among select powerful constituents, for example-- but that is another issue. And the members of Congress who are most in favor of imposing a monetary policy rule for the Fed are also, I suspect, the most inflation averse, so I find it hard to see how the potentially inflationary nature of rules is what would (a) make them politically unpopular and (b) lead Congress to thus restrict the Fed's independence.


  1. This comment has been removed by the author.

  2. The stagflationists:

    Don't know money from mud pie. Rates of change in money flows, volume X's velocity, equal RoC's in P*T (where N-gDp, R-gDp, and inflation are all subsets).

    The distributed lag effect of money flows have been mathematical constants for > 100 years.

    Thus we know several things, that money is robust, and that the RoC in R-gDp = exactly 10 months, trough to peak. And we know whether money is robust because we know when inflation accelerates relative to inflation, whose distributed lag is exactly 24 months, trough to peak.

    So we know when interest rates will respond to the monetary fulcrum, or inflation, based on money lags, and the "arrow of time".

    So to target N-gDp is stupid. It maximizes inflation and minimizes real-output. IF you can target R-gDp, why try targeting N-gDp? R-gDp accelerates before inflation. And we know when the teeter-totter tips.

    Inflation is the most destructive force capitalism encounters. What cost 1 dollar in 1913 costs $25.76 in terms of the CPI (which understates inflation).

  3. Chicken! Won't post: Price-level targeting? That sounds exactly like how Vladimir Lenin said he would overthrow capitalism, and presumably capitalists, by making their money worthless.


Comments appreciated!