It's NGDP growth expectations that matter, not inflation expectations
By Scott Sumner
For the past seven years, I’ve been trying to persuade my colleagues that inflation should be removed from macro models. Inflation is often used as a proxy for positive demand shocks, as when economists assume that higher inflation expectations are expansionary, because they reduce real interest rates. That’s a classic example of “reasoning from a price change”. Lower real interest rates are only expansionary if they are caused by monetary stimulus, i.e. faster NGDP growth. When inflation rises due to an adverse supply shock the effect is contractionary, even if real interest rates fall. Indeed just this morning I criticized an article by Narayana Kocherlakota on exactly this point.
Now there is a new study that provides empirical support for my claim:
In the mind of central bankers, keeping inflation expectations at or above their long-term inflation target is important for keeping the economy on track.
The thinking here is that expectations for future inflation will encourage consumers to spend more now in the face of a likely decline in purchasing power.
. . .
But research out of Credit Suisse published Wednesday argues that an increase in inflation expectations, which to a central banker would seem to be great news, actually curtails spending unless a corresponding increase in expectations for wage growth follows.
This research, led by Credit Suisse analysts Hiromichi Shirakawa and Takashi Shiono, looks at the difficulty the Bank of Japan is having is pushing economic activity higher with their aggressive monetary policy action (taking rates into negative territory and buying up a huge amount of assets each month).
“Inflation expectations appear to have weakened across the board of late, as real household consumption remains sluggish,” Credit Suisse writes.
“Lack of a robust transmission mechanism from rising inflation expectations to household consumption recovery seems responsible for the prolonged sluggishness of consumption and thus the failure of expected inflation rates to be held up steadily.”
Here’s the major passage (emphasis mine):
The key problem here seems to be the adverse impact of rising inflation expectations on real household consumption. Our regression analysis on household consumption function has actually indicated that a decline in the real interest rate driven by a rise in the expected inflation rate may actually have a negative impact on household consumption. An expectations-driven quickening of inflation appears liable to be a negative for real consumption unless households are reasonably confident in the prospect of faster wage growth.
So, backing this out into plainer English, Credit Suisse’s work finds that unless consumers think they are going to make more money to make up for a gap in their loss of purchasing power, near-term consumption will likely suffer.
My only quibble is that I’d replace “higher wages” with “higher incomes”. As a practical matter, however, demand stimulus leads to both.