Evan Soltas
Jan 29, 2015

The Deflation Coin Flip

By my reckoning, there is now about a 50-50 chance the U.S. sees deflation in 2015, as measured by the year-over-year percentage change in the headline personal consumption expenditures price index, the measure favored by the Federal Reserve.

Above, you see my point estimates from a basic econometric model -- I've just taken headline and core PCE and the monthly average Brent crude spot and put them into a vector autoregression. The forecast was done on monthly log changes and then summed up to get year-over-year results. (This is an update of my earlier forecast here.)

In the next few months, I expect to see an abrupt disinflation, as lower oil prices feed into the price index. That will be followed by a few months of hovering at zero, with a bottoming in the summer. And then we will have a few months of inflation rising quickly back to normal.

That is contingent upon oil prices staying about where they are. Remember: If it's just a change in the level of oil prices, we will see a U-shape pattern in inflation, as we saw in 1986, 2008, and 2011. One implication of this is that we should not treat the decline in inflation as highly relevant for macroeconomic policy.

Another implication, though, is that it will be critical to keep a focus on some measure of core prices. Anticipating some pass-through from energy prices, any substantial move in core inflation would be unexpected and concerning, particularly in light of the recent decline in inflation breakevens.