We see little reason for confidence in the Fed’s view that negative U.S. inflation surprises this year are temporary and transitory.

There are some genuine oddities in recent inflation trends (see chart). Medical care, one of the two biggest components of core CPI (Consumer Price Index) — the other is housing — stands out as an example. Medical care inflation has been moving around in puzzling ways, rising until mid-2016 and falling since then. However, given the enormous regulatory uncertainty in health-care policy, it is difficult to be confident about the future trajectory of medical-care inflation. With regard to housing, the CPI for the broadest measure is shelter, and it has three components: rent of primary residence, owners’ equivalent rent (OER), and lodging away from home. OER and rent of primary residence move together, since OER is derived from rental statistics. The CPI for shelter had been a consistent source of inflationary pressure since the recovery from the crisis, but it is now easing for two main reasons.

Rent and lodging prices slacken

First, rents for primary residences are weakening at the tail end of a long-term story. In the aftermath of the housing crisis and the recession, demand for housing available for purchase was weak. Credit was difficult to obtain, people were scarred by the fall in house prices, and the recovery in employment in the early years was sluggish. As the economy picked up, so did demand for housing, but it was heavily biased toward demand for rented housing.

The U.S. housing construction industry, dominated by small firms accustomed to building houses in the suburbs, was ill-equipped to respond to the demand for multi-unit rental buildings closer to cities. It took a while for new supply to be planned and built. The rising demand combined with sluggish supply to produce rising rents, of course, and OER rose in tandem. We are now closer to equilibrium in these markets; demand for housing for purchase has recovered, and new rental supply has become available. So, pressure on rents has eased, and indeed in some places rents are now falling. While this trend means little for the housing market as a whole, it has implications for inflation, since it is rents — not house prices or the cost of owning a home — that drive housing in the CPI.

As the economy picked up, so did demand for housing, but it was heavily biased toward demand for rented housing.

Lodging away from home has become a factor. This is not a large component of shelter, but it matters. Hotel occupancy is very high, so why are hotel rates not rising? The answer seems to be Airbnb and similar apps and websites. Some people describe this as Airbnb doing to hotels what Amazon has been doing to retail outlets, but I don’t think this is entirely accurate. Amazon offers low prices, better inventory, and convenience, and this has changed the distribution channel for goods and services, but Amazon has not really led to an increase in supply. Airbnb and similar enterprises, on the other hand, have increased the supply of rentable rooms quite dramatically. It’s as if a vast number of new hotels had been built very quickly; existing hotels not only must compete with the hotel across town, they must also now compete with the spare bedrooms in every apartment in town. This is a very powerful effect, and frankly it is hard to dismiss it as temporary and transient.

These trends will, we expect, cause inflation to move higher in the next few months, and this is pretty much baked in the cake. Indeed, the U.S. inflation forecast indicates a hurricane-Harvey-related spike in headline inflation during October, followed by a sharp fall into early next year. Even as headline inflation begins to fall, core inflation will rise in the next few months, conveniently for a Fed that wants to hike in December. Looking toward next year, however, we expect core inflation to remain in a narrow range, at a level that is consistent with the Fed’s target, which therefore may not warrant further tightening.

What can be expected of the Fed?

It is fairly clear that the economy is doing well enough, and the Fed is confident enough about the economy, that balance sheet reduction will begin in October. In addition, we think it is the Fed’s current expectation that, if the economy keeps ticking along and balance sheet reduction does not generate stress in financial markets, there’ll be another rate hike in December. We think that short-term inflation dynamics will allow this. At this point, however, the outlook for 2018 does not look strong enough to warrant the Fed moving further.

Next: North Korea's new kilotons

More from Macro Report
Download the Macro Report (PDF)

As the summer winds down, the outlook for Fed policy, the North Korea crisis, and the questions surrounding U.S. tax reform remain the chief areas of uncertainty for the global macro picture.