Inflation Wars: Top-Down Vs. Bottom-Up Data

|
Includes: BIL, CLTL, DFVL, DFVS, DLBL, DLBS, DTUL, DTUS, DTYL, DTYS, EDV, EGF, FIBR, FTT, GBIL, GOVT, GSY, HYDD, IEF, IEI, ITE, PLW, PST, RINF, RISE, SCHO, SCHR, SHV, SHY, SPTL, SPTS, TAPR, TBF, TBT, TBX, TBZ, TLH, TLT, TMF, TMV, TTT, TUZ, TYBS, TYD, TYNS, TYO, UBT, UDN, USDU, UST, UUP, VGIT, VGLT, VGSH, VUSTX, ZROZ
by: Principal Financial Group

By Robin Anderson, Ph.D., Senior Economist, Principal Global Investors

As the old saying goes, there are two sides to every story. This saying holds true for the latest inflation data. For instance, the latest Consumer Price Index (CPI) and Personal Consumer Expenditure (PCE) inflation data have fallen flat. But, the "macro" environment suggests that inflation should be picking up, not softening. The unemployment rate is declining, and surveys such as the National Federation of Independent Businesses' (NFIB) Small Business Optimism survey suggest that wage pressures are building. The dollar is even weakening supporting higher pipeline inflation for imported goods. However, the "bottom-up" inflation data are telling a different story than the "top-down" macro environment would suggest.

Core inflation (ex-food and energy prices) has weakened. Year-over-year core PCE inflation declined from 1.8% in January to 1.5% in April. Core CPI inflation fell from 2.3% to 1.9% over the same period.

Wireless plan pricing, the focus of last month's blog post, reared its ugly head yet again in April. New unlimited data plans and changes in how they are calculated had significant downward effects on both core PCE and CPI inflation over the last two months. Yet, idiosyncrasies in cell phone plan pricing were not the only problems with core inflation.

Declines were broad-based. Used car prices dropped the last four months. And the rapid rise in prescription drug prices subsided. Within the CPI, prescription drug price inflation slowed from about 7% year over year to a bit above 3%. PCE inflation showed a similar pattern; although broader medical care price inflation is calculated and weighted differently than in the CPI. Shelter, which includes both rent and owner-equivalent rent, was still a big driver of core CPI inflation (less so for PCE inflation). And it still grew much faster than overall core inflation. However, shelter inflation lost momentum. Quarter-over-quarter annualized price changes slowed since peaking in December in both indices.

Within the PCE inflation measure, the weakness in overall core inflation looked even worse when you sliced out imputed prices, like owner-equivalent rent and financial services furnished without payment. Market-based PCE inflation, which excludes imputed prices, increased only 1.4%. There is reason to think imputed prices increased too much in April.

Economists from Société Générale (SocGen) noted that half of the month-over-month increase in overall core PCE inflation came from gains in imputed prices. But, those gains may not be sustainable. This really gets into the weeds, but please bear with me. SocGen pointed out gross output from non-profit hospitals jumped the most since March 2005, and financial services provided without payment increased by the most since March 2015. If these gains were "one-offs," as gains are reversed, it could be negative for core PCE inflation in the coming months.

Economic intuition suggests that core inflation should be picking up, but it's falling flat. To make matters worse, looking at the "bottom-up" picture, the weakness is pretty wide-ranging, and just not from one or two factors. It's getting hard to argue, as May FOMC minutes suggested, the weakness in inflation is transitory. At some point, if the data do not fit the story, it may be time to change the story!