Uncategorized

Are wages and prices accelerating or merely catching up?

admin | September 13, 2019

This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.

Patterns this year in average hourly earnings and core CPI are strikingly similar: in both cases, readings in the first several months were softer than the 2018 trend, suggesting a puzzling slowdown. However, both wages and core inflation have picked up noticeably in recent months, bringing the year-to-date averages roughly back in line with 2018 performances. The key question is whether the swing to faster wage and core price increases is simply making up for a downside anomaly in early 2019, or a more persistent heating up of pressures in tight labor and product markets.

Early-2019 average hourly earnings

For years, wage gains had been sluggish. The inevitable result of a tight (and tightening) labor market finally showed through to wages in 2018. The average hourly earnings gauge accelerated from 2.7% on a December-to-December basis in 2017 to 3.3% in 2018. Heading into 2019, there was a widespread expectation that hourly pay would continue to pick up given the tightness of the labor market.

Instead, wage gains slowed up in early 2019.  In the first four months of the year, average hourly earnings posted a 2.4% annualized increase, a pace slower than the December-to-December rises in each of the past four years.

Bounceback in average hourly earnings

 Since then, average hourly earnings have bounced back vigorously.  In fact, the wage series has posted four consecutive monthly advances of 0.3% or more, a first for the measure (the data only go back to 2006).  The annualized increase over the past four months (May through August) was 3.9%.

The end result is that the annualized rate of increase on a year-to-date basis is 3.2%, roughly back in line with the 2018 performance (see Exhibit 1).

Exhibit 1: Average hourly earnings

Source: BLS

Early-2019 core CPI

Similarly, after the core PCE deflator finally reached 2% in 2018 after years of missing the Fed’s target, there was a hope that underlying inflation may have turned the corner.  Instead, core inflation slid in early 2019.  In the case of the core CPI, after a hefty January advance (0.25%), the core CPI rose by only 0.1% for four straight months from February through May.  As a result, the year-to-date pace of the core CPI dipped to a 1.8% annualized pace through May.

Bounceback in core CPI

Just when Chairman Powell began to give up hope that underlying inflation would return to target any time soon, the data came roaring back.  The core CPI posted back-to-back-to-back 0.3% increases in June, July, and August, the first such streak in almost 25 years.  This pushed the three-month annualized rise to 3.4%, also the highest since the mid-1990s.  As a result, the year-to-date annualized rise through August for the core CPI has rebounded to 2.4%, higher than any full year since 2007 (see Exhibit 2).

Exhibit 2: Core CPI

Source: BLS

Where do we go from here? 

There are two competing narratives to explain the nearly parallel movements in wages and prices so far this year.  The first story would be that the trends have been relatively steady, but there has been a heightened degree of statistical noise, reflecting seasonality or random fluctuations. Under this narrative, the swings should be ignored and we can take comfort in the fact that the year-to-date annualized gains through August are more or less in line with the 2018 results.

Alternatively, it may be that there were fundamental reasons that wages and prices moderated early in the year and a swing in those forces is also responsible for the pickup in subsequent months.  For example, it may be that the apparent tightness in the labor market over the past several years was mitigated by people who came off of the sidelines and back into the labor force, limiting the pressure on wages, but that the market has essentially run out of bodies in recent months and, as a result, pay hikes are gathering steam.

I lean toward the first explanation as the main driver of the fluctuations so far this year.  For the CPI figures, there is even a detailed story that supports that view.  Two key components within the core gauge drove both the weakness early in the year and the subsequent pickup in price hikes: apparel and used vehicles.  Apparel prices dropped sharply in March, which, at the time, was attributed to a change in the BLS methodology for collecting readings on apparel costs that was expected to result in a one-off, permanent drop in the level of clothing prices.  As it turns out, apparel prices on a seasonally adjusted basis declined in March, April, and May, by almost 3% cumulatively.  Apparel prices typically exhibit a high degree of seasonality, and economists would typically have been expecting a rebound after such a large dip, but with the change in methodology, it was unclear how apparel prices would behave going forward.  As it turns out, the spring plunge in apparel prices may have been mostly garden-variety noise rather than fallout from the new methodology, as clothing prices have risen for three straight months beginning in June, rebounding by 1.8% cumulatively.

Similarly, used vehicle prices posted declines in four straight months through May and subsequently rebounded noticeably in June, July, and August.  Used vehicle prices tend to swing up and down frequently, as the supply of used vehicles on the market tends to fluctuate substantially in a manner that dealers have limited control over (for example, as consumers return lease vehicles or trade in their autos) and the demand for used vehicles also swings around a lot, influenced by, among other things, the aggressiveness of dealer incentives on new vehicles.  The CPI component tends to track the data on wholesale vehicle auctions (where most used vehicles are acquired by the dealership that ends up making the sale to a consumer) with roughly a three-month lag, so that both the swoon early in the year and the bounceback this summer were not especially surprising.  The wholesale auction data suggest that used vehicle prices should continue to rise heading into the fall, but obviously the almost-15% annualized pace of increase seen over the past three months cannot be sustained for an extended period.

In my view, both wages and core prices should be on a gradually accelerating trend, reflecting the fact that job gains and real GDP growth has largely been faster than the economy’s long-run potential over the past several years, yielding tightness in labor and product markets.  However, the magnitude of the swings so far this year have probably been grossly exaggerated by the statistical noise in recent months in both series that I would view as mostly coincidence rather than evidence of a common underlying fundamental shift.

Given the extreme tightness in the labor market and the possible acceleration in productivity growth, a substantial pickup in wages is more plausible to me than a sharp acceleration in core inflation.  Having said that, I have been skeptical all along of the sustainability of the dip in core inflation in early 2019, and I do expect even the core PCE deflator, which has been an outlier to the downside among the various gauges of underlying inflation, to return to the 2% range on a year-over-year basis and probably even move above that mark by early 2020, when the low readings discussed above fall out of the 12-month window.