The U.S. economy continues to look fine, but there are signs of weakness in consumption data. Meanwhile, labor markets continue to improve, though not without some wage-related mysteries.
In the fourth quarter of 2016, U.S. GDP growth was 1.9% (quarter over quarter [qoq], seasonally adjusted annualized rate [saar]). This was a bit weaker than we anticipated. However, we are less concerned about the headline number than we are about the details in the underlying consumption data. While overall personal consumption growth, at 2.5% (qoq, saar), was down only a little from a strong third quarter, much of its strength was in the consumption of durables, in which auto purchases play a very large role.
Recent strength may have run out of gas
Auto purchases were strong in Q4, especially in December, but there aren’t many people who think this pace can be maintained. Indeed, January sales were 17.45 million (saar), down from 18.29 in December. Aggressive discounting by automakers has been important in keeping sales high, but most people seem to expect this will ease. We think sales in 2017 as a whole are unlikely to match the 2016 total.
Consumption spending on services, on the other hand, was quite weak. Health care is a large part of spending on services in the United States. But with all the uncertainties over health care, we think it is hard to see how this will help boost the economy while the various health-care proposals to “repeal and replace” the Affordable Care Act work their way through Washington.
The other big component of spending on services is housing, and part of the volatility here is linked to utilities, which are heavily influenced by the weather. More importantly, there are some signs that U.S. housing markets are beginning to cool. In part, this may be driven by mortgage rates, which have backed up a bit. And it may also be because the share of first-time buyers of housing has dropped, which may reflect some weakness in household formation.
A labor conundrum
Recent labor market data have been good in parts. Headline job creation in January was strong, but wage growth was weak. A recurring oddity is that when the survey week is early in the month, wage growth looks poor; when it’s late in the month, it looks stronger. Hours worked went up, but so did the unemployment rate. From a medium-term perspective, not much is new here. The U.S. labor market is improving steadily, and we don’t think it is inflected up or down.
However, when we look more closely at the interplay of unemployment data and wage growth, we see nothing so much as the abnormality of our current cycle. The accompanying chart illustrates this well. In past cycles, when unemployment fell, wages generally went up. In the current cycle, unemployment has fallen, but wages have remained stubbornly low.
There are many factors that might help explain this, including low labor participation rates, the impact of globalization, and the speed of technological change. But none of these factors is a smoking gun, in our view. In the meantime, we would point out that if wages should suddenly spike, that could push the Fed toward a more aggressive policy path.