By Paul Gomme and Peter Rupert
The BLS announced that payroll employment increased 216,000 between November and December, the private sector contributed 164,000 to the total. Private service producing jobs accounted for 142,000 of the total, despite a 33,300 decline in temp workers and a 22,600 decline in transportation and warehousing jobs. October employment was revised down 45,000 and November down 26,000.
Average weekly hours fell from 34.4 to 34.3 and with the 142,000 increase in private sector employment, total hours of work fell 2.0%.
Average hourly earnings increased 0.4%, from $34.12 to $34.27. Earnings growth has been outpacing price growth over the past few months with the big declines in inflation even though wage growth has slowed somewhat.
A different way to visualize what’s happening to hourly earnings relative to prices is to plot both in logarithms. For current purposes, there are two useful facts regarding logarithms. First, a straight line represents a constant growth rate. In other words, if the Fed hit its target for core PCE inflation, then when core PCE is plotted in logarithms, we would see a straight line. As shown in the chart below, core PCE inflation may have been constant between 2011 and 2015, and again from 2016 to 2020. The second useful property of logarithms is that a constant gap between two lines means that they are growing at the same rate. Of particular interest is comparing the gap between average hourly earnings and a price level since this gap represents real average hourly earnings (that is, after accounting for changes in the price level). In the chart below, the gaps between average hourly earnings and either the CPI or PCE were roughly constant from 2011 to 2013 which means that earnings grew at about the same rate as prices. Perhaps more interesting is these gaps are changing. By way of example, from 2015 to just prior to the pandemic, the gaps between earnings and the price levels increased meaning that real earnings (the quantity of goods and services that can be purchased) increased. During the pandemic-induced recession, real earnings blipped up as nominal earnings rose and prices fell. However, starting in mid-2020, prices grew faster than earnings: the gap between the two shrank. Then, starting in 2022, the gaps between earnings and prices started opening up again. In other words, despite all the concern about inflation, real earnings grew. Indeed, as measured against the CPI, real earnings in late 2023 were close to what they were just prior to the pandemic; against the PCE, real earnings are higher.
The household survey paints a much different picture of the December employment situation. The labor force participation rate fell from 62.8 to 62.5. Employment declined by 683,000 leading to a 0.3 percentage point decline in the employment to population ratio. The number of unemployed rose 6,000 and the unemployment rate rose slightly from 3.72% to 3.74%.
There are well known trends in male and female labor force participation, as well as by race. The chart below digs deeper into the overall labor force participation rate by looking at finer groupings. At the start of the pandemic, labor force participation fell across all groups. Looking since mid-2020, participation by white men has leveled off at a lower level; for white women, it’s hard to tell whether there has been a downward level shift. The participation rate for Hispanic men has trended down since 2007. More recently, their participation rate seems to be leveling off, with no obvious level shift. On the other hand, the participation rates for Hispanic women, Black women and Black men are all back to their pre-COVID levels.
Data for JOLTS (Job Openings and Labor Turnover Survey) were released on January 3. JOLTS gives information concerning worker flows in and out of jobs. As shown below, since 2021 the layoff rate has been fairly stable; since 2022, the rates for openings, hires and quits have trended down. Indeed, the hiring and quit rates are close to their pre-pandemic values.
JOLTS data is also useful because it allows us to plot the Beveridge curve, the relationship between vacancies and unemployment. Indeed, the Beveridge curve is central to Diamond-Mortensen-Pissarides search model of unemployment. In essence the Beveridge Curve is behind the notion of a “tight” labor market: high vacancy rates and low unemployment tend to lead to a growing, or healthy economy. In the figure below, we have color-coded points corresponding to: the Great Recession, pre-Great Recession, post-Great Recession, and since the onset of the COVID pandemic. Visually, the pre-Great Recession and Great Recession periods lie on a relatively stable Beveridge Curve. However, there is a shift in the Beveridge Curve after the Great Recession, and the pandemic period does not fit either of the previous periods. However, over the past couple of years, the unemployment rate has remained fairly stable while the vacancy rate has fallen (see also above). It’s hard to tell where the US labor market will end up, but recent observations are getting close to the post-Great Recession Beveridge curve.
Overall, the reports show a continued strong labor market. The latest reports shouldn’t change Fed calculus.
