What to watch on jobs day: Can wage growth normalize without substantially higher unemployment?

On Friday, the Bureau of Labor Statistics (BLS) will release its monthly report on the state of the labor market. In addition to top-line payroll employment growth and changes in labor force participation, probably the most anticipated measure is the pace of nominal wage growth.

Even with the recent contraction in gross domestic product (GDP), the labor market has been expanding at a steady rate and wage growth continues to fall short of inflation. Despite this, many remain worried that abnormally high nominal wage growth (relative to pre-pandemic) will prevent inflation from returning to more-normal levels in the year ahead. In this jobs day preview post, we take a closer look at wage growth using several different measures to gauge just how worried we should be that wage growth will not normalize in the coming year without aggressive policy measures that cause collateral damage (like higher unemployment) in the labor market. We find that most of these measures show decelerating wage growth in very recent quarters.

In Figure A, we report quarterly changes in wages (expressed as an annualized rate) using five measures: average hourly earnings for all workers from the Current Employment Statistics (CES); average hourly earnings for production and nonsupervisory workers from the CES; private wages and salaries from the employment cost index (ECI); private wages and salaries, excluding incentive paid, from the ECI; and private wages and salaries from the National Income and Product Accounts divided by a measure of aggregate hours that we construct. We average over quarters for CES measures, which are generally reported monthly. The data note at the end of this post provides more details on these series.

All series except the ECI ones show pronounced spikes in wage growth in 2020. These are the result of the well-known “composition effect” driven by job loss that was extremely skewed towards lower-wage workers. The ECI series are “fixed-weight” series that are not affected by compositional changes. In recent quarters, the effect of compositional changes in raising and then slowing wage growth seems well behind us. More importantly, four of the five series are showing wage growth that has not only plateaued but is decelerating recently, even as the unemployment rate remains quite low. This is obviously not dispositive evidence that wage growth will completely normalize to pre-pandemic trends, but it is comforting to see deceleration even as non-wage price drivers are keeping inflation very high and unemployment remains very low.

Figure A

Quarterly wage growth shows no sign of accelerating in the first half of 2022: Annualized quarterly wage changes, alternative measures, 2018–2022

quarterly changes AHE, total private AHE, prod/nonsuper ECI (private wages and salaries) ECI (private, excluding incentive paid) NIPA Wages and salaries/private-sector aggregate hours
2018Q2 2.8% 2.8% 2.8% 2.8% 1.8%
2018Q3 3.7% 3.3% 3.7% 2.7% 4.7%
2018Q4 3.6% 3.9% 2.7% 2.1% 1.3%
2019Q1 3.7% 4.0% 2.7% 3.6% 9.6%
2019Q2 2.3% 2.9% 3.0% 3.3% 2.1%
2019Q3 3.6% 3.9% 3.6% 2.7% -0.2%
2019Q4 3.1% 3.2% 2.6% 2.3% 5.3%
2020Q1 4.0% 4.2% 4.1% 4.4% 5.9%
2020Q2 16.3% 16.2% 1.4% 1.4% 24.8%
2020Q3 -3.0% -3.1% 2.3% 2.6% -3.8%
2020Q4 3.2% 3.4% 3.4% 2.6% 10.6%
2021Q1 4.2% 4.6% 4.8% 4.3% 0.0%
2021Q2 4.7% 6.4% 3.6% 3.9% 9.0%
2021Q3 5.7% 7.2% 6.5% 5.0% 8.2%
2021Q4 6.1% 7.3% 4.7% 5.3% 10.0%
2022Q1 5.2% 6.0% 5.2% 6.6% 7.7%
2022Q2 4.3% 5.4% 6.5% 5.4% 5.9%
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: NIPA wages and salaries are adjusted to hourly wages and salaries by dividing by CES aggregate hours (including imputed hours for agricultural and self-employed workers from the CPS).

Source: EPI analysis of Bureau of Labor Statistics' (BLS) Current Employment Series and Current Population Survey and Bureau of Economic Analysis' (BEA) National Income and Product Accounts.

Copy the code below to embed this chart on your website.

The data in this chart are more recent than most conventional reporting on wage trends, which often reports year-over-year measures of wage growth. But, comparing wage growth between, say, June 2021 and June 2022 makes it easy to miss key turning points. If wage growth was extremely rapid between June 2021 and December 2021 but then moderated significantly, year-over-year measures will generally just show very fast wage growth. Given that the Federal Reserve is trying to look ahead of the curve on determinants of inflation (like wage growth), more recent measures picking up current trends are valuable.

It is this desire to see more recent wage trends that leads us to think one common measure of wage growth—the Atlanta Federal Reserve Bank’s wage growth tracker (AWGT)—is too backward looking to shed much light on current policy debates. The AWGT looks at wage changes for specific individuals over a one-year span. Then, it takes a 3-month moving average of these year-over-year changes for its main reading. But this means that in the June 2022 data point in the AWGT, just under 60% of the monthly wage growth data that contributes to the final number happened in 2021.

In the coming weeks, we’ll keep looking for other wage growth measures that can be tracked on a timely basis. For now, most of the measures we’re tracking seem to lend some optimism to the view that wage growth can normalize without a large rise in unemployment.

Data Note

The two CES series are presented as reported by the BLS on an average quarterly basis. The first ECI measure is changes in the headline private-sector wage and salaries series. The second is a measure designed by the BLS that controls for wage volatility by excluding incentive paid workers. In the BLS’s own description, this series without incentive pay is similar to “core” measures of price inflation that strip out highly volatile prices (food and energy), potentially providing a better measure of underlying trends in wages.

The last series provide a measure of average wages economy-wide by using private-sector wages and salaries from the BEA NIPA data and dividing by a constructed measure of economy-wide hours. We construct this hours measure by adjusting the measure of aggregate weekly hours of private sectors provided monthly by the BLS to account for self-employed and agricultural workers. We make this adjustment by multiplying the aggregate weekly hours measure by a ratio of the sum of: private-sector employment plus self-employment plus agricultural employment divided by private-sector employment. This adjusts the hours measure to account for workforces not included directly but does assume that hours growth in the sectors not included mirror changes in the private sector. For aggregate weekly hours of non-farm private workers, and non-farm private employment, we use data from the Current Employment Statistics. For self-employed and agricultural workers, we use the Current Population Survey (CPS)—as these groups are not tracked in the CES.