Recent Trends in Individual Wage Growth

In a previous blog post, we discussed the recent evolution of wages in the U.S. The measure of wages most widely followed by analysts and policymakers is the average hourly earnings for all employees reported monthly by the Bureau of Labor Statistics. Between September and October 2022, average hourly earnings went up by an annualized rate of 4.4%.

It is important to note that this measure represents the change in the average wage in October relative to the average wage in September. While changes in this measure may reflect changes in workers’ pay, the average wage may also fluctuate due to changes in the composition of workers from one month to the next. For example, if the hourly wage rates for individual workers do not change, but an important number of new workers—each earning a low wage—enter the labor market, the average wage will fall.

An alternative measure of wage changes is to track the changes in the wage of each individual worker from one month to the next (or any other period of interest) and then take an average of individual wage changes. The Federal Reserve Bank of Atlanta constructs measures of individual wage changes over a year using data from the Current Population Survey. In this blog post, we use private high-frequency data to construct a similar measure but at the monthly level and at a higher frequency.

Constructing Individual Wage Changes

We use data from Homebase, a private company providing time-tracking and scheduling tools and payroll services to businesses. The Homebase data include average hourly wages for individual workers over time. There are some differences between the Homebase data and the overall U.S. workforce: Most of the Homebase data are centered in the leisure and hospitality sector, especially in food and drink. Homebase is also heavily weighted toward hourly workers and small establishments. The Atlanta Fed Wage Growth Tracker uses CPS data to measure the percent changes in the hourly wage of individual workers over a year, and reports the median of these changes for each month (smoothed). For our first set of results, we proceeded in a similar way in order to compare the changes in wages in the Homebase data with those in the CPS.

In the Homebase data, we matched a user’s monthly average hourly wage to the wage in the same month in the previous year based on user identification and company identification. We dropped any workers who were missing a wage value, had zero hours worked, or had an average hourly wage of less than $5. Then, we calculated each worker’s percent change in wages over a year and used the median of these changes for each month, smoothed. Since the set of firms and workers in the Homebase sample changed drastically when the COVID-19 pandemic hit in March 2020, we started our analysis with July 2021 data to ensure we had a large enough matched sample to compute year-over-year changes.

The two figures below compare the Atlanta Fed’s Wage Growth Tracker data for hourly workers and for job stayersIn the Atlanta Fed’s Wage Growth Tracker, a job stayer is a worker who remains in the same industry and occupation from a year ago and who has the same employer and job duties in the past three months. with the median percent change in wages in the Homebase data. The evolution of wage changes since July 2021 is similar between the two data sets, albeit the magnitude is smaller in the Homebase data, which may be due to differences in the composition of firms in Homebase relative to the U.S. economy and due to some components of pay, like tips, that are not included in Homebase’s wage measures. As the two figures below show, median wage changes have trended up since mid-2021, peaking in mid-2022 and gradually decreasing since then.

Median Wage Growth of Individual Workers: Hourly Workers

Median Wage Growth of Individual Workers: Job Stayers

SOURCES FOR BOTH FIGURES: Federal Reserve Bank of Atlanta, Homebase and authors’ calculations.

NOTES FOR BOTH FIGURES: Data are three-month moving averages and not seasonally adjusted. A job stayer is a worker who remains in the same industry and occupation from a year ago and who has the same employer and job duties in the past three months.

The Recent Evolution of Wage Inflation

Individual wage changes in the Homebase data set are well correlated with wage changes in the overall economy. The advantage of the Homebase data is that we can compute individual wage changes for different time periods and at a higher frequency than what is possible using the CPS. In particular, we calculated individual wage changes over four weeks for each week of data we have available. In a similar fashion as before, we matched users over a four-week window and calculated their annualized percent change in wages. Most workers in Homebase had no change in their wage over a four-week window, so rather than using the median, we computed the mean percent change in wages for each four-week window.

The third figure below shows the mean of the annualized changes. Since the calculation of the mean may be highly influenced by a few observations with abnormal values, the black line shows the mean of annualized wage changes over a four-week period, but trimming the top and bottom 1% of values from the calculation. The gray dotted line shows the mean for the whole sample.

Average Annualized Wage Change for Individual Workers

SOURCES: Homebase and authors’ calculations.

NOTE: The trimmed data excludes the top and bottom 1% of annualized wage changes.

A few patterns clearly emerge. First, the mean (gray dotted line) and trimmed mean (black line) follow similar patterns, but differ slightly in the magnitude, as expected. Second, wage changes are larger in January and July of each year, which suggests that, for most organizations, wage changes tend to occur at these times. Third, abstracting from the seasonal spikes, wages increased at a faster rate in 2021 relative to the pre-pandemic years. The trimmed mean of wage changes increased from 3.6% in 2019 to 6.2% in 2021, and more recently moderated slightly to 5.8% in 2022.

The next figure shows a different indicator of wage pressures: the share of positive wage changes, as opposed to zero or negative changes. In this way, this measure tracks how often wage changes occur, rather than their magnitude. With tight labor markets, wages tend to change more frequently, and this measure captures this. The share of positive wage changes has increased over time—from close to 6% monthly during 2019 to around 10% in 2022.

Share of Individual Wage Changes That Are Positive

SOURCES: Homebase and authors’ calculations.

NOTE: Data represent the percentage of individual wage changes that were positive relative to all possible changes, which includes positive changes, no changes and negative changes (pay cuts).

We also see that the seasonal spikes are also higher than before. This pattern indicates that in recent months, wages are changing more frequently than before the pandemic, although the frequency in the second half of 2022 is slightly lower than for 2021, which may indicate a gradual return to normal conditions in labor markets.

High-Frequency Data Can Provide Better Understanding of Wage Changes

Understanding current wage growth is important to creating and executing better monetary policy. Using high-frequency data, like those of Homebase, can provide a better and more timely understanding of wage changes and how different policy choices are impacting them.

Note

1. In the Atlanta Fed’s Wage Growth Tracker, a job stayer is a worker who remains in the same industry and occupation from a year ago and who has the same employer and job duties in the past three months.