Back to Blog

The Problem With Average Hourly Earnings

Sep 13, 2021

Executive Summary

  • Average hourly earnings is a poor measure of the labor market and inflation.

  • Average hourly earnings frequently rise during recessions because hours worked (the denominator) declines.

  • You should look at aggregate income, real aggregate income, or real disposable income per capita for a better understanding of wage dynamics in the economy.

  • The most important income metric is real disposable personal income per capita and it is on the verge of falling below trend.

The Problem With Average Hourly Earnings

Average hourly earnings "AHE" is the most commonly used measure for "wages" in the economy and is often cited as a reason for the Federal Reserve to tighten or ease monetary policy. 

The Bureau of Labor Statistics "BLS" publishes the popular measure of AHE in the employment situation report on the first Friday of each month. 

Despite the intense focus, the average hourly earnings measure is one of the worst metrics to use to gauge wage pressure, labor market conditions, or inflation. 

In this article, you'll see the problems with using average hourly earnings as an income measure and also better metrics to use when looking to assess economic conditions through private sector income. 

The Problem With Average Hourly Earnings

Average hourly earnings "AHE" is calculated as a ratio of average weekly earnings to average weekly hours worked. 

Given that we are working with two variables, weekly earnings and weekly hours, the relative changes impact the final result. 

For example, if average weekly earnings in the economy equal $100 and average weekly hours equal 10, then average hourly earnings are $10. 

If weekly earnings drop to $90 and average weekly hours also drop to 8, then average hourly earnings would increase to $11.25. 

When the employment situation report is released, the only focus is on average hourly earnings, not the relationship between the two variables. 

Should this increase in average hourly earnings be a reason for the Federal Reserve to tighten monetary policy, even if weekly earnings declined? 

Average Hourly Earnings: 

Source: BLS, FRED

As of result of weekly hours declining faster than weekly wages, the year-over-year change in average hourly earnings increases during recessions. 

The chart below shows a major spike higher in average hourly earnings growth in the middle of an economic crisis because hours worked fell faster than weekly earnings. 

Average Hourly Earnings: Year over Year 

Source: BLS, FRED

In 2008, a similar situation unfolded. Average hourly earnings growth rose in the middle of the recession, hardly the time for the Federal Reserve to raise rates, and then declined through the first four years of the recovery. 

Average Hourly Earnings: Year over Year 

Source: BLS, FRED

Today, despite the weakness of this metric on full display, investors are still citing "strong wages" as a reason for the Federal Reserve to tighten monetary policy. 

There is another failure of this metric when used as the sole measure of wages. The measure of average hourly earnings does not account for the number of employees. 

For example, if there are 100 workers in the economy, all earning $100 in weekly earnings, aggregate income is $10,000 per week. 

If the bottom 50% of wage earners lose their jobs, the economy may only have 50 workers but $125 in average weekly earnings. This is a 25% increase in average weekly earnings because the lower-wage employees fell out of the average. The economy would have "25% weekly wage growth," but aggregate weekly income would fall to $6,250. Should the Federal Reserve raise interest rates based on average hourly earnings or aggregate earnings? 

These are clearly extreme examples used to illustrate that it is inappropriate to use average hourly earnings or average hourly earnings growth as a sole measure of wages in the economy or as a catalyst for monetary policy changes. 

Charted below is the graph of aggregate weekly earnings or total employees multiplied by average weekly earnings. 

Aggregate Weekly Earnings: 

Source: BLS, FRED

When measuring total income through the lens of aggregate weekly income, we see a different story than the one being told of skyrocketing wages, inflation, and labor shortages. 

Aggregate income has surpassed the peak in aggregate income from the start of 2020, but almost 20 months have passed, so we have to account for continued growth. 

If we measure the current level of aggregate weekly income against the trendline from 2009-2020, we can see that aggregate income today has essentially returned to the pre-pandemic trendline. 

The composition of aggregate weekly earnings has simply changed. 

We have a higher average wage with fewer employees needed. 

Secondly, the composition between nominal and real wages has also changed. 

While nominal aggregate income is back to the pre-pandemic trendline, if we deflate aggregate weekly earnings by the CPI, we can see that most of the increase was from inflation, not real income. 

"Real" Aggregate Weekly Earnings: 

Source: BLS, FRED

Real aggregate weekly income is falling meaningfully below the pre-pandemic trendline, which means that consumers will start to feel a real income burden and reduce consumption unless offset by increased transfer payments. 

Below we'll take a look at better income metrics, some including government transfer payments, for a more comprehensive look at the income side of the economy.

Average hourly earnings are a valid metric when outlined in context weekly hours, the number of total employees, and the rate of inflation. 

Better Income Metrics

The Bureau of Economic Analysis "BEA" publishes an entire report on Personal Income and Outlays at the end of each month, but usually, only the consumption figures are analyzed. 

Despite the name of the report, investors use the employment report for wages and only focus on the consumption portion of the personal income and outlay report. 

In the BEA report, aggregate wages and salary disbursements are measured. The series is reported in nominal terms or inclusive of inflation. 

Compensation of Employees: 

Source: BEA, FRED

In the chart above, we can see that aggregate wages have returned to roughly the pre-COVID trend line in nominal terms. 

In real terms, however, aggregate wages and salaries growth has downshifted relative to the pre-COVID trendline.

"Real" Compensation of Employees: 

Source: BEA, FRED

There are many economic forces at work. 

The economy is generating nominal aggregate income at roughly the pre-COVID trend with millions of fewer workers. 

However, most of the income gains are eaten by rising inflation, and the result is real income growth that is materially below the pre-COVID trendline. 

Real disposable personal income per capita is the most informative income metric as it pertains to the overall standard of living, measured via income. 

Real disposable income includes all government transfer payments, including social security, unemployment benefits, and other sources of income but accounts for inflation and taxes.

Essentially, what is the average level of income left for each person after taxes and inflation?

Real Disposable Personal Income Per Capita: 

Source: BEA, FRED

While real aggregate income has fallen materially below the pre-COVID trendline, real disposable personal income is more erratic due to the successive rounds of stimulus payments during the pandemic. 

The large and timely transfer payments may have been effective at reducing the deflationary impact of the economic shutdowns, but the economy is now faced with a major income cliff, something that is not revealed when only considering the growth in average hourly earnings. 

The Income Cliff

From the trough in the economy through the peak just before the COVID recession, real disposable personal income per capita increased at a trend rate of 1.8%. 

Sub 2% real income growth was a weak trend relative to history, and sub-normal income growth led to a decade of sub-normal real economic growth.

After three successive bursts of stimulus clearly identified in the chart below, real disposable personal income is back to nearly the exact same trend line. 

Real Disposable Personal Income Per Capita vs. 2009-2019 Trend: 

Source: BEA, EPB Macro Research

Now with government transfer payments set to decline moving forward, the aggregate income picture will be more impacted by wages and salaries growth. 

Despite the supercharged incomes, the economy still has a major real GDP gap relative to the pre-COVID trend. 

Real GDP Gap: 

Source: BEA, EPB Macro Research

A large and persistent real GDP gap implies that the economy has excess capacity, likely in the form of labor, that will keep inflation suppressed after the current transient spike begins to fade through the end of this year.


Analyzing the labor market, wage growth, and the impact these metrics have on monetary policy requires more thorough analysis than the measure of average hourly earnings is capable of providing. 

While many Wall Street analysts are suggesting last months' wage figures are a reason for the Fed to taper their balance sheet, the reality is that the economy is on the cusp of a major income cliff. 

There have been major shifts in the economy after the COVID recession. Average hourly earnings may have increased permanently as lower-wage employees lost jobs. 

The economy is operating the same pre-COVID trendline with millions of fewer workers. 

Nominal income is keeping pace with the pre-COVID trendline, but most of the increase is coming from inflation. 

Real incomes are suffering, and real GDP growth will remain below the pre-COVID trendline, particularly as the income cliff is realized in the economy. 

When looking to analyze wages and income in the economy, consider aggregate income or measures like real disposable income per capita rather than average hourly earnings.



Join The Free EPB Macro Research Blog

New research, content, and special offers delivered straight to your inbox. 

We hate SPAM. We will never sell your information, for any reason.