Five Decades of Middle Class Wages: March 2016 Update
March 17, 2016
by Jill Mislinski
Here's a perspective on personal income for production and nonsupervisory private employees going back five decades.
The Bureau of Labor Statistics has been collecting data on this workforce cohort since 1964. The government numbers provide some excellent insights on the income history of what we might think of as the private middle class wage earner.
The first snapshot shows the growth of average hourly earnings. The nominal data exhibits a relatively smooth upward trend.
There are, however, two critical pieces of information that dramatically alter the nominal series: The average hours per week and inflation.
The average hours per week has trended in quite a different direction, from around 39 hours per week in the mid-1960s to a low of 33 hours at the end of the last recession. The post-recession recovery has seen a disappointingly trivial 0.7 bounce (that's 42 minutes).
What about inflation? The next chart adjusts hourly earnings to the purchasing power of today's dollar. We've used the familiar Consumer Price Index for Urban Consumers (usually abbreviated as the CPI) for the adjustment with a linear extrapolation for the latest month. Theoretically, the CPI is designed to reflect the cost-of-living for metropolitan-area households. We use a linear extrapolation for the latest monthly inflation data until the mid-month release of the official CPI data.
Now let's multiply the real average hourly earnings by the average hours per week. We thus get a hypothetical number for average weekly earnings of this middle-class cohort, currently at $718 -- well below its $830 peak back in the early 1970s.
Note that this is a gross income number that doesn't include any tax withholding or other deductions. Disposable income would be noticeably lower.
Latest Hypothetical Annual Earnings: $35,924, Down 13.2% from 44 Years Ago
If we multiply the hypothetical weekly earnings by 50, we get an annual figure of $35,924. That's a 13.2% decline from the similarly calculated real peak in October 1972. In the charts above, we've highlighted the presidencies during this timeframe. Our purpose is not necessarily to suggest political responsibility, but rather to offer some food for thought. We will point out that the so-called supply-side economics popularized during the Reagan administration (aka "trickle-down" economics), wasn't very friendly to production and nonsupervisory employees.
Footnote: Here is a slightly different look at the data. We've adjusted using the less familiar Consumer Price Index for Urban Wage Earners and Clerical Workers, the CPI-W, which is theoretically a better match for the production and nonsupervisory cohort. The index, among other things, assigns a slightly higher weighting for gasoline (e.g., longer drives to work and the grocery store). Also, this is the series the government uses to calculate Social Security Cost of Living Adjustments (COLAs).
Here is the real hourly history with this deflator.
Here is the real hourly data multiplied by the average weekly hours. The latest data point is 10.3% below the 1972 peak.
For additional perspectives on earnings, see our commentaries on household income.