On this page is an income adjusted by workforce calculator which allows you to adjust average incomes for the workforce age breakdown of past years. If you enter a year, the tool will take the weights per age of the United States workforce in that year and apply it to every other year, graphing the result.
When you are satisfied with the result, you can click “Download Current to CSV” and the tool will allow you to download a comma separated file containing your results.
Methodology and Source for the Income Adjusted by Workforce Calculator
Miriam King, Steven Ruggles, J. Trent Alexander, Sarah Flood, Katie Genadek, Matthew B. Schroeder, Brandon Trampe, and Rebecca Vick. Integrated Public Use Microdata Series, Current Population Survey: Version 3.0. [Machine-readable database]. Minneapolis: University of Minnesota, 2010.
The methodology is detailed in… detail at our article on importing this data into R to export it to a spreadsheet. There are no tricks here, only a single adjustment: when a year doesn’t have an age in the survey (for example, if a year’s survey data doesn’t have any 88 year olds), we take the weight of 88 year olds in the original data and divide it equally for every other age. This is barely a rounding error since all of the largest age groups in the workforce are represented, but to reproduce it exactly please follow the same step.
Here are some brief Q&As:
Why Average Income and Not Median?
If you have a way to adjust demographic sizes for medians, please let us know. As it is now, there isn’t a reversible way to add/subtract data points to get median numbers – but in a sufficient sample it’s safer to assume average stays about the same.
Does this Adjust for Male and Female Proportions in the Workforce?
No, this data is only adjusting for age. Age is a good proxy for experience (see our previous articles), so it goes a long way to making this data more sanitized.
How Do You Determine Who in the Workforce?
What Do You Mean By ‘Year’?
Surveys are conducted in March, so the data is for the previous year: ie, ‘2015’ means data from January to December of 2014. March is convenient because most folks have tax returns in hand and it is easier to get better numbers than in other months.
Adjusting Statistics for Changing Demographics is the Right Thing to Do
We like to talk about statistical aberrations like Simpson’s Paradox, and if you’re not careful you might walk into similar traps. See Calculated Risk for some excellent takes along similar lines, including this very detailed post on demographics and labor force participation rates from earlier this year.
There are a lot of scenarios other than the health of the economy that can drive changes in income numbers – and a lot of them aren’t accounted for in this post. One, as we first say applied to gini coefficients by Political Calculations is the need to adjust household income numbers for changing household sizes(!). That isn’t directly applicable to the individual income numbers of this post, but other variables which might this more useful are: gender, educational status, the types of jobs available, and countless others. (Of course, as you add more and more variables each slice gets smaller and smaller – and the conclusions we can draw are less precise.)
That said, it’s interesting to look at dates in the 70s – when there seems to be a bit of a demographic headwind as there were many young, inexperienced Boomers in the workforce. That cleared up through the 80s and 90s, and today there are many older, more experienced, highly earning Boomers in the workforce. As Boomers retire and more Millennials dominate the workforce, we might see headwinds on these numbers in the coming years – but please check your income expectations against the demographics like we do in our income brackets by age calculators.
Anything else interesting you see in the income adjusted by workforce calculator?