On Some Really, Really Misleading Income Inequality Statistics

Click here to read the original Cautious Optimism Facebook post with comments

The Cautious Optimism Correspondent for Economic Affairs and Other Egghead Stuff makes a hefty correction to some really misleading income inequality statistics.

Click photo to see detailed household income statistics

(Thanks to Russ Roberts of George Mason University's Mercatus Center for bringing these statistical adjustments to my attention)

2 MIN READ - Missing from the seemingly endless headlines bemoaning lopsided income inequality between the top quintile of American earners and the bottom quintile are, well, comparisons of actual incomes.

How can that be? After all the Census Bureau has reported that in 2016 the highest fifth of U.S. earners collected an average annual income of $213,941 versus a mere $12,942 for the bottom fifth. 

Hence politicians and the media have loudly lamented that the top 20% “earns seventeen times more” than the bottom 20%. Shouldn't we be alarmed?

Nope. Because they're not comparing actual incomes.

As they say statistics can be framed just about any way to tell just about any story. Problem is if it weren’t for a few critically missing mathematical adjustments we might actually believe the hype that income inequality is really that bad.

It’s not. A conveniently missing qualification from the press headlines is that their reported income numbers reflect household income, not individual income. And an even more critical omission is the average number of workers per household.

So 2016’s top income quintile households contained a mean of 2.04 workers (raw data in attached table). It's 2.04 because most households in the top 20% are married couples in their prime earning years, both working sometimes with a teenage child here or there earning a little extra income which pushes the household mean slightly above two workers. Those statistics are plainly obvious in the table data under “Marital Status.”

2016’s bottom quintile households contained a mean of 0.43 workers.

Yes, 0.43 workers per household.

Why less than one worker? Because the bottom quintile is overrepresented by single-parent households, the very young and the very old who are often retirees—also in the data under “Age of Householders.”

So what conclusion can be drawn or more precisely redrawn from these statistics? Well when dividing household incomes by the actual number of people earning an income, the average salary drawn for workers (workers, not households but actual workers) in the top and bottom household quintiles narrows to $104,873 vs $30,098.

That’s not seventeen times more. That’s 3.5 times more. Not quite the herculean gap the press and certain politicians dunk our ears in daily.

Furthermore, the share of top quintile households with no one working at all (zero workers) is 3.8%. In the lowest quintile it’s 62.6%.

Finally workers in the top quintile are 4.4 times more likely to have a college degree than workers in the bottom quintile. And stunningly, workers in the bottom quintile are 12 times more likely to have no high school degree than workers in the top.

So now a new, very different, and much more informed question has to be asked: Is it unfair for largely college educated workers in their prime earning years to make 3.5 times more than high school dropouts who are disproportionately very young or very old?

Of course some on the American Left won’t be satisfied until everyone makes exactly the same income regardless of education, skill, and work ethic. But is the American dream really undermined by divisive income inequality because a nurse, software programmer, or small business owner makes 3.5 times more than say, a fast food worker, custodian, or retiree?

Don’t expect to find the answer—or even the question—in the New York Times, on CNN, or from Alexandria Ocasio-Cortez.