Data in Policy Debate: The Shrinking Middle Class

Back on December 9th Pew Research released a new report on the status of the Middle Class in the United States, something that's a pretty big topic in the US presidential debate. 

The headline: "The American Middle Class is Losing Ground." Sounds pretty grim.

We noticed something curious on that day in our news feeds as a whole lot of different news sources syndicated the report. As is sometimes the case, we saw a few contradictory-sounding headlines:

The Sacramento Bee: "US middle-class families are no longer in the majority."

Daily Caller: "New Report Shows the Middle Class is Collapsing, But for the Best Possible Reason."

Washington Examiner: "Middle 'CRASH.'"

National Review: "Being a Member of the 'Hollowed Out' Middle Class Never Felt So Good."

And just to add to the confusion, a Pew article from February: "America's 'middle' holds its ground after the Great Recession.

As is often the case, we see that we can often take data and twist it to say whatever we want, or (more innocently) whatever our brains first process it as when we apply our own biased filters. 

So as a classic exercise, let's work through what's really going on.

Getting our Definitions Straight

As we know by now, we have to be very careful of vague definitions. First, how do we define "middle class?" In this case, it can't be a matter of "the middle 50%" or some similar measure (as some are), because its size wouldn't change. The fact that we have headlines like "middle class no longer in the majority" exist means we have a size-flexible definition here.

Some definitions could define the middle class as having a certain standard of living--perhaps "can buy A, B, C for a family of 4, but not X, Y, Z."

Others could tie the definition to some mathematical average, like +/- 25% from the median or mean income.

Pew's definition of the middle class is those who are between two-thirds of the median income and double the median income. 

Already we can see this getting somewhat complicated. To get an intuitive sense of Pew's definition requires being able to visualize some statistics. But we must also note that it's not an authoritative definition. Different definitions of the "middle class" will show different trends over time. 

Digging Under the Headline

So the middle class, by Pew's definition, is shrinking. Let's see what that means. 

So we see that over the past 44 years, 11% fewer households are in the middle class. Of those, 7% have moved into the upper class, and 4% have moved into the lower class. 

Worth considering whether this is a "good thing" or not. Would it be better if the 1971 distribution stayed the same? Consider what might make you answer one way or other other.

What Does This Mean for US Families?

Without other data, it doesn't mean a whole lot. It could be the case that the average national income is the same, but it's just shifted. It could be the case that everyone's making a whole lot more money, but the distribution has just changed a bit.

Because this controls for inflation ("in 2014 dollars"), we're seeing a significant real rise in income for everyone. 

The lower class has gained about 26% income, the middle class has gained about 34%, and the upper class  has gained 47%. 

So those facts together give us some very rough idea of what's going on.

Given all this: if you were to write a headline for this study, what would it be? Let us know in comments.


Erik Fogg

We do politics, but we don't do the thinking for you.