Middle of What?
February 14, 2013 11:51 AM   Subscribe

How have sociologists and/or economists historically defined or measured "middle class" in the United States or the developed world generally? I mean over the last 100-200 years.

I see this question but that's not what I'm asking. I don't mean this question as an opinion poll, I mean what objective measures have social scientists used to define "middle class" and how have those measures changed over time? I'm curious if measures have evolved in an analgous way to, for example, the "market basket" of goods chosen to make up the consumer price index. I think I'm looking for a quantitative answer but perhaps a qualitative one in the form of "the ability to sustainably afford xyz list of goods and services" would suffice.
posted by Wretch729 to Society & Culture (6 answers total) 3 users marked this as a favorite
There is definitely no one definition, but a number of fellow economists that I work with usually define classes as:

Upper class - top 20% of income earners;
Middle class - middle 60% of income earners;
Lower class: bottom 20% of income earners.

This allows you to port the definition from country to country and time to time, because it's not really about relative purchasing powers but strictly relative income-based.
posted by Rodrigo Lamaitre at 12:25 PM on February 14, 2013

Rodrigo any idea where on the Census Bureau's labyrinthine website the data to figure those percentages out is? I guess you can estimate it from the mean incomes for each quintile of the population which is fairly easy to find.
posted by Wretch729 at 12:34 PM on February 14, 2013

I am Canadian so unfortunately I dig into the Bureau of Labour Statistics and the Census very irregularly, sorry.
posted by Rodrigo Lamaitre at 12:40 PM on February 14, 2013

There is a bit of this sort of data in terms of what different classes of people could expect to afford in Schnitzler's Century: The Making of Middle-Class Culture 1815-1914, but it's primarily cultural history so it's not too heavy on that kind of quantifiable info. There might be additional sources in the notes or bibliography, though.
posted by scody at 2:58 PM on February 14, 2013

Possible sources of Census data on income levels:

U.S. Census definition of middle class. Follow the site breadcrumbs back to reports on income inequality.

The Statistical Abstract of the United States has a number of charts for family, household, and personal income.

You can also use American FactFinder to search a variety of Census data and reports.
posted by Boxenmacher at 3:11 PM on February 14, 2013

Alternate definitions I've seen used:

+/- 50% of median income. This is the one people typically use when they say the middle class is shrinking, since the middle 60% will always be 60% of the population. This is my pet peeve. Policy should be different if people are making too much money or not enough money to be middle class, yet the statistic treats those two folks are equivalents.

Only using the middle quintile (particularly when quantifying shifts in income). Since the Census provides quintile information, this is easier for quick analysis. Unpacking the data and repacking the middle three quintiles can be too much for a timely news feature.

If you're just putzing around the census data, this is probably what you want. There's no real definition of middle class, and doing the manual work to the census or labor data doesn't make sense for a casual user.

It's also worth finding out if the data set is using pre-tax/benefits or post-tax/benefits numbers. Much of Census data is consolidating wages into income, and doesn't factor in the progressive tax structure and safety net. This mostly warps the lower and upper quintiles, and not the middle class, but it still exists.

When you're looking at the data, make sure you don't conflate income and wage. Many low-wage workers live in a household with decent income, which makes income and wages very poorly correlated. There is also a lot of demographic and life cycle changes that makes it hard to make casual observations. As America gets older, we would want to see median income to decrease as folks retire and rely on their savings. If it didn't, we should worry people aren't secure in their old age.

I enjoy using FRED as a decent jumping off point to find various factoids. It'll let you know when you want to be digging in the BLS rather than the census. I haven't found any excellent non-academic accessible databases that readily break down the data in a way I feel doesn't distort the final result. But I am nitpicky, and just lazy enough I haven't committed to digging through the census microdata.
posted by politikitty at 4:54 PM on February 14, 2013 [1 favorite]

« Older Is it possible to ask a flight attendant to give a...   |   What should my idle phone be showing me? Newer »
This thread is closed to new comments.