Neil Hester

All poems © Neil Hester unless otherwritten

Location: North Carolina, United States

Friday, March 22, 2013

Wealth Disparity in the US

Take a moment to think: how is wealth distributed in the US? How much wealth do the poorest 20% of people control? The richest 20%? The richest 1%?

Now answer another question: How would you like for wealth in the US to be distributed (that is, what is your ideal)? Would you keep things the same? Shift some money from the rich to the poor? Set every population quintile at a solid 20% each (you damn Marxist!)?

A couple Fridays ago, I led a discussion with the Texas Tech Honors College to discuss wealth disparity in America. I started by collecting data from all of the people in the room by asking them to fill out this survey.

The idea was to do an in-house, 10-minute replication of a study conducted by Mike Norton and Dan Ariely (2011). Here are the results from our in-house surveying. Perceived distribution first, followed by ideal distribution:

(Note: My numbers this time are a little cleaner than they were during the discussion, since I knocked out a couple people who didn’t follow directions. There were still a couple people who didn’t add properly, though, which throws the numbers just a bit. Also, keep in mind that none of the data collected for this discussion were subjected to significance tests; they are for demonstration only.)

Perceived Distribution of Wealth in the US by Quintile (in percentage), Richest to Poorest

Top 20%:            57.50
2nd 20%:            19.56
Middle 20%:        12.21
4th 20%:             6.74
Bottom 20%:       4.00

Ideal Distribution of Wealth in the US by Quintile (in percentage), Richest to Poorest

Top 20%:            35.44
2nd 20%:            22.22
Middle 20%:        18.31
4th 20%:             13.83
Bottom 20%:       10.75


Y’know, the perceived wealth distribution looks pretty good, actually: sure, the rich people are really rich, but 4% of the wealth in the US is still enough for people to live comfortably. The ideal distribution reflects a desires to lower the wealth of the top 20% and boost everyone else’s wealth.

Even with a super-small sample size (20’ish people, compared to the thousands used in the model study), our data are roughly similar to Norton and Ariely’s data, and certainly reflect the same basic points: people perceive some notable, but not egregious, differences in wealth, and they would prefer a more even distribution of wealth. But, what of the actual distribution of wealth?

Watch this video (it’s worth it, promise!).


Ouch. In case you missed it (or skipped the video—shame on you), here’s the number breakdown of the actual wealth distribution in America (roughly):

Actual Distribution of Wealth in the US by Quintile (in percentage), Richest to Poorest

Top 20%:            84.5
2nd 20%:            10.7
Middle 20%:        4.5
4th 20%:             0.2
Bottom 20%:       0.1


So why the massive inaccuracy in judgment? Our discussion group pinpointed some solid possibilities:

~*We mostly hang out with people in our own socio-economic group, so we don’t have useful points of comparison
~*If you see that someone has a car and a house, just like you do, you don’t think about whether or not they had to take on major debt to afford these things
~*Equality is just a nicer thought!

Furthermore, for perceived and ideal distributions of wealth, Norton and Ariely found no significant differences between conservatives and liberals, poor people and rich people. You might think, Hey, wait a sec! If that’s true, then maybe everyone can agree for a change and work toward new policies!

...Not really. Finding out about this consensus on wealth distribution leads to almost no change in conversatives’ and liberals’ opinions on policy (they agree more on increasing educational opportunities and funding for kiddos, but that’s about it). Liberals want more government redistribution of wealth; conservatives want more opportunities for social mobility (changing your wealth by working hard). So, even if the what is the same between parties, the how stays quite different.


Let’s review the things we’ve learned so far:
  1. Rich people are really rich
  2. Poor people are really poor (some people at the discussion didn’t have a sense of the extent to which some people in the US lack food, shelter, clothes, etc.)
  3. Across political and socio-economic bounds, people generally agree on perceived and ideal wealth distributions 
  4. Most people have massively inaccurate perceptions of wealth distribution
Now, let’s zone in on this last point for a bit. Norton and Ariely asked participants to report their perceived and ideal wealth distributions in percentage values. Erikkson and Simpson (2012) wrote a response to Norton and Ariely, demonstrating that asking for people to report using percentages leads to anchoring effects and exaggerates the extent to which people mis-estimate the actual wealth distribution. To illustrate this point during the discussion, I had half of the people at the discussion use a percent measure and the other half use an average measure, in which they reported how much money (in dollars) the average household in each quintile earned. After converting these dollar values to percentages, I came up with these results (using another 20’ish responses):

Perceived Distribution of Wealth in the US by Quintile (in percentage), Richest to Poorest:

Top 20%:            65.88 (compare to 57.50 with the percent measure)
2nd 20%:            17.58 (compare to 19.56)
Middle 20%:        8.86 (compare to 12.21)
4th 20%:             5.06 (compare to 6.74)
Bottom 20%:       2.62 (compare to 4.00)

Ideal Distribution of Wealth in the US by Quintile (in percentage), Richest to Poorest: Average Measure

Top 20%:            48.05
2nd 20%:            19.67
Middle 20%:        14.13
4th 20%:             10.09
Bottom 20%:       8.05

These numbers are a nice demonstration of Erikkson and Simpson’s argument that using an average measure leads to more significantly different (and seemingly more valid) findings than using a percentage measure, but the basic points of Norton and Ariely seem to remain; people’s reports are just less inaccurate using an average measure.


So, where do we go from here? I’m not really sure, but here are a few questions for you all:
  1. Why else are people’s estimations of wealth distribution so inaccurate?
  2. Are there any good ways to reframe these findings to promote cooperation across political party lines?
  3. The bottom 40% think they possess a greater percentage of the country’s wealth than they actually do: is this a good or a bad thing?
Finally, a quick shout-out to everyone who helped me test my data collection procedure for the discussion—thanks for the help!

Take care,