Raphael Deketele, a student in my fourth year honours seminar, just unearthed a strange finding from the 2006 World Values Survey (WVS).
The WVS interviewed over 1000 Americans, and asked them:
Here is a scale of incomes on which 1 indicates the “lowest income decile” and 10 the “highest income decile” in your country. We would like to know in what group your household is. Please, specify the appropriate number, counting all wages, salaries, pensions and other incomes that come in.
Each income decile, by definition, contains ten percent of the population. In a randomly selected group of people, about ten percent should be in the first decile, ten percent in the second decile, and so on. The distribution of responses to the income decile question should be uniform, that is, flat.
In fact, it looks more like this:
Fewer than ten percent of respondents to the World Values Survey placed themselves in the top 30 percent of the income distribution. Twenty-five percent of respondents placed themselves right in the middle of the distribution, and just under ten percent placed themselves in the bottom twenty percent.
Now it could be that the WVS's sample is not representative of the US population; that the wealthy were under-surveyed and the middle class over-represented. One way of checking the representativeness of the WVS is to look at more objective indicators of socio-economic status, for example, income or education. The World Values Survey doesn't seem to be particularly biased against higher income earners: 25 percent of respondents have a bachelor's degree or higher; significant percentages work in professional or managerial positions.
I suspect people just live in their own little worlds. A typical respondent, when answering this question, probably reasons along these lines: "I'm not as well off as the people across the street with the slightly nicer car and slightly better job, but I'm slightly better off than the people down the road who seem to be struggling a bit financially, so I guess that puts me around the middle of the income distribution."
Such mistaken beliefs could have real impacts of public policy. If few people perceive themselves as being poor, then they will see programs to help the needy as benefiting others, not themselves. If taking action against the deficit requires sacrifice from those who are most able to pay, people who (incorrectly) perceive themselves as only average will refuse to chip in.
Still, the WVS results explain why politicans court the middle class: whether they're rich or whether they're poor, that's how most people perceive themselves.
Update: in the Canadian World Values Survey, the question was phrased slightly differently. People were asked:
Here is a scale of incomes [e.g. up to 12,500, 12,501 to 20,000,...]. We would like to know in what group your household is, counting all wages, pensions and other incomes that come in. Just give the letter of the group your household falls into, before taxes and other deductions.
To put these numbers into perspective, according to Cansim numbers in 2006 an household income of $101,000 (2010$) would put a two person household right in the middle of the fourth income quintile. The corresponding figure for a single person household would be $42,200.
The World Values Survey also asked respondents to identify, subjectively, their social class. I've reproduced the 2006 Canada and US results here, but they're pretty unsurprising.