Do Societies Differ in Their Degree of Self-Deception?

The case of self-inflation and income inequality

Posted Apr 11, 2012

I am often asked if societies differ in their degree of self-deception and I usually answer that, by logic, they are expected to differ but there is not much systematic evidence. We know that on average across the globe people tend toward self-enhancement. If asked whether they are in the top half of a desirable category (good looks, brains, leadership ability or whatnot) more than 50% reliably say they are. That this is not just our mouths “flabbing” but represents something deeper was shown beautifully for good looks in experiments run by Epley and Whitchurch (see p16 of my book, the Folly of Fools). Your photo is morphed to be better-looking or uglier than your true photo, and you are asked to pick your photo, as quickly as possible, out of a set of 12 matched for your age and sex. People spot 20% better-looking self 5% quicker than they spot true self, which they spot, in turn, 5% quicker than 20% uglier self. 

Using the simpler device of asking people to rate themselves as being in the top half of a desirable distribution some cross-cultural variation has been detected. Specifically Japanese—and some nearby Asian people, such as Korean and Chinese—appear to practice self-enhancement less than do those in the U.S. and Western countries more generally. Indeed, it has been suggested that they may compete to prove their modesty, i.e. tendency not to publicly self-inflate on other variables but possibly on degree of modesty. This is said to result from a more collectivist—instead of individualistic—ethos. An interesting difference is that people in far Eastern societies are also more likely to rate friend better than self (both of whom are inflated) while in the U.S. it is self over friend (again, both inflated).

Now comes dramatic evidence across 15 societies, including Japan, China, Germany, Hungary, Venezuela, the U.S., and South Africa. When Loughnan (and 18 co-authors) measured degree of self-inflation by a simple verbal self-assessment, they found a strong positive correlation between average levels of self-inflation in a country and degrees of income inequality, measured by the standard (U.N.-computed) economic Gini coefficient. Japan and Germany have highly equitable distributions of income and very low levels of self-inflation. The U.S. is intermediate, while Peru, South Africa and Venezuela show both highly inequitable distributions of income and high self-inflation. 

The authors correctly warn that the data for all countries (except the U.S.) came from college students and these are both a sub-population expected to be on the higher end of society already and variably so. That is, in highly unequal societies, those in college may be especially high in the system and prone to see themselves even more so. To attempt to correct for this bias, scientists asked students to compare themselves to other students. Nevertheless, whether people below college level are practicing self-deception differentially according to degree of income inequality remains unknown. This study presents tantalizing evidence that there is a real connection across societies between income inequality and self-inflation. Especially noteworthy in this regard is that the authors also included in their analysis data on the degree of individualism vs collectivism in their societies and only income inequality had a statistically significant effect, degree of individualism disappeared as a predictive factor.

Cause and effect can go in either direction, of course, but it is harder to visualize how self-inflation across a society will lead to income inequality while it is easy to imagine that when resources are already evenly distributed, there will be less pay- off to self-inflation, while inequality may foster more competition for resources and more self-inflation in hopes (often unconscious) of prevailing in the competition. This is consistent with data from Japan that when students are made to play a competitive economic game, their degree of self-inflation goes up. It would be very interesting to know if self-inflation has grown in the U.S. over the past 20 years as income inequality has, or to redo the work 20 years from now if Gini coefficients have changed in various countries in the meantime.

Epley, N. and Whitchurch, E. 2008. Mirror, mirror on the wall: enhancement in self-recognition. Pers Soc Psychol Bull 20: 1-12. [??]

Loughnan, S. et al. 2011. Income inequality is linked to biased self-perception. Psychological Science 22: 1254-1258.