Egos deflating with the Great Recession: Narcissistic traits rose and fell with the U.S. economy, incresed 1982 to 2008 and then declined; economic growth is linked to more narcissism and individualism.

Egos deflating with the Great Recession: A cross-temporal meta-analysis and within-campus analysis of the Narcissistic Personality Inventory, 1982–2016. Jean M.Twenge  et al. Personality and Individual Differences, Volume 179, September 2021, 110947. https://doi.org/10.1016/j.paid.2021.110947

Highlights

• Narcissistic personality traits increased 1982 to 2008 and then declined.

• Narcissistic traits rose and fell with the U.S. economy.

• Economic growth is linked to more narcissism and individualism.

Abstract: Scholars posit that economically prosperous times should produce higher individualism and narcissism, and economically challenging times lower individualism and narcissism. This creates the possibility that narcissism among U.S. college students, which increased between 1982 and 2009, may have declined after the Great Recession. Updating a cross-temporal meta-analysis of the Narcissistic Personality Inventory to 2013 (k = 164, N = 35,095) and adding two within-campus analyses to 2015 (Study 2: UC Davis, N = 58,287) and 2016 (Study 3: U South Alabama, N = 14,319) revealed a non-monotonic pattern, with increases in NPI scores between 1982 and 2008 and declines thereafter. The decline in NPI scores during and after the recession took narcissism back to their original levels in the 1980s and 1990s. Implications for the interplay between economic conditions and personality traits are discussed.

Keywords: NarcissismNarcissistic personality traitsBirth cohortTime periodRecession

5. General discussion

Across three separate studies, we identified a non-monotonic trend in narcissism scores over time, with scores increasing until the Great Recession and then decreasing during and after it. Consistent with previous research (Stewart & Bernhardt, 2010; Twenge et al., 2008; Twenge and Foster, 2008, Twenge and Foster, 2010, cf. Grijalva et al., 2015), narcissism increased among college students between 1982 and the late 2000s. Then, around the beginning of the Great Recession, narcissism scores began to falter, by 2013–2016 falling to the levels of the 1980s/1990s. This pattern appeared in both a nationwide cross-temporal meta-analysis of college student samples (Study 1) and within-campus analyses of students from University of California, Davis (Study 2) and the University of South Alabama (Study 3). In some analyses, years with higher unemployment and fewer young people participating in the workforce had lower narcissism scores. Thus, the Great Recession may have acted as a reset for the steady rise in narcissism between the 1980s and the 2000s.

These results are consistent with theoretical models that tie narcissism and related constructs (e.g., higher individualism, lower communalism) to economic growth and decline, especially employment (e.g., Bianchi, 2014, Bianchi, 2016; Greenfield, 2009; Park et al., 2014). It is also consistent with models that link higher socioeconomic status to higher narcissism and related variables (e.g., entitlement, antagonism; Piff, 2014; Piff et al., 2012).

Although we have explored economic factors as a potential cause of trends in narcissism, other causes are also possible. For example, narcissism began to decline when the nation elected its first African-American president, Barack Obama, who regularly spoke about the importance of empathy. In addition, the increasing popularity of social media may have played a role. In the years before 2010 when social media was less popular, these sites may have encouraged narcissism as they were an effective way to gain attention and followers (Liu & Baumeister, 2016; McCain & Campbell, 2018). Once social media became used by the vast majority of traditional-age college students after 2012, however, happiness and self-esteem – traits positively correlated with grandiose narcissism in young populations (Sedikides et al., 2004) – began to decline (Twenge et al., 2018), perhaps because social media leads to unflattering upward social comparison (Steers et al., 2014). The possible suppressive effects of social media on narcissism may be one reason why narcissism scores leveled off in Study 3 after 2013 and why economic factors were better predictors in analyses up to 2013 compared to those up to 2016. This suggests that other factors were lowering NPI scores after 2013. Research should continue to explore links between social media, narcissism, and poor psychological well-being.

The time-lag design of this study holds age relatively constant. Thus, age (i.e., being younger versus older) is unlikely to explain the results; not only would age have to differ systematically with year, but it would have to follow the same non-monotonic trend as narcissism to explain the results. However, this design cannot determine whether the shifts are due to cohort effects (which only affect young people) or time period effects (which affect people of all ages). If this is a cohort effect, early Millennials (those born between 1980 and 1988) reached all-time highs for narcissism and remained that way, while late Millennials (those born between 1989 and 1994) returned narcissism to the levels of the late Boomers (those born in the early 1960s) and will remain that way. If this is a time period effect, it would suggest that the narcissism of all generations deflated during and after the Great Recession.

As found in previous research, the change over time in narcissism is moderate at the average (around a third of a standard deviation), similar to many effects in social psychology (Richard et al., 2003). However, the effects are larger at the ends of the distribution. In 1982, about 19% of college students answered the majority of the NPI items in the narcissistic direction; by 2009 this was 30%, a 58% increase (Twenge & Foster, 2010). By 2013, it was back to around 19%, a 37% decrease. These changes are thus large enough to be noticeable, particularly if those scoring 20 or higher on narcissism cause issues in the classroom or workplace (Campbell et al., 2015).

5.1. Limitations and future research

This research is limited in several ways. First, the method of cross-temporal meta-analysis is limited to the available data. The samples taken each year are not random. Optimally, they are random with respect to the association of interest (i.e., narcissism and time) but that is not guaranteed. We partially remedied this by also examining samples from the same college campus in Studies 2 and 3. Ideally, future research will explore changes over time in other individual difference data sets that are differently constructed, include variables related to narcissism such as better-than-average ratings or values, and include relevant cultural products (e.g. song lyrics). Also, all three of these studies were limited to college students, who are a growing but select portion of young Americans. The conclusions are also limited to the U.S.; it is unknown if the downward trend in narcissism after 2008 also appeared in other countries.

Second, there is not an optimal economic measure to use in this research. We used the unemployment rate and the employment to population ratio because they have a long history of use and are linked directly to individuals' economic experience. The unemployment rate may not have a direct or immediate effect on college students via their job prospects, but may influence them through their parents' employment experiences and their sense of their own economic prospects in the future. Other measures of economic activity such as GDP may be less directly related to individuals psychologically, and price inflation/deflation is challenging to measure cleanly. For example, the consumer price index (CPI) often obscures the sources of inflation that dominate people's thinking on a day-to-day basis (e.g., education, housing prices, and medical care). Overall, there is a need for more sophisticated economic models and data in terms of psychological processes.