Happiness of economists

Abstract

We study the importance of economists’ professional situation towards their life satisfaction based on a unique survey of mostly academic economists. On average, economists report to be highly happy with life. Satisfaction is positively related to spending more time on doing research. The lack of a tenured position decreases satisfaction. However, the extent to which the uncertainty created by the tenure system affects satisfaction varies with the contract terms. The effect is stronger if the contract expires in the near future or cannot be extended. Publication success has no effect if it is controlled for academic rank and the contract duration. The finding suggests that publications are rather a means to an end, for example, to acquire a tenured position. While the perceived level of external pressure also has no impact, the perceived change of pressure in recent years is positively related to economists’ life satisfaction. An explanation is that economists have accepted a high level of pressure when entering academia but are not willing to cope with the recent increase.

Keywords:

• happiness
• academic labour market
• extrinsic and intrinsic motivation
• publish or perish

JEL Classification:

• A11
• I31
• J28

Acknowledgements

The authors are indebted to EEA, the German Economic Association, the French Economic Association and the Applied Microeconomic Congress who allowed us to use their members’ email addresses. The authors are grateful to Yvon Rocaboy for organizing the survey among French economists. The authors benefited from comments by Axel Dreher, Jürgen Hoffmeyer-Zlotnik, Lars Kaczmirek, Dominic Leiner, Natalja Menold, Andrew Oswald, participants of the European Public Choice Meeting 2013, the Beyond Basic Questions Workshop 2013 and the Meeting of the EEA 2014.

Notes

¹ Fifty-four participants indicated that they had already participated in an earlier wave (first question in 2nd and 3rd wave). To avoid duplicate participation, they were not shown the questions but immediately forwarded to the end of the survey.

² Details of the analysis of representativeness of the first two waves are provided in Necker (2012, 2014) (characteristics of the population of the third wave are not available). Various tests, for example, a comparison of sample and population statistics, a comparison of dropouts and those finishing the survey, indicate that the responses are not subject to unit nonresponse biases.

³ The number of response possibilities provided in the different large-scale surveys ranges from 3 to 11 (Dolan et al., 2008).

⁴ The means of life satisfaction in the different waves are 4.6 (EEA), 4.7 (German Economic Association) and 4.5 (French Economic Association). The difference between the latter two is significant.

⁵ It is controlled for the respondent’s country of origin if more than 20 observations are available. Otherwise countries are grouped by geographical proximity.

⁶ Much of the happiness literature treats the dependent variable as cardinal and uses ordinary least squares (OLS) regression analysis (Frey and Stutzer, 2001; Ferrer-i Carbonell and Frijters, 2004). The results are unchanged when the model is estimated using OLS.

⁷ The MI estimate of , the vector of parameters of interest, is . The variance–covariance matrix of is where is the within-imputation variance–covariance matrix and is the between-imputation variance–covariance matrix.

⁸ Results are largely unchanged if a complete case analysis is performed (available upon request).

⁹ We assign average salaries reported for full, associate and assistant professors from France, Germany, Italy, the Netherlands, the United Kingdom and the United States to respondents with the respective characteristics (results are available upon request).

¹⁰ As shown in Table 1, 18.6% report that their contract expires within the next year, 11.5% within 2 years, 11.5% within 3 or more years.

¹¹ While immediate expiration implies a decrease of 11ppts (sign. at 1% level), expiration in 2 years a decrease of 8ppts (sign. at 5% level), expiration in 3 years a decrease of 5ppts (insignificant). Results available upon request.

¹² Of those with a limited contract, 27.4% report that the contract can, 59.3% that it cannot be extended, 13.3% report that the question ‘does not apply’. Since we do not know the reasons for reporting inapplicability, the respective observations are dropped from the analysis.