The Origins of Happiness

Notes to Pages 68–83

17. The simple correlation is 0.12.


18. Kahneman, Krueger, et al. (2004).


19. Krueger (2007), Krueger, Kahneman, Fischler, et al. (2009). See
    also Krueger, Kahneman, Schkade, et al. (2009), tables 1.3, 1.6 and 1.9;
    and Bryson and MacKerron (2017).


20. Krueger, Kahneman, Schkade, et al. (2009), table 1.10.


21. Edmans (2011), Edmans (2012).


22. See for example Clark (2001), Clark (2010), OECD (2013b), Clark
    (2011), Lundberg and Cooper (2011), Robertson and Cooper (2011).


23. The effect of hours is swamped by the question on family life.


24. These actual effects are fairly consistent with what people in the
    BHPS say about what matters to them in a job— see Clark (2011).

Chapter 5. Building a Family

1. Office for National Statistics (2015).


2. Previous work on marriage suggested broadly complete adapta-
   tion in the SOEP (Clark, Diener, Georgellis, and Lucas 2008) and Lucas
   et al. [2003], BHPS (Clark and Georgellis [2013]), HILDA (Frijters,
   Johnston, and Shields [2011]) and Swiss Household Panel (Anusic, Yap,
   and Lucas [2014]). Adaptation is however only partial at best in Russia
   (Clark and Uglanova [2012]) and Korea (Rudolf and Kang [2015]). See
   also Qari (2014).


3. The small coefficients in Table 5.1 are an average of the big
   onset coefficient and subsequent adaptation.


4. There is a large literature on divorce, but this often happens
   some years after the worst period of separation. The existing literature
   finds full adaptation to divorce in the SOEP, BHPS, and HILDA, but
   only partial adaptation in the Swiss Household Panel (Anusic, Yap, and
   Lucas [2014]) and for Korean men (Rudolf and Kang [2015]).


5. See also online Annex 5.


6. See Clark, Diener, Georgellis, and Lucas (2008), Lucas et al. (2003).


7. See Frijters, Johnston, and Shields (2011).


8. See online Table A5.1. This shows that the happiness of those
   who are not partnered is lower by 0.27 points, when everyone else is
   partnered, compared to the situation when only 50% of other people
   are partnered.


9. People can also choose whether to partner. But not everyone who
   would like a partner gets one, while some 90% of people who want chil-
   dren get them.