Notes to Pages 62–67once. However the duration of unemployment is also influenced by
the income replacement ratio for the unemployed and the conditions
attached to the receipt of benefit. On these issues, see Layard, Nickell,
and Jackman (2005) and Pissarides (2000).
- On self- employment, see Blanchflower and Oswald (1998) and
Colombier and Masclet (2008). - For 30- to- 45- year- olds, see online Table A4.1.
- The results in the existing literature broadly find no evidence of
any such adaptation to unemployment. This holds for the analysis of
data from the SOEP (Clark, Diener, Georgellis, and Lucas 2008), BHPS
(Clark and Georgellis, 2013), Russian Longitudinal Monitoring Survey
(RLMS) (Clark and Uglanova, 2012), the Korean Labor and Income
Panel Study (KLIPS) for men (there are insufficient data points for
women to reach a conclusion) in Rudolf and Kang (2015), and HILDA
(Frijters, Johnston, and Shields [2011]). The research in Lucas et al.
(2004) suggests only partial adaptation to unemployment in SOEP
data, as does that on Swiss Household Panel (SHP) survey in Anusic,
Yap, and Lucas (2014), although these latter two papers use parametric
rather than nonparametric statistical methods. See also online Annex 4. - The regression is the same as that in Clark, Flèche, and Senik
(2014). - We exclude everyone for whom we lack any of this knowledge.
- The controls were marital status, children, income, age, age
squared, and regional and time dummies. - Strictly the variable is the proportion of waves when this oc-
curred. We also tried including this variable in panel regressions, but it
did not vary sufficiently for each individual for the analysis to produce
sensible results. (It is also of course somewhat arbitrary to take just five
years of previous experience.) - See Clark, Georgellis, and Sanfey (2001); Ruhm (1991).
- See online Full Table A2.2.
- See Clark (2003); Clark, Knabe, and Rätzel (2010); Powdthavee
(2007). It also of course expands the competition for jobs. - 0.138/0.70. This assumes that the labor- force participation rate
of the population under 65 is 70%. Di Tella, MacCulloch, and Oswald
(2003) also estimate that the employed bear three- fourths of the cost—
see online Annex 4. - The overall unemployment rate is determined by many other
factors that affect the level of the equation for individuals. See, for ex-
ample, Layard, Nickell, and Jackman (2005).