13
Employed respondents report very low satisfaction scores during all employment-related
activities. Working receives an average satisfaction score of only 6.69 and thus belongs to the
least satisfying times of the day. People report an equally low satisfaction level only while
commuting and feel even more dissatisfied only during job search activities (5.76) and while
doing household chores (6.45). Breaks during working hours seem to be enjoyable compared
to actual working time, but their satisfaction score of 7.77 does not come close to the
satisfaction values reached during most leisure activities. The low value of work corresponds
perfectly with the findings by Kahneman et al. (2004a), who also report that working,
commuting, and housework are the worst-rated activities among the employed.
Unemployed persons exhibit roughly the same ranking of activities according to their
episode satisfaction as that found for employed persons, but show lower episode satisfaction
scores in almost all activities (only when spending time with their children do the unemployed
report higher satisfaction scores than the employed). The negative difference in episode
satisfaction levels between the employed and the unemployed is statistically significant (at
least at the 10-percent level) for socializing, reading/listening to music, eating, relaxing,
watching TV, and during non-work trips. This corresponds to Krueger and Mueller (2008),
who also find that the unemployed are sadder and less happy than the employed while
engaged in the same type of activity, and that the largest differences occur during job search
activities (where the difference we find is not statistically significant, most likely because
only one percent of the employed in our sample actively engage in job search) and while
watching TV (which we also find in our data). This illustrates what we call the saddening
effect. Being unemployed reduces the satisfaction experienced during any specific activity.
Sorting the activities by their average net affect paints a very similar picture.8 Our ranking
of activities is thus in line with Kahneman et al. (2004a). Attending cultural events,
socializing, playing parlor and computer games, and pursuing one’s hobbies give the highest
net affect scores for the employed (between 6.53 and 9.01). The lowest scores are reported for
all work-related activities. For the employed, the actual working time is associated with the
lowest net affect of all activities (2.72). Job seeking and commuting give slightly higher net
affects between 2.78 and 3.08. For the unemployed, the ranking of activities is similar to the
one obtained by the episode satisfaction measure. They report the worst net affect for job
8 Spearman’s rank correlation coefficient between episode satisfaction and net affect is 0.94 for the employed
and 0.89 for the unemployed.