Chapter 4finding of previous literature,^7 and here we investigate it for
men using the household panel data.^8
We study those individuals in the sample who had at least
one spell of unemployment, and we observe their well- being
in the years before their first spell began and the subsequent
years (up till their next spell of unemployment).^9 For the
whole sample we now estimate a standard regression that
measures the level of life- satisfaction in these years, using a
fixed effect to remove any selection biases.^10 The equation
tells us how happy on average the sample were in the years
before unemployment set in, and then how happy those
people were who were still unemployed one year later, two
years later, three years later, and four or more years later. The
results are plotted in Figure 4.1.
As can be seen, in Britain and Germany the onset of un-
employment reduces life- satisfaction by nearly 1 point, and
life- satisfaction remains at least this low so long as the per-
son remains unemployed. In none of the three countries is
there any adaptation to unemployment.
What happens once a person is reemployed? Does the
experience of unemployment still linger, reducing the per-
son’s life- satisfaction? This is the issue of scarring. We in-
vestigate this by including in our regressions a variable
reflecting the amount of time a person was unemployed
previously. For the BCS we have a complete record at age
30 for the proportion of time the individual has been un-
employed since joining the labor force. When we introduce
this into our standard equation, it attracts a coefficient of
−1.47 (s.e. = .18). This means that each previous year of un-
employment is reducing current life- satisfaction by about
0.1 points (1.47/14)— one- tenth of the pain it causes at the
time.