Chapter 2At this point, one might of course ask: Does everyone care
equally strongly about how much other people have? Or do
these comparisons become more important when you are
well off, while poorer people worry more about their abso-
lute income? There is some evidence of this in Britain, where
the estimated coefficient on log comparator income is 0.05
points more negative at the upper quartile of income than at
the lower quartile.^26 The same is true in Germany. But the
implications are the same as those we have already noted.
Another issue is whether people compare themselves not
to the average of a comparator group but for example to
some particular part of the income distribution, like for ex-
ample top incomes. In this case we should have to include
some measure of inequality (like top incomes relative to
mean incomes) in the happiness equation. However, at-
tempts to disentangle the effects of inequality on individual
happiness have not been particularly successful, and a dis-
cussion of this issue is left to Chapter 8.^27
What of adaptation? This can only be studied by exploit-
ing the time- series aspect of the panel data. So we add to
a standard individual fixed- effects regression a variable for
comparator income plus another equal to the average of
own log income over the previous three years (see Table 2.4).Table 2.3. How life- satisfaction (0– 10) is affected by own income and
comparator income (household panel data) (pooled cross- section)Britain Germany Australia USALog own
income
0.16 (.01) 0.26 (.01) 0.16 (.01) 0.31 (.01)Log comparator
income
−0.23 (.07) −0.25 (.04) −0.17 (.06) −0.19 (.03)