Is Happiness Infectious?

Published date01 February 2017
AuthorJohn Knight,Ramani Gunatilaka
Date01 February 2017
DOIhttp://doi.org/10.1111/sjpe.12105
IS HAPPINESS INFECTIOUS?
John Knight* and Ramani Gunatilaka**
ABSTRACT
The paper uses an appropriate survey from rural China to answer the question:
is happiness infectious, i.e. does the happiness of an individual depend positively
on the happiness of his or her reference group? The evidence is consistent with
this hypothesis, but the challenge is to solve the ‘reflection problem’, i.e. is the
apparent effect of neighbours’ happiness on own happiness a causal one or
merely a reflection? A ‘quasi-panel’ approach is adopted, treating villages as
groups and individuals as multiple observations within each group, and using an
error components 2SLS estimator. The results suggest that the relationship
might be largely causal. The normative and policy implications are briefly con-
sidered.
How selfish soever a man may be supposed, there are evidently
some principles in his nature, which interest him in the fortunes
of others, and render their happiness necessary to him, though
he derives nothing from it except the pleasure of seeing it. Of
this kind is pity or compassion, the emotion which we feel for
the misery of others, when we either see it, or are made to con-
ceive it in a very lively manner. (Smith, 1759: 11).
II
NTRODUCTION
In this paper, we pose a question which economists have rarely tried to
answer empirically. Is happiness infectious? It is an oft-discerned and oft-
repeated notion that happiness is infectious. Adam Smith might well have
been the first economist to discuss the phenomenon, but it has attracted little
attention from modern economists despite the potential implications for eco-
nomic welfare.
We address this question by means of a national socioeconomic survey for
rural China that includes information on respondents’ subjective well-being.
One reason why the question has been neglected is that most data sets con-
taining information on subjective well-being have sample designs that do not
permit testing for externalities that might arise in small social networks
*University of Oxford
**University of Colombo
Scottish Journal of Political Economy, DOI: 10.1111/sjpe.12105, Vol. 64, No. 1, February 2017
©2016 Scottish Economic Society.
1
beyond the family. The particular advantage of this survey for our purpose is
that it contains information on close neighbours.
We do indeed find evidence consistent with this idea. Our methodological
task is to choose between this explanation for the positive relationship
between own happiness and neighbours’ happiness and alternative explana-
tions. We shall have to address the ‘reflection problem’ (Manski, 1993): is the
apparent effect of neighbours’ happiness on own happiness a causal one or
merely a reflection?
We explore the relevant literature in Section II, we explain the survey
instrument in Section III, and we set out the testing methodology in Sec-
tion IV. Section V presents and interprets the empirical results. Section VI
briefly considers the implications of our interpretation for welfare economics.
Section VII concludes.
II THE RELEVANT LITERATURE
There is a considerable literature, dating back to the classical economists, on
the effect of relative variables on peoples’ welfare. Thus for instance, Adam
Smith wrote, on the one hand, that satisfaction with income depends on such
things as the need for status and the avoidance of shame in society (Smith,
1776: 466) and, on the other, of there being a natural fellow-feeling for others
which, however, is weakened by social distance (Smith, 1759: 11, 156-8, 270).
Karl Marx (1849: 163) argued that we live in society and we take our values
from society, and that this can give rise to feelings of relative deprivation. The
concept and determinants of relative deprivation were explored sociologically
by Runciman (1966). Easterlin (1974) posed a puzzle: why did measures of
subjective well-being remain fairly constant over several decades in the United
States and Japan, whereas real incomes rose substantially? His solution to the
paradox was that peoples’ aspirations for income, based on the incomes of
people with whom they made comparisons, rose along with the average
income in society. Happiness is a positive function of own income but a nega-
tive function of aspirations, which depend on relative income: there is a ‘hedo-
nic treadmill’.
Economic research on subjective well-being has in some cases incorporated
relative variables into happiness functions (well surveyed by Clark et al.,
2008). A common finding is that subjective well-being varies inversely with
comparator incomes. The reference groups that are used vary from one study
to another and are generally inferred rather than identified directly. Thus, for
instance, Clark and Oswald (1996) found that job satisfaction of British work-
ers is inversely related to the wage rates of workers who have the same wage-
generating characteristics. Ferrer-I-Carbonell (2005), analysing a German data
set, found that the income of the respondent’s reference group, defined by a
combination of age, education and region, is roughly as important for happi-
ness as the respondent’s own income, but it has the opposite sign. Similarly,
McBride (2001), estimated a negative relative income effect for the United
States, based on the income of parents and of people in the same age group.
2JOHN KNIGHT AND RAMANI GUNATILAKA
Scottish Journal of Political Economy
©2016 Scottish Economic Society

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT