Does Social Identity Matter in Consumption Distance? Household-Level Evidence in Post-Reform India
Abstract
Does consumption distance as a measure of social alienation reveal the effect of social identity?<br />To characterize this, we examine the effect of different identities on their consumption<br />
level and distance measured as a z-score, in order to show whether the consumption level or<br />
distance across the distribution reveals any heterogeneous class behavior in consumption<br />
pattern. Using India's household-level data on consumption expenditure covering three major<br />
survey rounds since the inception of the reform period, we find that marginalized social<br />
groups and minority religious groups tend to be alienated across the distribution and over<br />
time, controlling for the effects of education, age, and household type. Since households<br />
with education tend to have higher per capita consumption relative to their group mean, we<br />
examine for the first time with NSS data the problem of endogeneity, using the number of<br />
schools at the district level as an exogenous instrument for educational attainment. Even<br />
with endogeneity correction, both marginalized social groups and minority religious group<br />
still remain alienated across the distribution. As a further robustness, whether their identity<br />
is the reason why these groups are alienated, we undertook a counterfactual decomposition<br />
by rewarding the socio-economic and demographic characteristics of one social group to another,<br />
and found that the estimated consumption difference or gap is observed to rise over<br />
time and across the distribution which also remain robust even in a matched sample of individual<br />
characteristics. Rise in such gap distributed heterogeneously across groups may lead<br />
to volatile social dynamics, with historically underprivileged classes being socially alienated,<br />
implying a possible trade-off between economic prosperity and social cohesion.