We examine whether income shocks affect a range of health outcomes and a
preventative behaviour. We instrument income with rainfall measurements by
matching satellite information on timing and positioning of 21 rainfall stations to
longitudinal data (1991-1994) of over 4,000 individuals in 51 villages of a North
Western region in Tanzania.
We find a pro-cyclical effect of income on health. A ten percent increase in
income reduces by 0.2 the number of illnesses. A further finding is the positive
effect on vaccinations of children under six: a ten percent increase in income
implies an increase of about one vaccination, from a mean of 2.3 per child, for the
four vaccinations of polio, tetanus, tuberculosis and measles. There is also some
evidence of a reduction in chronic malnutrition of children under six. Our results
suggest the income effect to offset the increased opportunity cost of time in this
data.