foods within the social card program - as proposed, for example, by the pilot project
of the state of California (Guthrie et al., 2007)16.
6 Conclusions
Cross-price elasticities estimated by demand systems provide a consistent framework
to evaluate substitution effects among goods. In this paper, we present estimates
of the long-run substitution effects of the categories of unhealthy and healthy foods
for Italy.
By showing the close pattern linking unhealthy foods with more energy-dense
foods, our findings suggest that the largest rise in healthy food prices, versus those
of unhealthy ones, have favoured the consumption of high-calorie foods. This result
matches the increase in body weight recently recorded in Italy.
In addition, to deal with the heterogeneous impact of food price changes on
food expenditure allocation, we extracted the advantages of using data surveys as
repeated cross-sections of time-series to assess the uneven substitution effects in
demographic and socio-economic groups. One peculiarity of our results is that, as
relative healthy food prices rise, individuals who are male, below the relative poverty
threshold, and those with lower education tend to substitute energy-light foods for
more energy-dense.
To sum up, the changes in relative prices, which have caused healthy foods to
become 10 basis points more expensive in Italy in ten years, have generated a mech-
anism of substitution towards foods which are cheaper and high in calories, affecting
mainly some disadvantaged groups. This calls into question the actions of policy-
makers and the strategies that should be used to reduce such inequalities. If com-
bating obesity is one objective of the government, then our results indicates that
a lack of intervention regarding the loss of power to purchase by some segments of
society may actually maintain the rate of overweight growing. Thus, among the
tools of policy-makers, it would seem potentially more efficient to use subsidies in
favour of the purchase of healthy foods, associated with an active welfare program
or more specific food programs.
An open question is to reconcile our estimated cross-price elasticities, according
16The Italian Minister for Economy and Finance stated that in 2007, there were about 1,300,000 indigent people
which were eligible for the ‘’social card”, and over 7,400,000 poor individuals were estimated to be below the poverty
line in the same year. A first ex-post evaluation reports that only 42 percent of indigent people complied with the
‘’social card” program.
23