Chapter 5


Compared to other approaches to targeted subsidies, additional costs accrue from the need to
set up and to manage a particular distribution system. Such costs are largely avoided in the
case of a food stamp programme, as discussed in the following section.

Box 5.6: Targeted food price subsidy scheme in the Philippines

This was implemented in three provinces using data collected by a detailed IFPRI study. The scheme consisted
of price discounts on rice and cooking oil and a nutrition education component. Targeting was geographical to
all households in villages with high incidence of malnutrition and poverty. Each household was issued with a
ration card indicating its monthly quota of rice and oil based on family size. The rice ration subject to price
discount was only about half the amount usually consumed by most of the households, but the oil ration
exceeded the amount usually purchased prior to the subsidy. Thus, consumer rice prices were not reduced at the
margin, unlike oil prices.

An evaluation of the project found that the subsidy component of the scheme caused a significant increase in
household food expenditures and calories acquired and consumed, as well as in calories consumed by most
individual household members. Although adults obtained the largest share, the average weight of pre-schoolers
also increased. The nutrition education component was found to have a small positive effect when accompanied
by the subsidy, but no effect could be detected when education was unaccompanied by additional purchasing
power. When education did have an effect, it was intra-household, resulting particularly in increased
consumption by children an indication that nutritional messages can improve childcare practices.

Findings indicated that consumers were more likely to increase their food consumption if food were subsidized
than if incomes were raised directly. Overall the evidence indicated that the subsidy had positive effects on both
households and pre-schoolers.

In terms of costs, 84% was the subsidy itself, 9% administrative costs, 7% incentive payment to ensure efficient
food distribution. The fiscal cost of each $1.00 transferred to participating households was $1.19; if only
transfers to malnourished pre-schoolers are considered a benefit, the cost increased to $3.61. Cost-effectiveness
is thought to be favourable. Costs were low, firstly because geographical targeting based on growth monitoring
costs less than targeting based on household income levels; secondly, because the use of existing private sector
retail outlets for the distribution of subsidized food costs less than a separate distribution network; and thirdly,
because the use and expansion of existing local bureaucratic structures costs less than the creation of a new and
independent structure.

Source: Gillespie and Mason, 1990, cited in: Relief and Development Institute, Guidelines for incorporating
food security and nutritional considerations into support for structural adjustment in ACP countries, London
1991.

4) Food stamps

Food stamps are a method of targeting subsidies through a coupon system. The coupons are
distributed to eligible groups and can be cashed as part or whole payment for purchases of a
predetermined set of commodities in specific retail outlets. The participating retail outlets can
be ordinary private shops which are refunded by government on presentation of the coupons.
Government is not directly involved in supply and retailing operations. Its involvement is
confined to administrative tasks such as the issue and distribution of the food stamps and
refunding the participating retailers. The beneficiaries have a choice among the range of
scheduled commodities. This gives the food stamps a "near-money" property. Food stamp
programmes are considered as consistent with the operation of markets, a cost-effective way
of food-mediated income transfer to low income households, and as an attractive alternative
to general food subsidies, largely in line with the objectives of budgetary savings and market
liberalisation under adjustment.


-210-