Do non-monetary prices target the poor? Evidence from a field experiment in India

Abstract

This paper uses willingness to pay data from a field experiment in India to study targeting health products to the poor, using monetary prices and non-monetary prices (time costs). I model demand for the product at monetary and non-monetary prices. The model illustrates that monetary prices screen out the poor and that whether non-monetary prices screen out the non-poor is theoretically ambiguous because of opposing income and substitution effects. I find monetary prices select richer households and non-monetary prices do not provide strong selection on income. Both the poor and non-poor appear very elastic in the non-monetary price because of the high value of time in home production. I evaluate the problem of a principal with a fixed budget whose objective places some weight on coverage and some weight on targeting. Despite better targeting with non-monetary prices, the principal optimally chooses a monetary price for a large range of parameters.