Modest and incomplete incentives may work: pricing plastic bags in Uruguay
Resumen:
We study the impact of voluntary prices on the demand for plastic bags, using two years of administrative data from a national supermarket chain in Uruguay. We find that prices of US$ 0.07 and US$ 0.10 per unit decreased the number of bags used by its customers in the range of 70% to 85%. We also find that in anticipation of the price, customers increased the use of bags by up to 38% in some branches before its implementation. This result has strong implications for future studies, regarding the need to use long pre-periods. According to our data, short pre periods could bias upward the estimation of the effect of the prices by up to 41%. Finally, we do not find evidence consistent with the supermarket suffering a loss of sales, due to clients moving to stores not pricing the bags. The fact that the supermarket is a discount chain (clients are more price sensitive) and that it seems to have chosen the cities and towns in which it had more market share to rollout the prices, may inform policy makers about the conditions under which incomplete and moderate incentives may work.
2021 | |
Plastic bags Price Consumer behavior Difference in difference Synthetic controls Anticipation effects |
|
Inglés | |
Universidad de Montevideo | |
REDUM | |
https://hdl.handle.net/20.500.12806/2442
https://doi.org/10.1016/j.jeem.2021.102525 |
|
Acceso abierto | |
Attribution-NonCommercial-NoDerivatives 4.0 Internacional |
Sumario: | We study the impact of voluntary prices on the demand for plastic bags, using two years of administrative data from a national supermarket chain in Uruguay. We find that prices of US$ 0.07 and US$ 0.10 per unit decreased the number of bags used by its customers in the range of 70% to 85%. We also find that in anticipation of the price, customers increased the use of bags by up to 38% in some branches before its implementation. This result has strong implications for future studies, regarding the need to use long pre-periods. According to our data, short pre periods could bias upward the estimation of the effect of the prices by up to 41%. Finally, we do not find evidence consistent with the supermarket suffering a loss of sales, due to clients moving to stores not pricing the bags. The fact that the supermarket is a discount chain (clients are more price sensitive) and that it seems to have chosen the cities and towns in which it had more market share to rollout the prices, may inform policy makers about the conditions under which incomplete and moderate incentives may work. |
---|