Linking a sugar-sweetened beverage tax with fruit and vegetable subsidies: A simulation analysis of the impact on the poor

Am J Clin Nutr. 2022 Jan 11;115(1):244-255. doi: 10.1093/ajcn/nqab330.


Background: US individuals, particularly from low-income subpopulations, have very poor diet quality. Policies encouraging shifts from consuming unhealthy food towards healthy food consumption are needed.

Objectives: We simulate the differential impacts of a national sugar-sweetened beverage (SSB) tax and combinations of SSB taxes with fruit and vegetable (FV) subsidies targeted to low-income households on SSB and FV purchases of lower and higher SSB purchasers.

Methods: We considered a 1-cent-per-ounce SSB tax and 2 FV subsidy rates of 30% and 50% and used longitudinal grocery purchase data for 79,044 urban/semiurban US households from 2010-2014 Nielsen Homescan data. We used demand elasticities for lower and higher SSB purchasers, estimated via longitudinal quantile regression, to simulate policies' differential effects.

Results: Higher-SSB-purchasing households made larger reductions (per adult equivalent) in SSB purchases than lower SSB purchasers due to the tax (e.g., 4.4 oz/day at SSB purchase percentile 90 compared with 0.5 oz/day at percentile 25; P < 0.05). Our analyses by household income indicated low-income households would make larger reductions than higher-income households at all SSB purchase levels. Targeted FV subsidies induced similar, but nutritionally insignificant, increases in FV purchases of low-income households, regardless of their SSB purchase levels. Subsidies, however, were effective in mitigating the tax burdens. All low-income households experienced a net financial gain when the tax was combined with a 50% FV subsidy, but net gains were smaller among higher SSB purchasers. Further, low-income households with children gained smaller net financial benefits than households without children and incurred net financial losses under a 30% subsidy rate.

Conclusions: SSB taxes can effectively reduce SSB consumption. FV subsidies would increase FV purchases, but nutritionally meaningful increases are limited due to low purchase levels before policy implementation. Expanding taxes beyond SSBs, providing larger FV subsidies, or offering subsidies beyond FVs, particularly for low-income households with children, may be more effective.

Keywords: SSB tax; combined policy; fruit and vegetable subsidies; heterogeneity; longitudinal data; price elasticity; quantile regression.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Computer Simulation
  • Consumer Behavior / economics
  • Diet, Healthy / economics
  • Family Characteristics
  • Female
  • Food Assistance / economics*
  • Fruit / economics*
  • Humans
  • Longitudinal Studies
  • Male
  • Poverty / statistics & numerical data*
  • Sugar-Sweetened Beverages / economics*
  • Supermarkets
  • Taxes / statistics & numerical data*
  • United States
  • Vegetables / economics*