Association Between a State-Level Fat Tax and Fast Food Purchases

Abstract

<p><strong>Importance</strong> India faces an increasing obesity problem, including in the Indian state of Kerala in which the fat tax was implemented but was nullified 11 months later. A fat tax, defined as a tax on unhealthy foods, may be associated with changes in food purchases and outcomes for multiple diet-related diseases.</p><p><strong>Objective</strong> To investigate the association between the state-level fat tax and fast food purchases in Kerala, India.</p><p><strong>Design, Setting, and Participants</strong> This cohort study analyzed a large-scale credit and debit card transaction data set and aggregated this sample at the account-year-month level of fast food purchases in Kerala state and 9 major cities in other Indian states (Ahmedabad, Bangalore, Bhubaneswar, Chennai, Delhi, Gurgaon, Kolkata, Mumbai, and Surat). Purchase records were obtained for January 1, 2016, to December 31, 2017. The association between the fat tax and fast food purchases was examined using the difference-in-differences method. This analysis was initiated on December 1, 2022.</p><p><strong>Exposures</strong> The exposure was the fat tax. Kerala was the exposed group, and 9 major Indian cities were the control group.</p><p><strong>Main Outcomes and Measures</strong> The main outcome was the fast food purchase ratio, defined as the proportion of fast food purchases of the total food purchases. Changes in the fast food purchase ratio were estimated in Kerala across the sample period and then compared with 9 major cities.</p><p><strong>Results</strong> The sample for analysis included 238 015 credit and debit card accounts, of which 36.7% were in Kerala and 63.3% were in 9 major cities. The cardholders included 191 603 males (80.5%) with a mean (SD) age of 36.6 (12.8) years. During the fat tax implementation (August 2016-June 2017), Kerala’s fast food purchase ratio decreased by 3.9 percentage points (β [SE], −0.039 [0.002]; 95% CI, −0.042 to −0.036), compared with 9 major cities. After the fat tax was nullified, the fast food purchase ratio reduced by 5.6 percentage points (γ [SE], −0.056 [0.002]; 95% CI, −0.059 to −0.052) compared with 9 major cities and using the pretax period as the benchmark.</p><p><strong>Conclusions and Relevance</strong> Results of this cohort study suggest that the Kerala fat tax was associated with fewer fast food purchases. Food tax policies need to have an elaborate design, and related issues, such as social inequality, nutritional deficiency, and political concerns, need to be evaluated in future studies.</p>