Changes in beverage purchases following the announcement and implementation of South Africa's Health Promotion Levy: an observational study

Abstract

Summary Background In 2016, South Africa announced an intention to levy a tax on sugar-sweetened beverages (SSBs). In 2018, the country implemented an SSB tax of approximately 10%, known as the Health Promotion Levy (HPL). We aimed to assess changes in the purchases of beverages before and after the HPL announcement and implementation. Methods We used Kantar Europanel data on monthly household purchases between January, 2014, and March, 2019, among a sample of South African households (n=113 653 household-month observations) from all nine provinces to obtain per-capita sugar, calories, and volume from taxable and non-taxable beverages purchased before and after the HPL announcement and implementation. We describe survey-weighted means for each period, and regression-controlled predictions of outcomes and counterfactuals based on pre-HPL announcement trends, with bootstrapped 95% CIs, and stratify results by socioeconomic status. Findings Mean sugar from taxable beverage purchases fell from 16·25 g/capita per day (95% CI 15·80–16·70) to 14·26 (13·85–14·67) from the pre-HPL announcement to post-announcement period, and then to 10·63 g/capita per day (10·22–11·04) in the year after implementation. Mean volumes of taxable beverage purchases fell from 518·99 mL/capita per day (506·90–531·08) to 492·16 (481·28–503·04) from pre-announcement to post announcement, and then to 443·39 mL/capita per day (430·10–456·56) after implementation. Across these time periods, there was a small increase in the purchases of non-taxable beverages, from 283·45 mL/capita per day (273·34–293·56) pre-announcement to 312·94 (296·29–329·29) post implementation. When compared with pre-announcement counterfactual trends, reductions in taxable beverage purchase outcomes were significantly larger than the unadjusted survey-weighted observed reductions. Households with lower socioeconomic status purchased larger amounts of taxable beverages in the pre-announcement period than did households with higher socioeconomic status, but demonstrated bigger reductions after the tax was implemented. Interpretation The announcement and introduction of South Africa’s HPL were followed by reductions in the sugar, calories, and volume of beverage purchases. Funding Bloomberg Philanthropies, International Development Research Centre, South African Medical Research Council, and the US National Institutes of Health.

Taxed and untaxed beverage intake by South African young adults after a national sugar-sweetened beverage tax: A before-and-after study

“…However, there were no differences in absolute changes between groups, likely because the sample as a whole is relatively low income. For comparison, a recent pre–post study using purchase data from a nationally representative South Africa sample found greater reductions in sugar consumption (−32.7%) in LSM 4–6—nearly the identical range to our sample—compared to the higher socioeconomic group of LSM 7–10 (−20.4%) [ 42 ].…”

South Africa's Health Promotion Levy: Excise tax findings and equity potential

“…The data imply that despite the beverage industry's resistance, they were able to respond to the HPL by lowering the sugar content of their beverages to lower their tax liability 23 . These studies also show that lower socioeconomic households and higher consuming or purchasers are indeed responding more in absolute terms in their reductions and thus address the health promotion goals of the HPL 22–24 . At the same time, overall sugar intakes from beverages and SSB purchases fell considerably; thus, addressing the health risk that the policy was targeting.…”

“…Urban household purchases of taxable beverages post implementation fell by 29% with sugar from these purchases falling by 51%. Importantly, lower socioeconomic urban households reduced their volume and grams of sugar from SSBs by 32%% and 57%, respectively 22 A detailed collection of daily beverage intake from young (18–39 years) adults surveyed in Langa, a Western Cape township, showed a 37% reduction in volume and 31% in sugar 23 .…”

“…The Langa paper separated these two changes for dietary intake in this one sample, but the national sample of food purchase data can be used to focus on this across South Africa 23 . One type of industry response includes replacing sugar with a variety of nonnutritive sweeteners (NNSs) as is the trend globally 22,26 . The longer term health implications of these NNS are inconclusive, and more research is needed, but there are concerns that early exposure to NNSs among children affects their sweetness preferences 27,28 .…”

Simulating international tax designs on sugar-sweetened beverages in Mexico

“…Available evidence suggests that these sugar-density taxes in ZA and the UK have led to product reformulation. In ZA, there is a strong indication that HPL resulted in reformulation given that the sugar purchase reduction (51% lower compared to counterfactual) was larger than the volume purchased reduction (29% lower compared to counterfactual) after the tax implementation [ 20 ]. In the UK, the share of drink products subject to SDIL was 33.8 percentage points lower (compared to a counterfactual) within a year of the SDIL implementation [ 19 ].…”