Sugar-Sweetened Beverages, Obesity, and the Case for Beverage Taxation in Africa

In 2022, Coca-Cola sold more than 1.3 billion unit cases of beverages across the African continent - a volume that had grown at roughly 4 to 6 per cent annually through the preceding decade, outpacing every other major global market except India. Nielsen retail audit data compiled for the same period estimated that carbonated soft drink sales volumes across Sub-Saharan Africa (SSA) had doubled since 2010, with the fastest growth concentrated in Nigeria, Kenya, Ethiopia, and Tanzania. These are not niche consumption patterns. They are the front edge of a nutrition transition whose health consequences are already measurable in clinical registries across the region.

Sugar-sweetened beverages - encompassing carbonated soft drinks, fruit-flavoured drinks, sweetened iced teas, energy drinks, and commercially produced flavoured milks - represent one of the most energy-dense, nutritionally empty components of the modern food supply. The epidemiological evidence linking their consumption to obesity, type 2 diabetes mellitus (T2DM), cardiovascular disease, and dental caries has accumulated steadily over two decades. The policy response, in the form of fiscal levies targeting SSB producers or retailers, has now been implemented in over 50 countries. South Africa became the first African nation to enact such a measure in 2018. Whether the evidence from that and other fiscal experiments is sufficient - and whether the distributional consequences of such taxes are acceptable - remains one of the most consequential debates in African public health policy.

This article reviews the evidence on both sides of that debate, drawing on published data specific to SSA market conditions, peer-reviewed economic modelling, and the emerging body of post-implementation evaluation literature.

The Nutrition Transition in Sub-Saharan Africa

To understand why SSB consumption has accelerated so rapidly in SSA, it is necessary to situate the phenomenon within the broader structural transformation that Popkin et al. (2012) termed the “nutrition transition.”¹ As urbanisation rates across SSA climbed - from roughly 27 per cent in 1990 to above 43 per cent by 2020 - traditional dietary patterns built around legumes, whole grains, and fresh vegetables began to erode. Urban wage labour compressed meal preparation time. Informal food vending expanded. Cheap, shelf-stable, and aggressively marketed processed foods and beverages filled the gap.

The economic logic for global beverage corporations is straightforward. SSA is one of the last large frontier markets where per capita SSB consumption remains well below the saturation levels seen in North America, Western Europe, and parts of Latin America. The average South African consumed approximately 254 serving-sized portions of carbonated soft drinks per year as of 2016 - a figure already considerably higher than the continental average, and roughly comparable to consumption rates in Chile in the early 2000s before that country’s own NCD epidemics became acute. In Nigeria and Kenya, per capita consumption remained lower but was growing at compound annual rates above 5 per cent through the 2010s.

Multinational corporations have invested heavily in cold-chain infrastructure, returnable glass bottle distribution networks, and targeted marketing to younger urban demographics. Advertising spend for SSBs in SSA consistently outpaces that for any other food category. This is not incidental: marketing to populations with rising disposable income but limited nutritional literacy represents a deliberate expansion strategy, and understanding it is essential context for evaluating fiscal countermeasures.

Health Effects: The Epidemiological Evidence

Obesity and Overweight

The relationship between habitual SSB consumption and excess weight gain is among the best-characterised diet–disease associations in modern nutritional epidemiology. Malik et al. (2010), in a systematic review and meta-analysis covering prospective cohort studies, found that individuals in the highest quantile of SSB intake had a 26 per cent greater risk of developing type 2 diabetes and considerably higher odds of weight gain and obesity compared with those in the lowest quantile.² The biological mechanisms are not difficult to identify: liquid calories from SSBs do not trigger the same satiety signals as solid food, and the rapid hepatic fructose metabolism associated with high-fructose corn syrup and sucrose promotes lipogenesis and insulin resistance.

In Africa, the double burden of malnutrition makes this dynamic particularly significant. It is now well established that populations transitioning from undernutrition can accumulate visceral fat preferentially when energy-dense, nutrient-poor foods are introduced - a physiological pattern sometimes described as the “thin-fat” phenotype. Data from the South African National Health and Nutrition Examination Survey (SANHANES-1, 2013) found overweight and obesity rates of 52.8 per cent among women and 31.3 per cent among men - figures that coexisted with persistent stunting in children in the same households. Urban SSA is not experiencing a simple transition from hunger to surfeit; it is experiencing a layered, simultaneous pathology.

Type 2 Diabetes Mellitus

Basu et al. (2013) applied an ecological analysis across 175 countries and found that for every additional 150 kilocalories per person per day from sugar - roughly equivalent to one 355 ml can of a standard cola - the prevalence of T2DM in the population increased by 1.1 per cent, independent of total caloric intake and obesity prevalence.³ The effect was sustained after controlling for physical activity levels, sedentary behaviour proxies, and urbanisation rates.

The IDF Diabetes Atlas (2021) estimated that SSA housed approximately 24 million adults with diabetes, with a projected figure of 55 million by 2045 - the fastest proportional growth of any global region. The majority of cases across Africa remain undiagnosed. Health system costs are already severe in countries where insulin access is intermittent and where ambulatory diabetes management competes with infectious disease priorities for constrained budgets. Afshin et al. (2019), in the Global Burden of Disease dietary risk analysis, estimated that suboptimal diet - including excess SSB intake - was responsible for 11 million deaths globally in 2017, with the highest attributable mortality rates in low- and middle-income countries undergoing rapid dietary transition.⁴

Cardiovascular Disease

The cardiovascular consequences of SSB consumption operate through multiple independent pathways: obesity-mediated hypertension, hypertriglyceridaemia, insulin resistance, and - for very high fructose intakes - direct hepatic lipid production. Prospective cohort data from the Nurses’ Health Study and the Health Professionals Follow-up Study indicated that two or more servings of SSBs per day were associated with a 35 per cent greater risk of fatal cardiovascular events among women, compared with one or fewer servings per month.

In SSA, cardiovascular disease has overtaken infectious disease as the leading cause of premature adult mortality in several countries, including South Africa, Ghana, and Nigeria. The Global Burden of Disease study estimates that ischaemic heart disease and stroke together accounted for approximately 1.1 million deaths across SSA in 2019 - a figure that has grown consistently since 2000, driven in large part by the same dietary and metabolic transitions that SSB consumption accelerates.

Dental Caries

The oral health consequences of SSBs are frequently underweighted in public health framing, yet they carry enormous economic and quality-of-life significance in low-resource settings. Dental caries - driven by the fermentation of sugars by oral bacteria and the consequent acid demineralisation of enamel - is the most prevalent non-communicable disease globally, affecting approximately 2.3 billion people. In SSA, where access to preventive dental care and fluoridated water supplies is limited, caries-driven tooth loss begins early in life and compounds social and nutritional vulnerability. Children with untreated dental pain exhibit reduced school attendance, reduced food intake, and documented cognitive performance deficits.

Market Data: SSBs in Sub-Saharan Africa

The scale of the SSB market in SSA is frequently underappreciated in global public health literature. Nakhimovsky et al. (2016), in one of the most detailed analyses of SSB taxation potential in African contexts, compiled national sales data for SSBs across multiple SSA countries and found that carbonated soft drink availability had grown at rates far exceeding population growth across the 2000–2015 period in most surveyed markets.⁵ Nigeria, with a population of over 200 million, represented the single largest SSB market in SSA by volume, followed by South Africa, Tanzania, and Kenya.

Several structural features of SSA markets are salient for policy design. First, informal retail channels - small neighbourhood kiosks, open-air market stalls, mobile vendors - account for a very large share of SSB sales, particularly in rural and peri-urban areas. In Nigeria, it is estimated that over 60 per cent of food and beverage retail transactions occur through informal channels. This has direct implications for tax administration, since standard point-of-sale collection mechanisms designed for formal retail chains may capture only a fraction of actual market volume.

Second, the market is dominated by a small number of global brands - Coca-Cola, PepsiCo, and their regional licensed bottling partners - alongside a growing tier of locally produced drinks that are often sweetened with sucrose rather than high-fructose corn syrup and sold at price points below those of major branded products. Policy design must account for this tiered market structure, since a levy focused on carbonated soft drinks may miss a substantial portion of free sugar delivered through uncarbonated juice-based drinks, which are sometimes perceived as healthier.

Third, income elasticity of demand for SSBs in SSA tends to be positive and relatively high, meaning that SSB consumption rises with income across the income distribution. This has important implications for long-run consumption trajectories as average incomes grow.

Fiscal Interventions: The Evidence Base

South Africa’s Health Promotion Levy (2018)

South Africa implemented its Health Promotion Levy (HPL) on 1 April 2018, charging 2.1 cents per gram of sugar content above 4 grams per 100 ml for beverages with more than 4 g/100 ml total sugar. The rate was later adjusted upwards in subsequent budget cycles. Stacey et al. (2017), in a modelling study conducted in anticipation of the levy’s introduction, projected that a 20 per cent price increase on SSBs in South Africa would avert between 220,000 and 658,000 cases of obesity and between 128,000 and 385,000 cases of T2DM over 20 years, depending on behavioural assumptions.⁶ The breadth of that range reflects genuine uncertainty about price elasticity parameters in the South African context, an uncertainty that post-implementation research has only partially resolved.

Early post-implementation data were encouraging. Treasury monitoring indicated that SSB sales volumes declined by approximately 29 per cent in the first year following levy introduction, though separating the levy effect from concurrent macroeconomic factors - notably a severe economic contraction and rising food price inflation - proved difficult. Retail audit data from Nielsen and Kantar suggested that consumers shifted in part towards still water, unpackaged beverages, and lower-sugar product formulations, the latter reflecting reformulation activity by several large producers ahead of the levy’s introduction. By 2022, more than 30 beverage products had been reformulated to reduce sugar content, a supply-side effect that may ultimately be more significant than the direct price-demand suppression mechanism.

The Mexican Experience

Mexico’s Impuesto Especial sobre Producción y Servicios (IEPS) SSB tax - a 1 peso per litre levy introduced on 1 January 2014 - generated the most detailedly evaluated natural experiment in SSB taxation prior to South Africa. Colchero et al. (2016) analysed household purchase data from the Mexican National Income and Expenditure Survey (ENIGH) and found that taxed beverage purchases declined by an average of 5.5 per cent in the first year and 9.7 per cent in the second year of the tax, with the reduction concentrated among lower-income households who showed the greatest price sensitivity.⁷ The same study found a 16 per cent increase in purchases of untaxed bottled water, suggesting meaningful substitution towards healthier alternatives in this population.

Critically, the Mexican evidence also demonstrated reformulation: manufacturers reduced sugar content in existing products across the board, a response that compresses consumer-level price increases (since lower-sugar products face lower levies) whilst still reducing population sugar exposure. The WHO has cited the Mexican data as the most robust real-world confirmation of the efficacy of SSB taxation in a middle-income country.

The UK Soft Drinks Industry Levy (2018)

The United Kingdom’s Soft Drinks Industry Levy (SDIL), introduced in April 2018 at tiered rates of 18p/litre for drinks with more than 5 g/100 ml sugar and 24p/litre for drinks with more than 8 g/100 ml sugar, provides a contrasting model: one oriented primarily towards incentivising reformulation rather than maximising fiscal yield. Briggs et al. (2013) had earlier modelled the projected health effects of a 20 per cent SSB tax in the UK and estimated approximately 180,000 cases of overweight and obesity averted annually, along with 270,000 fewer cases of T2DM.⁸

The SDIL’s most remarkable effect was pre-implementation: between the announcement of the levy in March 2016 and its implementation in April 2018, the proportion of full-sugar soft drinks in the UK market fell from 49 per cent to 15 per cent of sales volumes, as manufacturers reformulated to avoid the higher tax band. Total sugar sold through soft drinks fell by approximately 45 per cent over the same period. This is a powerful demonstration that fiscal signals - even when fully anticipated - can drive supply-side change at scale, and that the health benefit may be delivered primarily through reformulation rather than through consumer price responses.

Price Elasticity and Substitution Effects

Elasticity Parameters in SSA Contexts

Estimating price elasticity of demand for SSBs in SSA is analytically challenging. Nakhimovsky et al. (2016) compiled available elasticity estimates for SSA countries and found own-price elasticity estimates ranging from −0.8 to −1.3 for carbonated soft drinks, suggesting that demand is relatively elastic - that is, a 10 per cent price increase would reduce consumption by 8 to 13 per cent.⁵ These estimates are somewhat higher in absolute value than comparable figures from high-income countries (typically −0.6 to −0.9), which is consistent with the theory that lower-income populations - for whom SSBs represent a larger share of food budgets - are more price-responsive.

However, there is substantial heterogeneity across countries, income strata, and beverage sub-categories. Energy drinks, which are increasingly popular among young urban males in SSA and carry significant cardiovascular risk, appear to have lower price elasticity than standard carbonated soft drinks - possibly because of their positioning as aspirational commodities rather than everyday refreshments. Product-specific and population-specific elasticity modelling is essential before fiscal measures can be responsibly calibrated.

Substitution: Water, Juice, and Within-Category Switching

Substitution patterns matter greatly for health outcomes. If SSB taxation induces consumers to shift consumption towards unpackaged tap water or plain boiled water, health benefits may be substantial. If substitution is primarily towards other sugary products - commercial fruit juices, flavoured milks, or traditional fermented grain drinks that may contain significant free sugar - net health benefits could be considerably attenuated.

The evidence from Mexico suggested that water substitution was meaningful, particularly among lower-income households. South African post-levy data indicated both water and low-sugar drink substitution. However, the South African context is unusual in that municipal water quality and access are relatively high compared with most of SSA. In countries where safe piped water is unavailable or intermittently supplied - the reality for large majorities across rural SSA - the substitution pathway to water is effectively closed for many consumers, and the substitution question becomes more complex.

Equity Concerns and Distributional Effects

No serious policy analysis of SSB taxation can sidestep the equity question, and it is in this domain that the debate is most contested. SSB taxes are, in structural terms, a form of excise duty on consumption - and consumption taxes are regressive by nature, consuming a higher proportion of income among lower-income households than among wealthier ones. This is a genuine concern, not a rhetorical objection, and it demands a careful empirical response rather than dismissal.

There are at least three distinct dimensions to the equity debate. First, there is the question of the direct fiscal burden: if lower-income households spend more of their total income on SSBs, a specific levy on those products will take a proportionately greater share of their income. This is arithmetically straightforward and empirically confirmed in South African household expenditure data.

Second, however, is the question of health benefits. If the health harms from SSBs - obesity, T2DM, dental caries - are themselves concentrated among lower-income populations, which the epidemiological evidence consistently suggests they are, then the health gains from reduced consumption may also be concentrated in those populations. Colchero et al. (2016) documented that the largest absolute reductions in SSB purchases following Mexico’s tax were among lower-income households - precisely those with the highest price sensitivity.⁷ This creates a potential progressive asymmetry: the tax burden is regressive in fiscal terms, but the health benefit may be progressive.

Third, there is the question of revenue recycling. The South African HPL generated approximately R2.5 billion in its first fiscal year. If that revenue is hypothecated to fund health programmes, nutrition subsidies, or social protection measures that disproportionately benefit lower-income households, the net distributional effect of the policy package could be neutral or even progressive. This was the explicit policy rationale in South Africa, though ring-fencing of health levy revenue has proven politically difficult to maintain in practice.

The equity critique is strongest where it identifies genuinely regressive effects without compensating policy instruments. In settings where tax revenue enters general consolidated funds with no mechanism for redistribution, the fiscal burden on the poor is real and the benefit may not be commensurate. Policymakers who advance SSB taxes without simultaneously designing accompanying social protection measures are leaving the most vulnerable populations to bear costs without corresponding gains.

Conversely, industry actors who invoke equity concerns primarily as a strategic barrier to any fiscal intervention - while simultaneously deploying aggressive marketing to lower-income communities - are not acting in good faith. The equity argument must be engaged seriously, not weaponised selectively.

Limitations and Methodological Considerations

The evidence base for SSB taxation in SSA is growing, but it remains characterised by several important limitations that should temper both advocacy and opposition.

Attribution and confounding. Post-implementation evaluations of SSB taxes - including South Africa’s HPL - face persistent challenges of causal attribution. South Africa’s levy was implemented simultaneously with a period of significant economic contraction, rising inflation, and shifts in consumer confidence. Isolating the tax-specific effect on consumption from these concurrent macroeconomic shocks requires modelling choices that carry their own uncertainty. Difference-in-differences designs comparing South Africa with matched regional comparators are methodologically promising but depend heavily on the plausibility of the parallel trends assumption.

Informal market penetration. As noted above, the fiscal reach of formally administered SSB taxes in SSA is limited by the scale of informal retail. Conventional beverage sales data undercount actual consumption. Any evaluation that relies solely on formal retail scanner data will overestimate the behavioural response to taxation, because the informal market - which the tax may not reach uniformly - will provide a partial refuge for price-sensitive consumers.

Price pass-through. Whether and how fully manufacturers pass levy costs to consumers through retail prices is a critical empirical question that varies across market structures. In more concentrated markets dominated by a few major producers, oligopolistic pricing behaviour may result in partial pass-through, with producers absorbing some of the levy through compressed margins. In more competitive markets, full or even over-shifted pass-through is more likely. South African evidence suggests approximately 80 per cent pass-through on average, but this figure varies by retailer type and beverage category.

Long-term health outcome data. Most evidence linking SSB taxes to health outcomes relies on modelled projections from elasticity parameters rather than observed reductions in obesity or T2DM incidence post-implementation. The time lags involved - from dietary change to measurable clinical outcome - mean that real-world confirmation of projected health benefits remains limited even in Mexico and the UK, where taxes have been in place for the longest periods. SSA-specific clinical outcome data will not be available for many years.

Substitution data quality. Rigorous assessment of what consumers substitute for SSBs following a tax requires detailed, longitudinal individual-level dietary data. Such data are sparse across most of SSA, where 24-hour dietary recall studies and food frequency questionnaires are conducted intermittently and with variable methodological consistency. Understanding substitution patterns accurately - particularly the relative roles of safe water, fruit juice, and informal beverages - is essential for welfare analysis but currently under-evidenced in the African literature.

These limitations do not invalidate the policy case for SSB taxation in SSA. They do, however, argue for the kind of integrated monitoring frameworks that the evolution of public health monitoring in Sub-Saharan Africa has been building towards - longitudinal surveillance architectures capable of tracking dietary change, metabolic outcomes, and expenditure patterns simultaneously across representative populations. Without such infrastructure, the evidence for or against any dietary fiscal policy in SSA will remain permanently provisional.

The comparative analysis of food security frameworks across Sub-Saharan Africa similarly illustrates how the absence of standardised, continuous national data systems limits the precision with which any dietary policy - from SSB taxation to micronutrient fortification - can be evaluated against its intended outcomes.

Policy Recommendations

The weight of evidence available as of 2026 supports the following conclusions for SSA policymakers.

SSB taxes that are well-designed, transparently administered, and accompanied by parallel public education campaigns demonstrably reduce SSB purchases and - through the reformulation incentive - reduce the sugar content of beverages remaining in the market. The South African, Mexican, and UK experiences all confirm this, with the magnitude of effect varying by implementation model and market structure.

Tiered levy structures - charging higher rates on higher-sugar products, as in the UK SDIL - create stronger reformulation incentives than flat per-litre taxes. Where administrative capacity permits, this design is preferable.

Revenue allocation matters enormously for the distributional legitimacy of SSB taxes. Explicit hypothecation of levy revenue to fund universal health coverage expansion, school meal programmes, or water access infrastructure directly addresses the regressive incidence concern and builds political sustainability for the tax.

Informal market dynamics in SSA demand complementary interventions: front-of-pack labelling, public health communication campaigns, and community-level engagement with small retailers who are both the primary distribution channel and, in many cases, culturally significant health communicators. A fiscal measure operating only on formal channels is, at best, a partial intervention in the SSA context.

Finally, SSB taxes should be positioned as one component of a comprehensive NCD prevention architecture, not a single-lever solution. Their evidence base is strongest when combined with physical activity promotion, sugar content reduction targets in school feeding programmes, and restrictions on SSB marketing to children - measures whose synergies with broader food security policy are increasingly well documented.

Frequently Asked Questions

Has South Africa’s sugar tax actually reduced consumption? Post-implementation evidence from South Africa’s Health Promotion Levy suggests a meaningful decline in SSB purchases - some analyses indicate a reduction of over 25 per cent in the first year - though isolating the levy’s effect from concurrent economic contraction and food price inflation is methodologically challenging. The strongest documented effect has been reformulation: over 30 products were reformulated to reduce sugar content in advance of or following the levy’s introduction, with measurable reductions in total sugar sold through formal retail channels.

Are SSB taxes unfair to low-income consumers? The regressive incidence of SSB taxes in fiscal terms is real: lower-income households spend a higher share of their income on SSBs and therefore bear a proportionately greater tax burden. However, lower-income populations also bear a disproportionate share of the health consequences of high SSB consumption - obesity, T2DM, and dental caries all have higher prevalence among poorer communities. Evidence from Mexico indicates that the largest absolute consumption reductions following a tax were among lower-income households. Whether the net distributional effect is regressive or progressive depends critically on how tax revenues are deployed: recycling revenue into health programmes or social transfers targeting lower-income populations can convert a formally regressive instrument into a net progressive policy package.

Why does the UK experience matter for African policymakers? The UK Soft Drinks Industry Levy is the most thoroughly documented example of how a tiered, reformulation-oriented fiscal levy can drive rapid and substantial reduction in sugar content across a beverage market. The 45 per cent decline in total sugar sold through soft drinks between 2016 and 2020 - the majority achieved through pre-implementation reformulation - demonstrates that the supply-side response to fiscal signals can outweigh the direct demand-suppression effect. For SSA countries considering levy design, the UK model provides a strong empirical argument for tiered structures with a clear reformulation incentive, particularly where administrative capacity for consumer-level monitoring is limited.

What is the strongest argument against SSB taxes in the African context? The most substantive objection concerns the structural limitations on tax reach in SSA markets characterised by large informal retail sectors, limited formal registration of small beverage producers, and widespread use of unpackaged and home-prepared sweetened drinks that fall entirely outside any levy framework. If a tax reaches only the formal market - which may account for 40 per cent or less of actual consumption in some countries - its health impact will be considerably less than projected from formal sales data alone. This argues not for abandoning fiscal measures, but for designing them within a broader package that includes mandatory product registration, consumer labelling, and community-level health communication capable of reaching populations who primarily purchase outside formal retail.

References