Last updated: 10/09/2024
Expand the soft drinks industry levy
Expand the soft drinks industry levy to sweetened milk-based drinks
- Moderate impact on obesity
A percentage estimate of how much the policy would reduce national obesity rates
- Relative reduction in obesity prevalence: 0.6%
- High evidence quality
A rating of the strength of evidence, accounting for both reliability and validity of the evidence
- Reliability and validity rating: 4/5
- Low cost to governments
Cost to UK and devolved governments over 5 years
- Costs to governments over 5 years: £0.6m
- Benefit to governments per year: £0.4bn
What is the policy?
The Soft Drinks Industry Levy (SDIL), informally known as the ‘sugar tax’, applies a levy to the producers and importers of sugar-sweetened beverages based on the sugar content of the drinks. Currently, the levy only covers soft drinks with added sugar. Extending the SDIL to cover sweetened milk-based drinks would broaden the scope of the levy. The aim of this policy would be to encourage manufacturers to reformulate their products and nudge consumers to make healthier choices through price disparities in products.
Like the current SDIL, an expanded levy would likely be applied to producers and importers of the drinks rather than directly to consumers. Companies would then need to decide whether to alter the sugar content to fall below the tax threshold, or to pay the tax and potentially pass on the costs to consumers, retailing drinks at a higher price.
Recent context
As of 2022, 108 countries including the UK, had implemented a tax on at least some kinds of sugar-sweetened beverages. The SDIL in the UK was implemented in 2018 and applies a flat levy of 24 pence per litre (ppl) to drinks with more than 8 grams (g) of added sugar per 100 millilitre (ml), with a lower rate of 18ppl applied to drinks with more than 5g of added sugar per 100ml.
The implementation of the levy led to decreases of the amount of sugar in soft drinks by 2.7% per household per week and decreased incidence of acute childhood tooth decay.
The current levy does not apply to drinks that are:
- at least 75% milk
- a milk substitute
- an alcohol replacement
- made with fruit juice or vegetable juice and do not have any other added sugar
- liquid drink flavouring that’s added to food or drinks like coffee or cocktails
- sold as a powder
- prepared by mixing liquids and served in an open container, like cocktails
- infant formula, follow-on formula or baby foods
- formulated food intended as a total diet replacement, or dietary food used for special medical purposes.
While the SDIL was introduced UK-wide in 2018, the devolved nations have taken different approaches when it comes to taxes on sugary drinks and other efforts to reduce sugar consumption.
In Scotland, the Scottish Government has so far resisted calls to devolve the SDIL and set its own rates. However, it has implemented other measures like restricting the sale of sugary drinks in healthcare facilities. Scotland’s Diet and Healthy Weight Delivery Plan emphasises reformulation over taxation, working with food standards agencies to set sugar reduction targets.
Wales has been more open to utilising fiscal levers like taxation. In 2017, the Welsh Government was the first to vote in favour of a 20% sugar tax on sugary drinks, leading up to the UK-wide implementation of the SDIL.
Amid these differentiated policy approaches, calls have continued from health groups across the UK nations to review and potentially expand the scope of the SDIL to include beverages currently excluded, like juices without added sugar and sweetened milk-based drinks.
Case studies
Danone’s response to the sugar reduction programme
Since the legislation to extend to milk-based beverages is currently not in place, there are no UK-based case studies. However, in response to Public Health England’s 2015-2020 sugar reduction programme for yoghurts and fromage frais, which targeted a 20% drop in sales-weighted average total sugar per 100g, Danone reformulated products and reviewed its yoghurt offerings. This resulted in a 16% sugar reduction across their yoghurt range, exceeding the 13.5% decrease achieved industry-wide.
Despite the programme ending, Danone persisted in reducing sugar volumes in its products. As of 2022, the company realised a 21% total sugar reduction compared to 2015 baselines across its yoghurt ranges.
Moreover, Danone has pledged that at least 90% of its sales-weighted product lineup will not be considered HFSS as defined by the UK Government’s current HFSS regulations. Furthermore, it also aims to achieve a ‘healthy’ rating for over 90% of its portfolio under the Health Star Rating system.
Sugar tax, Norway
Norway had a clear goal of reducing sugar intake by 12.5% by 2021. In line with this, it has had a sugar tax in place since 1992. However, in 2017, the Government increased the tax on added sugar by 83%, taxing products like sweets and chocolates at £3.35 per kilo and on artificially-sweetened drinks by 42%.
Despite the significant tax hike, there was no change detected in volume sales of products. The Norwegian directorate’s annual report revealed that average annual sugar consumption had decreased from 43kg to 24kg per person between 2000 and 2018 – including a 27% reduction in the past decade.
It is to be noted however that the tax has since been repealed but many Norwegian institutions are working to reinstate it.
Considerations for implementation
Clearly defining what qualifies as a ‘sweetened milk-based drink’ and setting appropriate sugar thresholds for taxation levels will be crucial to implementing this policy. One option to increase the proportion of products which are reformulated is to have multiple tiers to the tax rather than one tax threshold, as per the existing SDIL. This could potentially motivate manufacturers with products very far from the targets to reformulate, as the cost of reformulation would be less than the tax. However it could have the opposite effect on those close to the thresholds.
Secondly, the economic impacts, such as increased costs for milk beverage manufacturing industries, potential changes in prices, sales volumes, and consumer purchasing behaviour, as well as administration and compliance costs of expanding the levy system, must be carefully assessed. Additionally, the disproportionate financial burden on low-income families if retail prices rise should be considered.
Thirdly, the implementation and enforcement aspects, including integrating new products into the existing SDIL taxation infrastructure, extending existing compliance monitoring and enforcement mechanisms, potential legal challenges from industry regarding definitions and sugar level limits, and obtaining legislative approval while considering government priorities, will need to be addressed.
Estimating the population impact
We estimated that this policy would reduce the prevalence of adult UK obesity rates by approximately 0.6%
Estimating the per-person impact
We estimated that this policy would reduce average daily calorie intake by approximately 2 kcal per person
We used data published in Dickson et al. (2023) and Griffith et al. (2021). Dickson et al. (2023) estimate that the SDIL removed 6,600 kcals per head per year in the UK. This is the equivalent of 18 kcals from an adult’s daily calorie intake. Griffith et al (2021) reported that the calories from sugar in milk drinks is 0.4%, while that from soft drinks is 1.2% (see Table 2.1 in this IFS report). The ratio of calories from sugar in milk drinks and soft drinks is approximately a third. We assumed then, that if a similar level of reformulation is achieved then we’d expect to see a reduction equivalent to one third that of soft drinks (ie, 18 divided by 3 kcals), resulting in approximately a 6 kcal reduction. And then as a conservative estimate, we took 50% of that (1/2 x 6). So extending the levy to include milk-based drinks would result in an additional ~3 kcals reduction per person per day. We assumed a compensation effect of approximately 23% (ie, 5 kcal) bringing the total calorie deficit to about 2 kcal per person per day.
Estimating the population reach
In our analytical model, we applied the effect sizes to people living with overweight or obesity. For adults, that is people aged 18 or above with a BMI of 25+. Modelling the effects of this policy for children is currently in progress and we plan to publish this once complete.
Changes in the prevalence of people living with obesity
| Adults (England and Wales) | Children (England and Wales) | Adults (Scotland) | Children (Scotland) |
| 0.6% | In progress | 0.9% | In progress |
Cost and benefits
Cost over 5 years
We estimated that this policy would cost the governments approximately £0.6 million over five years
We commissioned HealthLumen to estimate the cost of the policy to both industry and governments over a five-year period.
Table 2 below shows a breakdown of costs. The upfront costs to the government are estimated at approximately £0.6 million over five years for monitoring the implementation of the policy and its enforcement. The costs to the food industry are estimated at approximately £0.1 million over five years. In addition, governments are also likely to raise revenues of ~£97 million over five years. Please see cost calculations appendix for more details.
| Group affected | Cost | Horizon | Detail |
| Costs | |||
| Government | £0.6m | Annual (Five years) | Monitoring costs |
| Industry (Manufacturers) | £0.1m | One-off | Familiarisation costs |
| Government | £97.1m | Annual (Five years) | Revenue raised for treasury |
Total annual benefit
We estimated that this policy would have an annual benefit of approximately £0.4 billion
Using analysis conducted by the Tony Blair Institute and Frontier Economics we estimate this policy would result in benefits of approximately £0.4 billion per year. Approximately two-thirds of this saving would benefit individuals (via quality-adjusted life years, and informal social care). The remaining third relates to savings that benefit the state via NHS treatment costs, productivity and formal social care. See our Methods page for more information about the cost breakdowns. This does not include the revenue gained from businesses who choose not to reformulate and pay the levy.
Impact on disease incidence
We commissioned HealthLumen to report disease incidence avoided if the policy were implemented. These estimates do not represent the total health benefits. The specific diseases selected are those where there is good evidence that living with obesity is associated with the development of the disease.
Table 3 presents a summary of incidence avoided.
| Disease | Incidence avoided |
| Type 2 diabetes | Not statistically significant |
| Hypertension | 3,281 |
| Coronary heart disease | Not statistically significant |
| Colorectal cancer | Not statistically significant |
| Gall bladder disease | 2,673 |
| Ovarian cancer | Not statistically significant |
| Stroke | Not statistically significant |
| Liver cancer | Not statistically significant |
| Depression | Not statistically significant |
| Musculoskeletal disease | Not statistically significant |
Behind the averages: impact on inequalities
As a tax is levied at the same rate regardless of income, a tax on food or drink may have a disproportionate impact on those individuals in lower-income groups. If salt and sugar costs rise due to the levy, these products may become less affordable for those with limited financial resources. This could exacerbate existing inequalities, as studies show higher consumption of sugary drinks among children from disadvantaged backgrounds. While discouraging overconsumption has positive health outcomes, pricing interventions like broader sugar taxes tend to place a greater cost burden on lower socioeconomic groups as a percentage of income.
Rating the strength of evidence
We asked experts working in the fields of obesity, food, and health research to rate the strength of the evidence base for each policy, taking into account both reliability (size and consistency) and validity (quality and content) of the evidence. Policies were rated on a Likert scale of 1–5 (none, limited, medium, strong, and very strong evidence base). The Blueprint Expert Advisory Group rated this policy as having a Strong evidence base.
Introduce healthiness targets for large retailers
Regulate large retailers to change their sales-weighted converted NPM score to ≥ 69 across their entire food product portfolio