The SYNTAX project

Integrated evidence synthesis for joint appraisal of tobacco and alcohol tax interventions for harm reduction in the UK


Aim of the project

To provide evidence and tools to inform the potential scale of effects of taxation across tobacco and alcohol as an intervention to improve public health and reduce health inequalities in the UK.


Why this is important

Alcohol consumption and tobacco smoking are major modifiable risk factors for non-communicable diseases and people who both drink and smoke are at a further increased health risk. Taxation is recognised worldwide as an effective way to reduce alcohol consumption and encourage smokers to quit. However, there is a need to improve the evidence base upon which the joint effects of change to tobacco and alcohol tax will be calculated, e.g. so that the relationships among alcohol and tobacco use and socioeconomic status can be considered.

This research developed new evidence on current and potential future tobacco and alcohol pricing and tax policy in England. There were three key tasks:

  1. Understand the joint patterns of purchasing and use of the products across population subgroups.

  2. Use a consistent methodology to estimate the effects of industry influence on tax pass-through for off-trade alcohol, on-trade alcohol and tobacco.

  3. Produce more detailed evidence on how tobacco and alcohol consumption responds to price rises (the price elasticities of demand) and an understanding of the uncertainty in these estimates. For example, we need to better understand the extent to which consumers might quit smoking vs. switch to cheaper products.

Finally, the research plan recognised the need for impact assessment models capable of assessing the joint effects of tobacco and alcohol tax changes. This involves modelling the potential impact of combined alcohol and tobacco tax policy options on outcomes including changes in participation in drinking or smoking (any use vs abstinence), consumption levels, consumer expenditures, tax revenues to Government and net revenues to retailers, risk of harm related to 85 different clinical conditions, healthcare (NHS) costs, quality adjusted life years and mortality.

For information about the previous research that this project builds on see Partos et al. (2020) for tobacco and Meier et al. (2016) for alcohol.


Objectives

The objectives were designed to produce the evidence required for joint policy analysis of alcohol and tobacco tax policy changes.

  1. Qualitative investigation to understand through published evidence and stakeholder interviews the options for tobacco and alcohol tax policy changes, which options UK experts view as achievable and why.

  2. Analysis of social patterns in alcohol and tobacco spending to quantify the patterns of purchasing and consumption of across product types and price points for socio-demographic groups defined by age, sex, and socioeconomic status and for different smoker and drinker groups.

  3. Analysis of industry and consumer responses to tax changes to estimate the likely responses to tax changes in terms of:

    1. Alcohol and tobacco industry pricing strategy via modifying the extent to which tax changes are ‘passed through’ to product retail prices.
    2. Consumer demand for alcohol and tobacco products by estimating new price elasticities of demand for 10 alcohol product categories and two tobacco products in an integrated way.

  1. Population modelling to appraise the health and economic outcomes of tax interventions - To develop a health and economic model of the combined effects of changes to tax on alcohol and tobacco products, and use this model to appraise scenarios for changes to tax that are designed to support decision-making on tax policy.

Public involvement


The aim of the public involvement in the SYNTAX project was to provide a perspective from lived experience on the options for alcohol and tobacco tax policy changes, and how any changes might affect people differently according to their alcohol and tobacco consumption and socio-demographic characteristics. The idea was that the project team would then use their understanding of the public perspectives to guide the process of generating and interpreting the research findings.

Three public panels were involved:

  • The Sheffield Addiction Recovery Research Panel (support by the Clinical Research Office of Sheffield Teaching Hospitals NHS Foundation Trust). The group comprised former and current service users for drug and alcohol treatment and family and carers of people who have had experience of drug and alcohol misuse.

  • The Tobacco and Nicotine Discussion Group in Nottingham (funded as part of the UK Centre for Tobacco and Alcohol Studies (UKCTAS), and subsequently the SPECTRUM UK Prevention Research Partnership consortium – https://www.ed.ac.uk/spectrum/about/public-involvement). The Tobacco and Nicotine Discussion Group comprised current smokers or those trying to quit smoking who were experienced in discussing tobacco and new nicotine product use, tobacco control policy, approaches to smoking cessation and new developments in tobacco harm reduction.

  • The Alcohol and Food Discussion Group (also funded through UKCTAS and subsequently SPECTRUM) comprised members of the public who were occasional or regular drinkers and experienced in discussing alcohol consumption, public health policy relating to alcohol and related research ideas and current projects.

Each panel was visited by SYNTAX researchers during the project where panel members were guided through a structured discussion of alcohol and tobacco consumption behaviours, how consumption depends on product prices and available income, and their views on tax policy changes and its potential effects. These discussions were influenced by briefing information given to panel members including on the current rates of and revenues from alcohol and tobacco tax in the UK.

  • Click here to download the report of public involvement in the SYNTAX project.

Outputs


1. Options for tobacco and alcohol tax policy changes


We interviewed tax-expert stakeholders, sharing with them a rapid review of contemporary tax changes to elicit current alcohol and tobacco tax options. The stakeholders discussed the rationale for their choices of options. This set of tax changes informed the decision of what tax policy options to model later in the project.

This work identified common objectives between alcohol and tobacco tax policy – reducing harm, improving health and reducing consumption, and a common set of policy options – increasing duty rates, duty escalators, multi-rate tax structures, industry levies and revenue hypothecation.

First, a rapid scoping review of peer-reviewed and grey literature on UK taxation interventions (1997–2018) identified and categorised tax policy options and their underlying objectives.

  • Hatchard J., Buykx P., Brennan A. and Gillespie D. (2023) Options for modifying UK alcohol and tobacco tax: A rapid scoping review of the evidence over the period 1997–2018. NIHR Open Research, 3:26. doi: https://doi.org/10.3310/nihropenres.13379.3.

This review was used to produce a briefing on alcohol and tobacco tax policy options for the UK to give to policy stakeholders ahead of interviews.

Ten policy stakeholder experts (government, public body and advocacy group experts) were then interviewed in pairs, with subsequent transcribing and framework analysis to understand the technical specification of tax options, their policy objectives, and the factors affecting their implementation and outcomes.

  • Hatchard, J., Gillespie, D., Buykx, P. (2019) ‘Using framework analysis deductively: A case study from alcohol and tobacco tax policy and modelling research’, in Clift B, Gore J Bekker S, Costas Batlle I, Chudzikowski K, Hatchard J (eds) ‘Myths, Methods and Messiness: Insights for Qualitative Research Analysis: Edited Proceedings of the 5th Annual Qualitative Research Symposium’, Bath: University of Bath. https://researchportal.bath.ac.uk/en/publications/myths-methods-and-messiness-insights-for-qualitative-research-ana.

  • Hatchard J., Buykx P., Wilson L., Brennan A., Gillespie D. (2019) Mapping alcohol and tobacco tax interventions for health in the UK: a qualitative framework analysis. Poster presented at the Public Health Science conference, London. doi: https://doi.org/10.1016/S0140-6736(19)32848-X. See the poster here.

  • Hatchard J., Buykx P., Wilson L., Brennan A., Gillespie D. Mapping alcohol and tobacco tax policy interventions to inform health and economic impact analyses: A United Kingdom based qualitative framework analysis. International Journal of Drug Policy. 2023;122:104247. doi: https://doi.org/10.1016/j.drugpo.2023.104247.


2. Patterns of purchasing and consumption of alcohol and tobacco products


There was no clear picture of the current joint patterns of spending and consumption across alcohol and tobacco products, despite a multibehaviour approach being critical to reducing health inequalities and to considering the knock-on effects of tax rises to poverty.

To investigate patterns of use of tobacco and alcohol products across the population, we analysed repeat cross-sectional data from the UK Living Costs and Food Survey data (LCFS), correcting for reporting bias in the LCFS by calibrating prices and imputing tobacco pack sizes to match aggregated sales data from Nielsen/CGA.


(a) Spending patterns

How does the proportion of household expenditure allocated to alcohol and tobacco vary by socio-demographics?

This work found that dual purchasing households spent more on alcohol and tobacco than their single-purchasing counterparts, with the lowest income group being most exposed to potential increases in price.

  • Wilson, L. B., Angus, C., Pryce, R., Holmes, J., Brennan, A., Gillespie, D. (2021). Do dual purchasers behave differently? An analysis of purchasing data for households that buy both alcohol and tobacco in the United Kingdom. Addiction, 116(9), 2538-2547. DOI: https://doi.org/10.1111/add.15430


(b) Tax burden and exposure to future tax rises

How does price paid for each product category vary by socio-demographics and by level of smoking and drinking, and how does this translate to the combined distribution of tobacco and alcohol tax in the population?

The LCFS household data was also mapped onto individual consumption data for Health Survey for England participants. This provided individual-level estimates of consumption, spending and tax paid on alcohol and tobacco in England, enabling an investigation of tax burden by population subgroups.

This work found that the 25% of the population who drink more than the Chief Medical Officer guidelines consume 79% of all alcohol and contribute over 70% of alcohol tax revenues, whilst for tobacco it is the lowest income quintile that contributes the most tobacco tax revenues.

  • Wilson, L. B., Pryce, R., Angus, C., Morris, D. J., Holmes, J., Gillespie, D., Brennan, A. (in preparation). The targeted nature of alcohol and tobacco taxes in England: cross-sectional analysis of socio-demographic patterning in consumer expenditure and taxes paid.


3. Alcohol and tobacco industry and consumer responses to tax changes


(a) Industry responses to tax change

To what extent are alcohol and tobacco tax changes passed on to retail prices and how does this vary by product category and price point?

Cheap tobacco and alcohol products are persisting despite tax rises, in part due to industry strategies to maintain low prices. There was an urgent need therefore, for a better understanding of how industry is acting to modify tax pass-through to consumer prices for the three sectors of off-trade alcohol, on-trade alcohol and tobacco.

Building on the approach that Ally et al. (2014) used for off-trade alcohol, we used market research data to develop new evidence on tax pass-through to consumers for on-trade alcohol, and for tobacco.

On-trade alcohol

This work found substantial “under-shifting” (products go up in price by less than one would expect given the tax change) in cheaper beverages with retailers appearing to subsidise this loss in revenue with an “over-shift” (products go up in price by more than one would expect given the tax change) in the relatively more expensive products.

  • Wilson, L. B., Pryce, R., Angus, C., Hiscock, R., Brennan, A., Gillespie, D. (2021). The effect of alcohol tax changes on retail prices: how do on-trade alcohol retailers pass through tax changes to consumers? The European Journal of Health Economics, 22(3), 381-392. https://doi.org/10.1007/s10198-020-01261-1

Tobacco

This work found that most tobacco products are “over-shifted” (go up in price by more than one would expect given the tax change), and that the products at cheapest end of the spectrum are kept low in price.

  • Wilson, L. B., Pryce, R., Hiscock, R., Angus, C., Brennan, A., Gillespie, D. (2021). Quantile regression of tobacco tax pass-through in the UK 2013–2019. How have manufacturers passed through tax changes for different tobacco products? Tobacco control, 30(e1), e27-e32. http://dx.doi.org/10.1136/tobaccocontrol-2020-055931


(b) Consumer responses to price change

How does consumer demand respond to retail price changes for different alcohol and tobacco products?

To understand how tax changes on tobacco and alcohol might impact inequalities then there was also a need for comparable evidence on price elasticities across substances.

We developed new integrated estimates of own-price and cross-price elasticities for twelve products - ten alcohol categories (beer, cider, wine, spirits and ready-to-drinks, split by off trade and on-trade) and two tobacco categories (factory made cigarettes and roll-your-own tobacco). This separates the effect of price on participation in consuming the product from the effect of price on the mean weekly consumption of alcohol / daily consumption of cigarettes given that they are consumed.

  • Pryce, R., Wilson, L.B., Gillespie, D., Angus, C., Morris, D., Brennan, A. (2023) Estimation of integrated price elasticities for alcohol and tobacco in the United Kingdom using the Living Costs and Food Survey 2006-2017. Drug and Alcohol Review. doi: https://doi.org/10.1111/dar.13773.

For alcohol, we compared the modelled effects of using our new estimated price elasticities of demand to the modelled effects of using the previous University of Sheffield estimates by Meng et al. (2014).


4. Modelling for policy appraisal: The potential impact of alcohol and tobacco tax policy changes


What effects do identified tax changes have on consumer spending, consumption, health and inequalities, NHS treatment costs, and Government tax revenue, and how do effects differ by socioeconomics and levels of tobacco and alcohol consumption?

Previous UK population models that have been used to appraise tax have done so separately for tobacco and alcohol. We developed and ran a new version of the Sheffield Tobacco and Alcohol Policy Modelling (STAPM) called the Tobacco and Alcohol Tax and Price Intervention Simulation Model (TAX-sim), which incorporated the above evidence.

  • Morris, D., Brennan, A., Angus, C., Wilson, L., Pryce, R., & Gillespie, D. (2023). Tobacco and Alcohol Tax and Price Intervention Simulation Model (TAX-sim): full technical documentation. The University of Sheffield. URL: https://osf.io/nfa4v. DOI: https://doi.org/10.17605/OSF.IO/KR23Z

The model incorporates the probabilities of initiation, quit and relapse in tobacco smoking as well as participation and levels of alcohol consumption for population subgroups defined by age, sex and quintiles of the Index of Multiple Deprivation. It accounts for recent trends to generate a forecast of future participation and consumption.

The model examines two tobacco product categories (factory-made and hand rolled “roll your own” tobacco) and 10 alcohol product categories (beers, ciders, wines, spirits and ready to drink alcopops, further split by off-trade supermarket/shops and on trade pubs/bars/restaurants).

The analysis simulated the population of England aged 18 to 89 between 2017 and 2050. The main model outcome measures are: change in alcohol and tobacco attributable deaths, hospitalisations, savings in NHS costs, improvement in quality adjusted life years (QALYs), reduction in Years of Life Lost, change in alcohol purchasing and consumption, change in smoking participation including effect on meeting Government’s “smoke-free” target of 5% smoking prevalence in 2030, revenues to retailers (off-trade alcohol, on-trade alcohol, tobacco), duty and VAT revenues to Government, and the change in slope index of inequality between people living in most / least deprived areas of England.

The TAX-sim model was then used to analyse 33 different proposed tax policies in three categories:

  1. Implement one-off duty rises and escalators for alcohol and tobacco

  2. Tobacco escalator, Minimum Excise Tax (MET), and changing tobacco ad valorem policies

    1. Hypothetical alcohol strength tax policies

    2. Additional work was undertaken to estimate the impact of the change to the new strength-based alcohol duty system proposed in the Government’s October 2021 consultation.


Dissemination


Evidence to government (not publicly available)

  • Brennan A., Gillespie D., Morris D., Angus C., Pryce R., Wilson L., Henney M., Holmes J. Modelling potential impact of alcohol duty reform for the UK government consultation in January 2022. Response and modelling report submitted to the government consultation on the new alcohol duty system.

  • Brennan A., Gillespie D., Morris D., Angus C., Pryce R., Wilson L., Henney M., Holmes J. March 2022. Modelling potential impact of tobacco duty change to equivalise the price of factory-made and roll-your-own cigarettes for the Independent Review on Tobacco Policy. Invited advisory report submitted to the Office for Health Improvement and Disparities to inform the government’s independent review of tobacco control policy.

Academic conferences

  • Wilson, L., Pryce, R., Angus, C., Hiscock, R., Brennan, A., Gillespie, D. (2019). The effect of alcohol tax duty changes on on-trade retail prices: How do retailers respond? The Health Economists’ Study Group Winter 2019 Meeting, Centre for Health Economics, University of York (https://hesg.org.uk/meetings/winter-2019-university-york/).

  • Wilson, L., Angus, C., Pryce, R., Brennan, A., Gillespie, D. (2019). Socioeconomic variation in the relationship between household spending on alcohol and tobacco and overall household expenditure in England. Kettil Bruun Society conference, Utrecht, Netherlands (https://www.kbs2019utrecht.nl/).

  • Hatchard J., Buykx P., Wilson L., Brennan A., Gillespie D. (2019) Mapping alcohol and tobacco tax interventions for health in the UK: a qualitative framework analysis. Poster presented at the Public Health Science conference, London. doi: https://doi.org/10.1016/S0140-6736(19)32848-X. See the poster here.

  • Hatchard, J., Gillespie, D., Buykx, P. (2019) ‘Using framework analysis deductively: A case study from alcohol and tobacco tax policy and modelling research’, in Clift B, Gore J Bekker S, Costas Batlle I, Chudzikowski K, Hatchard J (eds) ‘Myths, Methods and Messiness: Insights for Qualitative Research Analysis: Edited Proceedings of the 5th Annual Qualitative Research Symposium’, Bath: University of Bath. https://researchportal.bath.ac.uk/en/publications/myths-methods-and-messiness-insights-for-qualitative-research-ana.


Funding

NIHR Public Health Research Board – project number 16/105/26. https://www.fundingawards.nihr.ac.uk/award/16/105/26


References

Ally, A. K., Y. Meng, R. Chakraborty, P. W. Dobson, J. S. Seaton, J. Holmes, C. Angus, et al. 2014. “Alcohol Tax Pass-Through Across the Product and Price Range: Do Retailers Treat Cheap Alcohol Differently?” Addiction 109 (12): 1994–2002. https://doi.org/10.1111/add.12590.
Meier, Petra S., John Holmes, Colin Angus, Abdallah K. Ally, Yang Meng, and Alan Brennan. 2016. “Estimated Effects of Different Alcohol Taxation and Price Policies on Health Inequalities: A Mathematical Modelling Study.” Journal Article. PLoS Medicine 13 (2): e1001963. https://doi.org/10.1371/journal.pmed.1001963.
Meng, Y., A. Brennan, R. Purshouse, D. Hill-McManus, C. Angus, J. Holmes, and P. S. Meier. 2014. “Estimation of Own and Cross Price Elasticities of Alcohol Demand in the UK—a Pseudo-Panel Approach Using the Living Costs and Food Survey 2001–2009.” Journal of Health Economics 34: 96–103. https://doi.org/10.1016/j.jhealeco.2013.12.006.
Partos, Timea R, Rosemary Hiscock, Anna B Gilmore, J Robert Branston, Sara Hitchman, and Ann McNeill. 2020. “Impact of Tobacco Tax Increases and Industry Pricing on Smoking Behaviours and Inequalities: A Mixed-Methods Study.” Public Health Research 8 (6): 1–174. https://doi.org/10.3310/phr08060.