Taxing Tobacco: the intervention that got away (happy World No Tobacco Day)
TL;DR
The death toll of tobacco dwarfs traditional EA global health focuses (e.g. Malaria).
Taxing tobacco is the most effective and yet most neglected form of reducing tobacco consumption.
Multiple EA orgs have ran cost-effectiveness numbers that would put a tobacco taxation NGO on GiveWell’s list (both in expected value and contingent on success) and even at the top in some cases.
Despite this, no enduring tobacco-taxation advocacy organization has emerged from EA (which would be the only advocacy organization in the world exclusively dedicated to tobacco taxation)
Join Joel Tan (Founder of CEARCH) and J.T. Stanley, who as incubatees of Charity Entrepreneurship closely examined tobacco taxation, on June 10th (Saturday) at 14:00 UTC for a virtual discussion about the intervention and what’s next to make it a reality.
Thanks to Moritz von Knebel for providing feedback on the draft.
Yes, I cite WHO a lot. Not all WHO citations are the same WHO link FYI.
Problem space
Tobacco kills over 8 million individuals a year—that’s 13x Malaria (WHO) (CDC).
Another way of framing it: annually more people are killed by tobacco usage than malaria, HIV, and neonatal deaths combined… twice over. And while the death toll of the latter three has been decreasing over time, death from tobacco is increasing.
Of the 8 million deaths, 1.2 million are bystanders killed from secondhand smoking (WHO).
Facts not related to death:
Tobacco consumption displaces productive forms of spending. Tobacco consumes 1.5 to 17% (with a rough median around 4.5%) of a person’s income depending on country and socioeconomic status (table of results) (de Beyer et al., 2001). Spending on tobacco typically displaces spending on education and nutrition in low-income households (John, 2008) (Nonnemaker and Sur, 2007) (Pu et al., 2008).
For example, smoking households spent 46% less on education than non-smoking households in surveyed townships in rural China (Wang et al., 2006). Another study states, “Average male cigarette smokers [in Bangladesh] spend more than twice as much on cigarettes as per capita expenditure on clothing, housing, health and education combined” (Efroymson et al., 2001).
Smoking decreases productivity (Tobacconomics, 2019) (Halpern et al., 2001). One study in the United States found that it cost 1,807 USD per year per smoker in lost productivity compared to non-smokers (Bunn III et al., 2006).
Tobacco increases individual healthcare costs (Tobacconomics, 2019). A study in China found that individual medical spending attributable to smoking increased the poverty rate by 1.5% in urban areas and 0.7% in rural areas (Liu et al., 2006).
Tobacco strains the healthcare sector and taxpayers foot the bill (Wunsch and Brodie, 2021) (Tobacconomics, 2019).
The annual costs of tobacco from healthcare expenditures and losses in productivity are 1.8% of the world GDP, breaking down to 1.1% to 1.7% of LMICs’ GDPs (Goodchild et al., 2018) (Tobacconomics, 2019).
One last fact: without intervention, tobacco is on track to kill a billion individuals during the 21st century (WHO, 2008) (Savedoff and Alwang, 2015) (Jha, 2012).
The intervention
The scientific literature on tobacco consumption is lengthy. The interventions that have the most significant effect on reducing tobacco consumption have been formalized in WHO’s MPOWER framework (Kaleta et al., 2009).
Of the MPOWER tobacco control measures, taxing tobacco has been demonstrated consistently in the scientific literature to the be the most effective intervention (WHO) (NIH).
Taxing tobacco has a price elasticity around −0.5, meaning that for a 10% increase in retail price of tobacco, consumption decreases by 5% in LMICs (4% in HICs) (NIH) (WHO)[2]
Despite being the most effective intervention, tobacco taxation is the most neglected (WHO) (World Bank).
Cost effectiveness
Giving What We Can explored a few estimates of the impact of reducing tobacco consumption and found it is plausible that the cost per life saved could reach 800 USD (Report) (Blog post)
Charity Entrepreneurship has calculated an expected value of 51 USD per DALY averted by a tax increase of 23.7% in Mongolia with 27.14% chance of success. (Spreadsheet)
Open Philanthropy’s BOTEC on Indonesia found 30 USD per DALY averted in expectation with a 10% probability of success, million-dollar campaign, 10% increase in retail price, and a counterfactual speed up of 3 years. (Spreadsheet)
Moritz von Knebel and I, Charity Entrepreneurship incubatees, found (Spreadsheet):
A policy victory would typically yield a DALY averted for single-digit dollars (9.49 and 1.01 USD in Nepal and Iran, respectively)[3] and even just cents in some cases (0.49 USD in Canada)[4] for just charity costs.
When accounting for government costs, the range was single digits (6.62 USD in Canada) to low three digits (155.67 USD in Mongolia)[5]. Estimates were most worsened (i.e. Kuwait, Mongolia) if we assigned a high probability that the government would increase the tax rate soon in the future independent our intervention.[6]
Accelerating Nepal enacting a tax increase of 24% by little more than half a year has an expected value of 59 USD per DALY averted with a probability of success of 27.5%.
EA’s journey with tobacco taxation
In 2013, Open Philanthropy conducts a shallow investigation of tobacco taxation in LMICs.
“A philanthropist could undertake any of a variety of strategies to attempt to ensure passage and enforcement of such [tobacco taxation] policies. We do not have a good sense of the likely returns to the different potential strategies.”
“Our research in this area has been relatively limited, and many important questions remain unanswered by our investigation.”
In 2015, Giving What We Can releases a report and blog post on tobacco control heavily featuring tobacco taxation.
“At the beginning of this post, I asked whether tobacco taxes can really claim to be one of the most—or even the most—cost-effective way of saving lives in the developing world. I think we have good reason to think they are, though we should ask for answers to the above questions—or at least better answers than I’ve been able to give.”
“Tobacco taxes appear to be a highly cost-effective way to save lives in the developing world. I suspect that further research will show that well-run tax advocacy organizations can save lives as cheaply as the charities currently touted by organizations like GiveWell and Giving What We Can. For now that research remains to be done.”
In 2016, Charity Entrepreneurship does shallow research on tobacco taxation.
“The evidence for lobbying succeeding in LMIC is far weaker than the evidence for the impact of successfully implemented tobacco taxation. It’s often difficult to credit responsibility for legislative change to particular players in advocacy campaigns because multiple campaigns operate simultaneously and it’s unclear whether the government would have changed anyway, absent lobbying efforts. While there is considerable evidence the government implementation of tobacco taxation is highly cost-effective, there is a dearth of analysis on the return of investment of actual lobbying campaigns.”
In 2017, GiveWell conducts three interviews on tobacco control. (The Union) (Campaign for Tobacco-Free Kids) (Author of the “The Single Best Health Policy in the World: Tobacco Taxes”)
In 2017, Charity Entrepreneurship recommends tobacco taxation as one of four recommended ideas in their first-ever charity incubation.
In late 2019, two tobacco-taxation charities emerge from Charity Entrepreneurship’s second-ever incubation program.
In 2020, both Charity Entrepreneurship tobacco-taxation charities shutdown largely due to COVID ramifications.[7] (Good Policies) (Policy Entrepreneurship Network)
In 2020, GiveWell conducts three interviews on tobacco control. (Vital Strategies) (Founder of Good Policies) (International Development Research Centre)
In 2022, Charity Entrepreneurship updates their research report on tobacco taxation and recommends it to another incubation round.
“There is very strong evidence linking tobacco to many negative health outcomes… There is also very strong evidence that increasing tobacco taxes reduces tobacco consumption… Our own cost-effectiveness analyses have also shown that tobacco taxation is an extremely cost-effective intervention… There is strong evidence that this intervention can be effectively delivered by the government… Regarding the probability of success of advocacy for this intervention, looking at a total of 159 case studies, we see an average success rate of ~27.14%.”
“Despite the promise of this intervention, there are some concerns. There are two main funders in the space, but they favor large, well-established organizations, so funding might be a limiting factor… Another concern about this intervention is neglectedness. There are a few organizations working in this space that appear to be quite successful, are well-funded, and are targeting high-burden countries. The key strategy might lie in focusing on countries in which other organizations are not currently working (which might not be the most burdened countries, but still high-burden).”
In 2023, Open Philanthropy publishes another shallow investigation on tobacco control featuring tobacco taxation.
“While these numbers are all very uncertain (and rely on several questionable assumptions), the main takeaway is that cigarette tax advocacy can be very valuable but probably only clears the 1,000x bar if it affects a large smoking population or can be implemented cheaply with a high probability of success.”
Where do we go from here
In my (current) opinion, the best reason to be pessimistic about this intervention is that it could be absurdly hard to beat the tobacco lobby. The tobacco control space receives substantial funding and has multiple established INGOs. The fact that tobacco taxation is neglected could be a reflection that tobacco control orgs have already reached cost-effectiveness optimization, and its settling for other tobacco control concessions short of tobacco taxation because the tobacco lobby will fight the hardest against taxation.
Yet, there are reasons to think that the organizations in the tobacco control space have not reached optimization and that the space is undersaturated (see this graph for example). But to be fair, I’ve only seen consistent indicators that the tobacco lobby will be aggressive in every circumstance possible.
Regardless, these hypotheses have not yet had the chance to be tested, and neither has a tobacco-taxation organization guided by the effectiveness mindset.
Join Joel Tan, Founder of CEARCH, and myself for a discussion of the promise of tobacco taxation and pathways forward.
June 10th (Saturday) 14:00 UTC
[1] Being in opposite time zones played into this.
[2] Results for HICs are tightly clustered whereas LMICs have more variation.
[3] Conditions: Nepal – 24% increase in tax, charity runs 4.75 years; Iran – 27.5% increase in tax, charity runs 6 years
[4] Conditions: Canada -- 25% increase in tax, charity runs 4 years
[5] Conditions: Mongolia -- 20% increase in tax, charity runs 5.5 years, 82.5% chance per year the government would enact that increase on their own
[6] The benefits of the intervention are current discounted over time by the odds per year that the government would enact the increase on its own.
[7] Also worth noting that seed grants for these orgs were 5X less than what CE seed grants currently typically amount to.
- Local charity evaluation: Lessons from the “Maximum Impact” Program in Israel by (27 Aug 2023 13:19 UTC; 110 points)
- [JOB] Opportunity to found Charity Entrepreneurship NGO (outside of the incubation program): Tobacco taxation advocacy by (31 Jul 2023 15:38 UTC; 59 points)
- CE-incubated tobacco & NCD policy Charity: updates, funding gap, and future plans for Concentric Policies by (29 Dec 2023 6:56 UTC; 54 points)
It might seem daunting for Global Health people to confront a big lobby while there are seemingly easier interventions with less resistance around.
From the vantage point of an animal activists it’s not that daunting though. I am relatively certain that it’s easier than e.g. to go against the factory farming lobby. At least it’s so much common sense that it should be possible to convince the rational parts of government—increasing both public health and tax returns. Concerns arise around corruption but if you encounter too much of that you could simply switch intervention countries.
Looking at scale and neglectedness, tobacco taxation seems just too important to shy away from the challenge.
Hi, J.T.,
As shown by the research conversation notes you link to, GiveWell lightly explored funding tobacco control policy advocacy in the past. Ultimately we decided to focus on other policy areas (alcohol policy, lead exposure, and self-harm from pesticide ingestion) that appeared more neglected. You can read more about our public health regulation research on this page—see “Cause areas we investigated at a shallow level and deprioritized,” and this spreadsheet, linked from footnote 24, which gives our estimates of how much is spent on tobacco policy (and other causes we looked into) per unit of burden.
Thank you for the thought-provoking post!
Best,
Miranda Kaplan
GiveWell Communications Associate
Thanks for those links; very interesting.
Is there a reason financial burden was only calculated for lead? This seems very applicable to alcohol and tobacco (as well as others on the list) which have studies (like ones I mentioned) that quantify things like lost productivity, income displacement, individual and societal healthcare costs, etc?
(not saying it would have changed GiveWell’s prioritization, but it seems like relevant info)
Hi, J.T.,
Sorry for the delay in getting back to you!
In an ideal world, we would have included financial burden for tobacco in these rough calculations (which were mostly intended to narrow down what we’d focus on in our initial foray into public health regulation grants). But essentially, the last line of your response is the answer: because we value children’s lives so highly, we expected that incorporating the financial burden of tobacco use would not be enough for us to prioritize it over lead. Lead specifically affects young children, and the gap in estimated spend per unit of burden ($0.07 for lead versus $0.24 for tobacco) is large. If we were doing a deeper investigation of tobacco policy as a possible funding area, we would aim to more thoroughly account for its harms.
Best,
Miranda
Great post J.T.! A few things I would like addressed for me to buy this more.
Effect on the healthcare system. Tobacco smoking probably saves a lot of healthcare dollars… by killing the person a bit earlier and thus they consume less healthcare over their life (particularly when they are no longer contributing to the tax base. https://www.kff.org/report-section/the-rising-cost-of-living-longer-introduction/). This isn’t super well accounted for in many sources
If they aren’t included in CEAs, I’d like to see some analysis done on the increase in black market tobacco sales that are likely. If you increase taxes enough, this often results in black market increase which I expect to be more prevalent in LMICs.
I would like to see an account of deaths that are caused by smoking vs. a contributing cause. The WHO and CDC will often fudge/juice numbers to advocate for certain outcomes. I expect this to be difficult to do.
Factor in that a death from smoking will be from someone who is older than someone who died of malaria (this is factored into DALYs but not the GWWC report for “lives saved”)
This will probably be minor but factor in the poorer economics of the country from the higher taxation.
Here are some quick things and not a full response:
That’s what you would think but I was actually reading today (link) that it has gotten more costly which has changed the calculus for some insurance companies and made them willing to divest from tobacco. Here is the trend: lung cancer used to be a quick death, but medical advances in palliative care now give patients the opportunity to live much longer whilst regularly undergoing intensive medical procedures (e.g radiation treatment)
So I think it’s an okay question on if black market effects have a significant enough effect that it should be modeled into the CEA. However, this is a different question than if there is reason to believe that black market effects would knock the intervention’s cost-effectiveness down a whole tier, which I believe the answer is no. The experts have been clear that no black-market effects come close to offsetting the benefits of reduced consumption from taxation. Also, taxes/price are not the primary determinant of black market (Savedoff citing others, 2015).
We could imagine a situation where the effects of the black market would not conform to those studies e.g. 300% price hike in a state with no import controls. However, I don’t think an EA tobacco taxation, or any other serious tobacco control org, would put itself in that position.
I think the concerns about black market effects are gonna be consequential, not in the worthiness of the intervention, but in which state to pursue. For example, we know that two variables that affect the prevalence of black market is the price differential between bordering states and how effective a state’s import controls.
If I had more time, I’d write about the value of getting these laws on the books in countries in SSA that are weaker states (but will hopefully develop state capacity over time) whilst tobacco companies aren’t as focused there and before these get a new generation hooked like they have in other emerging markets e.g. SE Asia
So if I understand you right, disentangle tobacco deaths from things that can be double counted like a smoker that died from COVID, which might get marked in both camps?
Yah I hear you on this. The GWWC number is exciting, but the intervention should be proven on a DALY standard, which I think it has notwithstanding someone finding a big hole in it.
I am not sure what you mean by this one. If you mean concern about it being a regressive tax, then I’d agree (more so than the black market concern) that this should be in a CEA. However, I again don’t expect it to drastically change the cost-effectiveness. Poorer people are more price sensitive and disproportionately reap more benefits from reduced consumption due to taxation than wealthier classes. I think the Liu 2006 study observes what is intuitive to us all—tobacco consumption increases poverty across a society.
If you are talking about the economy in general, then you’d need to elaborate more. Aside from tobacco being a substantial strain on the economy, the taxes on it generate great revenue for poorer governments.
Case study: Columbia tripled their tobacco tax from 2016 to 2018 and over that time consumption of tobacco fell by a third and their national health service doubled its funding with that tax.
I realize these thoughts were not as quick as I planned. But I hope that’s useful for a first pass. Thanks for bringing those points up
Great response J.T. This was me saying “I want these things thought about in a CEA that I suspect might not have been.” I actually think this is quite promising.
For 1, divestment in this case means nothing IMO. Divestment is mainly a signalling thing. It doesn’t really do anything to reduce stock price for example. https://forum.effectivealtruism.org/topics/divestment
For 2, I don’t think this is something to just dismiss without calculations
For 3, yes. If someone died of coronary heart disease, there was a decent chance that they would get it without smoking. Or they had lung cancer and diabetes, there needs to be some factor you attribute to each, not just “they died of lung cancer from smoking” which is what happens in these organizations.
For 5, I’m concerned generally that LMIC have corrupt governments who, when they get more money, is actually net negative. I’m concerned about this as well in developed countries if they use it to start wars, etc. as well.
On the whole, I think this is still likely to come out as a strong intervention. I’m still more skeptical of policy changes and I think we likely over-estimate their odds of success but this still seems well worth trying.
There is a net financial effect on public expenditures due to premature death from smoking, but it’s overwhelmingly to the old-age pension system, not to the healthcare system. It’s possible this is different in LMICs, though.
https://bmjopen.bmj.com/content/2/6/e001678
It seems unlikely that the decrease in tobacco-related harm is strongly proportional to the reduction in tobacco consumption. I base this on an article in Circulation about the adverse health effects of light or intermittent smoking. In particular:
(footnotes omitted).
Reduction in consumption can be both an individual and societal measure and measured via both outright quitting and reduction of cigarettes consumed per day (to a non-zero number). My focus here is on quitting at the societal level and I have been basing it on using price elasticity to measure taxation’s effects on population prevalence of smoking, which would be unaffected by your concerns.
However, I dug deeper into the mechanics of these price elasticities numbers, and it appears some can be measures of cigarette sales (i.e. general reduction of consumption) and others measures of smoking prevalence in the population (i.e. quitting or not initiating).
I am still working my way through this information, so I don’t have a full informed view yet, but here (at the bottom of this comment) are some things that caught my eye in the NIH’s Tobacco Control Monograph’s section on “Using Price Elasticity Estimates to Project the Future Impact of Tobacco Tax Increases.”
This has important implications because:
Charity Entrepreneurship’s model (which mine and Moritz’s is based off of) translates price elasticity into a reduction of consumption which is then translated 1:1 for reduction in prevalence.
Open Philanthropy’s model (which also uses hinges on price elasticity) assumes the reduction in consumption has 1:1 effect with reduction in DALYs.
So the CE model is solid so long as the median price elasticity used (for LMICs) reflects the average of price elasticities calculated on prevalence.
Both models are vulnerable regardless to overestimating the DALYs averted from prevalence change if the people that go onto quit/not initiate were disproportionately less of the DALY burden. A reason to believe that was the case is that smokers that smoke less cigarettes are more likely to quit. (This though is also reason to be less concerned with your original point about the nonlinear association between harm and consumption i.e. even if they don’t quit, taxes put them in a closer position to quitting in the future)
However, buffering the above for the CE model is that it’s not capturing DALYs averted for reduced consumption that is short of quitting—a source of underestimating the impact.
The Open Phil model has issues regardless because it applies reduction in consumption to 1:1 reduction in DALYs without mapping out quit/prevalence rates, which as you rightly point out are likely not interchangeable with just reducing the non-zero amount you smoke each day.
And both models might have significant overestimation if their price elasticity numbers are reflections of cigarette sales instead smoking prevalence.
Quotes of interest from the monograph
These and other projections generally start with the overall price elasticity of cigarette demand, typically obtained from econometric estimates based on tax-paid cigarette sales data, and assume that the tax increase being modeled will be fully passed on to consumers in the price paid for cigarettes.
It is also important to use both adult and youth prevalence elasticities in projecting the public health impact of a cigarette tax increase because youth and adults are not equally price sensitive.
Similarly, adult and youth prevalence elasticities are used in projecting the public health impact of a cigarette tax increase. Given the available estimates, many of these projections assume that the impact on adult prevalence is half of the overall elasticity and that youth uptake of tobacco use is two to three times as responsive to price. When projecting the impact on youth, these projection models often assume that young people will take up smoking at the same rate as adults or young adults have. Continuing the example above, if the state has one million adults and adult smoking prevalence is 20% (200,000 adult smokers), the US$ 1.00 tax increase will reduce the prevalence of adult smoking by 4%, or induce 8,000 adult smokers to quit smoking. If there are 500,000 young people in the state, and it is assumed that they will take up smoking at the same rate as adults have and youth smoking is twice as sensitive to price, then the US$ 1.00 tax increase maintained over time in real terms will prevent 8,000 young people from taking up smoking.
Econometric studies of the impact of tax and price on tobacco use employ two primary measures of tobacco use: (1) macro-level aggregate measures of consumption, such as country-level data on tobacco sales (this literature developed earlier, growing rapidly before the 1990s); and (2) household or individual-level data taken from surveys, such as national surveys of drug use or health risk behavior.
Additionally, studies have assessed the impact of tax and price on specific outcomes, such as prevalence of tobacco use, smoking cessation, initiation of smoking by youth, cross-price elasticity, and health outcomes.
I’m very excited about this intervention! Look forward to the discussion.
HI!
I have doubts about the following two points, I would love to hear your point of view.
1. Tobacco taxation leads to the diversion of significant sums of money to the State budget. We have no control over how these resources will be used, whether they will produce more good or evil.
2. For poorer households, tobacco taxes may not be significant enough to give up smoking, and the only thing they will lead to is a reduction in the budget for other more important areas (healthy eating, spots, drugs, etc.)