The majority of GiveWell's work focuses on identifying quantifiable, evidence-based giving opportunities that directly deliver goods and services to people, or provide technical assistance that supports government provision of goods and services.
In February 2019, we wrote a blog post about how GiveWell's research is evolving by exploring more "highly leveraged" giving opportunities. That is, opportunities that could have a disproportionate impact by influencing high-level decisions (e.g. advocacy for regulatory changes). The objective of this work is to increase the overall impact of our portfolio by finding more cost-effective giving opportunities, some of which may not meet our traditional criteria.
Our initial work to expand to new areas has focused on identifying opportunities to help governments implement evidence-informed public health regulations. This page is an update on our preliminary conclusions and grants we recommended.
We haven't written up our conclusions in the level of depth that we would for our top charities, but we wanted to publish a brief research note on our current thinking to ensure we're being transparent about our decision-making.
Published: August 2021
Summary
- Why public health regulation? Our work to date has focused on opportunities to help governments to implement evidence-informed public health regulations. This work could be highly cost-effective, and we believed that our experience evaluating health interventions would help us evaluate public health regulations more easily than other high-leverage areas. (More)
- Research conclusions. There are three areas we think are particularly promising: alcohol policy, pesticide regulation, and reducing lead exposure. Each of these areas receives limited philanthropic attention relative to the size of the burden it imposes, and can be at least partially addressed through a set of evidence-informed regulatory interventions. (More)
- Grantmaking (as of August 2021). We've recommended approximately $17 million in grants to four organizations working on public health regulation: (More)
- $8.5 million over two three-year grants ($1.3 million in August 2017 and $7.2 million in January 2021) to the Centre for Pesticide Suicide Prevention (CPSP), to help governments identify pesticides commonly used in suicide and advocate for their deregistration.
- $250,000 over three years to IPEN in January 2020 to advocate for lead paint regulation in Southeast Asia.
- $100,000 to Vital Strategies in January 2021 to develop a proposal for a $5-15 million three-year global alcohol policy program.
- $8 million over three years to Pure Earth in July 2021 to work on identifying and mitigating sources of lead exposure, including contaminated spices.
- Cost-effectiveness of our first grant. We believe that CPSP's work between 2017 and 2019 led to deregistration of pesticides accounting for approximately 55% of suicides in Nepal. We believe these regulations will prevent hundreds of deaths each year, although that will rely on the extent to which they are well-enforced, and which pesticides are substituted. We expect to find out the extent to which these regulations reduced suicides in the next 2-3 years. Our estimate suggests that our first grant to CPSP was substantially more cost-effective than additional funding to our top charities, although this estimate relies on a number of particularly uncertain assumptions that make us uncertain about this comparison. We do not yet have cost-effectiveness estimates for other grants we recommended. (More).
- Possible future research in public health regulation. There are four other areas we identified that may be promising areas for further research but are less clearly promising than the above three: mandatory micronutrient fortification, ambient air pollution, excessive salt consumption, and providing marginal funding to areas like tobacco control that already receive substantial funding. (More)
Table of Contents
Why public health regulation?
We started our exploration of high-leverage areas by investigating opportunities to contribute to improved public health regulations. Regulatory interventions have not been a focus for GiveWell in the past, but they are an important tool through which governments can affect population-level health.
We decided to focus on work to improve public health regulations because:
- We had the impression that certain regulations have a particularly strong track record of improving public health.1
- We believed that advocating for, and assisting in the design and implementation of public health regulations could plausibly be one of the most effective places for high-leverage philanthropy because of its potential long-term effect on health across entire populations.
- The emphasis on health and evidence-backed interventions meant that evaluating opportunities to improve public health regulation was adjacent to our current work, and therefore it seemed like a tractable place to start.
The downsides of working in public health regulation include:
- We are unsure how tractable it will be in any particular case to fund activities that will lead to policy changes in low- and middle-income countries. Powerful industry interests or inertia within government may make this challenging.2 In our rough cost-effectiveness models for improving different public health regulations, we rely on our best guesses of the chance of affecting regulation, which are inherently difficult to estimate. As a result, our models could substantially overestimate or underestimate cost-effectiveness in any given case.
- The costs and benefits of regulation are diffuse and can be difficult to estimate. For example, the enjoyment benefits of alcohol are highly subjective, and regulating highly hazardous pesticides may have an effect on agriculture, accidental poisonings, and ecology.3
- Our understanding is that regulation is often poorly followed and/or enforced, which would reduce the expected impact of regulatory changes.4 We expect that, for many regulations, supplementary activities will be required to make it more likely that the regulation achieves its goals.
For each of the grants we've recommended, we considered the positive case for the grant to outweigh these downsides.
Process
How did we identify grantmaking opportunities?
We were looking for problems that:
- (a) impose a large burden of disease
- (b) do not receive a high level of philanthropic funding
- (c) could be plausibly addressed through a set of evidence-informed regulations
We treated these criteria as proxies for ultimate cost-effectiveness, although we investigated questions we believed were important and specific to particular areas rather than sticking to these criteria closely.
The funnel included six stages. Stages 3-6 did not always occur sequentially.
- Identifying potential cause areas. We identified a list of potential cause areas by reviewing data on risk factors and causes of death that we believed might be addressable by regulatory solutions.5
- Narrowing the initial list. We selected nine of those areas based on a shallow review of the size of the burden and desktop research to understand whether regulatory interventions existed.
- Assessing importance, tractability, and neglectedness. For each of the nine areas, we tried to get a deeper understanding of: (i) the size of the burden, (ii) how much philanthropic spending there was in each area, (iii) potential regulatory interventions, (iv) which countries the disease burden was concentrated in, (v) whether there were potential funding opportunities, and (vi) the major open questions that we should investigate more deeply. We have not yet written up our initial investigations on each of the nine cause areas to a publishable standard.
- Reviewing key questions for the most promising areas. We investigated three areas that looked particularly promising in more depth: pesticide suicide, lead exposure, and alcohol policy.6
- Mapping the organizational landscape for promising areas. We spoke with potential grantees in the space and mapped other funders more carefully.7
- Investigating specific funding opportunities. We recommended grants to CPSP for work on pesticide suicide, Vital Strategies for work on alcohol policy, and IPEN and Pure Earth for work on reducing lead exposure (see more below).
Research conclusions and grantmaking
Research conclusions
Summary
- Alcohol policy stands out as imposing a very large health burden and receiving little philanthropic funding.8 We haven't yet completed cost-effectiveness analyses for opportunities in alcohol policy. This analysis could change our view on whether this is a promising area for further grantmaking.
- Pesticide regulation has the advantage of being a regulatory intervention with a potentially large effect size, although it addresses a smaller disease burden than alcohol policy.9
- Reducing lead exposure seems promising based on the potentially large negative effects that lead exposure may have on child development and because of its relative neglectedness.10 However, we have some remaining questions on the relative importance of various sources of lead exposure, and we plan to more deeply review evidence on the effect of lead on mortality, morbidity, and child development outcomes.
Analysis
See this spreadsheet for our complete analysis, including the calculations presented in the following tables.
The table below shows IHME estimates of the number of deaths, and morbidity burden (in terms of years lived with disability, or YLDs)11 in low- and middle-income countries for different cause areas we investigated. To compare the cause areas, we converted these estimates into GiveWell's moral weights, with each unit equal to the value of doubling economic consumption for one person for one year.
This means that, for example, we would weigh the benefits of preventing all deaths and disability from alcohol for one year equally to doubling the economic consumption of approximately 132 million people.
| Problem | Mortality burden (millions of deaths) | Morbidity burden (millions of YLDs) | Mortality burden (millions of units of value) | Morbidity burden (millions of units of value) | Total burden (millions of units of value) |
|---|---|---|---|---|---|
| Alcohol use | 2 | 13 | 103 | 30 | 132 |
| Pesticide self-harm | 0.2 | -12 | 11 | - | 11 |
| Exposure to lead | 1 | 4 | 23 | 9 | 9413 |
| Diet high in sodium | 2 | 4 | 54 | 8 | 62 |
| Tobacco | 7 | 25 | 230 | 58 | 288 |
| Transport injuries | 1 | 16 | 93 | 36 | 130 |
| Drug use | 0.3 | 8 | 21 | 19 | 40 |
| Micronutrient deficiencies (iron, zinc, vitamin A) | 0.1 | 29 | 11 | 107 | 118 |
| Ambient air pollution | 4 | 12 | 135 | 27 | 162 |
The table below shows our estimates of the amount of philanthropic funding spent on each of the cause areas relative to their burden. Based on this analysis, alcohol use, pesticide self-harm, and exposure to lead each appear to be highly neglected relative to their burden.14
| Problem | Total burden (units of value, millions) | Annual philanthropic spend ($ millions) | Spend per unit of burden ($ millions) |
|---|---|---|---|
| Alcohol use | 132 | $3 | $0.02 |
| Pesticide self-harm | 11 | $0.4 | $0.04 |
| Exposure to lead | 94 | $7 | $0.07 |
| Diet high in sodium | 62 | $6 | $0.10 |
| Tobacco | 288 | $70 | $0.24 |
| Transport injuries | 130 | $32 | $0.25 |
| Drug use | 40 | $10 | $0.25 |
| Micronutrient deficiencies (iron, zinc, vitamin A) | 118 | Unclear | Unclear |
| Ambient air pollution | 162 | Unclear | Unclear |
The above analysis was intended to give a rough approximation of which cause areas stood out in terms of importance and neglectedness, rather than providing a fully rigorous evaluation. In particular:
- We were not consistent across different cause areas when estimating the burden attributable to effects on child development. We included an estimate for lead exposure because we believe that its effect on child development is a significant part of the case for lead control, but we have not fully evaluated child development effects in other areas.15 We believe this inconsistency is unlikely to change our qualitative conclusions on which areas are most neglected relative to their burden, but our estimates should be interpreted as rough and directional.
- Calculations of philanthropic spending rely on a number of judgment calls, and so we view them as indicative but not conclusive. More information on these estimates is available here.
Due to the uncertainty around this analysis, we did not consider the differences in spend per burden between lead exposure and sodium consumption to be meaningful. We prioritized further work in lead exposure because efforts to reduce lead exposure in the USA appear to have been more successful than efforts to reduce salt consumption. We take this as suggestive evidence that reducing lead exposure in low and middle income countries will be more tractable than reducing salt consumption.16
Grantmaking
Pesticide suicide - Centre for Pesticide Suicide Prevention
In August 2017, we recommended a grant of $1.3 million over two years to the Centre for Pesticide Suicide Prevention to collect data on which pesticides were commonly used in suicide in India and Nepal, and to advocate for the deregistration of those pesticides.
We believe CPSP's efforts have contributed to policy change. Its work led to the deregistration of aluminum phosphide (in tablet form) and dichlorvos in Nepal in 2019. These two pesticides account for approximately 55% of annual pesticide suicides in Nepal. Our best guess is that the deregistration of these pesticides will prevent hundreds of deaths a year.17 On this basis, we estimate CPSP's work from 2017 to 2019 to be substantially more cost-effective than additional funding to our top charities. This estimate relies on a number of particularly uncertain assumptions, which make comparisons with our top charities challenging. For more details, see this page.
In January 2021, we recommended an additional grant of approximately $7.2 million over three years18 for CPSP to continue its work in India and Nepal, expand to China, and work with UN agencies and regulatory bodies to assist other countries in reducing pesticide suicide.
Lead exposure - IPEN
In January 2020, we recommended a grant of $250,000 over three years for IPEN, a network of in-country NGOs, to work on lead paint regulation in several countries in Southeast Asia. That grant was made through a Request for Proposals funded by Affinity Impact.19 Based on quarterly check-in calls, our understanding is that as a result of the work of IPEN and its local partner, Vietnam enacted a law banning lead in paint. Progress in other countries has been limited.20
We plan to do additional research to better understand the likely contribution of lead paint to blood lead levels before considering a larger grant to address this source of exposure.
Alcohol policy - Vital Strategies
As part of our exploration of alcohol policy, we have been investigating Vital Strategies, a public health organization that has previously worked on tobacco control advocacy and food policy (e.g., sugary beverage taxes), among other focus areas.21
In January 2021, we recommended a grant of $100,000 for Vital Strategies to convene a group of stakeholders to develop a proposal for a global advocacy campaign on alcohol control. When the proposal is complete, we plan to consider a grant (likely between $5 million and $15 million) to support this campaign.22
Lead exposure - Pure Earth
In July 2021, we recommended a grant of $8 million to Pure Earth over three years to identify the most important sources of lead exposure in low- and middle-income countries, implement interventions to reduce exposure from those sources, and monitor the effects of those interventions. The grant includes work to reduce exposure from lead-contaminated spices in India, and monitor reductions in blood lead levels after a previous intervention to reduce spice contamination in Bangladesh.
Pure Earth was the most promising opportunity we found to intervene to reduce sources of lead exposure while monitoring the effect of these interventions on blood lead levels. We placed particularly high weight on monitoring because of our relative uncertainty about the relative contributions of different sources of exposure to blood lead levels and the likely effectiveness of different interventions.
Future research in public health regulation
We are unsure how we will prioritize further research into public health regulation relative to our other priorities. We note some potential avenues for further research below.
Promising areas other than lead exposure, pesticide suicide, and alcohol policy
- Micronutrient fortification – Our understanding is that regulatory solutions (e.g., mandatory fortification laws) play an important role in improving coverage of micronutrient fortification.23 We don't yet have a good sense of whether there are promising funding opportunities in the space.
- Ambient air pollution – We may take additional steps in this area depending on the progress and results of an ongoing investigation by Open Philanthropy, an organization with which we work closely.
- Excessive salt consumption – We decided to prioritize deeper investigation into lead exposure rather than excessive salt consumption. Given the uncertainty of our burden and philanthropic spending estimates, and how close the two areas ranked on our "burden per million dollars of spending" framework, we may look into excessive salt consumption further in the future.
- Marginal funding to well-resourced cause areas – We deprioritized a number of areas because they already receive substantial funding (see below). However, it's possible that marginal funding to these areas could still be highly cost-effective.
Cause areas we investigated at a shallow level and deprioritized
Deprioritized primarily due to the high level of funding they receive relative to alcohol, lead, and pesticides:24
- Tobacco policy
- Road traffic injuries
- Drug use
Deprioritized early due to a relatively low burden:25
- Diets high in trans-fatty acids
- Occupational carcinogen exposure
- Residential radon exposure
Deprioritized early due to a lack of obvious and cost-effective regulatory solutions:26
- Intimate partner violence
- Other interpersonal violence
- Falls
- Drowning
- Venomous animal contact (e.g., snakebite)
Deprioritized early due to substantial overlap with ambient air pollution:27
- Household air pollution
- Ambient ozone pollution
Areas we haven't investigated yet
- Occupational injuries
- Other food policy (e.g., diets high in sugar, obesity, trans fat elimination)
Sources
- 1
This impression was informed by a number of sources, including:
- Case studies of public health regulations highlighted in the Center for Global Development's Millions Saved project, including a tobacco control campaign in Thailand and a comprehensive helmet law for improving road safety in Vietnam.
- A review of the benefits of tobacco taxation by Dr. William Savedoff and Albert Alwang at the Center for Global Development.
- A review of the evidence on the impacts of alcohol taxation by David Roodman at Open Philanthropy.
- Our review of evidence from Sri Lanka showing pesticide regulation coincided with large declines in suicide rates.
- 2
- "Despite a clear conflict of interest, alcohol corporations have positioned the industry as part of the solution to the harmful use of alcohol by advocating for responsible drinking. The significant funds spent by the alcohol industry on these lobbying activities have likely contributed to a lack of philanthropic spending on alcohol control." GiveWell's non-verbatim summary of a conversation with Professor Sally Casswell, April 1, 2019, p. 4.
- "The length of time required for a campaign to achieve a lead paint regulation depends on the consistency of funding. A multi-year grant, for example, significantly increases the probability of achieving regulation in a shorter timeframe by enabling continued momentum and awareness-raising. However, governmental decisionmaking processes—which IPEN is unable to control—may also speed or slow a campaign's ability to achieve regulation." GiveWell's non-verbatim summary of a conversation with IPEN, June 24, 2019, p. 2.
- 3
We discuss the various effects that pesticide regulation may have on this page.
- 4
- For example:
- A 2018 survey of the use of lead paint in India concluded: "The study concluded that even though the regulations were notified back in 2016 and came into force in November, 2017, the country has a long way to go in order to reach the desired compliance levels. High concentration of lead content indicates poor compliance on the part of SMEs to adhere to the regulation and inadequate monitoring system in place to enforce the standards." Toxics Link, Lead in paints in India: concerns and challenges, 2018, p. viii.
- "Dr. Sharma's PhD thesis concluded that 6.5% of the pesticides used in Nepal are purchased illegally, which suggests that enforcement of pesticide bans has been a challenge. Often, illegal pesticides are brought across the open India-Nepal border and sold directly to farmers. Government pesticide inspectors in every district have the authority to seize illegal pesticides and penalize people for selling or using them. In the past, the penalty was only 2,000 rupees (roughly $17). Nowadays, selling or using a banned pesticide can be subject to a 100,000 rupee (roughly $1,000) fine and a six-month prison sentence." GiveWell's non-verbatim summary of a conversation with Dr. Dilli Ram Sharma, December 21, 2020, p. 3.
- 5
We relied on estimates reported by the Institute for Health Metrics and Evaluation (IHME) in its Global Burden of Disease (GBD) tool. See this spreadsheet, sheet "IHME disease burden data."
- 6
Questions we investigated included:
Alcohol policy:
- How strong is the evidence base for alcohol control interventions?
- Are the disease burden estimates for alcohol consumption reported by the Institute for Health Metrics and Evaluation (IHME) reliable?
- How should enjoyment of alcohol affect our assessment of alcohol control policies?
As of August 2021, we had not yet completed our work on alcohol policy or written up these questions to a publishable standard.
Lead exposure:
- How strong is the evidence that lead exposure leads to cognitive deficits? (More)
- What are the most important sources of lead exposure? (More)
Pesticide suicide:
- 7
- See our rationale for selecting Vital Strategies as the best placed organization to coordinate a global alcohol control campaign on this page.
- The Centre for Pesticide Suicide Prevention is the only organization that we know of working on addressing pesticide suicide specifically. See this page for more details.
- See this page for an overview of organizations and funders working on reducing lead exposure.
- 8
See this spreadsheet, sheet "Summary," row "Alcohol use."
- 9
- See this page for our estimate of the effect size of pesticide regulation.
- For a comparison of disease burdens, see this spreadsheet, sheet "Summary," rows "Alcohol use" and "Pesticide self-harm."
- 10
See this spreadsheet, sheet "Summary," row "Exposure to lead."
- 11
"YLD is an abbreviation for years lived with disability, which can also be described as years lived in less than ideal health. This includes conditions such as influenza, which may last for only a few days, or epilepsy, which can last a lifetime. It is measured by taking the prevalence of the condition multiplied by the disability weight for that condition. Disability weights reflect the severity of different conditions and are developed through surveys of the general public." IHME, "Frequently asked questions"
- 12
We were not able to identify data on the number of YLDs in low- and middle-income countries due to deliberate ingestion of pesticides. We also expect harm from mortality to comprise the majority of the burden.
- 13
Our total burden estimate for lead exposure also includes an estimate of its financial burden due to lost IQ and earnings, a result of the negative developmental effects of lead exposure. See this spreadsheet, sheet "Summary," row "Exposure to lead" and column "Financial burden (units of value, millions)."
We included this estimate because we believe that negative effects on child development are central to the case for lead control.
- 14
We exclude micronutrient deficiencies and ambient air pollution, for which we don't have good spending estimates.
- 15
- "The case for lead paint regulation appears to center on its potential positive impact on IQ and earnings. We don't have a strong view on the true size of the effect of lead exposure on IQ and earnings, or how much lead paint contributes to exposure, but we think the qualitative case is fairly strong. The existing evidence is weak but consistent with the hypothesis that lead paint affects blood lead levels. Our impression is that there's a consensus that lead is a highly toxic metal and that exposure can have harmful effects on health, including on brain development." GiveWell, "Report on the 2019 GiveWell Grants for Global Health and Development in Southeast Asia and Bangladesh," 2020
- Our estimate of the burden of lead exposure on child development is on this spreadsheet, sheet "Calculations," section "Lost IQ and earnings from exposure to lead."
- These figures take IHME estimates for different areas at face value rather than discounting them depending on strength of evidence (which GiveWell typically does). We did this because this analysis was intended as a quick and high level prioritization device. For areas we have made major grants in, we reviewed the evidence in more depth and applied discounts to our estimates of burden. We do not believe these discounts would have changed the conclusions of our high level prioritization.
- 16
- "From 1976-1980 to 2015-2016, the overall estimated geometric mean BLL of the US population aged 1 to 74 years decreased from 12.8 to 0.82 μg/dL, representing a decline of 93.6% (CDC, unpublished data, 2018)." Dignam et al. 2019, Pg. 2.
- "A total of 47,509 individuals (median age = 44.0 years, 48.3% male) were included in the study. Median sodium consumption was 3232 mg per day (95% confidence interval [CI], 3210-3255), increasing from 3156 mg per day (95% CI 3,038-3,273) in 1999-2000 to 3273 mg per day (95% CI, 3218-3328) in 2015-2016 (P < .001)." Brouillard, Kraja, and Rich 2019, Abstract.
- 17
See this page for details.
- 18
The precise figure will depend on the exchange rate at the time funds are transferred.
- 19
"Affinity Impact, a social impact initiative founded by the children of a Taiwanese entrepreneur, approached GiveWell in late 2018 with an interest in providing support to excellent charities that work in southeast Asia and Bangladesh. As our top and standout charities primarily operate in sub-Saharan Africa, we agreed to run a grantmaking process to see whether we could solicit applications from promising opportunities in Affinity Impact's geographic area of interest." GiveWell, "Report on the 2019 GiveWell Grants for Global Health and Development in Southeast Asia and Bangladesh," 2020
- 20
- "On December 21, 2020, the government of Vietnam adopted its first lead paint regulation, which immediately restricts the content of lead in decorative paints to 600 parts per million (ppm) and gradually increases the restriction from 600 ppm to 90 ppm over the next five years...Vietnam's lead paint regulation was enabled by the work of IPEN's Participating Organization (PO, i.e., IPEN's local partner), the Research Centre for Gender, Family and Environment in Development (CGFED). Of the $250,000 that IPEN received from Affinity Impact in 2019 through the GiveWell Grants for Global Health and Development in Southeast Asia and Bangladesh, $80,000 was subgranted directly to CGFED for work supporting lead paint regulation in Vietnam...Without Affinity Impact's grant, the funding that IPEN could have disbursed to CGFED would have been limited and inconsistent, and a lead paint regulation in Vietnam would likely have taken significantly longer to be enacted." GiveWell's non-verbatim summary of a conversation with IPEN, January 28, 2021, p. 1.
- "IPEN believes national lead paint regulation in Indonesia is unlikely to be established within the next three months...Due to limitations caused by COVID-19, [IPEN's PO] has been unable to make significant additional progress on national lead paint regulation in Bangladesh...IPEN's PO in Cambodia is considering conducting a study on lead content in available paints. IPEN plans to investigate this opportunity more seriously once COVID-19 restrictions in Cambodia have been reduced and its PO is able to travel more freely." GiveWell's non-verbatim summary of a conversation with Dr. Sara Brosché, September 24, 2020, pp. 2 and 4.
- 21
"Our work has built population-level support for new evidence-based policies and enforcement of existing policies at the city, state, national and global level. For example: We have supported the adoption of national tobacco legislation in Turkey, Russia and Senegal (in Russia alone, the smoking rate dropped by 17 percent in the year after the legislation); smoke-free legislation in 10 cities in China covering 100 million people; and new taxes on sugary drinks in Mexico and South Africa that are already showing promise in reducing consumption." Vital Strategies, Capability statement 2019, p. 4.
- 22
The original grant was intended to support a proposal for $5-10 million. After discussion with Vital Strategies, we plan to consider a larger investment of $15 million.
- 23
"Fortification is in some places mandatory and in others voluntary. Where national mandatory fortification programmes have been implemented well and reached high coverage and quality, they have helped to significantly decrease micronutrient malnutrition among entire populations." The Global Alliance for Improved Nutrition, "Programmes: Large-Scale Food Fortification"
- 24
See this spreadsheet, sheet "Summary," column "Burden per unit of spent."
- 25
See this spreadsheet, sheet "Long list," column "Reason for deprioritization."
- 26
See this spreadsheet, sheet "Long list," column "Reason for deprioritization."
- 27
See this spreadsheet, sheet "Long list," column "Reason for deprioritization."