Introducing Good Policies: A new charity promoting behaviour change interventions

by calebp 3 min read18th Nov 20198 comments


Introducing Good Policies

I am excited to announce the launch of Good Policies, a new EA-aligned charity looking for the most cost-effective advocacy opportunities in global health, currently exploring starting new tobacco taxation campaigns in neglected regions. Good Policies is incubated under and developed in partnership with Charity Entrepreneurship.

The leading cause of preventable deaths in low and middle-income countries are non-communicable diseases, but designing cost-effective solutions can be challenging, as it often requires changing an individual’s behaviour over time [8]. Implementing certain public health policies at a state or national level can effectively change the behaviour of many people. Our aim is to systematically identify the most cost-effective policy windows, and then fund and provide technical assistance to new campaigns that advocate for evidence-based policies, starting with tobacco taxation.

Here, I aim to inform the EA community of my rationale for founding an organisation in the global health policy space, my plans for piloting to better understand the potential impact of this organisation, and where this effort could benefit from the resources of this community.

Why focus on noncommunicable diseases?

Noncommunicable diseases kill approximately 41 million people each year, which is about 71% of all deaths globally [1]. The four noncommunicable diseases with the greatest annual death toll are cardiovascular diseases (17.9M), cancers (9.0M), respiratory diseases (3.9M) and diabetes (1.6M) which between them account for about 80% of all premature noncommunicable disease deaths. These are all strongly linked to tobacco consumption [4].

The primary drivers of premature noncommunicable disease deaths are generally linked to behaviour, particularly unhealthy diets, physical activity, alcohol use and exposure to tobacco smoke [5][6]. Implementing policies at a national level that effectively change people’s choices can significantly reduce the premature death toll and improve the quality of life of the country’s citizens at scale.

Why Advocacy?

It is not surprising that changing systems is challenging, but there are multiple case studies of groups successfully advocating for better policies around tobacco control. For a relatively small amount of money, external actors can influence governments to implement policies that improve public health. This may be one of the reasons why GiveWell are building their capacity to investigate opportunities to influence government policy as outlined in a blog post earlier this year [2]. GiveWell spoke with Ms Yolanda Richardson from the Campaign for Tobacco-Free Kids about their work on tobacco control policy and have mentioned that they plan to or are in the process of investigating the cost-effectiveness of interventions related to tobacco control [3][7].

There are plenty of potential weaknesses to advocacy-based interventions compared to more direct interventions. One large concern we have is understanding the impact of organisations in this space. The gold standard for impact analysis is the randomised control trial (RCT). Unfortunately, it is difficult to effectively randomise with policy-based interventions as we are unable to control for the numerous factors that affect the impact of an advocacy effort. In fact, it is difficult to understand even the base rate of effectiveness of advocacy in this space, as there seems to be a selection bias in case studies (it is rare to read about unsuccessful campaigns).

Due to the all-or-nothing nature of lobbying, it is difficult to attribute impact properly, especially when there are multiple actors working towards a common goal. One approach to understanding our marginal contribution is selecting particularly neglected regions where there are no existing related campaigns but even then, it can still be difficult to determine how much your contributions sped up the policy being implemented.


In order to address some of these questions, Good Policies is roughly split between two tasks. The first is developing and working through its research agenda and the second is piloting a campaign for increased tobacco prices in Mongolia via taxation.


Our research focuses on systematically identifying cost-effective policy windows. It seems as though there are certain times and regions where an additional campaign might be particularly valuable. We are currently developing our model to more reliably estimate the cost-effectiveness of different policy windows. We are planning on publishing a more detailed research agenda in the near future, as well as cost-effectiveness estimates for various policy windows and the models used to compute these estimates and the model that we used to make this.


We are currently exploring the possibility of starting a tobacco taxation campaign in Mongolia, in collaboration with a local NGO (the National Cancer Council of Mongolia). We believe that this collaboration will be far more effective than working independently in this region. As there is no existing advocacy effort for increased tobacco taxes in this region (that we are aware of) it may be easier to establish our counterfactual impact more concretely than in regions with more actors. Mongolia also has a very high smoking prevalence and is not a priority country for any large donor interested in tobacco control which may mean it is disproportionately neglected.

In addition to impact, running the pilot will allow us to gather information to improve our model and make substantial progress on our research agenda. It is likely that in our current cost-effectiveness model we have neglected important factors that we think being up close during a policy window may highlight

Some of the key deliverables in the early stages of the pilot will be an expected cost-effectiveness analysis and a strategy for developing models that can be generalised to similar campaigns. If by our estimates our intervention looks less cost-effective than GiveWell’s current top charities then we will scale down operations and pivot to another intervention.

How you can help?


I am looking for a co-founder, ideally with experience lobbying LMIC governments or someone that can bring a similar skillset (see job description). I expect that I will also hire an operations lead and I have various internships on my website. All of these positions can be found here.


I am also keen to find advisors who are happy to consult with us whilst we continue to develop our vision. If you have experience in starting organisations, public health or lobbying/advocacy I would love to connect with you.


I am also looking for funding in order to raise money for our pilot. I have a fundraising proposal on my website so if you, or someone you know, is excited about funding a new organisation in this space I would love to hear from you. Additionally, if you are not in a position to fundraise but think you are well-positioned to advise us on securing funding, I am keen to get as wide a perspective on this as possible.


You can reach out to me directly to talk about any of the above or give feedback on the project at

Good Policies was incubated under Charity Entrepreneurship, an effective altruism organization, which provided our initial funding.