I explore two perspectives on how to think about resource allocation in trying to make an impact. The first perspective is objective, focusing on a utilitarian approach and cost-benefit analysis to maximise impact. The second is subjective, taking into account personal circumstances and comparative advantages, advocating for a context specific approach. Drawing from my own experiences in global health and development, I discuss my preference for the subjective perspective.
The question of how to use our resources, whether they be time or money, to contribute to the betterment of the world is difficult to answer. Any model which attempts to answer this question must consider a broad and diverse range of factors, such as unintended consequences, second-order effects and marginal impacts. For this piece I want to focus on two distinct perspectives on how one might most impactfully use $100 000 to make a positive impact on the world. I will use a recent personal example to flesh out each perspective.
Perspective 1: The Objective position
The question of how to most effectively use one’s money or time to positively impact the world is one that the utilitarian may feel they are particularly well positioned to respond to. This is especially so when the question is posed broadly without respect to, or knowledge of, any individual actors’ circumstances or comparative advantages, or specific consequences. The feature of agent neutrality is a strength for the approach. A strict cost-benefit analytic framework may be the preferred means of ranking practical solutions when optionality exists, with a project’s benefits per unit of cost calculated, and resources allocated to the top potential projects in order to maximise the resource’s impact.
This approach is attractive for a number of reasons. It offers quantifiability and reduces the impact of personal bias. If methodology is standardised, it ideally would remove differences of opinion or concerns for corruption from a funders point of view. It would also allow for better coordination and planning for the allocation of additional resources. More detailed analysis could be extrapolated, such as calculating the marginal returns of additional resource allocation, in order to further nuance resource allocation decisions.
Perspective 2: The Subjective position
Every individual actor necessarily is influenced by a unique set of circumstances, connections, skills, i.e. their comparative advantages. This fact is important as it augments the way in which a prespecified set of resources may be allocated or used. Any two individuals may have very different opportunities in most efficiently allocating resources. As an outside observer would not possibly be able to integrate the information required to appreciate the position of the individual, evaluation of the choices made by the individual are at best a loose approximation of the actual quality of the decision. Every actor may have very different choice sets available to them. Two individuals given the same resources will likely produce very different results, even if allocating these resources based on the same principles.
This approach is attractive in different ways to the objective position. In the same way the market benefits from individualised skill and the division of labour, the use of an individual’s resources and knowledge in a way most suitable to them diversifies the approaches to how good is done. It may produce unique approaches that an outside observer would not have been able to predict. I also believe this approach provides a more ‘organic’, or bottom up, alternative to the more prescribed utilitarian approach, in that an individual’s choice has considerations for second order or unintended consequences ‘baked in’ in a way that approaches founded on reasoning and logic alone struggle to integrate.
So what is it then that I would do?
After reading Singer’s Famine, Affluence and Morality and engaging with a chunk of Effective Altruism literature, for a long time I think my answer would have been to donate the money to the Against Malaria Foundation, or one of Givewell’s other top charities. This choice would have been closely aligned with the Objective option outlined above. However, over time, my approach to this question has slowly taken a different shape.
I have included below a quote from the conclusion of Hayek’s The Use of Knowledge in Society. The paper argues for the decentralised nature of knowledge and the importance of free markets and individual actors’ use of contextual knowledge in order to forward society, or at least that is what I had taken away from it. I think there are some strong parallels to the question posed here. Whilst the Objective approach may not explicitly claim to be acting as a central theory which distributes resources in the maximally optimised fashion for a society’s improvement, it does necessarily imply some form of ‘ranking’, or superiority of the programs it suggests over alternative approaches. Implicit in this is the idea that we can somehow integrate all the relevant information and draw conclusions on actions which are likely to do the ‘most’ good in a society. Whilst concession is often made by the Objective approach that ideas beyond what is currently known may trump the suggestions that are presently ranked highly, the very idea that enough information can be integrated over time and across contexts to meaningfully capture the true ‘impact’ of a specific program or intervention seem difficult to put much credence in.
“…there is something fundamentally wrong with an approach which habitually disregards an essential part of the phenomena with which we have to deal: the unavoidable imperfection of man’s knowledge and the consequent need for a process by which knowledge is constantly communicated and acquired. Any approach, such as that of much of mathematical economics with its simultaneous equations, which in effect starts from the assumption that people’s
knowledge corresponds with the objective facts of the situation, systematically leaves out what is our main task to explain. I am far from denying that in our system equilibrium analysis has a useful function to perform. But when it comes to the point where it misleads some of our leading thinkers into believing that the situation which it describes has direct relevance to the solution of practical problems, it is high time that we remember that it does not deal with the social process at all and that it is no more than a useful preliminary to the study of the main problem.”
The Use of Knowledge in Society, Friedrich Hayek
Working with a community based development organisation in rural Tanzania, we have directed funding toward two independent development projects which are highly contextual in their requirements. One is a maternal health project, specifically expanding the maternity services at the local health centre so that a full suite of antenatal and perinatal care, including surgical means of delivery, are available to women delivering in the local area. The second is a fish farming project utilising the vicinity of the project to a nearby lake and exploiting local marketable goods to generate jobs and skill acquisition for community members. Given available resources, I believe money directed toward these projects represents a ‘better’ allocation of resources than direction toward an established cost-effective intervention (such as those recommended by Givewell).
There is a higher level of certainty around the need for/likely success of these interventions
Interventions which flourish at scale require the property that they’re able to be successful independent of context. A preventative malaria campaign does require some sensitivity to local societal and cultural practices, but can quite easily be ‘layered on’ to an existing community to yield benefit. This is advantageous in that it is transferable across time and place. Interventions such as the maternal health, or fish farming project, outlined above however fall somewhat on the other end of the spectrum. They are highly context relevant. Without having formally investigated the questions, I would be surprised if attempting to scale up maternity centers to the level of being able to perform cesarean sections had anywhere near the (somewhat) uniform distribution of impact something like a preventative malaria campaign had. There is more likely to be a wider distribution of potential impact for such contextual interventions, that pivot on the need and current resource availability. The same applies for a development project such as the fish farming initiative. Would this be an appropriate intervention to apply to a wide range of developing communities? Highly unlikely. Is it worth piloting (see diversification of ideas below) in a fishing community with access to natural resources and a market for its products? It probably is.
The current emphasis for scalable programs misses potential opportunities for effective programs specific to specific communities. Of course, the main caveat to the claim of greater certainty or impact is the limitations in terms of scale. Current global-health interventions delivered at scale have an enormous positive impact on the world, but perhaps considering applying this context-specific approaches could offer a strong, complimentary, range of development interventions, which address specific needs and strengthen communities.
Diversification of ideas
Communities across the globe have diverse needs and capabilities. A focus on implementing context specific interventions is likely to derive a highly diverse set of interventions. Whilst these interventions may be highly context dependent, there is also a possibility that lessons or features can be drawn that are relevant to the development community more broadly. Perhaps an approach which seeks to address neglected and tractable issues at the community level, rather than issues which are important, neglected and tractable at the global scale, would offer greater opportunities for cost effectiveness. Generalising this principle to a variety of communities may offer an alternative approach to doing good at scale.
Opportunity to build capital/experience/skills in development work
This point is far more selfish but is a relevant consideration if one’s goal is to ‘do good’ over the course of their entire career. By utilising the $100 000 to implement specific projects, rather than making a donation to an already functioning organisation with a track record of doing effective good, there is the opportunity to be involved in and learn from a series of experiences that would otherwise be unavailable to me. The process of thinking through project implementation, monitoring and evaluation, as well as more pragmatic considerations of infrastructure development and local policies and procedures are all important and generalisable skills for a career in development.
Limitations
In writing a piece such as this there are a number of biases which ought to be highlighted and additional considerations which are important in thought around the question of an ‘objective’ vs ‘subjective’ use of resources.
Levels of intervention
The Objective approach is likely to be more useful at a high level, whilst the Subjective has its strengths in the local level where contextual knowledge is very important. Policy decisions necessarily need to consider the needs of a group rather than the individual, and thus taking a ‘subjective’ position may not be of most benefit under these circumstances. This is not, however, to say that the ‘subjective’ approach has no utility, interest groups and those who lobby policy makers on behalf of groups of individuals apply the ‘subjective’ approach to higher level decision making, and via a so called marketplace of ideas, along with representatives from other interest groups, contribute to broad policy decisions being tailored to the groups which are represented in the populous. This line of argument itself is open to all of the criticisms which are applicable to any process of democracy, but it is nevertheless an important line of thought.
Potential for corruption
In taking a subjective viewpoint and going away from more objective and evidence-driven approaches, the subjective position opens itself to the bias and faults of human decision making. This includes the propensity for individuals to become misaligned from working toward the interests of the group they seek to help, whether this may be with or without intention. Utilising the context specific personal knowledge characteristic of the subjective approach makes outside evaluation more difficult, and choices in resource allocation less transparent. This is an issue not only directly, in the sense that it is difficult to know if resources are being allocated effectively, but also indirectly, in the sense outside actors are less likely to be inclined to fund activities whose decision making is not transparent.
Difficulty in obtaining proof-of-concept
The gold standard in evidence for any intervention in the social sciences and medicine is that of the randomised controlled trial. The RCT is dependent on taking groups with similar baseline characteristics and varying only the intervention. This can mean that the effectiveness of interventions which leverage the context specific knowledge relevant to a particular community will be difficult to demonstrate using the RCT approach. Or that conducting RCT level analysis costs more than the proposed program itself. Lower levels of evidence, such as before-and-after studies may be the only reliable means of such data-gathering. Consider the maternal health and fishing interventions mentioned above. A RCT assessing the impact of holistic maternal healthcare services would be constrained by the resources required to build enough units to reach a level of statistical significance in analysis. A RCT assessing the effectiveness of a fish farming program would be difficult to interpret, unless all included study sites had similar sections of their economies dedicated to fishing. This does not however mean these programs should be abandoned, just that care and nuance needs to be applied in evaluating their impact.
Closing thoughts
I wrote this as a reflection on my own personal experiences with fundraising and involvement with global health and development work. For some time I had been convinced allocating one’s own resources, and attempting to direct others toward the most cost-effective interventions was the best means of doing good given its objectivity and perceived certainty. However as I became involved, and spent time living and working with a specific community organisation in Tanzania, I began to re-evaluate my previous framework of thinking about ‘doing good’.
I had spent some time fundraising and talking to friends and colleagues about effective giving, and had some, but little, traction. As is the way with human emotions, when I began speaking about the specific health centre I had worked in, the specific patients I had treated, and the specific issues that needed addressing, people were far more interested and excited. This directly and indirectly led to us at the Cedar Foundation raising sufficient funds for the two projects in Kamanga, Tanzania.
Specifically in regard to the maternal health and fish farming projects, the given funds were raised from philanthropic sources who made donations with less data driven and outcome contingent processes (as opposed to more evidence driven EA or EA-adjacent funders). It is unlikely these resources would have been made available to us if they had not been for projects with an attractive narrative. Considering the counterfactual scenario that these projects did not exist, the money would have likely either not been donated charitably at all, or to a local charity in a high-income setting, where these donors had previously allocated funding. I believe by helping to direct these resources into exploratory, context-specific, global health projects the potential impact may have benefits which are highly valuable for the community and us as implementors. These benefits are likely to be both direct, in treating disease and increasing economic activity locally, and indirect, in strengthening health systems and teaching us important and generalisable lessons about project implementation.
My argument here isn’t that everyone should drop the principle of cost-effectiveness, choose a community and attempt to max it out on wellbeing, but more that for an individual working toward doing good, taking a narrower frame of reference can offer opportunities and lessons that may not otherwise be available. These opportunities and lessons can both be valuable in themselves, and leveraged into further opportunities in the future. Whilst the question of what one would do with $100 000 is slightly provocative, for those interested in a career outside of earning to give, thinking laterally beyond what we see as the best buys in global health can have value well beyond the immediate impact of the project, value which should be strongly considered in making these decisions.
What would you do with $100 000?
Link post
Summary
I explore two perspectives on how to think about resource allocation in trying to make an impact. The first perspective is objective, focusing on a utilitarian approach and cost-benefit analysis to maximise impact. The second is subjective, taking into account personal circumstances and comparative advantages, advocating for a context specific approach. Drawing from my own experiences in global health and development, I discuss my preference for the subjective perspective.
The question of how to use our resources, whether they be time or money, to contribute to the betterment of the world is difficult to answer. Any model which attempts to answer this question must consider a broad and diverse range of factors, such as unintended consequences, second-order effects and marginal impacts. For this piece I want to focus on two distinct perspectives on how one might most impactfully use $100 000 to make a positive impact on the world. I will use a recent personal example to flesh out each perspective.
Perspective 1: The Objective position
The question of how to most effectively use one’s money or time to positively impact the world is one that the utilitarian may feel they are particularly well positioned to respond to. This is especially so when the question is posed broadly without respect to, or knowledge of, any individual actors’ circumstances or comparative advantages, or specific consequences. The feature of agent neutrality is a strength for the approach. A strict cost-benefit analytic framework may be the preferred means of ranking practical solutions when optionality exists, with a project’s benefits per unit of cost calculated, and resources allocated to the top potential projects in order to maximise the resource’s impact.
This approach is attractive for a number of reasons. It offers quantifiability and reduces the impact of personal bias. If methodology is standardised, it ideally would remove differences of opinion or concerns for corruption from a funders point of view. It would also allow for better coordination and planning for the allocation of additional resources. More detailed analysis could be extrapolated, such as calculating the marginal returns of additional resource allocation, in order to further nuance resource allocation decisions.
Perspective 2: The Subjective position
Every individual actor necessarily is influenced by a unique set of circumstances, connections, skills, i.e. their comparative advantages. This fact is important as it augments the way in which a prespecified set of resources may be allocated or used. Any two individuals may have very different opportunities in most efficiently allocating resources. As an outside observer would not possibly be able to integrate the information required to appreciate the position of the individual, evaluation of the choices made by the individual are at best a loose approximation of the actual quality of the decision. Every actor may have very different choice sets available to them. Two individuals given the same resources will likely produce very different results, even if allocating these resources based on the same principles.
This approach is attractive in different ways to the objective position. In the same way the market benefits from individualised skill and the division of labour, the use of an individual’s resources and knowledge in a way most suitable to them diversifies the approaches to how good is done. It may produce unique approaches that an outside observer would not have been able to predict. I also believe this approach provides a more ‘organic’, or bottom up, alternative to the more prescribed utilitarian approach, in that an individual’s choice has considerations for second order or unintended consequences ‘baked in’ in a way that approaches founded on reasoning and logic alone struggle to integrate.
So what is it then that I would do?
After reading Singer’s Famine, Affluence and Morality and engaging with a chunk of Effective Altruism literature, for a long time I think my answer would have been to donate the money to the Against Malaria Foundation, or one of Givewell’s other top charities. This choice would have been closely aligned with the Objective option outlined above. However, over time, my approach to this question has slowly taken a different shape.
I have included below a quote from the conclusion of Hayek’s The Use of Knowledge in Society. The paper argues for the decentralised nature of knowledge and the importance of free markets and individual actors’ use of contextual knowledge in order to forward society, or at least that is what I had taken away from it. I think there are some strong parallels to the question posed here. Whilst the Objective approach may not explicitly claim to be acting as a central theory which distributes resources in the maximally optimised fashion for a society’s improvement, it does necessarily imply some form of ‘ranking’, or superiority of the programs it suggests over alternative approaches. Implicit in this is the idea that we can somehow integrate all the relevant information and draw conclusions on actions which are likely to do the ‘most’ good in a society. Whilst concession is often made by the Objective approach that ideas beyond what is currently known may trump the suggestions that are presently ranked highly, the very idea that enough information can be integrated over time and across contexts to meaningfully capture the true ‘impact’ of a specific program or intervention seem difficult to put much credence in.
Working with a community based development organisation in rural Tanzania, we have directed funding toward two independent development projects which are highly contextual in their requirements. One is a maternal health project, specifically expanding the maternity services at the local health centre so that a full suite of antenatal and perinatal care, including surgical means of delivery, are available to women delivering in the local area. The second is a fish farming project utilising the vicinity of the project to a nearby lake and exploiting local marketable goods to generate jobs and skill acquisition for community members. Given available resources, I believe money directed toward these projects represents a ‘better’ allocation of resources than direction toward an established cost-effective intervention (such as those recommended by Givewell).
There is a higher level of certainty around the need for/likely success of these interventions
Interventions which flourish at scale require the property that they’re able to be successful independent of context. A preventative malaria campaign does require some sensitivity to local societal and cultural practices, but can quite easily be ‘layered on’ to an existing community to yield benefit. This is advantageous in that it is transferable across time and place. Interventions such as the maternal health, or fish farming project, outlined above however fall somewhat on the other end of the spectrum. They are highly context relevant. Without having formally investigated the questions, I would be surprised if attempting to scale up maternity centers to the level of being able to perform cesarean sections had anywhere near the (somewhat) uniform distribution of impact something like a preventative malaria campaign had. There is more likely to be a wider distribution of potential impact for such contextual interventions, that pivot on the need and current resource availability. The same applies for a development project such as the fish farming initiative. Would this be an appropriate intervention to apply to a wide range of developing communities? Highly unlikely. Is it worth piloting (see diversification of ideas below) in a fishing community with access to natural resources and a market for its products? It probably is.
The current emphasis for scalable programs misses potential opportunities for effective programs specific to specific communities. Of course, the main caveat to the claim of greater certainty or impact is the limitations in terms of scale. Current global-health interventions delivered at scale have an enormous positive impact on the world, but perhaps considering applying this context-specific approaches could offer a strong, complimentary, range of development interventions, which address specific needs and strengthen communities.
Diversification of ideas
Communities across the globe have diverse needs and capabilities. A focus on implementing context specific interventions is likely to derive a highly diverse set of interventions. Whilst these interventions may be highly context dependent, there is also a possibility that lessons or features can be drawn that are relevant to the development community more broadly. Perhaps an approach which seeks to address neglected and tractable issues at the community level, rather than issues which are important, neglected and tractable at the global scale, would offer greater opportunities for cost effectiveness. Generalising this principle to a variety of communities may offer an alternative approach to doing good at scale.
Opportunity to build capital/experience/skills in development work
This point is far more selfish but is a relevant consideration if one’s goal is to ‘do good’ over the course of their entire career. By utilising the $100 000 to implement specific projects, rather than making a donation to an already functioning organisation with a track record of doing effective good, there is the opportunity to be involved in and learn from a series of experiences that would otherwise be unavailable to me. The process of thinking through project implementation, monitoring and evaluation, as well as more pragmatic considerations of infrastructure development and local policies and procedures are all important and generalisable skills for a career in development.
Limitations
In writing a piece such as this there are a number of biases which ought to be highlighted and additional considerations which are important in thought around the question of an ‘objective’ vs ‘subjective’ use of resources.
Levels of intervention
The Objective approach is likely to be more useful at a high level, whilst the Subjective has its strengths in the local level where contextual knowledge is very important. Policy decisions necessarily need to consider the needs of a group rather than the individual, and thus taking a ‘subjective’ position may not be of most benefit under these circumstances. This is not, however, to say that the ‘subjective’ approach has no utility, interest groups and those who lobby policy makers on behalf of groups of individuals apply the ‘subjective’ approach to higher level decision making, and via a so called marketplace of ideas, along with representatives from other interest groups, contribute to broad policy decisions being tailored to the groups which are represented in the populous. This line of argument itself is open to all of the criticisms which are applicable to any process of democracy, but it is nevertheless an important line of thought.
Potential for corruption
In taking a subjective viewpoint and going away from more objective and evidence-driven approaches, the subjective position opens itself to the bias and faults of human decision making. This includes the propensity for individuals to become misaligned from working toward the interests of the group they seek to help, whether this may be with or without intention. Utilising the context specific personal knowledge characteristic of the subjective approach makes outside evaluation more difficult, and choices in resource allocation less transparent. This is an issue not only directly, in the sense that it is difficult to know if resources are being allocated effectively, but also indirectly, in the sense outside actors are less likely to be inclined to fund activities whose decision making is not transparent.
Difficulty in obtaining proof-of-concept
The gold standard in evidence for any intervention in the social sciences and medicine is that of the randomised controlled trial. The RCT is dependent on taking groups with similar baseline characteristics and varying only the intervention. This can mean that the effectiveness of interventions which leverage the context specific knowledge relevant to a particular community will be difficult to demonstrate using the RCT approach. Or that conducting RCT level analysis costs more than the proposed program itself. Lower levels of evidence, such as before-and-after studies may be the only reliable means of such data-gathering. Consider the maternal health and fishing interventions mentioned above. A RCT assessing the impact of holistic maternal healthcare services would be constrained by the resources required to build enough units to reach a level of statistical significance in analysis. A RCT assessing the effectiveness of a fish farming program would be difficult to interpret, unless all included study sites had similar sections of their economies dedicated to fishing. This does not however mean these programs should be abandoned, just that care and nuance needs to be applied in evaluating their impact.
Closing thoughts
I wrote this as a reflection on my own personal experiences with fundraising and involvement with global health and development work. For some time I had been convinced allocating one’s own resources, and attempting to direct others toward the most cost-effective interventions was the best means of doing good given its objectivity and perceived certainty. However as I became involved, and spent time living and working with a specific community organisation in Tanzania, I began to re-evaluate my previous framework of thinking about ‘doing good’.
I had spent some time fundraising and talking to friends and colleagues about effective giving, and had some, but little, traction. As is the way with human emotions, when I began speaking about the specific health centre I had worked in, the specific patients I had treated, and the specific issues that needed addressing, people were far more interested and excited. This directly and indirectly led to us at the Cedar Foundation raising sufficient funds for the two projects in Kamanga, Tanzania.
Specifically in regard to the maternal health and fish farming projects, the given funds were raised from philanthropic sources who made donations with less data driven and outcome contingent processes (as opposed to more evidence driven EA or EA-adjacent funders). It is unlikely these resources would have been made available to us if they had not been for projects with an attractive narrative. Considering the counterfactual scenario that these projects did not exist, the money would have likely either not been donated charitably at all, or to a local charity in a high-income setting, where these donors had previously allocated funding. I believe by helping to direct these resources into exploratory, context-specific, global health projects the potential impact may have benefits which are highly valuable for the community and us as implementors. These benefits are likely to be both direct, in treating disease and increasing economic activity locally, and indirect, in strengthening health systems and teaching us important and generalisable lessons about project implementation.
My argument here isn’t that everyone should drop the principle of cost-effectiveness, choose a community and attempt to max it out on wellbeing, but more that for an individual working toward doing good, taking a narrower frame of reference can offer opportunities and lessons that may not otherwise be available. These opportunities and lessons can both be valuable in themselves, and leveraged into further opportunities in the future. Whilst the question of what one would do with $100 000 is slightly provocative, for those interested in a career outside of earning to give, thinking laterally beyond what we see as the best buys in global health can have value well beyond the immediate impact of the project, value which should be strongly considered in making these decisions.