TLDR: GiveWell’s moral weights can include additional categories, based on intended beneficiaries’ and prospective funders’ preferences.
Introduction
GiveWell’s cost-effectiveness analyses can comprise a greater scope of moral weight categories. Currently, preventing death at different ages and increased income/consumption are the only considered criteria (pp. 5–6). Intended beneficiaries may have a complex set of priorities, which can resonate with prospective funders.
Covering a broader range of non-profit goals can increase international cooperation and safeguard peace. A caveat is that current power dynamics among nations may be strengthened, although individuals, globally, presented with a greater variety of lifestyle choices.
Sufficiently disaggregated national statistics can be used to estimate the impact of programs on different moral weight categories. World Bank data aggregates these statistics.
With additional moral weights, the impact of individual programs will likely be nominally smaller. This could discourage funders looking to save a large number of lives and to increase others' income relatively highly from donating. However, a well-developed set of weights can better showcase the strategic importance of specific funding focus as well as the need for philanthropic coordination.
My specific recommendations rely on a biased interpretation and selection of evidence. Research that mitigates human biases can be conducted.
GiveWell should research intended beneficiaries’ and prospective funders’ preferences robustly, if it can thus maintain its competitive advantage over larger charity assessors. Alternatively, GiveWell can seek to cooperate with other assessment organizations that define development more broadly.
Intended beneficiaries’ preferences
Based on in-person interactions with globally poor individuals in Asia, Africa, and Latin America, as well as international development undergraduate and graduate coursework, IPA and J-PAL resources, and other studies and expert insights, I conclude that many intended beneficiaries have, at least some of, the following priorities:
- Prospects for themselves and family
- Ability to afford emergency healthcare
- Mutual respect within community
- Adequate rest and physical health
- Enjoyable living environment
- Healthy family relationships
- Safety from conflict and financial risks
This perspective suggests that moral weights should include some of the following categories:
- Employment and underemployment
- Education rates and quality
- Preventive healthcare and insurance
- Emergency healthcare availability and affordability
- Community epistemics and relationships
- Healthy and strenuous exercise
- Rest, sleep, noise levels
- Environmental sustainability
- Spousal cooperation and relationships
- Treatment of children
- Civil and international conflict
- Local crime rates
These categories can be understood as examples of what intended beneficiaries may value and what could be measured to resolve and prevent underlying issues. An extensive study should be conducted to understand these preferences better. Alternatively, existing studies can be synthesized by an impartial team or a software. Enumerator bias should be controlled for, including by non-leading survey design and impartial and trusted enumerator engagement.
Funders’ and prospective funders’ values
The values commonly presented in the Giving Pledge letters and by the top 100 Forbes billionaires, in my perspective, include:
- Family and business success
- Healthcare innovation and longevity
- Technological innovation and competitive advantage
- Meaningful AI advancement and safety
- Education and upskilling
- Market growth and brand recognition
- Environmental sustainability
- Economic inclusion (mainly within the US)
- Art, culture, heritage
This would suggest that GiveWell moral weights should include:
- Family cooperation and relationships
- Assuming that prospective funders’ would be willing to support also other families’ happiness
- Global value chain (GVC) participation
- Healthcare systems advancement
- Assuming the willingness to share innovations abroad
- Life expectancy at different ages
- Supplier efficiency increase
- Education rates and economic relevance, job training
- Total and brand product consumption
- Climate change mitigation, adaptation, and preparedness
- Impact of industrialization on nature and animals
- For example, industrialization can prevent the suffering of wild animals
- Income equality
- Assuming preference for income equality also abroad
- Artistic expression, cultural celebration, heritage preservation
- Assuming preferences for such also in other nations
My perspective may be biased by the set of resources that I reviewed and their order as well as the frameworks that I used to filter and synthesize prospective funders’ preferences. GiveWell should dedicate significant effort to understand its customers better, in order to present a competitive product.
International cooperation, security, power distribution
Programs with a broader set of objectives can have wider impacts on international cooperation, security, and power distribution.
Job training and education
More inclusive job training and education can prevent the rise of elites abroad, which can disempower these nations compared to those where elites have already emerged. However, greater economic inclusion and GVC integration can safeguard global peace: countries that trade together have more to lose in a conflict.
Family and community relationships
Positive family and community relationships can influence constituents’ and leaders' preference for a peaceful conflict resolution. Cooperative norms among nations with weaker institutions are increasingly more important as destructive technologies become more affordable.
Health
Health can have little direct impact on international security, although indirectly, increased economic integration due to healthier workforce can prevent the use of force in conflict resolution.
Environment
The profitability of environmental commitments and green technology adaptation should advantage nations with the greatest potential to implement relevant changes and develop technologies. Climate change mitigation, adaptation, and preparedness can prevent disputes related to environmental migration. Considering nature and non-human animals can increase societal empathy, which can motivate peaceful conflict resolution.
Art, culture, heritage
The impact of the advancement of art, culture, and heritage can have a range of impact, depending on the sentiments that it inspires. For example, a gallery that presents family farm pictures can have a widely different impact from a place that glorifies medieval torture instruments.
World Bank data
The World Bank periodically collects thousands of indicators relevant to various moral weight categories and collates additional thousands of datasets. GiveWell should peruse these datasets to find the most relevant and complete and least biased statistics.
While the DataBank databases display national-level data, national statistics may exist at the individual, household, and other levels that present sufficient disaggregation and sample sizes for statistically robust studies.
Examples of metrics relevant to moral weight categories preferred by both beneficiaries’ and prospective funders include (World Development Indicators):
- “Labor force with basic education (% of total working-age population with basic education)”
- “Primary education, pupils (% female)”
- “Firms offering formal training (% of firms)”
- “Children in employment, unpaid family workers (% of children in employment, ages 7-14)”
- “Women participating in the three decisions (own health care, major household purchases, and visiting family) (% of women age 15-49)”
- “Births attended by skilled health staff (% of total)”
- “Number of people spending more than 25% of household consumption or income on out-of-pocket health care expenditure”
- “CPIA policy and institutions for environmental sustainability rating (1=low to 6=high)”
- “Total fisheries production (metric tons)”
Nominal values, strategic importance, philanthropic cooperation
A greater variety of moral weight categories will make the contribution of individual programs nominally smaller. For example, a chlorine dispenser program can greatly improve an aspect of the WASH metric, which falls under the health moral weight category as well as community relationships and culture (chatting during water collection). However, the program can have no impact on education, environment, global value chain integration, other aspects of health, and art. Thus, the impact of this globally top program can look relatively small. This small value could discourage a prospective donor from selecting the program.
Businesses, however, can see the strategic importance of supporting specific aspects of global development. For example, a company that supplies clean water in rural areas can be interested in early brand recognition in an emerging market by donating chlorine dispensers, in addition to its philanthropic contribution.
With multiple moral weight categories, it will become apparent that an individual, or a single foundation, cannot resolve all global issues by itself. This can motivate philanthropic coordination.
While GiveWell can facilitate this philanthropic coordination by implementing additional categories in its own analyses, it can also cooperate with other assessors that evaluate charities based on a broader range of impact criteria.
Note on personal biases
I am biased by the selection of resources that I engaged with, their order, and the frameworks that I reviewed them with. It is possible that intended beneficiaries and prospective funders have different priorities. Research that mitigates human biases can be conducted.
Conclusion
I suggested that GiveWell should include additional moral weight categories in its cost-effectiveness analyses or cooperate with charity assessors that use broader impact criteria. I hypothesized intended beneficiary and prospective funders’ preferences. Further, I briefly discussed the effects of a broader set of philanthropic objectives on international cooperation, security, and power distribution. Relevant World Bank data series and metric examples were overviewed. The impact of nominal changes in GiveWell’s analyses on the global philanthropic landscape was debated. I concluded with a note on personal biases.
Strong upvote! I came here to say something similar. One of your most compelling points is addressing the needs and wants of the intended beneficiaries, in contrast with pursuing the most economically efficient cause area. I think there is significant moral weight in ensuring people have what they want and need, which cannot be commodified.
Thanks!