Worldview Investigations

Measuring Non-Health, Non-Pecuniary Benefits

Open Philanthropy’s mission is to help others as much as we can with the resources available to us. Much of our human-focused Global Health and Wellbeing grantmaking aims to save lives, improve health, and increase incomes. However, we recognize that there are many ways for a person’s life to go better or worse that aren’t fully captured by changes in health and income. Lives full of injustice and discrimination are the worse for it; lives full of empowerment and freedom are the better for it.

To carry out our mission, we need to be able to compare different giving opportunities. Our current framework evaluates interventions using their effects on beneficiary incomes and beneficiary health states, including both length of life and quality of life. (We aim to include both direct and indirect beneficiaries in our evaluations, but recognize that it is not always possible or worthwhile to measure all spillover benefits.)

To incorporate non-health, non-pecuniary benefits into our evaluation framework, we must first adopt a suite of metrics for measuring those benefits.

Health metrics, such as the disability-adjusted life year (DALY), allow us to compare disparate health states, such as pneumonia, dementia, and multiple sclerosis. They also allow us to compare premature deaths to years lived in less than perfect health. With these metrics in hand, we can efficiently allocate resources in our portfolio by converting different potential intervention effects into their DALY impacts, then funding those interventions that avert the most DALYs per dollar spent.

We are looking for essays that propose an equivalent metric for a hard-to-measure state (such as freedom, empowerment, injustice, or discrimination) that we could ultimately use to undertake similar comparisons. A DALY-like metric for female empowerment, for instance, might allow us to compare the effects of increased literacy for girls in Ghana to the effects of improved political representation for women in Guinea. The ideal metric would be reliable, feasible to measure, and applicable across a wide range of circumstances and populations of interest.

We are not looking for essays that compare the value of, e.g., empowerment, to the value of health and income. While we would need to make such a comparison to optimize our full grantmaking portfolio, that exercise is outside the scope of this prompt.


If you are interested in this prompt, we suggest that you:

  • Read about our Global Health and Wellbeing grantmaking (here and here) and evaluation framework (here).

  • Reflect on potential benefits that may be systematically under-captured by our current evaluation process.

  • Explore the literature on measuring things like freedom, injustice, empowerment, and discrimination.

  • Propose one or more practical metrics we could adopt and explain why adopting it would improve our grantmaking.

    • Describe the types of circumstances or philanthropic interventions that might be relevant for that metric. For example, if your metric focuses on female empowerment, you might consider how to measure the empowerment resulting from literacy for girls vs. political representation for women.

    • Consider the ways the metric might be used in practice. (E.g. “Here are a few social science papers on female empowerment, and here is how I would use them to inform or construct my metric.”)

    • Provide evidence for why the proposed metric might be preferred to possible alternatives.

Suggested Reading:

This prompt is meant to help you get started, but we are very open to different approaches to answering this question.

Indirect and second-order effects

We are interested in a framework that advises us on which indirect and/or long-term effects are worth accounting for in the back-of-the-envelope calculations we use to prioritize our grantmaking.

To efficiently allocate our resources, we must compare different giving opportunities. When the comparison is done well, we can select interventions that have the highest expected impact per dollar spent and thus optimize our philanthropic portfolio. However, the total impact of a given intervention is not always exhausted by its direct, short-term consequences. Indirect and/or long-term effects (also known as “flow-through”, “knock-on”, or “ripple” effects) may represent a significant portion of the total impact.

If we were to exclude these consequences from our evaluation framework, we would almost certainly misestimate (perhaps to a large degree) the value of many giving opportunities. Unfortunately, it is difficult to incorporate indirect, long-term effects in a principled and practical way.

For example, when considering whether to recommend a grant to the Against Malaria Foundation, a narrow approach would consider how many deaths such a grant might prevent, while a broader approach might also take into account the lower morbidity from malaria, various economic spillovers, and many other second-order effects as diagrammed below.

Ignoring all indirect effects is obviously unsatisfying, since in some cases these effects might be determinative of which of two interventions is more impactful. At the other extreme, including all indirect effects seems infeasible, since they are so numerous and hard to estimate. We need to find some useful shortcuts for more easily measuring indirect effects in our analysis.

For instance: in the example above, should we attempt to capture the benefits and costs in the grey boxes? If so, what time horizon should we use? What near-term proxies could we use to estimate the long-term effects? Under what conditions do the indirect effects likely swamp the direct effect, and is this a reason for caution? If so, how should we act on that caution?

If you are interested in this prompt, we suggest that you:

  • Read about our Global Health and Wellbeing grantmaking (here and here), our evaluation framework (here), and the back-of-the-envelope calculations we use in our grantmaking (here).

  • Reflect on potential costs and benefits that may be systematically under-captured by our current evaluation process.

  • Propose one or more general principles for analyzing indirect, long-term effects and explain why adopting those principles would improve our grantmaking.

  • Give examples of which second-order effects your framework would advise us to account for, and which (if any) it would recommend ignoring. Some interesting second-order effects you might consider:

    • Antimalarial bednets → lower anemia → higher fertility → very long-term increase in human population 

    • Increasing US R&D spending for one year → increased frontier productivity gains for one year → slightly better GDP trajectory for centuries

    • More economic growth → more meat consumption → more animal suffering

    • Relaxed zoning laws → reduced housing prices in coastal US cities → different population growth patterns → different election outcomes

Suggested Reading:

This prompt is meant to help you get started, but we are very open to different approaches to answering this question.