Summary
It’s difficult for aspiring economists to understand how they can contribute to global priorities research. This post shares a syllabus and extended literature for an introduction to global priorities for economists, provides background on the development of the syllabus, its use for a reading group, and advice to others who want to use it. The syllabus was designed with graduate students in mind, but advanced undergraduates will likely be able to get the relevant points from almost all papers. We hope this will be useful for aspiring economists who want to contribute to global priorities research but don’t know where to start.
For those who are just interested in economics literature relevant to global priorities research just follow the link to the syllabus above and ignore the rest of the post. If you want more information about what purpose the syllabus serves, how it came to be, and advice for using it for your own reading groups, read on. It probably makes most sense to read the syllabus alongside the 'Using the syllabus' section. I also recommend the generic Reading group guide for EA groups. Please contact me in the comments section or privately if you have any questions about running your own reading group based on this syllabus.
Thanks to Aaron Gertler, Rossa O'Keeffe-O'Donovan, Philip Trammell, and Duncan Webb for feedback. Opinions and errors remain my own.
Motivation
Global priorities research (GPR) is concerned with the question ‘If our aim is to do the most good possible, from a totally impartial perspective, with limited resources, what should we do?’. Two fundamental fields in this endeavour are economics and philosophy. Since its inception, the Global Priorities Institute (GPI) has made progress in GPR research itself and the instrumental goal of developing GPR as an academic field.
However, this progress is currently lopsided, with much more progress being made in philosophy. Most staff at GPI are philosophers (but they have recently hired their first two postdoctoral economists). All of the papers currently in the GPI Working Paper Series are philosophy papers (but they aim to add ~5 economics papers soon). A comment in a relevant EA Forum post: “One economics student told me that when reading the GPI research agenda, the economics parts read like it was written by philosophers”, also suggests that the economics side of GPR is less clear than the philosophy side (but the research agenda is currently being refreshed). Numerous academic EA/GPR courses have been run in philosophy departments, but as far as I can tell, none have been run in economics.
The result of this state of affairs is that the budding philosopher is easily able to understand what the core texts, seminal papers and cutting edge articles are for philosophical global priorities research, while the budding economist is left with not much to go on. The philosopher has plenty of senior role models and potential supervisors to look to while the economist is unsure who they can speak to about their wacky cause prioritisation ideas. The subfields which GPR philosophy mostly relies on are ethics, decision theory and epistemology, while GPR-relevant economics is spread across a wide variety of subfields, relying on theory and empirics, and normative and positive approaches.
The aim of this syllabus is to help partially solve these issues, by providing a framework and literature to introduce people to global priorities research through published articles in and around the economics literature. It is not a reading list for GPI’s current research priorities. It has a broader scope than GPI’s research agenda and is aimed to be a general introduction to academic research in topics that could contribute to GPR.
Development of syllabus
Matthias Endres and I started a global priorities/effective altruism reading group at the Paris School of Economics (PSE) in the spring semester of 2019. We were both visiting GPI that summer so wanted to dig into some of the GPR literature before our arrival. We mostly read blog posts and articles chosen on an ad hoc basis. It was enjoyable and interesting enough, but did not prepare us particularly well for our time at GPI as we missed most of the literature that GPI researchers were working on.
During our time at GPI we decided that we would like to rerun the reading group with more structure, so we could continue discussing GPR, introduce it to new people at PSE and generate new research ideas. We brainstormed topics and papers, assembled them into a structure, got feedback from economists and visitors at GPI and iterated, settling on 12 core papers.
We ran reading group 2.0 in the spring semester of 2020 at PSE. We initially didn’t include the sessions on biosecurity or population ethics, but due to interest from attendees we added them. There were typically 6-8 attendees each week, and more than that likely would have been too many.
Using the syllabus
In this section I describe why we chose the topics and papers in the structure we did. The syllabus is currently split into two parts plus an introduction. The first part looks at the case for a number of different short and longtermist causes, while the second part takes a more theoretical look at the issues that arise when taking seriously a longtermist perspective.
The introductory week is Cotton-Barratt’s Prospecting for Gold. This is a great introduction to the idea of global priorities research and borrows many concepts native to economics. The hope was that this article would provide tools for us to refer back to in future discussions, and I think this worked fairly well.
The first part (weeks 2 to 5) look at a number of possible causes that both economists and the effective altruism community often consider; poverty, immigration, climate change and biosecurity. Toby Ord with The moral imperative toward cost-effectiveness in global health is used to introduce relevant fat-tailed distributions and the importance of cost-effectiveness. Michael Clemens on Economics and Emigration: Trillion-Dollar Bills on the Sidewalk? is next as a reminder about the importance of scale when doing cause prioritisation and an alternative approach to tackling global poverty. Next we begin the transition to more longtermist causes with a summary of the Stern and Nordhaus debate on climate change which raises questions about discount rates and risk in a familiar setting. These same issues arise in Millett and Snyder on Existential risk and cost-effective biosecurity. There is significantly more work in economics on pandemics now than there was six months ago, so there may be good, new alternatives by economists here.
The second part of the syllabus (weeks 6 to 14) is on longtermism and the issues that arise when you take it seriously. The first paper Policymaking for posterity by Summers and Zeckhauser is an excellent summary of the theory behind longtermist economic policy, and sets the stage for the following weeks. We follow this in weeks 7 to 9 with three normative papers to help sort out what we value: Greaves with a survey of the discounting for public policy literature, Dietrich and Jabarian introducing normative uncertainty into economics, and Broome on population ethics and welfare economics. These papers are different from standard economics fare, so it was definitely helpful for us to have some members who had a background in philosophy as well. For discounting, the Discounting disentangled paper by Drupp et al. is also a good survey of what economists actually think about discounting.
Next is week 10 and 11 on low probability but high impact events. First is catastrophes in general with Martin and Pindyck on Welfare Costs of Catastrophes: Lost Consumption and Lost Lives. This was the most technically challenging paper we read, and as it overlaps significantly with biosecurity, it’s probably the first one we would drop. Second was existential risk as a more extreme version of catastrophe, with Matheny’s Reducing the risk of human extinction.
We finish with a focus on uncertainty. Although there is plenty of literature on risk and ambiguity in economics, there is nothing specifically about cluelessness so for this we read Greaves’ seminal work on the topic. The final two weeks are on resolutions to uncertainty; forecasting and persistence. For forecasting we are spoilt for choice with possible papers, but we focus on Predicting history by Risi and co-authors. For persistence we use the classic by Acemoglu and Robinson, Persistence of power, elites, and institutions.
Reflections
One obvious absence is a session on animal welfare, which we have a section for later in the syllabus. We avoided this partly because work on animal welfare is limited in economics and partly because we were based in France where our impression was that people would not be particularly interested in this topic. In hindsight, both these justifications are probably wrong since there is enough good animal welfare work in economics to discuss, and our selected sample of international graduate students were likely to be interested enough.
The structure of the syllabus worked well for our group which had a mixture of people with and without previous involvement in GPR. Because of this, backloading the things that are outside mainstream economics and common sense morality made more sense. If you have people who are more familiar with GPR and EA, it might be better to frontload the foundational topics currently in weeks 1, 6, 7, 8, 9 and 12, and then move onto different possible causes once you have a more robust framework in place for evaluating them.
The reading group sessions were semi-structured which also worked well. I usually acted as a moderator for the group. Each week we started with a 3-5 minute summary of the paper, collected discussion points from the group and then worked our way through these. We did this both in person and over Zoom. We had a couple of sessions where the author joined us in the discussion. These were a nice change but I felt having an expert there somewhat muted the exploratory nature of the discussion.
Papers that worked best for us were survey articles, e.g. weeks 3, 4, 6 and 7, as they reviewed a large body of literature, and allowed us to see a wider variety of viewpoints and more easily spot where relevant gaps in the literature were. Our group was mostly applied economists so we typically found applied papers using econometric methods more engaging than theoretical papers, but I definitely would not want to neglect theory as a lot of GPR research, especially work related to the future, has very little data available from which we can draw empirical conclusions. For many standard GPR topics, we could not find papers clearly in the economics literature, i.e. by economists published in economics journals. To some extent this is unavoidable as GPR is necessarily an interdisciplinary endeavour, but I think this also highlights the possibility of importing useful concepts from other disciplines, such as normative uncertainty and cluelessness, into economics.
I think the reading group was valuable for a number of reasons. It allowed deep and substantial engagement for all of us with global priorities research on a number of different topics. We generated a number of different research ideas in these areas. After we finished the original 14 weeks we had more ad hoc sessions throughout the summer during which some of us presented more developed research proposals, which we all found useful. We increased the number of people exposed to and interested in GPR at our university and developed a sense of community among those people. Finally, it currently seems pretty difficult to learn about academic work on global priorities research without being located in a couple of buildings in Oxford. Hopefully, we have shown that this model is a reasonable way for other people to learn about GPR at other institutions across the world.
I love that the syllabus includes a section on generalizability. A related topic is how well lab experiments generalize to real-world field experiments, as in John List’s work on baseball cards.
A related topic is replicability: internal validity rather than external. If you do exactly the same study twice, how likely are you to get a similar result?
I believe replications are highly socially valuable but not well-rewarded in econ academia, since the job market tends to reward originality and technical skill, especially at the junior level. This makes replication work relatively neglected, tractable, and a good choice for EAs who are not bound by career concerns. It’s also valuable to understand why replications fail if you want to make sure your own results replicate.
Papers on replicability in economics:
Papers on this outside economics:
Thanks for the paper suggestions! Most of my own research is on internal validity in the LaLonde style so I definitely think it is important too. I'll add a section on replicability to the syllabus.