The Power of Effective Agenda-Setting
Tom Kalil
Renaissance Philanthropy believes in the power of effective agenda-setting. In this Q&A, Tom Kalil discusses what we mean by that, how agenda-setting can create positive, self-fulfilling prophesies, and the role of field strategists in agenda-setting.
Caption: Edwin “Buzz” Aldrin stepped onto the moon on July 20, 1969, eight years after President John F. Kennedy announced before a special joint session of Congress the dramatic and ambitious goal of sending an American safely to the Moon before the end of the decade. Source: NASA
Q: What do you mean by agenda-setting?
Agenda-setting usually involves:
Making the argument that a particular problem or opportunity deserves more attention and resources;
Identifying a goal or desired end-state, and intermediate milestones towards that goal; and
Identifying the public and private actions that would have the biggest impact on achieving the goal.
Agenda-setting also benefits from describing the context needed to understand the core ideas. This might include (1) facts about the world; (2) analytical frameworks or models that are critical to understanding the idea (e.g. the private sector will underinvest in X because of “market failure” Y); (3) normative beliefs (e.g. achieving X would be a good thing); (4) predictions (e.g. in a business as usual scenario, X is likely to be true in Y years); and (5) theories of change (i.e. if X does Y, then Z is likely to occur).
Q: Why do you think that the potential payoff associated with agenda-setting has increased, particularly in the context of scientific research or innovation?
There are now many more philanthropists and foundations who can support scientific research. In the U.S. alone, there are 2,000 families with $0.5 billion or more in wealth, and their current philanthropy as a percentage of their wealth is 1.2 percent. If one assumes that they are doing as well as an index fund, they could dramatically increase their philanthropy with no reduction in their wealth. Also, U.S. billionaires have $6 trillion in wealth.
Many of today’s wealthy individuals were successful tech entrepreneurs, so often have an appreciation for the importance of science and technology. But the opportunity is not just the aggregate level of resources that is potentially available to support science and technology – it is the flexibility that philanthropists have.
Q: Why do you think flexibility is a philanthropic “superpower?”
For several reasons. First, in the U.S., we have prematurely standardized on a handful of mechanisms to support scientific research. One example is that we use peer review to allocate small grants to professors at universities for the purposes of supporting their graduate students and postdocs. Philanthropists are beginning to back novel approaches to funding and organizing research. Patrick Collison supported Fast Grants, an effort to make decisions on research proposals related to COVID-19 in 14 days or less. Over 30 philanthropists and foundations are supporting Focused Research Organizations incubated by Convergent Research – science moonshots organized as time-bound research non-profits with a CEO and a well-defined “North Star.” The existence of FROs as a funding mechanism is increasing the level of ambition of the research community. Researchers do not devote much time and energy to thinking about ideas if they do not see a legible and repeatable path to getting them funded. Prior to the emergence of FROs, researchers were not asking themselves, “What key bottleneck to scientific progress could I address with the team of 20-30 scientists and professional engineers, all rowing in the same direction?”
Second, in addition to traditional grant-making, philanthropists are also in a position to advocate for policy changes, build coalitions of the willing and able, and invest, but with a longer time horizon or motivated by solving a problem. For example, some disease-oriented philanthropists are making strategic investments in VCs, with the goal of treating and curing neurodegenerative diseases as opposed to maximizing financial returns. Not many philanthropists are thinking about how to effectively combine two or more of these approaches to problem-solving.
Finally, it is often the case that there is a large gap between what we are currently doing with respect to a given problem and what we should be doing. For example, the private sector underinvests in public goods, fundamental research, research that is beyond the time horizons of VCs, or innovations that primarily help low-income communities. Federal research funding can be risk-averse and incremental. The gallows humor in the research community is that they have to do the experiment before they write the grant! Philanthropists with the ability to identify and address these gaps can have a large, “but for” impact.
Q: What types of agenda-setting are you particularly excited about?
I think that agenda-setting can create positive, self-fulfilling prophesies by identifying an ambitious and important goal, and combining that with a “why now” story. What has changed about the world that makes the previously impossible now within reach? In science and innovation, it could be a new fundamental insight, a novel combination of technological building blocks, a new business model, or a change in public policy. The “why now” could also relate to an increase in the urgency of the problem. The combination of an important and ambitious goal and the “why now” – if credible, can serve as a magnet for people and resources. As President Kennedy noted, “By defining our goal more clearly, by making it seem more manageable and less remote, we can help all peoples to see it, to draw hope from it, and to move irresistibly towards it.” If we do this in many areas, it could create an ambitious and optimistic culture, in which more people believe our best days are yet to come.
Q: What are some other high-level observations you have about agenda-setting?
It’s hard to do well. Not everyone can analyze a complex scientific and technological landscape and identify key leverage points that can move an entire field forward.
It is inexpensive relative to the potential payoff. It would be inexpensive to identify dozens of transformational opportunities for science and technology philanthropy in different fields, and if this led to just one major gift or grant, this exercise would pay for itself many times over.
It should be done in an iterative fashion with the relevant decision-makers. For example, researchers shouldn’t spend a year working on a detailed scientific roadmap if there are no funders who have expressed interest in supporting the research.
It can be useful to know both (1) which ideas have broad support in a given scientific community, and (2) which ideas are creative and ambitious researchers most excited about, even if these are non-consensus ideas.
We could get better at it. For example, Ed Boyden and Adam Marblestone have described some techniques they use to identify and address key bottlenecks to scientific progress. An example is the Boyden lab’s use of “tiling trees” – the identification of mutually exclusive and collectively exhaustive options for solving a given scientific problem.
We could also expose people involved in agenda-setting to a broader range of tools and techniques for solving problems. They don’t have to be experts in all of them – they just need to know that they exist, and to have some heuristics for when and under what circumstances one might use them.
Economists can play a role by estimating the prospective value of an innovation. For example, I have become an advocate for increased research to discover and validate biomarkers after reading a paper by Heidi Williams et. al, in which they estimated that the social net present value of having better biomarkers for cancer is over $2.2 trillion!
Agenda-setting can occur at different levels of specificity, depending on the audience. I call this “forest, trees, leaves, and chlorophyl.” For example, a growing number of funders are aware that, used thoughtfully, and in conjunction with other computational and experimental approaches, AI can accelerate the pace of scientific and technological progress. They may also know that training data and the right benchmark is important. Organizations such as Align Bio are pursuing more detailed agenda-setting by answering questions such as:
What are transformative prediction and design tasks in a given scientific field or problem that AI could help solve?
Is there a dataset or other shared resource that could accelerate progress?
What assay could generate the data?
Is there an investment in platform technologies that could lower the cost of collecting the relevant data?
Q: What is a “field strategist,” and what role could they be playing in agenda-setting?
Most researchers are understandably focused on the next project they want to do. But some scientists and innovators can identify ideas that have the potential to move an entire field forward, or maybe create a new field, and can:
See something important about what can and should be done before it becomes the “conventional wisdom.”
Identify a bottleneck that is holding back progress in the field – e.g. the ability to measure and perturb a complex, three-dimensional biological system at different length scales.
Work backwards from that bottleneck (e.g. a new tool or technique, a key dataset) to identify the research directions and other activities that should be pursued to address the bottleneck.
Combine insights, methods and tools from multiple fields for potential novel solutions.
Determine whether the timing is right for a big push in pursuit of a given goal, or whether it is premature.
Objectively evaluate the merits of other people’s ideas or determine who is likely to be in a position to do so.
Identify the team that would need to be assembled or the portfolio of mutually reinforcing projects that would need to be supported to achieve the goal.
Determine the optimal (existing or new) funding mechanisms, incentive structure, and institutional setting.
Propose metrics and milestones for measuring both intermediate progress and the success or failure of an initiative, even if these are provisional and subject to change.
Help lead and manage a team or even a “team of teams.”
Q: How might philanthropists partner with field strategists?
Field strategists could serve as intermediaries between (1) philanthropists and foundations; and (2) the researchers and innovators who are interested in solving a particular problem or answering an open scientific question. For example, field strategists could help design and lead a fund, supported by multiple philanthropists and foundations. This could be particularly helpful for philanthropists that don’t want to create a large organization with highly technical staff, but still want to “give with confidence,” and to more established foundations who realize they will not be able to solve a given problem by themselves.
Q: What are some examples of questions and thought experiments that you pose to scientists and innovators, and that are relevant to agenda-setting?
In addition to a very broad question, such as “what are we not doing that we should be doing,” or “what are we currently doing that we should stop doing” – I find it useful to ask some more specific questions.
If you could call anyone on the planet (and it can be a conference call), who would you call and what would you ask them to do?
What platform technology could have a large impact on science, similar to the $100 genome or the electron microscope?
What dataset would accelerate progress in AI for Science, in the same way that the Protein Databank led to Nobel Prize-winning advances in protein structure prediction and protein design? More broadly, if the predictions made by the CEOs of AI companies are accurate, what are the complements to frontier models for AI for science that we should be investing in, such as cloud labs, or self-driving labs?
What technologies would lower the time and cost, and increase the success rate, of some key end-to-end innovation process? For example, cloud computing and open source software lowered the cost of launching an Internet or software startup. DARPA’s MOSIS program democratized access to semiconductor fabrication for academic researchers and startups.
What are plausible candidates for the next metascience experiment? An example might be supporting the emergence of new fields.
Which sectors have an important “public good” that we are under-investing in? A historical example is wind tunnels for aircraft design, supported by the predecessor to NASA. For fusion, it might be a Fusion Prototypic Neutron Source. This replicates the high-energy neutron environment that fusion reactors will experience, enabling the testing and validation of materials, components, and technologies needed to withstand these conditions.
What are classes of important innovations that VCs do not invest in, such as new antibiotics or diagnostics? Why don’t they? What would it take to change this, or what alternative financing mechanism should we experiment with?
What is an example of a critical institution that (1) operates under some important constraint; and (2) that constraint could be relaxed by a third party, such as a philanthropist? An example is long-term investors, which have $87 trillion in assets under management. Some are pension funds governed by state legislatures, and therefore may be operating under constraints which prevent them from living up to their full potential.
Imagine that a federal agency that currently has little or no capacity to support research and innovation grew such a capability, and you are the head of the research and innovation arm. What goals would you set, and what projects would you support to achieve those goals? For example, the Department of Labor has a very modest research budget, and can’t harness advances in science and technology to increase the impact of workforce development programs on wages.
What are strong candidates for the next private sector-led “buyer’s club” – like Frontier Climate, which has made a $1 billion commitment to purchase permanent carbon removal?
What federal market-shaping initiatives could have the impact of Operation Warp Speed or the NASA COTS program, which used milestone payments to help finance the development of SpaceX’s Falcon 9 rocket?
Imagine that you have been given the job of proactively identifying top scientists, engineers and founders from around the world to contribute to the U.S. economy, and that you have a commitment from the U.S. Government to support this initiative. What would you do?
What problems can and should individuals and organizations with longer time horizons work on? Examples include bootstrapping a solar system civilization – using more of the mass and energy in space to expand the sphere of human activity beyond Earth, and projects that increase civilizational resilience in the face of natural and man-made threats?
Well-run companies are good at scale but are unlikely to address problems that have high social returns and low private returns. Non-profits are good at working on those kinds of problems but struggle to scale their interventions. What models would allow us to get the best of both worlds?
