Questions to ask about funding models

by Jason Crawford · April 24, 2020 · 2 min read

Previously: the importance of funding models, types of funding models.

To compare funding models for science and technology, and to understand their pros and cons, it will help to look at them along multiple axes. Here’s a list of questions it can be useful to ask about each model, system, or program, loosely grouped in a few categories:

Resources

• How much total funding can the model aggregate and concentrate on a program?
• How scalable is the funding? If the research is promising or productive, can funding grow to meet demand?
• How reliable is the funding? Is it uncertain or subject to high variance?
• How well can the model attract talent (via compensation, career path, autonomy, prestige, etc.)?

Scope

• What is the model’s time horizon? How soon does the research have to deliver results? Are there short-term pressures?
• Is the research highly goal-directed? Must projects be justified as contributing to a defined objective? How much undirected exploration is supported?
• Related, how much of the direction is top-down from the management or administration, vs. bottom-up from the researchers themselves?
• Is the research otherwise constrained by a defined theme?
• To what extent does the model support basic vs. applied research? Does the research have to be justified by visible or near-term practical applications?
• Does the value of the research need to be captured by the organization (through intellectual property, commercialization, process improvements, etc.)?

Efficiency

• Can progress and results be objectively identified? Measured? Optimized?
• Is the system able to identify and eliminate waste?
• What pressures, incentives or mechanisms exist to drive efficiency?
• What feedback loops exist, or are missing, for iterative improvement?

Output

• If the research can be commercialized, how likely is it for that to happen?
• Are the results of the research shared openly? Does the model encourage or even allow publishing?

Diversification

• Is the model risk-tolerant, or conservative? Does it support high-risk, high-reward bets?
• Does the model reward being right early? Are rewards proportional to risk?
• To what extent are funding decisions centralized vs. decentralized?
• To what extent are decisions made by consensus, vs. by individuals?
• Does the system encourage participants to differentiate from one another?
• Does the system support contrarians? Does it actively promote them?
• Overall, to what extent does the system lead to a diversified portfolio of bets?

Social pressures

• Are there pressures to focus on socially-approved goals, e.g., those seen as humanitarian or otherwise noble?
• How sensitive are funding decisions to ingroup politics?
• To national politics? Public opinion in general?

More broadly, we can simply ask: what are the pressures and incentives in the system, and what do they lead to?

---

Broadly, when I look at the different institutional funding models, my impression is:

• Private non-profit models are the most able to be long-term and to do undirected/exploratory research, but are limited in scale and subject to social pressures

• Government models have an advantage in scale, but are the most subject to politics and are often risk-averse (military might be an exception)

• For-profit models have an advantage in efficiency, risk tolerance, and overall diversification; but they are limited by time horizon and the need to capture value, and are thus better suited to applied vs. basic research

I’m treating these as guesses for now.

Comment on Reddit