Institutional Design Under Scientific Uncertainty: Lessons from the Fauci Case
- 1 day ago
- 2 min read
Updated: 1 day ago
Arguments about Anthony Fauci almost always become arguments about Anthony Fauci. Was he right, was he honest, did he deserve the trust he was given? Those are legitimate questions. They are also the least durable ones, because the next pandemic will put a different person under the same pressures.
Here's a more relevant question:
How should democratic societies govern when consequential decisions must be made before the evidence is complete?
There are three surprising parts to the answer:
First, the familiar trade-off in crisis communication — project confidence to secure compliance, or admit uncertainty and lose it — is largely false. Experimental work by van der Bles and by Petersen and colleagues, covering nearly 19,000 participants, found that transparency about uncertainty costs very little trust, while vague reassurance costs trust and fails to improve compliance. Ambiguity was the worst option available.
Second, the comparison across countries does not vindicate any leader. Sweden's approach was nearly the opposite of the American one, and the United Kingdom used a collective advisory structure rather than a single visible expert. All three systems erred. What varied was not expertise but the price each attached to changing its mind in public.
Third — and this is where the argument touches my own work — credible review cannot be improvised. Institutions command confidence when their procedures are fixed before anyone knows whose conduct will be examined. That is the same principle underlying every well-designed negotiation process: an agreed method, settled in advance, produces outcomes the parties can accept even when they dislike the result. Fair process is not a substitute for good judgment. It is what makes good judgment survivable when it turns out to be wrong.
The full argument, with sources (8 pages):
Comments