I was involved in pandemic preparation work before COVID.
I raised the alarm. I got laughed out of the room. Then COVID hit, and it was exactly the mess I expected, for exactly the reasons I’d laid out.
I’m telling it because it’s the clearest example I have of the thing this book is about: the information is nearly always available and almost nobody in a position to act on it does.
Why did nobody act on pandemic warnings?
Because the incentives run the wrong way and everybody involved was behaving rationally.
Preparing costs money now, attention now, and political capital now, against a risk that’s diffuse and might arrive during somebody else’s tenure. The person who raises the alarm pays a reputational cost immediately, in the room, in front of colleagues. The person who dismisses it pays nothing until the event, and by the time the event arrives, the connection between the dismissal and the outcome is no longer obvious to anybody.
That asymmetry is a structure, and it produces the same result with different people in the chairs.
The outrage is that this structure is well understood and nobody with the power to change it bothers. The individuals in the room were behaving rationally. The people who could make a dismissal cost something, and never have, are the ones I hold responsible.
Watch: The Coronavirus Explained & What You Should Do, Kurzgesagt – In a Nutshell
Why write about pandemic preparation angry?
I wrote this book because I’m angry. Angry that the information in these pages is available to anyone who looks for it and yet almost nobody in a position of authority acts on it. Angry that the tools to solve these problems exist and aren’t being used.
I could have written a cooler version. Plenty of people do, and the cool version reads as more credible, and that’s the problem. There’s a tone of writing about serious risk that’s so measured it communicates that nothing much needs doing, and that tone is the house style for a great deal of expert commentary. The anger is the real response to the situation and I decided not to launder it.
Do books about pandemic preparedness go out of date?
The details do, immediately, and the book says so on the first page. The Iran war broke out while I was writing this and reached a ceasefire before I finished, and the facts on the ground shifted daily the whole time. Casualty figures, oil prices, political developments will all be outdated by the time anybody reads it.
That doesn’t make the argument wrong. The specific numbers are illustrations of a structure, and the structure is what I’m claiming. If the figures move and the argument still holds, the argument was about the right thing. Books that avoid this problem by never touching specifics are safer and considerably less useful.
Do pandemic preparedness predictions hold up?
The book makes predictions and some of them will be wrong.
Getting an institution to move is a problem about institutions, and people who treat it as a problem about information are chasing a dead end.
I say that in the text instead of in a footnote, because a book that predicts without admitting the error rate is doing something other than analysis. The conclusions are mine, the errors, if any, are mine, and the opinions are definitely mine. What I’d defend is the reasoning instead of the individual calls. If the reasoning is sound and a specific prediction fails, that’s useful information about the world. If the reasoning is unsound, the book is worthless whether or not it happens to be right.
What does the pandemic example teach?
That being right in advance is worth almost nothing.
That’s the uncomfortable lesson and it took me a while to accept it. I’d assumed that the value was in the analysis and that the analysis would eventually be recognized. It wasn’t, and being able to point at the record afterward changed nothing, because the preparation had to happen before anybody believed it.
Which means the useful question is how a warning gets acted on. Plenty of people see things coming, and seeing it was never the hard part. Getting an institution to move is a problem about institutions, and people who treat it as a problem about information are chasing a dead end.
My take
Correct warnings are worth almost nothing on their own, and I learned that the hard way. I warned about pandemic preparation, got laughed out of the room, and pointing at the record afterward changed nothing. Institutions took far more work to move than being right about the risk did. If you’re the one with the warning, stop polishing the analysis and work on who has to act and why they won’t.
What is worth doing when nobody listens to a warning?
Assume people will ignore the warning and plan for that. Anything that depends on being believed in advance will fail.
Prepare at whatever level you control. If the institution won’t act, the household and the business still can.
Watch the structure and let the headline go by. The specifics are stale on arrival and the forces producing them aren’t.
Say what you think and put your name on it. The measured version of a serious warning communicates that it’s not serious.
Publish the error rate. A prediction without one is entertainment.
I got laughed out of that room and I’d go back into it. It wouldn’t go differently, and I’d still go, because the alternative is to know something and say nothing, and I’ve watched where that leads.
| Rule | Why it matters |
|---|---|
| Assume the warning will be ignored | Anything that depends on being believed in advance will fail, so plan for that. |
| Prepare at the level you control | If the institution won’t act, the household and the business still can. |
| Watch the structure, not the headline | The specifics are stale on arrival; the forces producing them aren’t. |
| Say what you think and sign it | The measured version of a serious warning communicates that it’s not serious. |
| Publish the error rate | A prediction without one is entertainment. |
The book on this: Stuck in the Middle is the full argument, written while the facts on the ground shifted daily, with the conclusions, the errors and the opinions all mine.
