Thread by @jasoncrawford

OK I'm going to try this again. Let's talk about feedback loops. In:

• business
• science
• charity
• government

https://twitter.com/jasoncrawford/status/1196157228098605056

I'm taking a sort of ecological perspective here. That is, if you think if these things almost as biological systems, what are their life processes, how do they self-sustain, what evolutionary pressures do they face?

A business is sustained by generating its own profit, by pulling in revenues greater than expenses, over the long term. It does this by creating value for customers at a price > costs of production. If it fails in this, it runs out of money and shuts down or sells itself off.

How can it avoid this feedback loop, and for how long?

• Investment: sell promise of the future to replace profits

• Fraud: trick consumers into thinking they're getting value when they're not

• Government: legal monopoly or regulatory capture

Investment is limited; investors won't prop up a failing business for more than a matter of years

Fraud can go on indefinitely but could be exposed at any time. And note there is an active incentive from competitors among others to expose the fraud

Legal monopoly or other government intervention can go on indefinitely, unfortunately. (Which is one argument for eliminating those things)

Science has different feedback loops. First, there is the epistemic feedback loop of observation and testable hypotheses. This is great! It is why science makes progress. As long as the community is intellectually honest and committed to the growth of knowledge, this works.

But what about the ecological feedback loop? How is science sustained as a human activity? Researchers can do their work as long as they continue to be funded to do it. How tightly connected is funding to actual progress towards truth and understanding?

I submit that the connection here is not as tight as the connection in business between profit and generating economic value.

Much science funding comes from government & universities. Lots of room for influence of prestige & status to creep in.

To the extent that funding correlates with actually advancing knowledge, the feedback loop works. Good researchers get more funding and bad ones fail and get different jobs.

But when this correlation breaks down you get things like the replication crisis in psychology.

It's worse with charitable organizations. Like universities and other non-profits, they're sustained by outside donations. But unlike science, most charities don't—and aren't required to—prove their effectiveness through any objective means.

A charity can go on indefinitely doing no actual good for the world, or even doing harm, as long as it can convince donors to keep funding it. And in charity, issues of prestige, status, moral signaling, etc. loom even larger than in science.

Another way to look at it: where can people best make sure their dollars are spent wisely?

a. buying stuff for themselves
b. giving to other people to do science
c. giving to other people to do “good”

Re both *ability* and *incentive* to make good choices, I submit: a > b > c

What about government? Hoo boy.

Government sustains itself by staying in power. It's funded by tax revenues that it can collect because of that power. The power ultimately comes from popular approval. But there's a lot of friction in that mechanism.

First, how closely connected is popular approval to the government that we have? How much does government give the people what it wants?

Key question: If the people don't like the government, how much ability do they have to change it?

For most of history, this was possible only through violent revolution! Not only did that suck in itself, but it meant you had to be *really* upset to change anything.

Democratic republics much better in this regard! But many people still feel powerless to effect much change.

Second, how much do the people even know what's good for them?

Ideally the feedback loop would lead to government getting better over time, by the standard of prosperity, health, happiness of the people. But for this to happen, people have to demand the right *root causes*

Unfortunately, people's political choices are often driven by mistaken ideologies or simple tribalism. And the consequences of their choices are long-term and diffuse, so it's hard to self-correct even if they wanted to.

(I'm realizing now that a full analysis of this should probably look at *both* epistemic and teleological feedback loops in each area. But that will have to wait for another thread)

Bottom line, I think the feedback loops are the least functional in government, and most functional in business, with science and charity being in between (science probably closer to business, and charity closer to government).

Again: anything that *can* be for-profit, should be