Mr. Harford is an economist, writes for the FT, fellow at Oxford & somewhere else, author of a few books.
The main idea of this talk is that trial and error is the best way to solve complex problems. An advantage of markets is they allow trial an error. Think "limited liability company". Compare (US) west coast "fail faster" and east coast "too big to fail". Which adds more value to the economy?
The key component is error. Many things will fail. People are bad at predicting the future, even experts are usually wrong.
A better word is experimentation. Have several different approaches tested out. He is very fond of Cochran & evidence based medicine-- conduct randomized trials.
Mr. Harford is also concerned that governments are really bad at this stuff. People don't like leaders who are not confident that they have the solution. School lunches, it is ok for Jamie Oliver to do it, but no way can Blair run experiments on school food.
The coke analogy. People want the same standard everywhere. No post-code lottery. My hospital should be just as good as the other hospital. This standarization is the death of experimentation. You cannot try out different things if everything has to be the same.
Right. So my take:
The experimentation takes place on the level of the society. An individual, or smaller group, depends on a vision and absolute conviction to that vision. The trial does not succeed without the individual vision pushing it.
Example from this talk. Army colonel in Iraq sees how to fight that thing, and does it. Universally scorned by his commanders because he isn't doing it there way. Risking his men's lives. But also, he was right, or rather, his method was hugely successful.
Point from the discussion. The importance of luck, both lifting one up and pulling one down.
No comments:
Post a Comment