Insects and Economics
Mar. 2nd, 2007 10:07 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
dickensI just finished a fascinating book called Butterfly Economics by Paul Ormerod.
The point of the book is that many economists try to treat their field like they are describing a machine which has predictable outputs based on the inputs. (Though they would probably be okay with the suggestion that it is an enormous black box that they need to know more about.) What is missing from this model, and what Ormerod discusses, is the fact that individuals can affect the behavior of other individuals.
It really ought to have been titled Ant Economics, but there was probably an editor suggesting that wouldn't sell as well. Ormerod discusses how the economy is more like a living system and uses some experiments with ants to make the point.
Give a colony of ants two identical food sources (or two identical paths to the same food source), ants can go to either, or be influenced by other ants to change their behavior. The system never settles into a stable pattern of X ants going to one and Y ants going to the other.
Then he discusses what this means for economic policy. Politicians want to know 'what to do to get a desired result', and the random nature of the economy makes it impossible to do that in the short term.
However, based on the model, there are some things that you can model... Market volatility for example: If ants are generally unlikely to change their behavior the ratio of ants going to 1 pile or the other will swing wildly. (That may seem counter-intuitive, but in this case a couple very 'convincing' ants could change the behavior of the entire colony and there would be little change until another 'convincing' ant arose that influenced the colony in the other direction.) If ants change their behavior easily, there will be smaller ratio changes because it is harder to maintain influence.
Okay, now think of stocks. A well known company's stock will have lots of influences, both positive and negative, working most of the time. It's price will be reasonably stable. Now consider a penny stock scam. One scammer (influential ant) convinces a lot of people to purchase a stock that doesn't have a lot of activity. The price rises as more people buy in, until the scammer sells their stock and the price crashes again.
The point of the book is that many economists try to treat their field like they are describing a machine which has predictable outputs based on the inputs. (Though they would probably be okay with the suggestion that it is an enormous black box that they need to know more about.) What is missing from this model, and what Ormerod discusses, is the fact that individuals can affect the behavior of other individuals.
It really ought to have been titled Ant Economics, but there was probably an editor suggesting that wouldn't sell as well. Ormerod discusses how the economy is more like a living system and uses some experiments with ants to make the point.
Give a colony of ants two identical food sources (or two identical paths to the same food source), ants can go to either, or be influenced by other ants to change their behavior. The system never settles into a stable pattern of X ants going to one and Y ants going to the other.
Then he discusses what this means for economic policy. Politicians want to know 'what to do to get a desired result', and the random nature of the economy makes it impossible to do that in the short term.
However, based on the model, there are some things that you can model... Market volatility for example: If ants are generally unlikely to change their behavior the ratio of ants going to 1 pile or the other will swing wildly. (That may seem counter-intuitive, but in this case a couple very 'convincing' ants could change the behavior of the entire colony and there would be little change until another 'convincing' ant arose that influenced the colony in the other direction.) If ants change their behavior easily, there will be smaller ratio changes because it is harder to maintain influence.
Okay, now think of stocks. A well known company's stock will have lots of influences, both positive and negative, working most of the time. It's price will be reasonably stable. Now consider a penny stock scam. One scammer (influential ant) convinces a lot of people to purchase a stock that doesn't have a lot of activity. The price rises as more people buy in, until the scammer sells their stock and the price crashes again.
