Should We Trust GDP Prediction?

From an article published on eToro on November 2021.

The track record of economists’ forecast compared to reality is quite bad.. (see image attached) and to the surprise of nobody (is well documented in the works of Kahneman and Montier among the most famous..).

But why is that?

It maybe because we use rotten math models to make those predictions.

Indeed the basic assumption is that risk is normally distributed, when in fact economics is a “complex system”. As other complex systems, it is poorly explained by Bell curves (Gaussian) and best explained by power laws.

In short, a power law is a mathematical relationship where the frequency of some event varies as a power, or exponent, of some characteristic of that event. It’s also know as “fat tail” system.

Enough talking, let’s view an example: Black Monday (1987). The probability of Black Monday

• under a power law distribution is more like 10^-5 (meaning a decent chance we’ll experience that type of event every 100 years),
  
• under a normal/gaussian distribution are 10^-148 or essentially impossible in the lifetime of the universe.
  

(graph from: amp.theatlantic.com/amp/article/250211/ ).