Hacker News Reader

Comment by maccam912 | Hacker News Reader

When it's a survey of many economists I tend to put stock in what comes out. In this case if 80% say there is a 25% chance of a second dip, thats a much more informed opinion than I could ever have on the subject. I now know it isn't likely so don't plan my life around waiting for the second dip, but also don't be too surprised if we do see a second dip.

View on HN

How accurate were those same economists with their previous predictions? Were they more accurate than random people off the street?

Yes, they’re more accurate than random noise. Unfortunately like with any science which seems impactful on people’s lives, there is a lot of drama around people who try to become famous by making predictions and then failing. But the science itself is on firm ground.

Do you have actual data on prediction accuracy for economists versus non-economists? I'd be interested to read the research.

> When it's a survey of many economists I tend to put stock in what comes out. In this case if 80% say there is a 25% chance of a second dip, thats a much more informed opinion than I could ever have on the subject.

Who do you think has a higher chance of being right:

* macroeconomists when predicting the future of the economy

* fund managers when predicting the future of stocks

Just because people are smart or know a lot of things doesn't mean that that knowledge translates into useful ability.

> Economists are as effective as astrologists at predicting the future.

That's unfair to astrologists, economists are not going to be that good. :-)

To be fair, the better economists work by creating models based on reality. But we don't have a rich dataset about modern economies responding to pandemics. It's not that they are stupid, it's that economies are difficult to predict in general, and we don't have the data sets to give any confidence for this case.

I don't think the term "economists" refers to an easily generalizable group. In my experience there is more world-view variation among economists than one might expect looking at the differences between bayesians vs. frequentists, etc. Some economists don't embrace modeling as a means of arriving at policy recommendations at all, for instance. Others do of course, but influence doesn't seem as tied to quantitative assessments as other fields.

It's true that some use quantitative models, and some don't. But all experts use past experience as a guide. As the present becomes more different from the past, it becomes more uncertain that past experience is relevant.

Reply on news.ycombinator.com