Hacker News Reader

Comment by ryebit | Hacker News Reader

IMO, economics is an area that will continually be hampered by models that don't fit, and increasing complexity.

Any model that predicts things simply and well, will be gamed; thus requiring a more complex model.

Or as Goodhart's Law[1] has it: "When a measure becomes a target, it ceases to be a good measure."

That's presuming we never reach some economic state so perfect, that even with an accurate model, no one can find an extra advantage, nor can it be disrupted by natural events.

[1] https://en.m.wikipedia.org/wiki/Goodhart%27s_law

View on HN

Economics like psychology aren't true sciences yet because they don't have a repeatable theoretical model. They're like Alchemy before it evolved in to Chemistry or Astrology before it evolved in to Astronomy. How can you get a repeatable theoretical model? imo you would need to model a world simulation near or at the level of atoms. The more abstraction and the more assumptions the simulation has, the less useful for the social "sciences". imo this is what Hari Seldon's "equation" is in the Foundation series; a simulation

'Repeatable theoretical model' is not an episemological criterion of a science, usually for empirical sciences what is needed are fasifiable predictions.

Falsifiable predictions are generated by theories. You don’t just throw out predictions haphazardly and test them, they always originate from a theory.

Note that even with Seldon's predictions, a requirement for their fulfilment was that the predictions not be made public so that they would not be altered. The difference in economics and psychology, unlike in chemistry or astronomy, is that these fields are self-reflective. They have an "awareness" of themselves as disciplines and can change and act on information of themselves to in turn gain new "awareness" and behaviour.

Economics is even worse, as it does hold to some fundamental aspects of finite resource allocation and supply/demand in addition to the self-reflectance. This means its model is probably some hybrid of physics-like fundamentals and the weird self-awareness of the "soft science" disciplines.

> IMO, economics is an area that will continually be hampered by models that don't fit, and increasing complexity.

Psychohistory

https://en.m.wikipedia.org/wiki/Psychohistory_(fictional)

The relationship is circular. Asimov's story was of course based on emerging notions of economics, which had become increasingly mathamaticised and grounded in statistics. His writing also affected and influenced economists, and at least one, Paul Krugman, has stated this specifically:

There are certain novels that can shape a teenage boy's life. For some, it's Ayn Rand's Atlas Shrugged; for others it's Tolkien's The Lord of the Rings. As a widely quoted internet meme says, the unrealistic fantasy world portrayed in one of those books can warp a young man's character forever; the other book is about orcs. But for me, of course, it was neither. My Book – the one that has stayed with me for four-and-a-half decades – is Isaac Asimov's Foundation Trilogy, written when Asimov was barely out of his teens himself. I didn't grow up wanting to be a square-jawed individualist or join a heroic quest; I grew up wanting to be Hari Seldon, using my understanding of the mathematics of human behaviour to save civilisation.

https://www.theguardian.com/books/2012/dec/04/paul-krugman-a...

Haha

I studied Physics because of reading Asimov as a child, and discovered it was way more difficult and boring in real life. Re-read Foundation when I heard the TV show was coming out, and discovered that it was a terrible book. Asimov, you old deceiver

What made you think it was terrible? I recently reread it too, and while the writing isn’t earth-shatteringly good, I definitely appreciated it for being the novel idea it was at the time.

It's not really a story at all, just an exploration of an idea. The plot is very predictable, and the characters are just the thinnest of veils for whatever point the author is trying to make.

First time I've ever heard of this (always preferred Heinlein to Asimov) and I'm glad I'm not the only one who immediately drew connections to Historical Materialism and Marx.

I have thought this for years. It’s like they don’t realize they are studying a living system, and even worse one that is actively trying to invalidate their models.

I think most economists understand the adversarial nature of an economy, which is why a lot of their models are built on game theory.

I think the bigger issue is that people are not as rational, nor as consistent, as the models treat them. The fundamental premise of a lot of economic theory is that people have consistent, ranked, preferences, and that people work to maximize their preferences. Most people don’t fit that.

My sister is a political science professor, and she says this is an open secret in political science. The assumption is that people have consistent preferences (e.g. someone who is pro gun control is always pro gun control), but research has shown people are fickle, and will give different answers at different times.

Living organisms also adapt to their environments and models which understand them.

If you want to get to base levels, look at metabolism, energy inputs, conversion efficiencies, productive activity, and immunity / defence / risk dynamics.

Economic analysis arose out of cost accounting in which those inputs were specifically discounted, under-priced, and/or externalised.

But that quote is literally from an economist... as is Lucas' critique?

I don't think that invalidates the quote. It's how I came across it in the first place.

I think economists are well aware of the nature of the beast they're studying.

And even if doing the task is daunting, and the landscape changing as it's described, it's still worth it.

Especially if anyone finds fixed points which do have the property that describing them doesn't invalidate them.

Reply on news.ycombinator.com