[fpadilha]

This article confuses how top-level economists (who are de facto politicians) that attribute sometimes questionable decision making to models (most of the times due to political pressure) with a honest and very complex (at the edge of academic research) profession. It points out the it got so "mathy" that it kind of decoupled from non-technocrats but that can easily be said about literally any field today. What amazes me most is that the author complains about how that happens in the US, the most transparent and actually research-based application of economics in the world. I would invite you to go to some other less developed countries to see how "wonky" economics couples with intelectual dishonesty tries (and many time succeeds) to fool the many in favor of the few. And again, using wonky arguments to fool people is a practice that happens in any field, dismissing an entire profession who actually brought many advances to modern societies is ridiculous. The advance in statistics and econometrics is what made machine learning possible... I have to say, it has been a while since I've read such nonsense trying to pass as a good analysis/critique.

[Pamar]

This happened in USA, right?

http://www.newyorker.com/news/john-cassidy/the-reinhart-and-...

[fpadilha]

Cherry picking a controversy doesn't do much to generalize an entire field (if you want I can find some Freud controversies and say that psychology is hocus pocus). By the way, I'm not american. Funny enough, that one you picked is exactly the opposite from here. The authors were harshly criticized for NOT applying "mathy" rigor. The conclusion from "This time is different" is exactly what common sense would tell you, don't get too much debt or else you might grow less in the future and they got crushed because they did not entered in the extreme math/econometrics analysis. What the article criticizes is exactly what "This time is different" did, it applied less data rigor to point out almost a common sense and yet it backfired, so what is it? Should economists apply more or less "math"? I guess either way, this misses completely the point about economics and dismissing the entire field because of it is ridiculous.

[Pamar]

I am not American either. I pointed this out because you seemed to imply that USA was the best example of how to properly use a mathematical approach.

Also, I think it is appropriate: what we had here was someone "demonstrating" why a specific policy works: "it works because our math model proves it works". So instead of going for an empirical proof everybody was expected to be ok wih this because, look, we have math. Also, if it is not working for you it's because you are not doing it right.

In a proper scientific discipline you design experiments to verify if your formulas are correct: if you send a rocket to Venus and it crashes back on Earth you can't just say "you didn't it right, I have an Excel sheet that says it should have reached the target"...

[robert_foss]

> the US, the most transparent and actually research-based application of economics in the world

An assertion based on what (apart from American exceptionalism)?

[fpadilha]

Sorry didn't mean it like that. I meant one of the most. Again, I'm not american.

[fiatjaf]

> the US, the most transparent and actually research-based application of economics in the world

Exactly. And if even with all the "research" all the predictions are wrong all the times and all the political measures based on economics studies give yield bad results all the times that makes the argument correct, right?

[fpadilha]

I think you are confusing what economics is then. Economics is not only about predicting, actually there are various fields in economics focused in different kind of predictions techniques - including such ones that (coupled with the field of statistics/mathematics) resulted in big data analysis and machine learning. In one of many definitions that you can find in Wikipedia definition of its "Economics" page:

"Economics is a study of man in the ordinary business of life. It enquires how he gets his income and how he uses it. Thus, it is on the one side, the study of wealth and on the other and more important side, a part of the study of man.[21]" Lionel Robbins (1932) developed implications of what has been termed "[p]erhaps the most commonly accepted current definition of the subject":[22]

Economics is about understanding how your everyday actions impacts the whole, it is a study that allowed whole societies to organize and plan at the beginning and now it dares, through data, to predict events. And believe me, if companies and people couldn't have a minimal idea of the inflation rate by the end of next year, investments wouldn't occur as much (because the interest rate would be higher) and everyone would be worse off. Obviously, trying to predict the future is almost nonsensical, and yet those methods apply to every field nowadays. Marketing without data is nothing, does a marketing analysis always result in a successful prediction? No, and yet if you don't actually engage in trying you don't have a clue of what works and is ultimately worse off. Again, there are predictions for every possible outcome. There are researchers that say is no environment crisis and use data to back that up. The policy that ends up being made is about the study, but about the elected by the people that chooses that prediction it think it suits the society that he represents better, doesn't mean that aren't very serious people doing work that benefits us all doing those predictions.

[elmar]

"economics is the study of the use of scarce resources which have alternative uses." Thomas Sowell

https://en.m.wikipedia.org/wiki/Thomas_Sowell

[fpadilha]

typos in last phrase: The policy that ends up being made is NOT* about the study, but about the one who was elected by the people thinks suits the society that he represents better, doesn't mean that aren't very serious people doing work that benefits us all doing those predictions...

[_ptgt]

Economics may not be about predicting, but if economic theories can't predict well (in the statistical sense, meaning the generated data should fit your predicted probability distribution), then the economic theory doesn't reflect reality. And thus has questionable use in policy decisions.