Hacker News new | past | comments | ask | show | jobs | submit login

Economics could (and should) be much more of an engineering science than it is. The math and theories are rather beautiful, but do not have much to do with the real non-linear chaotic world in which we live. For some reason, no one expects that economic theory be predictive in the same sense that we expect of theories in the physical sciences.

It seems to me that a bit of data and a little bit of modeling would have greatly improved the opinion piece this document describes.

As for Economics, it is really time to rethink the field. I rather liked Eric D. Beinhocker's book, The Origin of Wealth (http://www.amazon.com/Origin-Wealth-Evolution-Complexity-Eco...). See also his Wikipedia bio: http://en.wikipedia.org/wiki/Eric_Beinhocker.




It _can't_ be totally an engineering field anymore than language communication can. Trade and marketing are not about maximizing profits or reducing costs, though those are important aspects. It's as much politics and power and sex and violence as it is anything to do with prices and shipping costs. No matter how "good" your models are, people will get funny ideas in their heads and do something completely unexpected.

When Asimov posited his Pyschohistory in the Foundation books he indicated that it could only be statistically useful on populations measured in the _trillions_. And even then it could only work if the methodology were kept absolutely secret from everyone in the modeled society.


What passes today as economic analysis is, in truth, modeling in which the assumptions and the data are implicit and only the conclusions are expressed. Moreover, projections are frequently based upon simple extrapolation of a constructed macroeconomic indicator and rather than exploring the underlying structure. It seems to me that "politics and power and sex and violence" and resource depletion,innovation, FUD, and all the rest is amenable to modeling and that a composite economic model based on systematic analysis is likely to better reflect reality than the process used now.

Asimov's Foundation Series books are enlightened works of science fiction, but they are fiction, not fact. I find is strange that you cite them as if they were seminal works in societal modeling.


I said Asimov "posited", which seems pretty clear to me. FWIW, Asimov has a certain cachet these days among some Keynesian economists.

When you are able to usefully, and predictively model "politics and power and sex and violence" then you will be doing extraordinary work and I'll look forward to you getting your well deserved Nobel Prize. :)




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: