In this world of quant-bots, I don't know how anyone can argue that we have a fully rational market.

My understanding is that much of the progress in economics has been merging economics with psychology to identify rational failures.

"Homo economicus" is still really important for macroeconomics. The reason for this is pretty simple: designing rich, large models is still hard to do and the practical limitations introduced by basing your assumptions on the idea that people act irrationally instead of rationally can make things too complicated to be of practical use.

So in microeconomics or small models, people can practically accept and implement these (pretty obviously true) ideas that people don't behave rationally. But in large scale macroeconomic models it's hard to do. It would certainly be a lot easier if people just acted like computers...

The issue is not just that models are hard. The claim of some economists is that while people behave irrationally, not everyone behaves irrationally all the time, so the issue is self-correcting as long as you have enough liquidity and participants in the market.

It's like the law of large numbers; while a single transaction may have a completely wrong price, a sufficiently large number will average the irrationalities out.

If people make biased (http://en.wikipedia.org/wiki/Bias_of_an_estimator) guesses you will still have a problem. I think that's often what's meant by "rational": that people, in aggregate, make unbiased guesses.

The truth is not so - people get irrationally exuberant or are afraid to cut their losses, etc. These are problems of statistical bias of their estimations - and the opposite is assumed in many (most?) economic models.

Most of the time it's like that, but sometimes you have a situation like nominal loss aversion/sticky wages where humans are very consistent about their irrationality and that irrationality has very important effects.

http://en.wikipedia.org/wiki/Sticky_wages

http://econlog.econlib.org/archives/2013/09/why_dont_wages.h...

Simplifying models has always struck me as a particularly dangerous idea in economics, since a lot of the time it's actually profitable for economic actors to deliberately exploit the difference between your assumptions and reality.

I believe the argument (massively simplified) goes that if the market were truly irrational someone would be able to reliably beat it and there's not much evidence that that has happened.

Quant bots make the market better by increasing liquidity.

I've always thought that's a sketchy claim. The market would have sufficient liquidity without the bots.

More importantly, they're a huge waste of resources that produces nothing of economic value.

There's no such thing as sufficient liquidity.

Which one claims that all actors in a market have perfect information?

I'm guessing parent's talking about Eugene Fama, who shared the prize with Robert Shiller.

He doesn't actually claim the markets have perfect information, just that the price always reflects all available information. In essence, you can't "beat the market" consistently, assuming you have the same information.

> assuming you have the same information

Which you don't, since Goldman Sachs is always a few milliseconds ahead of everyone else. Given that almost everyone else out there is going to be trading on information that has already been consumed and acted upon by privileged parties, the markets may as well for all intents and purposes be irrational.