= 27d567bc2d48d9f40e9ccc20ea370b15)
falenas108 comments on Is Politics the Mindkiller? An Inconclusive Test - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (276)
I always wondered why the Less Wrong community was so "libertarian" (US-style, ie, pro-free market).
It seems at odds to me with LW views on other topics. Free market is akin to evolution : it's at optimisation process which, given enough time and space, will end up finding local maxima, but it's a blind, uncaring force that doesn't care about the sufferings it produces, that has no long-term vision. It's Azathoth. The same way that good engineering is more efficient than evolution (show me a bird flying as fast as a plane), wouldn't a good partially planned economy be better than free market ?
Or if you look at it from a CS view, especially with the SIAI view on AI (which is not shared by all Less Wrongers, but by most) : we use Azathoth-like solutions (neural network, genetic algorithms, ...) when we don't have a classical engineering solution. Shouldn't we do the same in economy ? Try to have more "engineered" solution when we can do so, and resort to the "free market" as a suboptimal but working default when we don't have an engineered solution ? If you look at EDF or SNCF (french electricity and railroads), it seems there are domains in which the "engineered solution" works well.
It would seem more coherent with the rest of the LW view to support things like Cybersyn rather than Azathoth.
Also (but my comment is already too long so I won't elaborate that one), a rational view on human psychology and cognitive biases should tell us that Homo Economicus just doesn't exist, and things like "consent" and "free will" are always a bit fuzzy, humans are prone to error and manipulation, so we should have safeguards to ensure the errors done by individual don't completely ruin their (or others) life, which argue for a strong social safety net.
At least one problem with this is that any attempt to actually control the market will almost definitely get sidetracked by politics instead of what works. With lobbyists involved, I wouldn't trust the government to do what's best for the country. See farm subsidies for an example.
I understand the issue, but I'm at odd with it for three reasons :
If the problem is lobbying and corporate corruption of the government, I don't see how getting rid of the proxy and putting directly the corporations in charge will make anything better. Regulations may be imperfect and biased by lobbying, but having the corporations directly in charge seems even worse to me.
It seems to me by looking around the world than when a reasonably democratic government starts providing real services to the population (universal healthcare and education, social safety net, ...) the people become less apathetic towards the government, and will get more involved with how the government is runned. It also seems to me that countries with higher wealth redistribution, like Scandinavian countries, have lower corruption.
This is a kind of defeatist arguments. Here at Less Wrong, we speak of defeating death itself, conquering the stars, breaking the FAI problem, getting to the "level above" in understanding of the world, and yet, on this specific issue of politics/economics, we concede defeat so easily ? There are countless ways to "actually control the market" that we could imagine. Shouldn't we try to find a political system that ensures the market is controlled in a reasonably efficient way, rather than giving up ? Doesn't sound harder than solving the FAI problem. Corruption and lobbying ? What about making a jury trial for every law after the Parliament voted it, with 20 randomly selected citizen, held isolated from pressures like in normal jury trial, decide if the law goes through or not ? That's just one random idea in the enormous space of possible mechanisms. Why do we give up so easily ?
I wonder about cause and effect here. I would trust the government more to redistribute wealth fairly if it weren't so very corrupt and incompetent.
I can think of several objections. Some laws are very complicated and require a lot of staff work by experts to formulate, and the jury won't be able to do a good job. Or did you want to require the jury to spend six months listening to testimony before delivering their verdict?
Do you mean an American or British-style jury, that requires unanimity? If so, you will reliably get hung juries on any controversial law.
There's a lot of laws that are time-sensitive. If the appropriations bills don't pass, the government shuts down. And we don't want to allow indefinite delays in restarting the government until the jury reaches consensus. (In America, there is probably 10% of the country who would reliably vote for shutting down the government, so you really cannot keep the country running if they can derail the jury.)
I agree that putting the corporations in charge of government would be bad. That's why libertarians oppose crony capitalism.
Eliezer would probably argue that it's more-or-less equivalent to the FAI problem. Personally, depending on what one means by FAI I think it may well be harder. Specifically it may well be possible to create at FAI capable of managing an economy composed of humans, said FAI would not be capable of managing an economy composed of AIs of comparable complexity to itself, more or less due to the pigeon-hole principal.
That's the thing, it wouldn't be corporations in charge of setting prices, there wouldn't be anyone setting the prices. Except in the case of monopolies, it would be the combined market.
Okay, if that's true then that's a good argument for those forms of government control. But, that doesn't argue for involving the government in the other parts of the market.
As to the third point, I'm not sure we know how to reliably make changes to the market that results in positive changes. I'd appreciate the input of an economist here, but from the basic econ I've learned, except in the cases of monopolies or other failure modes of the free market, government intervention mathematically always results in a net loss. (http://en.wikipedia.org/wiki/Deadweight_loss)