= 27d567bc2d48d9f40e9ccc20ea370b15)
Stuart_Armstrong comments on How should negative externalities be handled? (Warning: politics) - Less Wrong Discussion
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 (131)
Generally economists recommend pricing these externalities into the system - I think the american proposal was called "cap and trade" (it died). The EU has a similar system running. If these are properly implemented, you get a more efficient market than you had before, because the externalities now carry a price, and consumers and producers can react to the price accordingly.
If that's impossible (because of difficulties in getting international binding targets, for instance), then subsidies, rather than regulations, seem the most efficient way. I'd personally prefer to subsidise solar rather than nuclear, but this is mainly because I believe that solar power follows a Moore's law and that the long-term costs (including decommissioning) of nuclear are unclear and probably underestimated (I'm open to being persuaded on both counts).
I think taxes works better. Cap and trade is good if there's a specific amount that's perfectly fine, but anything beyond that is horrible. In practice, it's generally going to be that the costs are fairly linear. If you increase CO2 emissions by x, it will cause a cost of kx, for some k. Just tax people by kx. In order for cap and trade to work, you'd have to know in advance how much CO2 emmissions is ideal.
Also, from what I understand, they are giving people a certain amount of CO2 emissions or whatever it is they are trading, rather than just auctioning it off to begin with. This causes people to try and fight over who gets them to begin with, since you're essentially just giving them money for no good reason.
With cap and trade you're directly controlling the quantity of emissions and allowing the market to price emissions subject to that constraint. With a carbon tax you're directly controlling the price and allowing the market to determine quantity subject to that constraint. In both cases you need to make some advance determination about the "ideal" level -- of emission quantity in one case and emission price in the other -- so I don't see any obvious disadvantage to cap and trade in this respect.
In fact, the disadvantage may go the other way. After all, the direct aim of both cap and trade and a carbon tax is to control the quantity of emissions. We are only interested in controlling the price in so far as that effects the quantity demanded. If it turned out that the price elasticity of carbon emissions was effectively zero, for instance, so that changes in the price had no effect on the quantity of emissions, then I'm assuming you would regard a carbon tax as pointless. But if your interest is controlling quantity, why not control it directly rather than via control of another variable whose relationship to quantity you're uncertain about?
Incidentally, there is some evidence that the price elasticity of oil is quite low (not zero, of course), even over the long term. This is especially true in some developing countries, where any effect of realistic increases in oil prices on demand is swamped by the effect of rapidly rising incomes.
I do agree, though, that any actually implemented cap and trade system will fall prey to all kinds of jiggery-pokery due to corporate influence on the government, and will be far from the idealized system that many economists envision. Perhaps a carbon tax will be better in this regard. In any case, either policy would be a significant improvement over the status quo (in America).
Ideal emission quantity is a function of cost. Since cost is itself a function of quantity, it all gets very complicated. Ideal price, so long as the cost is an approximately linear function of quantity, which it will tend to be, is much simpler.
You basically have to know the ideal price and the demand curve to figure out how to cap it, but you only need to know the price to find the the tax rate.
If it was something where giving off more than X emissions would cause a runaway greenhouse effect and anything less than that is okay, then you'd cap and trade. But it's generally not like that.
True, but that would also mean that anything that does work would have a cost far higher than we are willing to pay. The beauty of taxing it is that it makes us decrease CO2 emissions conditional on it being worth while.
I never said that. As it is, we're giving away emissions for no good reason, but we don't have to. We can let the government have them to begin with, then sell them.
Of course, these kinds of jiggery-poker are a large part of why we used this system. If lobbyists can change who you're giving emission rights to, they can also convince you it's a good idea to give people emission rights in the first place.
The situation is entirely symmetric. If quantity is a linear function of cost, then cost is a linear function of quantity. I could easily flip your claim around and say, "You have to know the ideal quantity and the demand curve to figure out the optimal tax rate, but you only need to know the ideal quantity in order to determine the optimal cap." So I don't see how this is an argument for the claim that determining ideal price is simpler.
The question is, which is more appropriately regarded as the dependent variable, ideal price or ideal quantity? Of course, neither is completely independent of the other. This is not an acyclic graph, unfortunately. Still, it does seem to me that our notion of the ideal price of emissions is (or at least should be) largely determined by a prior estimation of the quantity of sustainable emissions. It does not seem to me that our notion of ideal quantity should be fixed by some prior estimation of what the ideal price of emissions would be. How would we even come up with such an estimate?
My understanding is that climate change does involve tipping points. Of course, it doesn't follow that "anything less than that is okay", but I don't see why cap and trade advocates would have to be committed to that claim. When individuals make decisions about carbon emissions in either the tax or the cap systems, I don't see their incentives as being significantly different. After all, if an individual (or individual firm) does not use up all of its credits, it could sell them. So in so far as you think taxes will disincentivize individual emissions linearly down to zero, cap and trade would have the same effect.
The difference arises in the aggregate. A capping system allows you to be sensitive to tipping points in a way that a tax does not, at least not without performing the extremely complex task of figuring out what tax rate will ensure that we do not cross the point while at the same time avoiding unnecessary inefficiencies.
I should also note that capping and taxing are not mutually exclusive, though the hope of both being passed in the current American political climate is laughable.
Given that humans irrationally discount future goods (not to mention that a large number of them have false beliefs about climate change), I don't think willingness to pay is a useful proxy for genuine worth in this domain. (Of course your point would still hold if the price elasticity of emissions were actually zero, which is why I specified effectively zero. I meant to suggest that price changes do have some effect on consumption but that the effect is negligible when we consider feasible price changes).
But how sure are you of where they are? How sure are you of how much others will pollute? If you're not very sure, you're looking at a linearly increasing probability of hitting a tipping point as the amount of emissions increase.
They do interface a bit strangely. You'd end up with a linear cost (the taxes) until you run into the cap, then the trading will take off and nothing more will be emitted.
I thought about this a bit more and came up with another idea. You could sell the emission rights, and change the price as you go. This would allow you to match it to the actual costs.
You will have to be careful to make it so you're not practically giving money to the first people that come. You could offer someone a fraction of the revenue to try to maximize the revenue minus the cost curve, so they'd guess the final value and sell it at that uniformly. You would have to be careful about bribery and such.
You could also probably use some kind of prediction market.
There are opportunity costs. If you can only get a 3% return on investment by reducing emissions, but you can get a 5% return on investments somewhere else, you'd be a fool to reduce emissions.
This does bring up the question of how to best encourage investment. The obvious method is to subsidize it. The government could also take a more direct approach, and get rid of the deficit. Once they pay back all their loans, they could actually start investing. This does have a problem though. Once the US has significant investments, whoever controls where they're invested will be very, very powerful.
We could also legalize long-term investments. Just let a few people set up trusts, wait a few generations, and they will have lots of money they're investing.
Also, while I personally disagree with the idea of time discounting, it's not strictly irrational. Caring less about the future is a perfectly valid utility function. Even hyperbolic discounting is. It's just not the same utility function at every moment in time.
If it were actually zero, making it illegal wouldn't work, because the cost of going to jail wouldn't be enough to dissuade people from using oil. If the effect is just negligible, and we have to multiply the price many times over to get the necessary change, it's probably not worth it. If we weren't planning on this from the beginning, then it's almost certainly not worth it.
Hyperbolic discounting leads to preference reversal, which makes the discounter vulnerable to money-pumping. That's usually taken as a symptom of irrationality around here. Synchronic inconsistency is not necessary for irrationality, diachronic inconsistency works too.
Two different agents can hold two different utility functions. It's not irrational for me to value different things than you, and it's similarly not irrational for past!me to value different things than future!me.
I think it's a mistake to always treat distinct temporal slices of the same person as different agents, since agency is tied up with decision making and decision making is a temporally extended process. I presume you regard intransitive preferences as irrational, but why? The usual rationale is that it turns you into a money pump, but since any realistic money pumping scenario will be temporally extended, it's unclear why this is evidence for irrationality on your view. If an arbitrageur can make money by engaging in a sequence of trades, each with a different agent, why should any one of those agents be convicted of irrationality?
Anyway, the problem with hyperbolic discounting is not just that the agent's utility function changes with time. The preference switches are implicit in the agent's current utility function; they are predictable. As a self-aware hyperbolic discounter, I know right now that I will be willing to make deals in the future that will undo deals I make now and cost me some additional money, and that this condition will persist unless I self-modify, allowing my adversary to pump an arbitrarily large amount of money out of me (or out of my future selves, if you prefer). I will sign a contract right now pledging to pay you $55 next Friday in return for $100 the following Saturday, even though I know right now that when Friday comes around I will be willing to sign a contract paying you $105 on Saturday in exchange for $50 immediately.
You can make the decision to consider the options and let future!you make a better-informed decision.
If you prefer paper to rocks, scissors to paper, and rock to scissors, that can be taken advantage of in a single step. If your preferences change, you don't have intransitive preferences. You do have to take into account that an action changes your preferences, and future!you might not do what you want, as with the murder pill.
They are predictable, but they are not part of the agent's current utility function. It's no more irrational than the idea of agents caring more about themselves than each other. An adversary could take advantage of this by setting up a prisoner's dilemma, just as past! and future! you could be taken advantage of with a prisoner's dilemma. You might use a decision theory that avoids that, but that's not the same as changing the utility function.
Yes, Cap and Trade is a political fight over the vested goodies of your allocated cap. I, and I believe most free market types, believe taxes are the best solution for infringement on a commons.
Some, such as me, would extend that principle to natural resources in general, particularly land, and make the taxes revenue neutral, requiring reimbursement of the taxes to the general public.
In this case, the problem that needs addressing is the quantity of carbon emitted. Cap and trade addresses that number directly. A carbon tax addresses it indirectly: the gov. has to figure out the right level of tax to get the right level of emission reduction in an open market, and continually play with that number to keep it well calibrated. That seems the level of skill that governments (or anyone for that matter) don't possess.
The idea that the government can figure out the right level of emissions to begin with assumes a level of skill that nobody possesses.
Taxes have four major advantages over cap and trade schemes - first, they're less prone to crony capitalism; preexisting companies don't get a major advantage over new entries. Second, they don't disrupt the existing market; you don't hit a wall in which half the power plants in the country are simultaneously shut down. Third, they allow flexibility. Fourth, they're incremental; they -continually- encourage reduction of carbon emissions; see Europe where the market for carbon credits has effectively collapsed to see what happens when carbon drops unexpectedly; instead of a smooth reduction as possible, it's jagged and arbitrary.
Why is this a problem? If the emissions are being reduced, then the gyrations of the carbon credit market are irrelevant. If the prices have collapsed, it means that carbon reduction is turning out to be much easier/cheaper than expected. Yay!
I calculate the cost of this carbon reduction to be $3.3 trillion per year across all of Europe. That's how much smaller their economy is as a result of the recession for which the drop in CO2 was merely an externality. I'm not sure celebrations are in order.
Some studies (which I don't have to hand immediately) suggest that the recession was only partially responsible for the reduction.
But the price of carbon still remains irrelevant. As long as the reduction happened, the price could be zero for all the difference it makes. We aren't trying to punish people for emitting too much carbon, making them pay the moral price of their erroneous ways. We're simply trying to reduce carbon emissions.
I don't see any drawback to the price of carbon having collapsed.
If your goal is long-term reduction in CO2 emissions, you've introduced market volatility. The goal of any such measure is to reduce CO2 emissions, but the -mechanism- by which it does so is encouraging research into alternatives. If the market is volatile, the value of any such research is called into question; tomorrow it might be valuable, it might be worthless. A tax, by comparison, has a fairly static value. The cap-and-trade measure, through its volatility, increases the risk of investment into reducing carbon emissions; the value of your investment isn't determined by the degree to which you can reduce carbon, but by the amount of carbon emitted in the market as a whole.
Unless we assume some level of carbon emission is better than no carbon emission, the tax scheme is better.
Yeah, a quota is absolutely inelastic supply, giving wildly erratic price. Pricing incremental emissions gives incremental feedback.
I don't think this one is hard. In either case, you have to assume an measurement by which you're accounting for emissions.
Yes, you won't exactly hit your targets, but then you adjust the price and move on. We're talking about accumulations that matter over a period of decades. Consumption varies, but it's not all over the place.
In other words, let the government sell it to the public.
Since they're taxing the public anyway, they could reimburse it just by letting the public pay that much less taxes. This is exactly the same as what would happen if they used it for revenue, and decreased taxes because they don't need as much revenue.
More of a leasing scheme with respect to land. You get it as long as you pay your taxes on it. Which is largely the way it is now.
The difference isn't about stopping global warming or funding government, it's about justice. If there is a commons that everyone has equal right to use, those that do use it owe compensation to those who don't, not the government. The funding mechanism is a separate issue from this basic issue of justice. The government is collecting what is owed by some parties to others; that doesn't give it a license to keep as much of it as it wants.
It works out the same either way. Loans and investment act as a way to exchange present money and future money, so charging present money vs. future money makes about the same difference as charging USD vs. euros.
If they're charging enough taxes to offset the value of the land, then that means that the price of buying land would be zero. You will pay the full cost with taxes.
But there is no difference. The money gets passed around exactly the same in either case. Is it really more just to give someone a reimbursement then immediately take it back as taxes than to do nothing?
It has the license to just take what it wants in the form of taxes. I would expect that to work as a license to keep what it wants.
No, the money does not get passed around the same way.
The homeless guy living under a bridge has no property, Homeless guy doesn't get taxed on land, Bill does, divide Bill's taxes between Bill and homeless guy, and homeless guy ends up with a check that he doesn't get now.
Homeless guy has no income and no wealth to tax for the support of government. I guess with a head tax, we could make the result the same. We could find ways to make the result the same, but we'd really have to be going out of our way to do it.
Perhaps from you, but not from me. That is largely what political arguments are about in the US these days.
If you're assuming that taxes are otherwise just a constant portion of your income, or really any tax system that's specified to be the same in both cases, then there is a difference. However, the two systems are the same modulo tax method. In other words, in one case the homeless guy gets money from Bull's land, and in the other he gets money from welfare.
I don't understand.
Nope, not the same with the addition of welfare either. People on welfare often face effective tax rates in excess of 100% - if they start working and making money, they lose benefits worth more than their incremental income.
Perhaps the government has such license from you, but they don't have it from me, or a great many other people.
That is a problem with how welfare is currently done. It would still be welfare if they didn't do it like that. I'd just implement it as taxes going into the negatives.
So, they're not allowed to collect income tax from you?
One major problem is the practical effect of adding new taxes. Adding a tax creates a group of people who benefit from the tax, either a part of government that gets to decide how to spend the tax money, a recipient of tax money, or both, and it is in that group's interest to let the size and purpose of the tax slip beyond its original intention. A carbon tax might be a good idea if it is revenue neutral, but there's no way you can make one that is guaranteed to stay revenue neutral.
You can't guarantee anything because of political will. But take in taxes, count total, divide by eligible recipients, electronic transfer money. It's not complicated to make it revenue neutral if you have the will to do it.
Cap and trade means the government can fix the right figure for carbon emissions (at least somewhat plausible). Carbon taxes means the government can fix the right price to get the right figure for carbon emissions (very implausible).
There is no intrinsic right figure for carbon emissions. There is a cost that's incurred by carbon emissions. The more you emit, the higher the cost. The right figure is when the cost to emit more is equal to the benefit. You could find out the variables you need and solve the equations, or you can just work out the cost, charge people that much, and let the market deal with it.
Estimating the costs of carbon emissions involves estimating the temperature rises AND the effects of these on agriculture and other parts of the economy AND the likely costs of disasters AND the potential benefits AND the effects on the natural world AND how we should price these effects (which no-one would agree on). And also implement these calculations properly within a political system.
If we want to stabilise temperature rises at a particular point (which would be "good enough" according to most preference systems), you only need to estimate and implement the first term.
(ok, technically you need to be reasonably sure that temperature rises would be a net negative and keep an eye on carbon credits in case their prices get extraordinarily high, but these require muuuuuch less work than a proper pricing; the second thing will happen naturally within the political system anyway)
If there's an approximately linear cost for carbon emissions, then estimating a good enough value will be worse than estimating the cost. You have all the original error from guessing the cost of emissions, and now you're guessing the demand curve for emissions licenses as well.
If there is a point that will cause sudden problems, then capping and trading would work better, assuming you know where that point is. As you've pointed out, it's hard to tell. As such, it's better to treat it as a linearly increasing probability of hitting that point.
That doesn't seem to be a reasonable assumption at all - expected damage seems to escalate faster at the extreme end.
In any particular area, the cost will be approximately linear. If all you're looking at is the emissions in one state caused by one group of things (like power plants) over the course of one year, then it's approximately linear. Also, depending on how accurately you can guess at what the market equilibrium will be, you can narrow it down further so the linear approximation is still more accurate.
There is a principle in economics called Theory of the Second Best, whereby if a market failure cannot be corrected, it may be better from a welfare and systemic hazard perspective to substitute it for a satisficing but suboptimal mechanism.
Examples would include ham-fisted government intervention, which are strictly worse than an idealised market solution, but generally better than what a realistic market solution will actually provide.
Do you think this is a good case for that kind of interventions?
Or are we thinking "second best" just because the first best options are not politically viable?
I am not remotely qualified to say whether or not it's a good case, but it is an extremely unusual situation which deviates from micro market assumptions along numerous dimensions.
When it comes to things like civil ordinance and domestic tomato produce, standard micro analysis is pretty good at capturing the salient characteristics of the situation and proposing remedial policy. In the case of high-profile issues, you're selecting for aberrant situations with unusual characteristics. In those cases, it's worth giving more consideration to ugly and distasteful interventions as well as smart and elegant ones.
Subsidizing efficiency improvements sometimes leads to a net increase in pollution, and so this is more appropriate for things like subsidizing disruptive technologies than subsidizing improving existing coal plants.
I'm under the impression that nuclear is one of the few where decommissioning costs are baked into the operating costs, and thus comparisons between it and other technologies are biased against nuclear.
This is primarily when the efficiency improvements make a previously niche technology viable over a much broader range of cases. Already-dominant technologies don't suffer from this to nearly the extent. Cases where the efficiency is not the limiting factor on usage do not suffer this effect at all.
Like, if cars took a tenth the gas they do now, people wouldn't drive ten times as much. They'd drive some more, and buy more SUVs and winnebagos, but unless a colossal number of people are held back from using a winnebago by gas prices, that would only partially compensate for the efficiency increase, because people are already doing close to as much driving as they care to. Back in the 1970's this was not the case at all.
This isn't just the Jevons paradox, though that does amplify this. It's also the claim that while a tax and subsidy are equivalent in the short-run, in the long-run the tax decreases the profitability of the polluting industry while a subsidy increases the profitability of the polluting industry, which may lead to increased pollution.
The subsidy itself will also make the technology viable over a broader range of cases, yes... should have mentioned that part. Tried to get too clever.
But only by granting higher standards of living, which are also a goal of the government.
Not necessarily; the subsidies have to come from somewhere. (I should also point out that I'm making a somewhat technical point, here, and so unless you can sketch the curves involved we may be talking about different things.)
Yeah, we're talking about very slightly different things - I was ignoring the opportunity cost of the subsidy in my mental model. Whether or not subsidizing an efficiency improvement benefits standard-of-living (compared to alternative uses of that money) depends entirely on the relative cost of that subsidy.
That does seem like a better idea, ignoring issues of price setting. Unfortunately, nation states are extremely bad at game theory, and it's difficult to achieve international agreement on these issues, especially when it will impact one nation disproportionately (China would be much harder hit, economically, by cap-and-trade legislation than the US).
I'd disagree pretty strongly with the energy issue, at least for now - but that's a discussion for another time. In politics, as in fighting couples, it is crucial to keep your peas separate from your pudding - one issue at a time.
Without wanting to start a fight, which half do you disagree with? The Moore's law or the nuclear estimate? I'm personally more confident about the first than the second.
Regardless of OP's objection, there's a strong counter to the assertion of solar power following a Moore's Law trajectory. Solar irradiance at ground level has a fairly hard limit of < 1200 watts/m^2. Even in the upper atmosphere it's not much more.
So solar cells may get more efficient, but their output isn't going to get exponentially greater over time. They may also become considerably cheaper, but the price of land isn't going down, and will remain a non-reducing term when calculating implementation costs.
It could be that you're referring to some other feature of Moore's Law I'm not considering, but in the intuitive sense of "my phone has more computing power than the whole of the 1960s", gains of that magnitude are simply not possible.
The Moore's law seem to be in cost, not in efficiency (though efficiency is also improving): http://blogs.scientificamerican.com/guest-blog/2011/03/16/smaller-cheaper-faster-does-moores-law-apply-to-solar-cells/
And if we get to the point where land is the bottleneck, well, I'd say we'd be doing fantastically well at that point :-)