Fiscal Policy, the Environment, and How Not to Solve Policy Problems

Canada and by extension the Canadian government faces a number of challenges. There are two in particular I find fascinating:

1. A shortfall in federal government revenue to finance policy changes. Some of us are advocating corporate tax cuts.  Some advocate pharmacare programs and early childhood education initiatives, while others advocate eliminating the deficit as quickly as possible.  Plus we need stadiums - lots and lots of stadiums.  But where will the money come from? We can't both raise and lower corporate income taxes at the same time. Raising personal income taxes will lead to slower economic growth (relative to raising other taxes) and raising payroll taxes on firms would deter hiring. Economically speaking raising the GST would make the most sense, but we're told it's politically infeasible since when you raise the GST you raise taxes with Hitler (or something like that). So where is the money going to come from?
2. We need to reduce various types of emissions. Whether it be toxic (and often carcinogenic) pollutants from London, Ontario or greenhouse gases linked to anthropogenic climate change there is near universal agreement that we ought to investigate ways to reduce these emissions.  But how to do so in a cost-effective way with minimal economic damage?

You probably see where I'm going with this. The solution is simple - we can use Pigovian taxes or cap-and-trade with auctionable permits to reduce emissions (problem 2) and the revenue can be used to fund the initiatives we like (problem 1).  As a pollution control measure, the solution is rather elegant.  We don't need an army of bureaucrats to try to figure out which emissions can be reduced most cost-effectively.

Sounds great in theory, but would a cap-and-trade system really be a low-cost way to reduce emissions in the "real world"? Fortunately cap-and-trade isn't just some textbook theory - it's been used to great success in North America:

"Cap and trade" harnesses the forces of markets to achieve cost-effective environmental protection. Markets can achieve superior environmental protection by giving businesses both flexibility and a direct financial incentive to find faster, cheaper and more innovative ways to reduce pollution.

Cap and trade was designed, tested and proven here in the United States, as a program within the 1990 Clean Air Act Amendments. The success of this program led The Economist magazine to crown it "probably the greatest green success story of the past decade." (July 6, 2002).

As an aside, I've often wondered if the Clean Air Act's cap-and-trade was almost too successful, in the sense that people seem to have concluded not that "cap and trade does wonders" but "boy, that whole Acid Rain thing was overblown".

Of course, since this policy would be straight forward and effective and help boost government revenues, it has no chance of happening.  As Dan Gardner points out, the Conservatives have decided to take their policy cue not from Arthur Pigou, but rather from Karl Marx:

“Climate change is one of the most serious environmental issues facing the world today,” declared Peter Kent, the new environment minister, in a recent speech before the Economic Club of Canada. Kent didn’t wink when he spoke those words. He didn’t smirk. He was emphatic and, apparently, sincere. The Conservative government “is determined to do our part for the planet,” he said....

“Achieving our objectives requires a systematic approach of regulating GHG emissions sector by sector,” said Peter Kent. Translation: Some bureaucrat in Ottawa will decide who will reduce emissions and how they will do it.

This is old school command-and-control regulation. Order the Vladivostok Power Plant to install a new type of scrubber. Direct the October Revolution Factory to retrofit with pipe insulation of a particular specification. This Kremlin-knows-best approach is what made the Soviet Union such a model of efficient production. Left shoes? We have eight for every person in the Motherland. Right shoes? Maybe in the next Five Year Plan.

And it’s not just inefficient. It also opens the way to abuse, even corruption. Say you’re the boss of People’s Shoe Factory #17 and you really want to screw People’s Shoe Factory #18. You could try to out-produce them. Or you could try to convince the Kremlin that #18 should be subjected to a burdensome new directive that you, naturally, would be exempt from. In the old Moscow, that sort of thing was done with party connections. It’s much easier in Ottawa. Look under “L” for lobbyist. Just be sure to pick one with party connections.

If the Conservatives are uncomfortable with taking policy advice from Arthur Pigou, they could always choose Milton Friedman instead.  From Free to Choose (1980):

Most economists agree that a far better way to control pollution than the present method of specific regulation and supervision is to introduce market discipline by imposing effluent charges.  For example, instead of requiring firms to erect specific kinds of waste disposal plants or to achieve a specified water quality in water discharged into a lake of river, impose a tax of a specified amount per unit of effluent discharged.  That way, the firm would have an incentive to use the cheapest way to keep down the effluent. (pg. 217)

He goes on to add:

Fuel from shale, tar sands, and so on, makes sense if and only if that way to produce energy is cheaper than alternatives - account being taken of all costs. The most effective mechanism to determine whether it is cheaper is the market. If it is cheaper, it will be in the self-interest of private enterprises to exploit these alternatives - provided they reap the benefits and bear the cost...

Private enterprises will bear all the cost only if they are required to pay for environmental damage. The way to do that is to impose effluent charges - not to have one government agency impose arbitrary standards and then set up another to cut through the first's red tape. (pg. 221)

If "have one government agency impose arbitrary standards and then set up another to cut through the first's red tape" isn't an accurate description of the Conservative position on the matter, I don't know what is.

So we can agree the Conservative position on this issue is, well, dumb.  And the Liberal position?

Liberal environment critic Gerard Kennedy said the Harper government’s failure to regulate the oil industry’s emissions has left the sector subject to international criticism and rising concerns among institutional investors.

“It does seem that the government is tethered to the oil sands in a way that even the operators of the oil sands understand is dangerous,” Mr. Kennedy said. “They’re getting international criticism … simply because there isn’t a credible government regulating and setting up the predictable environment from them.

In other words, the Conservative position isn't dumb enough.

I am quite certain the response to this post will be a litany of political reasons why the Liberals and Conservatives cannot propose sensible policies.  I understand but I don't care. My hope is that if enough of us keep advocating sensible public policy it may become popular enough to become politically feasible.  I want to be part of the solution rather than part of the problem.

 

Comments

[Joel Wood]

Interesting post Mike.

For current air pollution in London Ontario (and the rest of Southern Ontario), Cap and Trade and even pigouvian taxes may not be an effective solution. The major problems in London are groundlevel ozone (O3) and fine particulate matter (PM25). O3 is not directly emitted, it is a chemical reaction between NOx, VOCs and intense sunlight. The majority of NOx in Ontario is produced by transportation (diffuse source pollution, so cap and trade isn't an option). Due to the seasonal nature of O3 formation, you would need to levy the env tax only around hot, sunny days. My PhD advisor, Ross McKitrick at Guelph, has proposed an env tax levied on gasoline that takes effect when hot sunny weather is forecasted, however, I would expect this to really increase implementation and enforcement costs.
The other problem with controlling O3 in Southern Ontario is transboundary pollution. It is often the case that the O3 precursor pollutants are blown in from multiple states in the US (e.g. Ohio, Michigan, and Pennsylvania).

But on the positive side, O3 and PM2.5 ambient concentrations in Ontario and Canada are generally decreasing. The current combination of the certificate of approval process for point source polluters (and its variants), vehicle emission standards, technological progress, and economic growth have been effective at making large improvements in Canada's air quality over the last 30-35 years. Is this regulatory/econ growth mix cost-effective? Not in a first best world, but it may be in a second best world.

[Stephen Gordon]

I've often wondered if the Clean Air Act's cap-and-trade was almost too successful, in the sense that people seem to have concluded not that "cap and trade does wonders" but "boy, that whole Acid Rain thing was overblown".

I think you're on to something there. Pretty sure that public opinion polls would put that second thing in front.

[Joel Wood]

Acid Rain was overblown! further developments in the science showed that acid rain leads to acidification of lakes, but does not have the earlier expected effects on forests and rivers. There is a good review of the literature in Bjorn Lomborg's The Skeptical Environmentalist.

[Mike Moffatt]

Hi Joel,

"The major problems in London are groundlevel ozone (O3) and fine particulate matter (PM25). O3 is not directly emitted, it is a chemical reaction between NOx, VOCs and intense sunlight. The majority of NOx in Ontario is produced by transportation (diffuse source pollution, so cap and trade isn't an option). Due to the seasonal nature of O3 formation, you would need to levy the env tax only around hot, sunny days. My PhD advisor, Ross McKitrick at Guelph, has proposed an env tax levied on gasoline that takes effect when hot sunny weather is forecasted, however, I would expect this to really increase implementation and enforcement costs."

That's really interesting - thanks for sharing! A not-as-efficient solution would be to have a higher gas tax in the summer months. You probably wouldn't want to do it nation wide, but perhaps one in SW Ontario, with the revenue generated given to municipal governments.

[Joel Wood]

Maybe use the revenue to reduce municipal property taxes (a distortionary tax if I've ever seen one).

[Kevin Milligan]

Joel--why do you think property taxes are excessively distortionary? To the extent they tax unimproved land they are almost perfectly non-distortionary. And, at least in Vancouver, it seems to me that property taxes are mostly a tax on land not on improvements.

[Patrick]

Acid rain hasn't gone away. Nitric oxides from tail pipes is still a problem.

Just because Kent was a good enough actor to sound sincere doesn't mean that the Conservatives really accept that climate change is a looming catastrophe. They're taking a page out of the old Liberal play book: say all the right things and do nothing. Canadians don't seem to care about actually doing anything about climate change. They just want politicians to sound serious about climate change.

The one and only policy concern of the Conservatives is getting a majority. I suspect they are well aware that GST cuts, CIT cuts, stadiums, super sci-fi jet fighters, and jails! jails! jails! are all policies that wreck Canada's fiscal position. But they don't care. They're all popular policies.

If they get a majority, they might try some spending cuts - but not to programs baby boomers really like. Given the conservatives idealogical bent, I doubt they'd ever implement a new tax - even if it was offset by cutting or eliminating some other tax.

[Craig Burley]

"My PhD advisor, Ross McKitrick at Guelph, has proposed an env tax levied on gasoline that takes effect when hot sunny weather is forecasted, however, I would expect this to really increase implementation and enforcement costs."

The biggest problem, even more than enforcement, is that it's harder and more costly for people to adapt to taxes that swing wildly, just as high-volatility costs of any kind impose disproportionate transactional and adaptation costs.

I know what I would be tempted to do if faced with wildly swinging gas taxes that were seasonally predictable - repurpose a couple of fuel oil tanks into home gas tanks. That is something you REALLY don't want to be encouraging people to do.

[Andrew F]

Increasing taxes around hot days would likely lead to hoarding in jerry cans. You'd need to use something like a toll to control emissions proximate to the hot sunny days.

To deal with diffuse emissions from transportation, you just need to levy the tax on the product of refineries and on fuel imports.

Isn't the reason why effluent charges are controversial that it is hard to determine the efficient rate at which to set the charge?

[Joel Wood]

Kevin,
Point taken. I was thinking off the top of my head and had misremembered that from Public Finance class.

[Joel Wood]

Patrick,
"Nitric oxides from tail pipes is still a problem."

Nitrogen Dioxide concentrations have been significantly decreased in Canada over the past 30 years (Niric Oxide emsisions transform to NO2 in the atmosphere). There is no reason to expect this trend not to continue. Vehicle emission standards continue to be more and more stringent and the switch from coal to natural gas and renewables in Ontario will result in large improvements as well.

[Patrick]

"Vehicle emission standards continue to be more and more stringent"

There are more cars though.

[sl]

Your point on advocating enough that eventually it becomes less of an issue is a good one. I think Toronto may be at that point soon with respect to road tolling.

[Joel Wood]

Patrick,

Despite there being more cars, ambient concentrations of NO2 are declining.

[RSJ]

"account being taken of all costs"

At least with prescriptive rules, we can get a scientific consensus as to what levels of pollution constitute a health hazard -- very roughly. But trying to determine the full costs of the health hazard, including medical expenses, lost productivity of those that become sick, loss of income earning opportunities from tourism, cost of remediation, etc -- that's much less scientific and open to manipulation and corruption.

Only by assuming the can opener (e.g. "all costs are objectively known") can you conclude that Pigovian taxes are less intrusive or liable to corruption than prescriptive rules and fines. But Pigovian taxes are more liable to corruption and manipulation, because they are trying to measure something that is hopelessly vague and open to endless re-interpretation, particularly as new science comes out, new remediation techniques are discovered, and new economic uses for resources are deployed.

[Determinant]

OK Mike, the problem with Pigouvian taxes, which are merely an excise tax dressed up with some fancy moral arguments about "Social Good" have a fundamental problem which I call the Pitt Conundrum.

To wit, in 1793 Great Britain declared war on Revolutionary France, beginning 20 years of war against France. This entailed a massive and expensive increase of the Royal Navy and Army and the payment of extremely large subsidies to foreign allies, mostly Prussia, Austria and Russia.

The British Treasury, then funded mostly with excise taxes and duties was strained to the breaking point. Soldiers starved from lack of supplies and the Royal Navy had mutinied at the main fleet bases of Portsmouth and Chatham in 1797, asking an increase in pay (the first in 50 years).

From 1792 to 1798 Pitt's Ministry increases excise taxes again and again, on ever more items, but it still wasn't enough. In 1795 he instituted a stamp duty on Hair Powder, then a necessity in that bewigged age. Instead of increasing revenue it only served to kill off hair power and wigs as a fashion item. This is the conundrum of excise taxes designed to tax an "evil". You can levy a tax to produce revenue, you can raise a tax to correct a social ill, but you can't have the same tax do both things efficiently at the same time. If something taxes an "ill" to produce sufficient revenue it often causes the tax base to decline. That is exactly the dilemma Pitt was faces with and he had tried to put an excise tax on every good possible.

In 1798 Pitt was forced by Britain's deteriorating financial state to levy the first Income Tax of the modern era.

In a monetary economy it is only a flexible and broad definition of "income" measured in money that produces a reliable tax base.

[Mike Moffatt]

"At least with prescriptive rules, we can get a scientific consensus as to what levels of pollution constitute a health hazard -- very roughly."

Okay. So we have to figure out the best way to get there. Got it. Then why not set a cap-and-trade limit to that level? Why is that somehow worse than setting rules and regulations.

[Jeremy Fox]

"As an aside, I've often wondered if the Clean Air Act's cap-and-trade was almost too successful, in the sense that people seem to have concluded not that "cap and trade does wonders" but "boy, that whole Acid Rain thing was overblown"."

I think you're on to something there. I suspect many would draw the same conclusion if Pigovian taxes or cap-and-trade were used to successfully limit global warming.

[Mike Moffatt]

" You can levy a tax to produce revenue, you can raise a tax to correct a social ill, but you can't have the same tax do both things efficiently at the same time."

Complete and utter nonsense. The Feds raise 5 billion currently in excise taxes on things such as gasoline, alcohol and tobacco.

Since nobody is suggesting the government raise their entire revenue from excise taxes, I'm not sure what on earth your point is.

[Mike Moffatt]

"I think you're on to something there. I suspect many would draw the same conclusion if Pigovian taxes or cap-and-trade were used to successfully limit global warming."

Agreed. I wonder how wide spread this effect is - it could help explain, in part, the backlash against vaccinations.

[Determinant]

Not nonsense. Duties and excises are a sideshow compared to Income Taxes and GST. They will not fully pay for any of the programs you stated as costs, not Pharmacare, not corporate income tax cuts, early childhood education.

You can levy a Carbon Tax as a sin tax and try to coerce behaviour, but the main cost of the initiatives you mentioned and I just repeated will be borne by the Income Tax/GST base.

Discussion of Carbon Taxes does not belong in the same conversation as policy initiatives.

[Mike Moffatt]

Absolute nonsense. Even a modest carbon tax would raise tens of billions. Do you have any evidence for your claims whatsoever?

[Mike Moffatt]

Andrew: "Isn't the reason why effluent charges are controversial that it is hard to determine the efficient rate at which to set the charge?"

Then set the quantity through cap-and-trade and let the market decide the price.

I know, the response is going to be "very funny, smart guy.. how do we set the quantity?"

Guess what - that's how we do it *now* through regulation.

In any set of regulations there's always a regulatory impact statement that gives this info. For instance, VOC regulations, that give the total expected reduction in kilos released, how many of those pounds will come from the changes to hairsprays, how many will come from the changes to aerosol paint, etc.

It's not a particularly efficient way of doing things. Though it has been pretty lucrative for me personally.

[Joel Wood]

Mike, I haven't had time to read it yet, but it looks like Cap and Trade loses its shine if you cannot auction permits: http://www.rff.org/Publications/Pages/PublicationDetails.aspx?PublicationID=21470

[Determinant]

You claim it would raise tens of billions. I am refuting the assumption you implicitly make that we can: (a) tax carbon and rely on that revenue to fund programs and (b) use carbon taxes to reduce carbon emissions.

I'm saying that a level of taxation to optimize (b) reduces the tax and base thereby erodes (a). A level of taxation consistent to achieve the goals of (a) means that you have to compromise (b). It isn't consistent to have a tax that seeks to erode its own base of taxation and also produce the same level o revenue with a constant level of production. That's why I called it Pitt's Conundrum.

Five billion dollars is a drop in the fiscal bucket.

Any evidence for your claims that a carbon tax will raise tens of billions *without* raising them to a level that causes substitution away from carbon fuels?

[Joel Wood]

Determinant, the efficiency gain from a carbon tax is not entirely from the revenue recycling, it is also from the efficiency gains from reducing other more distortionary taxes. See the rff paper I linked to in a comment above, it covers carbon taxes too.

[Mike Moffatt]

"I am refuting the assumption you implicitly make that we can: (a) tax carbon and rely on that revenue to fund programs and (b) use carbon taxes to reduce carbon emissions."

Except we do this *right now* with cigarette and alcohol taxes. So what's your point?

Again. Evidence. Do you have any?

"Any evidence for your claims that a carbon tax will raise tens of billions *without* raising them to a level that causes substitution away from carbon fuels?"

Of course there would be some substitution, just as there are with cigarette and alcohol taxes. Again, what is your point?

[Joel Wood]

Isn't substitution the goal of a env tax?

[Mike Moffatt]

Also, if 5 billion dollars is 'no big deal', then what possible objection could you have to a 3-4 billion dollar corporate tax cut?

[Mike Moffatt]

I find this idea hilarious. 'What if we reduce pollution so much we can't raise any money from it and we have to tax something else?!?'. Oh man, I'd *hate* to leave such a country to my daughter. How horrible.

[Determinant]

I'm fine with a goal of substitution. Program spending, OTOH, has to be done after tallying up all other revenue in the Consolidated Revenue Fund.

A goal of "Carbon Taxes for Pharmacare" is a false construction.

[RSJ]

"why not set a cap-and-trade limit to that level? Why is that somehow worse than setting rules and regulations."

It's not, and I'm not against cap and trade per se -- the argument was with your framing of the issue.

Think of it this way:

We know pollution has costs, but we don't know what those costs are. Our ability to determine the economic costs of pollution is much less than our ability to come to an agreement on a policy of "how much of chemical X can we allow in drinking water before we deem it unsafe". The reason why one is feasible whereas the other is not is precisely the issues of power politics and bureaucratic arbitrariness that you mentioned.

We can imagine, ideally, some curve where X units of pollution causes f(X) health effects. Pigovian taxes would be to impose a fine of f(X) whenever X of pollution occurs. You can see the problem with this: Not only is f(X) is unknown, but with many producers, each producer is not paying the marginal cost of his marginal pollution, he is paying the (industry) average cost -- that is the only way you are going to get f(X) in fees for X units of pollution.

Assume, quite naturally, that f(X)/X is an increasing function. Now imagine a small business that emits a constant stream of pollution each period, and that other (larger) business are increasing their pollution with time. Under a pigovian system, the small business would need to pay more and more each period and would effectively be subsidizing the growing business. In fact, the small business would be subject to enormous uncertainty, as his costs would be determined by the actions of other businesses -- they would not be internalized.

Now change this to assume that both businesses are growing, but one is substituting away from pollution at a faster rate. The more efficient businesses subsidize the polluting businesses.

In other words, in an environment with many non-identical producers and increasing externalities to scale, Pigovian taxes are unworkable, and we are left with the arbitrary bureaucrat option -- i.e. with either prescriptive rules or cap and trade.

Given the bureaucrat option, cap and trade is better in some sense, but not because of any argument advanced by Milton Freedman or Arthur Pigou.

It's better because selling pollution permits raises more revenue than fines, but it is more arbitrary, as those most able to bear the costs are the ones that get to pollute, while those less able to bear the costs are squeezed out. This is not the same thing as saying that each producer is paying their fair share. You can imagine, for example, that utility companies or those who are better able to pass on the costs to consumers will be the ones that will purchase most of the permits, while those firms that must internalize the costs the most will be the ones to purchase the fewest permits. But there is no a priori relationship between price elasticity of demand for a good and the carbon footprint of that good.

The reason why the EPA's cap and trade is a bad example here is that it primarily hit electricity utilities that all roughly had the same pricing power, and sufficient vouchers were handed out that the revenue effects were miniscule, with roughly a billion or so dollars raised per year, in a multi trillion kw market. I.e. the sector has about half a trillion in revenue.

If you try to generalize this model to a situation in which every producer pays based on their usage of carbon, and you want to raise large amounts of revenue, then these elasticity and uncertainty issues will start to bite. Some firms will be squeezed out simply because they are unlucky enough to be in a market with higher price elasticity, or that require more certainty about their production costs, say due to longer supply chains or the underlying market structure. None of that has anything to do with that firm's carbon footprint. I don't think the success of the EPA cap and trade program is a good model to go on for carbon trading, but it may be a good model for other types of pollution control.

[Mike Moffatt]

"I'm fine with a goal of substitution. Program spending, OTOH, has to be done after tallying up all other revenue in the Consolidated Revenue Fund.

A goal of "Carbon Taxes for Pharmacare" is a false construction."

Then you're okay with treating a corporate tax cut completely separately from program spending? That is, I could advocate eliminating corporate taxation altogether and arguing that it wouldn't affect our ability to fund social programs at all and you'd agree with such an argument?

[Mike Moffatt]

RSJ - you do realize your argument comes straight from Karl Marx, correct? Watch:

Assume, quite naturally, that the price of a good is an increasing function of demand. Now imagine a small business that uses a constant amount of the good each period, and that other (larger) businesses are increasing their use of the good with time. Under the price mechanism, the small business would need to pay more and more each period and would effectively be subsidizing the growing business. In fact, the small business would be subject to enormous uncertainty, as his costs would be determined by the actions of other businesses -- they would not be internalized.

Now change this to assume that both businesses are growing, but one is substituting away from the good at a faster rate. The more efficient businesses subsidizes the business using more of the good.

In other words, in an environment with many non-identical producers and increasing economies of scale,the price mechanism is unworkable.

[RSJ]

Mike, the requirement of convexity for general equilibrium is not a Marxist idea, it's a mainstream idea. But just because you can assume diminishing returns does not mean that you can assume diminishing harms.

The whole point of allowing X_0 units of pollution in water is that for just a little bit of pollution, the water is drinkable, but when you add more, the health effects start to snowball and the water stops being drinkable.

One can easily imagine a "zone" of increasing harms that relatively good ecologies are in, and by the time you reach the point where dumping more stuff in the water doesn't make much of a difference, then you are already at the point where the planet is unlivable.

At least, you need to make a case for why negative externalities would satisfy the convexity properties necessary in order to conclude that Pigovian taxes will allow each producer to internalize their externalities.

[Determinant]

Sure, eliminate Corporate Taxes, but if you want to fund the same spending, you'll have to find the revenue elsewhere. Carbon Taxes alone won't do it. That's about as realistic as the "National Sales Tax instead of Income Tax" camp in the US.

This is an instance of Nick's Theory of "Money is Special". Income, Corporation and Consumption taxes are very broad nets that are difficult if not impossible to substitute away from. That's how they are designed. They do not tax the good, they tax the other half of the transaction, the monetary side. It's the monetary side that is always there. In a monetary economy with n goods there are 2n markets, goods for money and money for goods. Everything has to be sold (income, corporate) or purchased (consumption). Corporate, Income and Consumption Taxes are invariable to substitution effects. In a dynamic market economy where the government needs X dollars in revenue to fund expenditures, taxing money is the only way to go. Who can predict which of the n markets will be hot or not at any given time? Certainly not the Ministry of Finance.

A Pigouvian tax, furthermore is extremely susceptible to substitution effects. It's tax base is the good, not the money part of the transaction. That's what Pitt realized way back when. In five years if carbon consumption is reduced 20% through conservation and improved technology what do you do, take a huge slice out of some other sector of the economy? Now we're back to Mr. Arbitrary Bureaucrat again. The tax base is too narrow.

You can make a case for Carbon Taxes on ethical or policy grounds, but not revenue grounds. And yes, Alcohol and Tobacco are special because they are addictive and the alternatives are illegal. But they aren't a large source of revenue either.

Your argument Mike subtly falls for the old "Righteousness" fallacy of taxation. Many people, including you in this post, want a tax that a person doesn't *have* to pay, that is you can avoid it through righteous behaviour. That's why income taxes and consumption taxes are so unpopular, you can't avoid receiving income or spending it on consumption.

[anon]

RSJ, my understanding is that Pigouvian taxes are set at marginal social cost. This means that the overall level X is taken as given: what matters is only the additional harm created by the marginal unit of pollutant. Thus, if the pollutant only has a threshold effect (such as fouling a source of drinking water), the optimal Pigouvian tax is probably very low, since the marginal polluter cannot do much to alter the amount of damage.

[hosertohoosier]

"My hope is that if enough of us keep advocating sensible public policy it may become popular enough to become politically feasible. I want to be part of the solution rather than part of the problem."

Spoken like a true economist... If you think being a broken record is the solution, then you ARE the problem. The popularity of environmentalism is not actually a good predictor of policy. For instance, lets look at support for increased taxes to curb pollution in the late 90's, as Kyoto began to be implemented (or not implemented). The countries that implemented Kyoto did so against public opposition, while the laggards generally acted against widespread public preferences:

Germany: 29.7%
France: 37.6%
Britain: 49.9%
China: 74.3%
Australia: 69.2%
Japan: 62.2%
USA: 60.9%
Canada: 58.1%

Source: World Value Survey

National interests are a far better predictor of environmental policy. It explains why Scandinavians and Canadians took the lead on Acid Rain. It explains why the US and Canada were leaders on the Ozone layer (they didn't produce that many CFCs), and also why the discovery of the Ozone layer changed the calculus overnight (the costs no longer outweighed the benefits). It explains why oil importing European countries and small pacific islands supported and implemented Kyoto.

So it isn't just firms that don't account for the negative externalities of C02, there is also a system of states that fail to contribute to the global public good that is C02 reduction (unless doing so is in their interest). Just as a carbon tax would encourage the entrepreneurial energies of firms to find the most energy efficient way to deliver the goods, the motive of national aggrandizement could one day drive competitive investments in green technology... probably around the same time they develop solar-powered weapons.

[Andrew F]

"So it isn't just firms that don't account for the negative externalities of C02,"

Apparently Shell evaluates projects assuming a cost of CO2 emissions of $40/tonne. So, not quite.

[K]

Mike: "Then you're okay with treating a corporate tax cut completely separately from program spending?."

That doesn't sound like the same thing at all. The point is that the carbon tax ought to exist no matter what the revenue needs happen to be. If no revenues are needed it should still be levied and redistributed, e.g. as a citizen's dividend. Income and consumption taxes have no such necessary role. They should all be eliminated if Pigovian and land taxes suffice.

Joel:"Isn't substitution the goal of a env tax?"

Not necessarily. So long as it is set at the marginal social cost, we don't care if you pollute or not.

[RSJ]

It boils down to whether you believe that, by levying the taxes (and distributing them as a social dividend), you are un-doing the damage to the environment, or whether you are compensating the public for the loss of utility from a harmed environment. The notion of Pigovian taxes allowing for free disposal is the former, but the latter notion is more applicable to the environment, particularly to relatively irreversible effects of pollution (e.g. death, carbon in the air, etc.)

If the latter case is true, then the only affects are by encouraging substitution to less polluting techniques. Imagine if only one technique was available, then the government taxes producers and gives the money to consumers who turn around the spend the income on producers and there is no change. You don't have less pollution than if you had not levied the tax. Nobody earns any more or less money. The tax is invisible and the environment suffers as before. All you have done is decreased the proportion of earned income and increased the proportion of transfer payments, but its the same budget constraint in real terms, and the same amount of pollution will be created. No one will be better off as a result of that type of tax.

But if the economy contains two techniques that produce similarly useful output, but one of these techniques pollutes less, then there is an effect from substituting to the more efficient technique.

Now assume that one of these techniques is more costly to adopt, then if the (one-time) cost of adopting the new technique is equal to the NPV of the taxes saved, then you are indifferent to using either technique. If you then keep raising the taxes so that the NPV of the additional taxes paid in case you use the old technique is equal to to the additional social costs of using the old technique relative to the new technique, then you have an efficient tax. But note that only relative social costs and relative tax burdens count here. The absolute social cost or absolute tax rate doesn't, which is why there is no effect if there is only a single technique.

But all of that assumes that the two techniques produce equally useful output. You can have perverse effects if they do not, and this is what I was trying to point out. I.e. suppose taxes levied when using the more polluting technique raise the cost of a producer by 1%, but a 1% increase in price corresponds to only a .5% decrease in demand, for a given income. The taxes paid when using the more efficient technique raise the cost of a producer by .5%, but a .5% increase in price leads to a 2% decrease in demand for that good, for a given income. As the proceeds of the tax are mailed out and realized as additional spending on both goods, the effect of the tax would be to shift production towards the more inefficiently produced good, thereby increasing pollution (and total social costs). So again, at the macro level, it is all relative. You care about the ratios of the costs of switching techniques, the ratios of social costs, and the ratios of tax burdens.

Now I am wondering where, in the above calculus, I could be wrong. I don't see it, but perhaps someone else does.

[Mike Moffatt]

"A Pigouvian tax, furthermore is extremely susceptible to substitution effects. It's tax base is the good, not the money part of the transaction. That's what Pitt realized way back when. In five years if carbon consumption is reduced 20% through conservation and improved technology what do you do, take a huge slice out of some other sector of the economy? Now we're back to Mr. Arbitrary Bureaucrat again. The tax base is too narrow."

Again, you're assuming that I'm advocating switching the majority (or a significant majority) of taxation from income taxes to excise taxes. I'm not. While I agree there's a limit to how far you could push this, you haven't shown that anything I'm advocating is anywhere near that limit, while dismissing existing billions of dollars of real-world revenue as 'a drop in the bucket'.

And I'd love to live in a world where a 20% reduction in CO2 emissions through 'conservation and improved technology' is considered a 'problem'. If that's a problem it's a hell of a nice problem to have.

But let's suppose you're exactly right. In 2012 we set up a carbon tax similar to the one in the Green Shift, and it has a collection rate of 75% (that is 25% of emissions go untaxed). The tax revenue from that would be:

772 million tonnes of carbon dioxide equivalents * $40 per tonne * 75 percent collection rate = $23.1 billion.

The plan is so successful, these emissions drop by 20% by 2017. 20% reduction of 772 leads to 612 million tonnes of CO2 equivalents.

612 million tonnes of carbon dioxide equivalents * $40 per tonne * 75 percent collection rate = $18.3 billion.

Revenues have fallen by $5 billion, which we'd have to make up somewhere. Few different ways we can do it - like say a 1% point rise in the GST.

In the meantime we've only managed to hit our original Kyoto Protocol target (well, close, target was 571 million tonnes) in 5 years after 22 years of neglect.

With your assumptions on reduction and this program, the government of Canada has raised 100 billion dollars to put towards program spending, tax cuts, the debt, etc. It's managed a dramatic turnaround on CO2 emissions. But it's somehow a problem because we have to raise the GST a percentage point in year 6?

[anon]

"Imagine if only one technique was available, then the government taxes producers and gives the money to consumers who turn around the spend the income on producers and there is no change. You don't have less pollution than if you had not levied the tax."

If the tax has literally no effect on the amount of pollution, this still makes it a very efficient source of revenue. Most taxes (including income, labor, consumption and corporate taxes) have large excess burden effects which reduce economic activity.

[Oliver]

The solution is simple - we can use Pigovian taxes or cap-and-trade with auctionable permits to reduce emissions (problem 2) and the revenue can be used to fund the initiatives we like (problem 1)

I haven't read through all the comments yet, so someone may have brought this up already. But if Pigovian tax consists of taxing emissions with the aim of minimizing them, then that's the same as saying you want to minimize tax revenue. So you're trying to minimize the funding of initiatives you like... I'd say the incentives aren't aligned.

Can't we just agree that the problem itself is political and not economic in nature and that any mechanism to tackle it will have to take that into account? Cap & trade seems more apt in my mind: cap = politically set goal, trade = economically determined distribution.

[K]

Oliver: Cap and trade makes sense if the marginal social cost of increased emissions is infinite.   And if the marginal benefit of reduction is zero. I.e. You refuse to accept *any* increase in emission no matter what the cost of refusal and you have no interest in rewarding any global marginal reduction. The logic is deeply flawed.

"So you're trying to minimize the funding of initiatives you like... I'd say the incentives aren't aligned."

You do need to read the comments.  The objective should be to internalize externalities - not to raise revenues.  A carbon tax, for example, should be implemented independent of revenue requirements.  The revenues ought be paid to those who are the victims of emissions.   If that includes everyone then it is appropriate to distribute the revenues equally in the form of a citizen's dividend or to cut other, more distortionary, taxes.

[Oliver]

I agree, K. Mike's comment just sounded like the reduction should be absolute (reduce emissions!), not marginal and used to raise revenue with which to fund other things - hence my reaction. But I'll read though the comments.

[Mike Moffatt]

"I haven't read through all the comments yet, so someone may have brought this up already. But if Pigovian tax consists of taxing emissions with the aim of minimizing them, then that's the same as saying you want to minimize tax revenue. So you're trying to minimize the funding of initiatives you like... I'd say the incentives aren't aligned."

I see what you're getting at, but it isn't the same as 'minimizing revenue' (since our revenue by not taxing them is zero).

K's explanation is excellent: "If that includes everyone then it is appropriate to distribute the revenues equally in the form of a citizen's dividend or to cut other, more distortionary, taxes."

[Joel Wood]

K: "Not necessarily. So long as it is set at the marginal social cost, we don't care if you pollute or not. "

But the implementation of the tax presumably reduces the polluting activity to the optimal level right? Unless you are considering a world with no substitutes, then consumers will be consuming something else as well. In a dynamic framework with an uncertain stock pollutant, IMO we would definitely want substitution away from the activity, or technical innovations that make the activity less polluting especially if intergenerational effects are involved. An env tax provides relatively strong incentives for both of these forces.

[Oliver]

Right, read the comments. I see determinant has been down my road.

reduces the polluting activity to the optimal level

what's the optimal level of pollution? 0, I would have thought.

It boils down to whether you believe that, by levying the taxes (and distributing them as a social dividend), you are un-doing the damage to the environment, or whether you are compensating the public for the loss of utility from a harmed environment. The notion of Pigovian taxes allowing for free disposal is the former, but the latter notion is more applicable to the environment, particularly to relatively irreversible effects of pollution (e.g. death, carbon in the air, etc.)

I would have thought the idea of any environmental policy is to avoid damage in the first place.

[Joel Wood]

Oliver:

The optimal level of pollution is where marginal damage from pollution equals marginal benefit from pollution (i.e., opportunity cost of reducing pollution either through technical solutions or reduced output). This could of course be zero, but is generally not zero for most pollutants.

[Gregory Sokoloff]

Mike, one question I have for advocates of Pigovian taxes to reduce greenhouse gases from transporation: exactly what level would the tax have to be to cause the automotive industry to produce zero emission vehicles? For decades the price of gasoline in Europe has been over twice what it is in North America, and yet Europeans did not develop and market electric or hybrid cars. It is true that Europeans engineered fuel efficient diesel engines, but this path would never have lead to a ZEV. So what level of gas tax would? Ten times the current level? Would this be in the realm of the politically feasible?

By contrast, blunt force regulations propounded by the state of California, and fought tooth and nail by all sectors of the auto industry, nevertheless have resulted, at least in part, in things like the GM EV1, the Tesla, and the Nissan Leaf.

There is also a third approach: large scale government investments in research to develop technologies that are critical for a cheap ZEV. For example, if there were breakthroughs in the nano-engineering of the battery, it could completely change the economics of the electric car.

It is not clear to me that these three approaches are in any way mutually exclusive, and I'm surprised that advocates of one or the other behave as if they are.

[ChrisFahlman]

The Conservative government will refuse a carbon tax or a cap and trade system, because that would provide an industrial boost, in Canada, to concerns centered in Ontario (for instance, electric cars, wind turbines, electric power equipment, design and expertise, and, especially, the nuclear power industry), rendering that province's future job growth more independent of growth in the oil sands industries. This is also why they are intransigent internationally on this issue, and why they are shrinking the box into which AECL may grow, and shrivelling the nuclear industry from an international concern (in a sector poised for growth) to a Canadian rump.

The greater Ontario's economic jobs and growth is linked to the western provinces' growth, the more Ontarians will not want to disappoint the political sympathies of the westerners providing their supper. Keep them lean, hungry, and knowing who their masters are.

It's important to note that any hypothesized boost to Ontario from any sort of proper Canadian carbon accounting would not likely come at the economic expense to the Alberta or Saskatchewan. Even oil sands interests have called for a carbon tax (I don't think insincerely). Their major growth product, oil, for transportation and plastics, has a global market broadly separate from energy from electricity. It is a pet peeve of mine to emphasize this, since many left or centrist commentators assume that Harper, in rejecting any sane policy to account for climate change, is doing the bidding of oil companies. Not so. The simplest explanation of what Harper is doing is to assume he is just doing what is best for Harper.

A carbon tax or cap and trade regime could greatly affect the internal Canadian political scene. This, rather than economic concerns, is what is driving this issue now. Harper isn't a Marxist, but he'll make the trade of the country being worse off, long term, if it improves his personal fortunes today. Just like any run-of-mill controlling autocrat.

[Joel Wood]

Gregory:
Why would zero emission cars be optimal? The goal of an env tax is to set marginal damages equal to marginal abatement costs, it in no way suggests that the optimal level of emissions is zero. Also, electric cars are usually not absolutely zero emissions. You must take into account the generation of the electricity.

Chris:
Do you have evidence that a carbon tax would generate net econ benefits for Ontario? I would think that it would impose serious short term costs on the current Ontario manufacturing industry. Please do not cite FIT subsidized "green" manufacturing and "green jobs" as evidence since many of these projects would not be undertaken under a carbon tax and no FIT program (they would need an excessively high carbon price). Don't jump to conclusions that a carbon tax would be the end of Alberta. I believe that Pembina estimated that oil sands development has a value/t CO2e over $500.