February 28, 2006
Just how implausible is a gasoline tax?
New poll results on energy
Responses to my previous posts here and here regarding the advisability and (political) feasibiity of energy taxes have met with some skepticism. Today's New York Times report "Americans Are Cautiously Open to Gas Tax Rise, Poll Shows", casts a slightly different light on the issue:
Americans are overwhelmingly opposed to a higher federal gasoline tax, but a significant number would go along with an increase if it reduced global warming or made the United States less dependent on foreign oil, according to the latest New York Times/CBS News poll.Here's the graphic:
Eighty-five percent of the 1,018 adults polled opposed an increase in the federal gasoline tax, suggesting that politicians have good reason to steer away from so unpopular a measure. But 55 percent said they would support an increase in the tax, which has been 18.4 cents a gallon since 1993, if it did in fact reduce dependence on foreign oil. Fifty-nine percent were in favor if the result was less gasoline consumption and less global warming. The margin of sampling error is plus or minus three percentage points.
In related news, the FT reports:
President George W. Bush yesterday stepped up his rhetoric about US dependence on oil from the Middle East, warning about the dangers of being dependent on countries where "tyrants control the spigots".
In remarks at the National Governors Association meeting, Mr Bush said: "I spend a lot of time worrying about disruption of energy because of politics or civil strife in other countries - because tyrants control the spigots. And it is in our national interest that we become less dependent on oil."
Posted by Menzie Chinn at February 28, 2006 07:56 AMdigg this | reddit
I doubt that the government would be able to efficiently mark the proceeds from this tax and direct it to programs that either reduce oil dependence or decrease global warming.
The poll also assumes that the US can actually have a significant impact on reducing global warning.
Posted by: cole at February 28, 2006 12:57 PM
This is some of the strangest polling data I've ever seen.
Do you like cheeseburgers?
What if I told you that cheeseburgers are actually quite delicious? Then would you like cheeseburgers?
Posted by: Ben at February 28, 2006 02:06 PM
Ben, that's the funniest thing I've read all week. Nice work.
Posted by: Rick at February 28, 2006 02:41 PM
It isn't very transparent to me as to how an increase in tax could result in decrease in dependance on foreign oil or for that matter decrease in global warming. Can someone explain the rationale ?
The following random thoughts cross my mind though they don't seem to indicate anything to me as they only bring up more questions. Is the idea that the additional tax generated would be used as revenue to fund projects for alternative fuel vehicles ? How does one decide the amount of tax per gallon ? Surely an increase in tax would reduce the sales and result in lower revenues, right ? So, how does one decide the turning point in the curve ? How is it assured that the additional tax will be used for such projects and not any other purpose (say an emergency were to arise) ? How does an additional tax relate to decrease in global warming. Clearly this is something which is not as easily controlled as the US only partly influences the climate of the earth, right ? In fact, for a large fraction of the world's population, the possibility of an adverse impact on the planet earth by means of global warming is an insignificant concern when compared to thier other basic concerns (like hunger, shelter). How is the additional tax going to translate into something which would convince others to spend time and money on the problem of global warming when they have more basic priorities to address ? And finally, what is the government proposing to do which would change with additional funding ? These problems have been around for a fair amount of time and by now, there should be a reasonable amount of prototyping work to convince others of a direction to proceed in.
Posted by: Vikram Asrani at February 28, 2006 05:12 PM
1. Yes, the increased tax would lower consumption, but only slightly, because energy demand has a low elasticity. However, the philosophy here is that "every little bit counts", as there is not much time left. Over long periods of time, more chance would occur, in response to price pressures, of which gas price would be part.
2. Yes, the idea is that the gubb'mint would spend the extra tax money on The Best Alternatives(tm). The theory here is that thousands of years of tradition of government blowing the vast majority of its tax revenues would suddenly be cast off and you could sleep easy knowing that clueless politicians are disbursing your hard-earned dollars better than nobel prize winning energy physicists (or so the government says and wants you to assume).
3. Any reduction in consumption of energy will reduce greenhouse gas emissions, presumably in a gradually-increasing fashion. I would argue the goal should not to be to "reverse global warming", but to make man's impact on climate-sensitive things negligible. Then, no matter which way things go, we are blameless.
Posted by: Aaron Krowne at February 28, 2006 06:36 PM
Like the thin man that allegedly lurks within every fat man, isn't there a fix for many American problems through making an increase in gas taxes part of a broader shift to comsumption taxes? Even adopting the dreaded VAT?
Might there not be a cross party deal in eliminating the AMT, removing taxes on savings, yet keeping some sort of basic income tax with a progressive structure?
This deal could include increasing state consumption taxes to address the problem, for example, of financing schools on too narrow a property tax base.
And there would be a lot of bones to throw around like keeping the lower dividend tax in exchange, perhaps, for keeping the death tax. Lots of fun negotiating to keep Congress busy. The end result would undoubtedly still be a complex mess but one which actually helped the country through lower gas consumption, higher savings and some rationality in the tax code.
Posted by: C Thomson at February 28, 2006 06:46 PM
I don't see the market failure here. In a free market the price of an exhaustible resource will climb at a steady rate as the resource is consumed, thereby maximizing the net discounted social value of the resource. In such a situation, consumers will adjust their use of that resource based on price signals, again maximizing the net social utility of the resource. Adding a gas tax is only going to reduce consumption below the optimal level and reduce total social welfare.
Posted by: Hal at February 28, 2006 07:16 PM
Cole: Since the U.S. is a major generator of green house gases. In 2002, it generated 23.3% of world emissions, according to the December 2005 EIA report. So action in the US would have an appreciable impact on the world (note: China emitted only 2/3 of what the US emitted).
Hal: Remember the basic micro teaches us that a free market is not the same as a competitive market (or better yet, see Hal Varian's graduate micro textbook for a technical exposition). So, combustion of fossil fuels for transportation purposes lead to at least the negative externality of pollution. If used to fuel private motor vehicles, there is an additional congestion cost. The current Federal tax might or might not fully internalize these costs.
In addition, if dependence on oil imports subjects the country to oil shocks that combine inflationary price impulses with aggregate-demand reducing import value increases, then public policy might be profitably aimed at reducing dependence, because the market price of oil does not represent the marginal social cost of using oil. In instances where the externality is internalized by a tax, welfare improves.
Vikram: Please see my post on tackling oil addiction. There I discuss the price elasticities in short and long run used to estimate the changes in behavior arising from a gasoline tax increase.
Posted by: menzie chinn at February 28, 2006 08:56 PM
The daily left-wing drivel from this site makes me laugh: Cite with a straight face a NYT poll showing that Americans are in favor of more government (higher taxes) to "save the world". That is the best way to market higher taxes to pay for left-wing "economic" professors on the government dole. Actually the NYT left off the results from one poll question: 91% of liberal yuppies are in favor of higher taxes if it will get the other poor scum off of the road and out of their way so that they can get to their work and daycare more quickly. Liberals...Uggghh. Tune in tomorrow when our favorite liberal professors post an article citing the need for a "U.S. Industrial Policy" - so our economy can be as dynamic and robust as Europe's...bwwaaaaaaaa, ha, ha..!
Posted by: Jack Morrison at February 28, 2006 09:27 PM
What's so strange? The logical conclusion would seem to be that most of the people polled don't believe that a gas tax would reduce dependence on foreign oil, reduce oil consumption, or help with global warming. I think all those things are probably false, if the tax were big enough, but I don't think it's so obvious that no one can reasonably be skeptical about it.
Posted by: SqueakyRat at March 1, 2006 04:48 AM
Ben (and SqueakyRat),
I think the logical inference is that people at first assume that the tax is just a revenue measure, say to reduce the deficit. When people are given a good purpose for the tax, then they see it very differently.
I think if people would be even more in favor of such a thing if we were to eliminate the question of deficit reduction from the gasoline demand reduction question by making it revenue and tax progressivity neutral - a FICA credit, or an income tax rebate, or even a rebate to states to reduce sales taxes. Or possible 90% neutral, and 10% to energy investments.
Maybe apply 75% of the revenue to an income tax cut, and 25% to subsidies for new car purchases (to help out GM & Ford, and lower income groups that need to replace their cars). Another alternative tax cut would be revenue sharing to the states, earmarked to reduce sales taxes, which would have roughly the same impact on lower income groups as the gas tax. Maybe earmark 5% to cover health care costs for domestic car manufacturers (a cost which foreign car makers don't pay, with their national health systems) - that might generate $5B per year, more than enough to help out GM & Ford.
Anyway, give it a good purpose, and make it not a tax increase, and I think people will go for it.
Posted by: Nick at March 1, 2006 10:02 AM
I'm tickled by the fact that 47% of people seem oblivious to the notion that taxing gasoline would reduce energy consumption and greenhouse emissions. And furthermore that by merely suggesting a connection you can change their stance on an important piece of policy. It shows just how ineffective economists have been in making their case for gas taxes.
Posted by: Ben at March 1, 2006 10:51 AM
Two of the most profoundly disrespected MSM organizations show, yet again, why they are profoundly disrespected.
The follow-on questions are akin to "If I promised you 72 virgins when you die, would you blow yourself up to kill innocent men, women, and children?" A matter of degree difference only, the illogic remains.
I'm with you, Jack, what a hoot.
Posted by: Joseph Somsel at March 1, 2006 11:05 AM
Here's a hoot: the reliably conservative MSM organization Fortune magazine supports a gas tax of $1 as being "economically optimal"!
Posted by: Nick at March 1, 2006 01:06 PM
We can argue about imposing a gasoline tax and we can argue about slanted, misleading opinion polls.
Fortune Magazine's support for former does not imply an endorsement of the latter.
Posted by: Joseph Somsel at March 1, 2006 03:15 PM
Why do you feel the poll is slanted, or misleading? It appears to suggest that people support a gas tax if they feel it would have a productive goal. That seems reasonably straightforward..
Posted by: Nick at March 2, 2006 08:50 AM
This is being discussed at The Oil Drum.
My comment in that thread:
Re: the gas tax, a commonly expressed opinion is that whatever the level of the tax, the proceeds should be "earmarked" for highway construction and maintenance.
I want to turn that around and ask, how much general tax revenue is diverted to supporting the infrastructure that makes it possible to actually get anywhere by car in the first place?
I have this book, The Elephant in the Bedroom: Automobile Dependence & Denial: Impacts on the Economy and Environment by Hart and Spivak. It states (p. 44): "For every dollar motorists pay to local governments, the governments spend eight dollars to provide them with these essential services." (The services are: "Traffic signals, traffic engineering services, police and fire protection for motorists, traffic control, auto theft control, street lighting, street repair and maintenance, flood control, parking facilities, paramedics, courts, hospitals, air pollution control and related services.")
I have no idea how accurate that 1 to 8 ratio is. (The book doesn't cite any sources to justify the numbers.)
So any ideas about what the real ratio is? (With evidence?) Are motorists subsidized, and if so by how much?
It seems to me that if there are major subsidies to motorists, that should be the main argument for raising the gas tax. Let drivers just pay the costs of their driving. Why do they need government handouts?
Posted by: mikey at March 2, 2006 07:08 PM
china,the frugal ,only contributes 2/3 the co2 done by the US....and only 1/10th the gdp??
when china doubles its gdp to 1/5 the US's ,it will contribute 4/3 the co2 the US does...how's that for a bargain..
Posted by: embutler at March 3, 2006 07:33 AM
In one sense the number is probably accurate. Figure out the dollar cost of the lives lost or crippled due to road accidents (in terms of lost GDP) alone. Car crashes disproportionately kill the young, middle class and highly productive.
Another big factor is time lost due to congestion-- a potentially very large cost to the economy.
The number I have seen for the UK (where gas taxes are c. $3/gal and there is an annual 'Road Tax' which every car pays based on its CO2 generation capacity) is that for every £ the government recoops, it spends £4.
In another sense you would have to redesign a whole economy *without* the car, and do a like for like. Which is impossible, or at least difficult to imagine. I remember Robert Fogel did this 'counterfactual' with regard to the US economy- the US without railroads (so the opening of the West was via waterways). It becomes a pretty theoretical exercise.
Posted by: John at March 3, 2006 09:56 AM
Generally on gas tax, we know that:
- demand for gasoline is price inelastic, so that a big increase doesn't cause a big fall in gas consumption (at least in the short run)
Therefore gasoline taxes are a great way to create government revenue (the least distorting taxes are those which don't cause big changes in behaviour: it is those big changes in behaviour which are welfare-losing).
So you can't argue that a gas tax is good for making a more fuel efficient, CO2-reduced economy, *and* argue it's a good source of revenue.
Either the tax is a good source of revenue, *or* it causes a big change in long run behaviour (more fuel efficient cars, commuting shorter distances, etc.).
It can't be both.
UK experience ($6/gal gas, $3/gal+ taxes) is that it doesn't cause huge changes in behaviour. We drive more fuel efficient cars than you do (an average of about 10 mpg I think) and *far* more diesels (and we have fewer cars per capita). But mileage driven each year keeps creeping up, in line with GDP as does the number of cars per 1000 population. And our oil consumption per head is not so different from our GDP per head (relative to the US). We have c. 50% of the oil/head (in a smaller country with fewer extremes of climate) and about 60%/head the GDP.
The big break on driving in the UK is congestion, now the worst in Western Europe.
So the lesson for the US may be to discourage driving, stop building roads. It is the road building programmes which cause urban sprawl.
Posted by: John at March 3, 2006 10:01 AM
John: The CBO reports estimates for US short run price elasticity of between -0.22 and -0.31, and long run price elasticity between -0.8 and -1.01. Things may be different in the UK.
embutler: The idea that the pollution per unit output might decline with income is behind the "environmental Kuznets curve". (see Copeland and Taylor for a discussion).
Posted by: menzie chinn at March 4, 2006 12:45 PM
Thanks for that. If I read you correctly, a 50% rise in the price of gasoline, in the US, has a long term reduction of 40 to 50% on gasoline consumption?
In which case, such a tax would be revenue neutral?
We haven't abolished gasoline consumption in the UK, with gas prices 2 times yours. We are about 30% more efficient (about half of new cars sold are diesel engined) on fuel economy, and we probably drive fewer miles (but more miles of congested roads, so the actual fuel consumption per driver may not be as much less).
I guess I am confusing Substitution Effects and Income Effects here? What I note is that even in a smaller country, with gas prices twice as high, the UK's citizens appear to consume about as much petroleum product as the US's citizens, *if adjusted for* GDP per head.
I guess my basic point is a gas tax is a great way to raise money, because gasoline demand is so price inelastic.
As a way of helping the environment or meeting peak oil it fails.
I suspect these problems (of income effect overwhelming price substitution effect) are so large that it we get serious about global warming, we are talking about *really* significant taxes on carbon generation.
Something of an aside-- London now seems full of small electric cars-- I would guestimate 0.25-0.5% of all the cars I see (1 in 200). The trick being you can save £8/day in congestion charge (the $14 charge to drive into the centre of town 7am to 7pm) *and* parking (as much as £25/day), if you drive an electric car.
So that is a measure of what it takes to force a driver to switch to an ultraeconomical (but not carbon neutral) mode of transport - say £30/day or $50-55/day or c. $11k pa (40 wks per year, 5 days/ wk).
Now *that* is inelasticity ;-).
Posted by: John at March 6, 2006 06:50 AM
John: Fuel costs are only part of total costs of operating a motor vehicle, and as income rises, the benefits of operating a motor vehicle may rise (with traffic congestion operating in the opposite direction). In my mind, no particular reason why people in different countries should have the same elasticities (just a little time doing cross-country regressions on various issues will convince you of that).
In any event, the 1.01 elasticity is at the high end. The mid-point of the range is 0.9. The revenue implications at the long horizon are small, but given secular increase in consumption, the level of revenues will be higher than w/o the tax increase. And as economists, we should be most concerned with internalizing the externalities, rather than the revenue implications.
Posted by: menzie chinn at March 6, 2006 11:11 AM
You are an economist, I am just an innocent bystander ;-).
Still it is an interesting point. The case for the gasoline tax is usually made around saving energy, but actually we are not expecting that to occur (income effect will substitute over substitution effect).
It seems to me that the long run efficiency gains of a gas tax might be quite small. Certainly the claimed energy efficiency/ national security benefits don't look huge.
It also strikes me that you *might* get some very big efficiency gains if you used the revenue generated to lower the *employer* share of social security. A direct tax on using labour, and a direct tax on using the *lowest paid* labour. I have this suspicion (not backed up by reading) that a wage-linked, regressive tax on employment like social security on employers is about the most damaging tax you can levy.
I would be very surprised if the elasticities re gasoline between the UK and US are radically different *except* that we are a much poorer country (GDP/ head). What happens in the US happens in the UK, albeit 10 years later eg obesity is now beginning to be a significant concern here.
The point being similar, for 90% of UK car journeys there is no easy alternative-- public transport isn't that great outside of central cities. And as you say, gasoline is a relatively small proportion of total cost of ownership v. depreciation, road tax, and insurance.
It's a warning if we ever get serious about carbon taxation. The levels necessary to achieve the kinds of carbon usage reduction we need will be very high-- not 30 euros per tonne (the current EU market price for industries in carbon trading) but 300 euros/ tonne. At which point, a long haul flight to Australia is going to be $1000 more expensive (but still cheaper than it was in 1965-- another problem, the income effect again). I am guessing here that if the developed world cuts CO2 emissions by 50% pa, we can just about hold average wrold temperature increases this century to the 6 degree centigrade level (that's 14 fahrenheit to you Yankees ;-).
I'll ramble on and say that another intuition I have is that the consumer will wear that 30 euro or 100 eur/ tonne cost and still go on package holidays to Spain or Florida BUT industry, when faced with that kind of carbon cost, will discover a multitude of ways to reduce carbon emission. If you think the average business makes a 10% net margin, a 1 or 2% reduction in carbon costs is a 20% increase in net profits.
Posted by: John at March 6, 2006 02:27 PM
The poll is slanted by its false premises or at least unsupported ones. Given the vigorous debates on the efficiacy of a gasoline tax, it remains a controversial issue and a poll that uses the unsupported premise is a form of "push polling."
These questions have two functions: to create a headline supportive of the editors' agenda and to insinate the truth of the unsupported premises.
Would this question be valid?
"Would you support or oppose another hidden tax on a vital economic commodity that would do nothing about oil imports with proceeds to be used for more pork barrel projects?"
Of course, it probably wouldn't have gotten much lower assent than the original and straightforward "Would you favor or oppose increased Federal taxes on gasoline?"
Posted by: Joseph Somsel at March 9, 2006 09:16 AM