21 November 2008

In Defense of Ethanol

Two days ago, I criticized the corn ethanol program for increasing food prices (I also dislike it because it subsidizes one crop with a heavy environmental footprint to produce fuel that's probably doesn't even reduce overall carbon emissions.)

In response to the points in that post and a conversation we had last weekend, Ben Ho, an assistant professor at Cornell and former economist (2006-2007) at the Council of Economic Advisers, wrote this post.
Ethanol is an immensely complicated issue that requires a detailed institutional background to fully appreciate, and those with a little bit of knowledge often jump to wrong conclusions.

On the Benefits of Ethanol.
The environmental benefit of ethanol has been hotly disputed over the years. Various studies have found various things over the years as is typical in science (a recent study finds that 1/3 of the most cited publications in a top medical journal are refuted in just a few years). A meta-analysis of all of these papers shows a clear time trend. Newer studies show greater benefits. The reason for this is simple. While the environmental benefits from corn-based ethanol are arguably small today, they will only grow in the future. Corn yields have been increasing steadily for decades, and there is every reason to expect that to continue. Thus we should get more ethanol per environmental cost each year. Furthermore, new enzymes are being developed again to get more ethanol from the same amount of corn, thereby further increasing the amount of ethanol per given input.

All of this also does not account for the possibility of more dramatic technological shifts that subsides make possible. Current car engines are tuned for gasoline from oil. However, engine tuned for gasoline from ethanol could lead to 25-30% increases in engine efficiency. Furthermore, the infrastructure being developed is helping to pave the way for new biofuels, like biodiesel and cellulosic ethanol that have substantial and already demonstrated environmental benefits.

Finally, environmental analyses ignore the other benefit of ethanol, which is a diversification away from oil consumption. Much of the problem in the oil markets comes from the monopoly power of the oil industry given the lack of substitutes for oil. By helping develop a viable substitute, there are immense gains beyond just the environment, for example, dampening the geopolitical power of OPEC, and alleviating the resource curse in the Middle East.

On the Ethanol Subsidy.
Many people wrongly assume that the government specifically subsidizes only corn based ethanol. In fact, regular ethanol gets a 51 cent per gallon subsidy, biodiesel gets around 80 cents, cellulosic ethanol gets over $1. Other fuels get subsidized in their own way. And in fact a system of mandates and other tax credits makes total subsides substantially higher and substantially more complicated.

This is not an ideal system. A carbon tax would be much simpler that would replace this myriad of complicated and overlapping policies. But we do not live in an ideal world. So on the simple question of whether we should repeal the ethanol subsidy, we go back to the tools of cost benefit analysis (something I used to teach back at Stanford).

What are the costs of a subsidy? The subsidy is effectively just a transfer from one American to another, so the cost is only the cost associated form the distortions to the economy from raising that 51 cents. A large literature estimates the cost of raising 51 cents to be around one quarter of that, or around 12.5 cents. Is that worth it?

It is first worth noting that mathematically, a subsidy on a good works exactly the same way as a tax on a bad. So if you believe we should increase gasoline taxes by 51 cents, then you should support a 51 subsidy for ethanol.

The benefits of an extra gallon of ethanol are hard to quantify, but if you add it up (the small environmental benefit, the impact of geopolitics, the fostering of technological innovation, the support of an agricultural lifestyle) it is actually fairly easy to find benefits far outweighing 12.5 cents.

Again, subsides for ethanol is messy ugly kludgy policy. But absent a comprehensive gasoline/carbon tax, repealing the subsidy could be even worse.

On the Ethanol Tariff
Many people (including free-market economists) decry the 54 cent tariff on imported ethanol, without really understanding the institutional background for the tariff. First of all, most imported ethanol falls under various free trade agreements that makes them exempt from the tariff. Second of all, while it is true that we place a tariff on Brazilian ethanol, Brazilian ethanol also enjoys the 51 cent subsidy, hence leaving Brazilian ethanol producers no better or worse off than in a free market environment. The main purpose of the tariff is just to make sure only American ethanol producers get the subsidy.

Whether this is fair from a global justice perspective is debatable but from a standard economic cost-benefit analysis, it is eminently reasonable. Recall that a transfer between two Americans is not considered an economic cost from the point of view of America since it has no net effect on American social welfare (only the dead weight loss of raising the taxes is considered a cost). However, a transfer from American tax payers to a Brazilian producer is considered a cost.

On Ethanol and Food Prices.
It is incredibly difficult to estimate the impact of biofuels on food prices. This is why most respectable economists decline to give exact numbers. I was asked to do such an analysis for the White House in 2007, and found that drought and crop failure combined with rising energy costs (a major input in production) and increased meat consumption in places like China, account for a far larger share of the price increase. This may have changed in 2008 (though I doubt it), though either way, the argument seems strange to me.

I find it incredibly odd that the same economists (like those at even Oxfam) who have for years been decrying farm subsides for depressing food prices and therefore hurting developing world farmers are today decrying biofuel subsides for increasing food prices. It is hard to say whether the developing world would prefer lower or higher food prices, though one would think that since developing countries tend to have a comparative advantage (note I said comparative, not absolute advantage) in agriculture (seen by the fact that they devote most of their labor force to agriculture), then increasing food prices leads to a favorable shift in their terms of trade, which means it should be good for developing countries overall (yes it hurts the non-farmers in these countries, but it helps all of the farmers who tend to out-number the non-farmers, leading to a net gain).

32 comments:

Anonymous said...

What is this? Please don't publish every piece of propaganda that is passed on to you, David.

Seriously.

Anonymous said...

....Unless this was all just a ruse to get us talking again about how terrible ethanol is... in which case... well played.

dWj said...

> It is first worth noting that mathematically, a subsidy on a good works exactly the same way as a tax on a bad. So if you believe we should increase gasoline taxes by 51 cents, then you should support a 51 subsidy for ethanol.

?! Is he assuming that any increase in ethanol use due to the subsidy is a substitution out of gasoline use?

The public seems more accepting of subsidies than of taxes, probably because they don't realize that they have to pay for them. It seems to me that because of this we end up putting a lot of our environmental strategy into the form of subsidies for things that aren't as bad as their substitutes. Public transit produces carbon emissions, but less than if those people drove. Accordingly, we subsidize transportation overall -- reducing the appropriate incentives for living near work, biking, etc. It probably reduces carbon emissions versus not doing it, and certainly has other benefits, but it's not equivalent to charging cars for their externalities.

Aslihan said...

The last point on farmers vs. non-farmers is wrong too. It's net-sellers vs. net-buyers, in which case the claimed net gain does not happen...

Philip said...

While I remain a skeptic, I think this fellow made a well reasoned and rational appeal. I would not lump it in with the mindless boosterism the industry typically spouts. The estimable Giannini Foundation released an interesting take on ethanol here: http://www.agecon.ucdavis.edu/extension/update/articles/v12n1_1.pdf
It is hardly a rousing defense of ethanol, but it is another example of the kind of thoughtful analysis that trumps a priori yakking every time.

Anonymous said...

How can a fuel (ethanol) that takes as much as 500 gallons of water to produce a gallon of output be even remotely sustainable? Leave aside the debatable energy balance, the fact that you cannot pipeline the stuff (you will have to build a whole new infrastructure!), the impact on food production and land prices, the temptation to cut down rainforests to grow the feedstock, the mountains of glycerine that will result as a byproduct from many processes...the water consumption issue is the clincher, - isn't it?

jwt said...

The most important thing omitted from the ethanol rant is that U.S. agriculture subsidies, of which ethanol is a large part, destroyed the Doha round of trade talks. Even the nasty protectionist French were willing to give ground. There is an overwhelming amount of evidence that world free trade is good for everybody.

Damian said...

"... if you believe we should increase gasoline taxes by 51 cents, then you should support a 51 subsidy for ethanol."

No, I do not, because a tax on gasoline encourages a reduction in use as well as a move to substitute to other things. A subsidy for ethanol on the other hand specifically encourages more fuel consumption and is not the opposite of a gas tax. The real opposite of a gas tax is impossible to identify in one piece--in reality its something like a combo of walking, biking, alternative fuels (yes incl ethanol), better engines, different commute choices, etc etc. To subsidize all of these I guess could be equivalent to taxing gasoling. But to just subsidize ethanol is most certainly not, and for him to say this makes me seriously question his other statements.

"The benefits of an extra gallon of ethanol are hard to quantify, but if you add it up (the small environmental benefit, the impact of geopolitics, the fostering of technological innovation, the support of an agricultural lifestyle) it is actually fairly easy to find benefits far outweighing 12.5 cents."

Possibly, but I take issue with these benefits. Geopolitics? What about the trade distortions and animosity that develop from a foolish obsession to get away from foreign oil? Commerce yields peace, some say. Furthermore, spending more on something that is cheaper elsewhere makes us poorer.

Fostering of Innovation? How about stifling of innovation? How much waste has gone into ethanol research when it may have been much more valuable to spend time researching other alternatives? You cannot pick one and not the other.

Family Lifestyle? This is awful. Who are you to judge that a corn farmer's lifestyle is better than my own? And if not, why are we going to subsidize it? Yes it has benefits to some people, but so does my own lifestyle.

All of your arguments are either looking at half the picture, or are just wrong. Government should not be in the business of picking winners.

Finally, the idea that a 51 cent subsidy is not a cost, just a transfer, is kind of true, but foolish. The diversity in this country is extreme, so the costs are paid by a specific few (corn users, all who pay taxes) and the benefits are reaped by a specific few (corn farmers, Archer Daniels Midland). To say that America is better off as a result glosses over the details which are very important here.

HoBs said...

To DAMIAN

"Government should not be in the business of picking winners. "

Agreed. I said that too. But given that we live in an imperfect world, sometims we have imperfect policy.

"All of your arguments are either looking at half the picture"

Fair enough. I figure others on this site happily provide the other half, so I left that to thme.

"Who are you to judge that a corn farmer's lifestyle is better than my own?"

It is not me to judge. It is for the democratic system. Who are you to judge that the decisions made by a fairly elected legitimate process are wrong?

"Fostering of Innovation? How about stifling of innovation?"

Possible. But give me numbers. Not jut speculation.

"Possibly, but I take issue with these benefits."

You are welcome to. I spent much of 2006 along with the help of an RA and several interns trying to calculate the benefits, and found them to be positive. But I still wouldn't be too much credence in them.

"No, I do not, because a tax on gasoline encourages a reduction in use as well as a move to substitute to other things."

Touche. You caught me. Yes, that was a gross oversimplification. I should have been more careful there.

Empirically though, the the impact of a tax on gasoline, has a very small impact on reduction in use. (Parry and Small have a very compelling paper on this). I should have said, the tax and the subsidy are identical, if you subsidize ALL alternatives to gasoline (which was actually the proposed policy), and you ignore the reduction in use, the "intrinsic margin" vs the "extrinsic margin"

HoBs said...

to JWT

"The most important thing omitted from the ethanol rant is that U.S. agriculture subsidies, of which ethanol is a large part, destroyed the Doha round"

I think that is not quite true. First of all, the US is not alone in its biofuel subsidies. Brazil, and all of western europe all subsidize their own biofuels just as much if not more.

Second of all, I don't think it was US intransience that blocked Doha. It was more European and Chinese/Indian it seemed to me. Pretty much everyone I knew in the administration was all for reducing US farm subsides. The only reason we kept them was as a bargaining chip.

HoBs said...

To Anonymous 10:30

"How can a fuel (ethanol) that takes as much as 500 gallons of water to produce a gallon of output be even remotely sustainable? Leave aside the debatable energy balance, the fact that you cannot pipeline the stuff (you will have to build a whole new infrastructure!), the impact on food production and land prices, the temptation to cut down rainforests to grow the feedstock, the mountains of glycerine that will result as a byproduct from many processes...the water consumption issue is the clincher, - isn't it?"

Well the nice thing about economics and cost benefit analysis is that you can try to put a dollar value on each of those things you mention. We tried (albeit very imprecisely) and found that it was probably still worthwhile. But I'd be happy to see a better measure.

As an aside, the pipelinging issue is resolvable. Amyris and other firms are developing enzymes to convert ethanol into a form that is pipelinable.



To PHILIP

Thank you.




To ASHILAN

"The last point on farmers vs. non-farmers is wrong too. It's net-sellers vs. net-buyers, in which case the claimed net gain does not happen..."

I agree that net-sellers and net-buyers is more precise terminology. Farmers are net-sellers, I defy you to find statistics that show they are mostly net-buyers.


To DWJ

"?! Is he assuming that any increase in ethanol use due to the subsidy is a substitution out of gasoline use?"

Yes. See my comment to DAMIAN.

"The public seems more accepting of subsidies than of taxes, probably because they don't realize that they have to pay for them"

That is the trick. As I said in the post, I agree completely that taxes are a better way. But given political realities, sometimes you have to accept second best.

David Zetland said...

Ben,

I think that many of your responses to others' comments (e.g., doha, innovation, ethanol DISPLACING gas, etc.) lack substance, but I won't bother to lay out the obvious counter-arguments.

Instead, I will take one of your claims (subsidy of ALL fuels is equivalent to a tax on gas), combine it with the FACT that calibration of such subsidies is impossible (and far more complicated than a simple gas/carbon tax), and note that subsidies direct money from the many to the few (whereas a gas/carbon tax would move it from the users to the taxed).

Given these factors and political economy (bribes from ag processors to congress), your amazingly shallow excuse ("given that we live in an imperfect world, sometims we have imperfect policy.") does nothing to persuade me against concluding that ethanol is merely a program for sending OUR money to THEM -- with much harm and waste on the way.

Your "arguments" have done nothing to change my view that the ethanol program is worse than ZERO action and far inferior to a carbon/gas tax.

Aslihan said...
This comment has been removed by the author.
HoBs said...

To David:

You are certainly all entitled to your opinion, so long as you are comfortable with the fact that yours are based on gut reaction and faith rather than empirical analysis.

Recall that I said from the beginning I think a gasoline/carbon tax is better policy (despite their being legitimate political economy reasons why that would be a bad idea), and I encourage you to try. Carter and Clinton both tried and failed. Every political expert I've talked to laughs at the thought. But it is possible that Obama could get it done. Recall that even Obama pandered to his coal constituents during the campaign though.

But that was not the question. The question was ethanol subsidy vs not.

I fully admit that the empirical analysis I have done requires heroic assumptions. That is why at the end of the day, I don't stand strongly behind it. But it is the best I know of.

You can go on and on, rattling off a list of costs. But until you give me a different quantification of these costs vs the benefits, then your argument is based entirely on faith and instinct. Which is fine, but those who deny global warming exists also base their reasoning on faith and instinct. You should at least acknowledge that your opposition to ethanol does not stand on firm empirical footing.

Ben

(My responses were superficial, given space constraints, but here are more details. It is hard to imagine that ethanol derailed Doha (again, a little bit of knowledge is a dangerous thing. Ethanol is a "green box" subsidy (vs amber or blue), and thus was not really even under contention during negotiations. And again, it is hard to figure out who is responsible for a break down in negotiations. But it is a far stretch to blame ethanol.

On innovation, I agree, I don't have a good answer. I spent a lot of time surveying the research on innovation, and good numbers don't exist. Best estimates seem to be that government spending has a 0-50% return rate on average, and that on the whole, most government spending on new technology actually is worth it. But really, you can argue the money on ethanol was poorly spent, but that is an argument that needs to be backed up by data.

It is also rather clear if you look at the data that ethanol displaces gasoline 1 for 1, at least until ethanol makes up 10% of supply. Over 1/3 of all gasoline sold in the US contains ethanol (I don't think most people are even aware it is there), and if you look at consumption data, you see that proportion ethanol has no noticeable impact on fuel consumption suggesting they are in face near perfect substitutes.)

Aslihan said...

See table 3 page 33 of "The High-Level Conference on World Food
Security: The Challenges of Climate Change and Bioenergy"
(www.fao.org/foodclimate/hlc-home/en/) for the proportion of net food
seller households in a sample of countries from the three main
developing regions.

HoBs said...

To Ashilan:

I think you meant Table 13.

Ok, fair point then.

So does that mean you think we need more crop subsides then?

Though, percentage of households doesn't give relative magnitudes. And focusing only on staples does not seem appropriate (a farmer that buys wheat to raise cattle still stands to benefit). I suppose what you'd want to know is net agricultural imports.

David Zetland said...

Ben -- "gut reaction and faith" is pretty funny. Let me direct you to the FACT that the Farm Bill completely fails to strengthen our food production system, that small farmers receive few benefits, and that consumers have access to poorer quality food.

In fact, take a look at the Soviet Union if you need to understand the source of my faith.

By let's put that aside. You say "the empirical analysis I have done requires heroic assumptions. That is why at the end of the day, I don't stand strongly behind it"

Good -- I certainly don't agree with some of your assumptions and agree that you can give it (at best) weak support. Since I come from the "first, do no harm" school of public policy, I'd say that -- analysis or not -- the ethanol program should not exist.

You point to "political necessity," but I point to cowardice and corruption. Perhaps I think that leaders should lead instead of cowering to talk show idiots and sucking on lobbyists', um, talking points.

Ethanol subsides are just another STUPID idea, and OUR JOB as economists is to point that out, not look for reasons or offer weasel words in defense of it.

All of my thoughts are based on opportunity costs, and this program's opportunity costs are much larger than its trivial at best/probably negative "benefits".

HoBs said...

To David:

"this program's opportunity costs are much larger than its trivial at best/probably negative "benefits"."

And that's where we will have to disagree. My (admittedly imperfect) analysis says that's not true. Your faith says it is. We can leave it at that.

As an aside, if you do believe that energy security and climate change are crucial problems. Do you have any proposed policy that that can address them aside from higher taxes?

I'm not sure if any policy can pass your "do no harm test."

I can even make a compelling argument that carbon taxes wouldn't necessarily pass a cost benefit test. The costs of higher taxes are easily estimable, and could be in the trillions. The benefits are less certain. Real change will depend on innovation, and the evidence is scant that taxes alone can create the necessary innovation (that would overcome the underinvestment inherent to R&D even in functioning markets).

David Zetland said...

Ben -- I just wrote a long reply and then lost it (my own blog!)

In it, I listed the many harms from ethanol (listed in other posts) and the zero/negative "benefits," i.e., no reduction in world oil use, no reduction in carbon output, increases in corruption and lobbying and no reduction in US consumption.

I also note that you do not understand cost accounting, and that your "cost-benefit" analysis is invalid without considering (at least) secondary impacts of ethanol.

I also think your "my analysis, your faith" sophism is not only invalid and bankrupt but insulting.

Alternatives to carbon taxes (you mean even MORE ideas that are better than ethanol?), I offer you these:
1) Don't invade countries or support dictatorships to "secure" oil. We've NEVER won that one.
2) Don't subsidize harmful activities.

As my final word (on your last post), I'll note that your final comment (higher taxes will not increase innovation) is so wrong that I LoL'd. Haven't you ever heard of that annoying tendency of innovation occurring where there are profits to be made? (Higher taxes, higher prices, higher profits.) If not, check in with the folks at Exxon, google, IBM, etc. (Just don't check with the government, and esp. not Dept of Ag or Energy).

ps/I'll return Obama's calls to be SecEnergy.

HoBs said...

So I apologize for being insulting. Such exchanges tend to easily get out of hand in faceless blog exchanges. Saying I do not understand cost-accounting doesn't help. I will try to be more respectful though.

I should just drop it, but felt compelled to correct some factual errors:

On reduction of oil use. Each gallon of ethanol pretty much reduces oil demand by one gallon. There are some second order price effects to work out after that, but the reduction is pretty much one for one. See for example, US gasoline consumption has been mostly flat, even as ethanol has come to make up 5% of it, meaning that gasoline consumption from oil is down roughly 5%.
http://tonto.eia.doe.gov/dnav/pet/hist/mgfupus2a.htm

On reduction of carbon output. Again, there are many studies. I have read many of them. They mostly say a small reduction yes. But as I said, they do not account for change in technology, which can only improve the carbon balance.

On the increase in corruption. Hard to find evidence for. Many political scientists have tried and failed. Furthermore, the increase in corruption is the main reason why people oppose a carbon tax; because trillions of dollars in additional government revenue is a dangerous thing.

On the secondary effects, they are there and are hard to quantify. My colleague at CEA (a Democrat) handled a lot of water projects for EPA, but still believed the costs were not enough to outweigh the other benefits.

And finally, I know that econ 101 says taxes should spur innovation, but honestly, I've talked to dozens of CEOs of alternative energy companies, and most say that carbon taxes wouldn't be enough. Of course, they are full of shit. But then you go to the literature, and you find that there is scant evidence either way that it does. Europe and Japan have had massive gasoline taxes for decades (Parry et al., AER 2006 says they are way too high actually), yet they have not produced any magical innovations. Certainly none of the innovations that would lead to an 80% reduction in GHG as the IPCC wants. The one country that did do something arguably transformative is Brazil, who had decades of equally derided subsidies to their own ethanol industry.

I respect your opinion, I really do. If I were given a magic wand and could use it to abolish the subsidies, I'm not sure what I would do. Or if I were asked to advise a future DOE Secretary, I'm not sure which way I would advise. I was just questioning your certitude.

And when I say faith, I don't mean to be perjorative. Just something I've observed about economics as a whole. We have faith that markets work. We couch in scientific terms, but really, the empirical grounding for markets is often tenuous. The theoretical grounding certainly is. Country case studies often fair no better. The Soviet Union failed under communism, but was failing under the Washington Consensus until oil prices increased. Market reforms have failed in as many countries as they have helped. About all we know about what leads to development, depends on the virolency of mosquitos, yet economists still have faith that markets work. I still have faith in markets, but at the same time, I recognize much of it is simply faith not science.

Ben

David Zetland said...

Ben,

I didn't mean to insult you, but I was referring to the cost and benefits not included in your analysis. Omitted factors (like variables) can invalidate any analysis, and I think you are taking results seriously that probably shouldn't be.

Re: reduction of oil use.

I think that you should consider
worldwide demand for oil/gas. In the GE analysis ethanol dampens price and results in higher Q_d elsewhere.

RE: reduction of carbon output.

"Subsidize me today so I can innovate tomorrow" is a fallacy oft invoked by industry (e.g., fuel cells, etc.) As soneone else said, it's even worse b/c the gov't is bad at choosing technology.

RE: increase in corruption. C&T will (and has) delivered far more corruption than taxes will because C&T is not transparent. (See Russia wrt Kyoto.) For the connection between C&T and corruption, see WV coal and SO_2 permits.

RE: secondary effects.

No idea what your colleague found wrt water, but I've seen nothing but bad news on water quality wrt greater corn/ethanol production (read past ethanol posts here).

RE: econ 101 and innovation. Beside undermining your own argument (subsidize to get technological innovation), I'd say that our greatest technologies and innovations are in front of our faces: high MPG cars, living closer to work, etc. Look up per capita consumption of energy in Europe (control for average temperature, of course), and you will see lower numbers there that are probably the result of culture and expensive energy/water.

RE: If could use it to abolish the subsidies, I'm not sure what I would do.

Start there and then do nothing.

RE: advise a future DOE Secretary/certitude.

We already have a gas tax. We've seen the effect of $4 gas. If the goal is to reduce gas consumption (or carbon), then raise taxes so we have $4 (or $8) gas again.

If people are just talking shit (eat cake and have it), then point out their hypocrisy.

RE: We have faith that markets work.

No -- we've seen them work, repeatedly. There are problems, there's imperfect competition, and there are distribution problems, but markets are far more robust than command and control.

I agree with you that most models are BS. That's why I do institutional and experimental economics.

RE: The Soviet Union failed under communism, but [Russia?] was failing under the Washington Consensus until oil prices increased.

No idea what you're talking about, but few oil rich countries stay that way (Norway is the obvious exception). Russia is close to a cliff RIGHT NOW because it's leadership is so corruption (cf. Venezuela)

btw -- I have nothing against "socialism", I just dislike totalitarianism that often results from "clever" bureaucrats trying to subsidize "correct" technologies, etc. (that includes those who want to tax carbon and direct the $$ to favored industries; I am for tax and rebate)

RE: Market reforms have failed in as many countries as they have helped.

Not if you exclude places Sachs "helped" :)

Seriously, you can't have markets (or democracy) without civil liberties, and big bang stuff is often seriously deficient. It takes years (decades) to implement/evolve markets...

HoBs said...

"In the GE analysis ethanol dampens price and results in higher Q_d elsewhere."

Sure it does. But that effect is certainly second order. Again, ethanol is a perfect substitute. You are basically saying that discovering an additional gallon of oil in the US has such a large price effect that it leads to more than an additional gallon of oil being discovered somewhere else. Perhaps theoretically possible, but nonsense.

"I've seen nothing but bad news on water quality"

ok. you know this better than me. Give me numbers. dollar values.

"our greatest technologies and innovations are in front of our faces"

Energy efficiency is useless. It will do nothing at all to move toward 80% reductions. If the whole world were as efficient as the Europeans (and again, Parry et al finds they are *too* efficient) you might get 10-20% reductions in emissions (not 80%). Again, I want numbers. Small and Van Dender (2007) do some fancy 3SLS and find that doubling the price of oil might get you 6% reductions in oil use. Yay. Helpful, but that's no solution.

The goal is not to "reduce" oil consumption,the goal is to essentially eliminate oil consumption.

"but [Russia?] was failing under the Washington Consensus until oil prices increased"

Post communist russia had plunging life expectancies, and economic ruin.

"you can't have markets (or democracy) without civil liberties"

again, that's faith. China and Singapore defy that argument.

"we've seen them work, repeatedly"

And we've seen them fail repeatedly. I don't want anecdotes. I want data.

The institutional economics literature still has very little to say as to whether markets work. i started grad school to do that, but gave up on it given its futility.

David Zetland said...

Ben,

Perhaps theoretically possible, but nonsense.

Now I don't know what you are talking about. I am extending your claim (more ethanol in the US lowers demand for gas in the US by 1:1) to include the world, which will face a lower price for gas (via the residual supply curve) and thus consume higher Q_d.

As to your litany of "give me facts," etc., you are just using a rhetorical tool without being able to show that I am wrong.

Since ethanol is a new program, I have property rights (in a Coasian sense) that demand that you show me (us) how ethanol is better. Claiming that markets fail (sometimes) and then giving examples of two-non-democracies (Singapore having rather good human rights) as your counterpoint does not take care of that requirement. I can give you many examples of where a lack of civil liberties is correlated (and probably causal) wrt market development. You can give me far fewer counter examples.

But, really, it seems like you've managed to select your opinion, find some doubt in theory (and no evidence) and then conclude that it's correct, since noone can argue against your theory or provide enough data (give me more) to show that you should change your mind.

Perhaps we will agree to disagree on this, but I am sure that I do not want to live in a world where the things you claim to be true are -- not for idealism, but because few things make sense (economically, politically) in your world.

What I am saying (ethanol is bad for the environment, leads to corruption, benefits the few at cost to the many) is so much more obviously a a description of the world we LIVE IN that I cannot help but wonder where where you've lived all these years?

HoBs said...

Easy stuff first:

"Since ethanol is a new program"

Ethanol subsidies have been around for over 30 years! What do you mean, a new program?!

"Now I don't know what you are talking about. I am extending your claim (more ethanol in the US lowers demand for gas in the US by 1:1) to include the world, which will face a lower price for gas (via the residual supply curve) and thus consume higher Q_d."

Ok. Fine, let's do this carefully. (This is an approx but close enough)
Let x be the percent increase total fuel from ethanol.
E_s = elasticity of supply (about 0.3)
E_d = elasticity of demand (about 0.3)
Dd = change in fuel demand
Dp = change in price of fuel
The global change in price to fuel is Dp = x / E_s
The global change of increased demand is Dd = Dp * E_d

Thus, Dd = x * E_d / E_s

In this case, every gallon of ethanol perfectly displaces one gallon of oil. Because yes, more ethanol means lower price which means more oil demand. But it also means lower price which means less oil supply. Since the estimated E_s = E_d, those effects are equal. Actually, E_s might be a bit higher, in which case, more ethanol could actually have an add-on effect by further discourging oil supply.


"I can give you many examples of where a lack of civil liberties is correlated"

Look examples is not evidence. You are falling into rhetoric not science. I want (at the very least) regressions. And a massive literature of people have run the regressions. And I am teling you, the regressions are inconclusive. And yes, the regresssions are usually poorly done because there are only like 30 observations, but they are more compelling than just throwing out anecdotes and examples.

"But, really, it seems like you've managed to select your opinion, find some doubt in theory (and no evidence) and then conclude that it's correct, since noone can argue against your theory or provide enough data (give me more) to show that you should change your mind."

Actually, I'd say it is exactly the opposite.

I haven't selected an opinion at all. As I said, I do not have strong feelings either way. And ex ante, I was opposed to ethanol subsidies, since that seemed to be consensus amongst the cognoscenti. I had been since at least the 2000 campaign, as candidates were taking the ethanol pledge and so I looked it up then albeit superficially.

You on the other hand, sound like you have very strong opinions that you are ethanol is bad, but don't really have much hard evidence beyond impressions and examples.

You have to at least give me credit for probably having spent more time collecting evidence than you have. This was a job for me. And there was a team of 4+ people (not sure if interns count as a full person) collecting evidence. And before you say this is all politically motivated, it wasn't. The CEA was mostly Democrat, and in fact, few people in the White House were fans of the subsidies and would have been happy to abolish them, but decided not to touch it because of the way Congress and democracy works.

I'm happy to work with you to try to get numbers for the effects you worry about. I admit we didn't try to quantify the water effects because my colleague thought it was too small to affect the calculation, but we can do it now if you'd like. I can change my mind. But I need numbers. Not just your assertion that the effects are big.

I can summarize more of the evidence and give you citations if you want more. I haven't put them here out of laziness. (I have somewhere the WHite House memo I wrote on the subject, but I don't thnk I'm allowed to circulate that). The calculations are based on putting dollar values for national security benefit of oil displacement (heroic yes, but positive), as well as taking the DOE/EPA estimate for carbon savings, (and yes, there are new papers on this, and I read them, and if you read them carefully, you find they don't really contradict the DOE/EPA estimates) and then adjusting for technological change. You can find a lot of the citations in Parry's JEL 2008 (maybe it was 2007) article.

My opinion is that it is a hard question. We don't know for sure. And thus you shouldn't be so certain of your opinions. I have said over and over, that I am not sure of mine.

David Zetland said...

If the ethanol program is 30 years old, it hasn't been relevant until recently and the targets recently enacted by the Congress are certainly 1) larger than ever before and 2) explicit targets that need to be met.

(That's like saying the Volt is not new since electric cars have been around since 100 years...)

I have no idea how you've managed to convince yourself that overall supply will stay the same when new, cheaper supply is added. Given elasticity of supply of less than 1.0, net supply will increase (at same/lower prices) and thus Q_d will increase.

As for regressions, I'll just agree with you that they are crap.

As for priors, prejudice, etc. and your claim that you are "just looking at the numbers" whereas I have strong opinions, it seems that you are telling me

"Don't believe your lying eyes."

In some ways, I do not need numbers on ethanol in the same way I don't need numbers on gravity. The political economy of the ethanol program is so obviously corrupt (and the ecology so obviously terrible), that I take my conclusions from the "obvious regression".

[I guess the reason that this is so obvious to me is that i spent a lot of time on US sugar production, and the parallels are 100% correlated...]

Even if we disagree on "obvious," let me try a different tactic: I'm not going to argue with you -- I'll bet you (a la Simon and Erlich): If ethanol subsidies are replaced by a carbon tax (and production externalities are priced), you appear to believe that ethanol production will stay the same or increase. I'll take the other bet (it'll fall).

Given that this result seems hopelessly optimistic, I'll give another version: More evidence will come out that supports my POV OR congress will cancel the program as the economic/ecological disaster that it is. If the program expands AND a majority of evidence supports your POV, then you win.

So -- put down your $20 and let's see what happens in the next congressional session.

... and I'll end by adding to your conclusion "we're not so sure"... **so we shouldn't do anything.**

Philip said...

Boy, (David & Ben) I'd love to see you two on a panel somewhere. Thanks for actually wrestling over a thesis, instead of just trying to holler louder than the next fellow.

HoBs said...

"If the ethanol program is 30 years old, it hasn't been relevant until recently and the targets recently enacted by the Congress are certainly 1) larger than ever before and 2) explicit targets that need to be met."

Sure it has been relevent. People have been campaigning about it at least since Clinton ran the first time. It was 40 cents a gallon in 1978. It is 51 cents today. But it has been around.

The mandates are relatively new, but if you read the legislation (which I helped redraft) the mandates are not for ethanol alone but for *any* renewable fuel. And in fact, fuels like biodiesel and cellulosic ethanol count for extra (2.5x for cellulosic).

(The version we sent to Conress called for making the multiplier proportional to carbon footprint, and adding a safety valve that would effectively act as a tax on gasoline, but that got stripped out by Congress.)

"I have no idea how you've managed to convince yourself"

What's wrong with my math?

How can supply of oil go up if the price of oil goes down? It is really as simple as that. If the oil supply curve (note I said oil, not fuel) does not change, the if price of oil goes down, then no matter what happens to demand, Q_s of oil MUST go down. No?

Q_d of fuel where fuel = oil + ethanol (might increase if the elasticities are different, but more likely to decrease given my math in the previosu post), but Q of oil by itself must go down. This is freshman econ david. Surely we should be able to agree on this.

"political economy of the ethanol program is so obviously corrupt "

Again, the point of economics and science is to quantify these things. Read Ansolahbahere, De figuereido, Snyder runs the regressions, and does not find the corruption you seek. As does whoever tested Grossman-Helpman. Sure they did so in different contexts (ADS on contributions in general, grossman-helpman on protecitonism) but I'd imagine corruption is not as big as you think. Give me citations to back up your claims, or I cannot believe them.

Sure Iowans are getting benefits from the subsidies, just like Alaska gets benefits from federal bridge spending, and New York gets benefits from federal wefare benefits. But I wouldn't call all of that corruption. And I wouldn't say that all welfare and all infrastructure projects are obviously corrupt.

Gravity is a good example. Everyone from Aristotle to whoever was right before gallileo said it was obvious that heavier things fall faster than lighter things (in pretty much those words). They were all wrong.

"If ethanol subsidies are replaced by a carbon tax (and production externalities are priced), you appear to believe that ethanol production will stay the same or increase. I'll take the other bet (it'll fall)."

That's not what I'm saying at all. I'm saying that ethanol subsidies (especially in the absense of a carbon tax) are closer to welfare maximizing than not. (and as for your hope of getting rid of ethanol, Obama took the ethanol pledge. McCain never did, but he lost).

Honestly. a carbon tax that's anywhere close to what people and models suggest (see IPCC estimates for $10 to $20 per ton CO2 for example) would increase gasoline prices by 10-20 cents per gallon. Given that Europe and Japan have had gasoline taxes of $2-$4 per gallon which have done nothing at all to encourage development of oil alternatives.

Look, I've tried to cite and quantify assertions that I make (other than ones that are easy to check or I'm too lazy to remember, or at least give you a sense of where to find the citations). I have yet to see a single citation from you.

Here's a new tack.

Thinking about this more, I actually see taxes as more and more futile. We are trying to spur innovation. In a perfect econ 101 market with perfect intellectual property rights and zero transaction costs, an optiomal carbon tax would be enough. But given that an innovator tends to only capture 0-10% of the surplus from a new invention (especially since in the case of climate change, most benefits occur long after we've died), even with an optimal carbon tax, most innovators wouldn't be willing to pay the costs unless the costs were ridiculously low, which in the case of energy projects (as opposed to software) is never the case. How would you solve this problem?

"**so we shouldn't do anything.**"

Most people argue against any carbon regulation using exactly that logic. I don't entirely disagree. This was why Bush adopted a wait-and-see approach to climate change. There is a compelling theory paper that defends that point of view.

HoBs said...

to clarify, on the citations point. i don't expect you to go out there and look up citations for every line. my point is just that i wouldn't put so much faith in things that appear to be "obviously" true, unless there is at least some empirical evidence to back the claim up.

David Zetland said...

The mandates are relatively new

...and that's all that matters these days.

What's wrong with my math?

There's nothing wrong with your math. I think I was unclear. The reduction in demand for oil in the US *will* reduce global demand, but (my point) not 1:1, i.e., if US demand drops by x bbl, then world demand will drop by less than x bbl, so your 1:1 claim about ethanol displacing oil is not true on a global (relevant to me) level.

Ag and Campaign Contributions

Lopez, R. A. Campaign Contributions and Agricultural Subsidies Economics and Politics, 2001, 13, 257-279

Abstract: This article examines the influence of campaign contributions on agricultural subsidies. Empirical results revealed that rent-seeking works, i.e. contributions, influence agricultural subsidies in the manner they best serve contributors' economic interests. Eliminating campaign contributions would significantly decrease agricultural subsidies, hurt farm groups, benefit consumers and taxpayers, and increase social welfare by approximately $5.5 billion. Although contributions are not the only determinants of agricultural subsidies, investment returns to farm PAC contributors are quite high ($1 in contributions brings about $2,000 in policy transfers). In fact, the results are in sharp contrast to the "truthful contributions" assumption of the Grossman-Helpman model.

I'm saying that ethanol subsidies (especially in the absense of a carbon tax) are closer to welfare maximizing than not.

1) Ethanol doesn't change behavior of consumers (drive same distance, perhaps more, since it's subsidized) and it doesn't help (and may hurt) the environment.

2) I prefer a carbon tax as a means of "leveling the field" among energy sources.

3) Given that ethanol doesn't change behavior, I don't need to show that a carbon tax would, but I *am* in favor of a tax that captures externalities (30-40c/gal) and -- even higher -- to affect behavior.

4) So, yes, I advocate taxes that will change behavior and lower carbon output. That's much more different from what you advocate (using ethanol instead of gas for national security reasons or what?)

Solve the Problem

If there's a tax and someone invents something that delivers energy and avoids the tax, then he can make money on the difference between the cost of the invention and market price. There's no need to capture all the rents (Bill Gates never did), so why is this not a good idea?

don't do anything

no -- I do not support the "Bush Doctrine" (lower taxes and leave big business alone). I am in favor of a carbon tax (and rebate) for the same reason that everyone else is -- the weight of evidence (citation: IPCC) points to carbon (and equivalents) resulting in harmful changes to the atmosphere. It's a simple step to go from harm to Pigouvian tax.

Keep trying to justify ethanol. It's just wrong (or inefficient if you like...)

HoBs said...

"The mandates are relatively new
...and that's all that matters these days"

They really aren't. The mandates were set so low, they aren't even binding, and even in projections were never expected to be binding. In the 2007 Act, they made them likely binding in around 2015, but that's far in the future, and most don't expect the law to still be intat by then.

"There's nothing wrong with your math. I think I was unclear. The reduction in demand for oil in the US *will* reduce global demand, but (my point) not 1:1, "

No, I was unclear. My math says that global demand for oil will change by x E_d / E_s. Given that E_d roughly equals E_s, then it is 1:1 (or close to it)

" It's a simple step to go from harm to Pigouvian tax."

It is simple. But the optimal pigouvian tax is not significantly different from zero (Tol, 2005 in a large meta-analysis). So given your aversion to doing anything when there is uncertainty, then you shoudl agree with the Bush Climate Policy.


As for "Lopez, R. A. Campaign Contributions and Agricultural Subsidies Economics and Politics, 2001"

That is more like it. Though the papers I cite which find negative results are in ReStat and JEL so without reading more carefully, you might think their analysis is more credible. (using journal ranking snootiness as a measure)

But let's do this properly and read it or at least skim it. Skimming it now, one reason why their results aren't credible is they find that $1 in donations yields $2000 in returns. which suggests that people are under-donating. OR that they don't properly solve the endogeneity problem. (that people donate to politicians that have similar values, while the money itself has no real influence) In fact, I'm not sure if they address endogeneity at all. Furthermore, they find about $600 million in welfare costs from corn subsides. Not sure if that is huge, considering that with about 9 billion gallons of ethanol produced last year, and arguably $9 billion of social benefit externality, the $600 million is dwarfed.

"1) Ethanol doesn't change behavior of consumers "

I don't care about behavior of consumers. I want innoation. Behavior of consumers has minimal effect on innovation.

Small and van Dender (2005) estimate that even if you doubled gasoline prices (a $2 tax), you might get a 6% reduction in gasoline. Ethanol has already achieved that and could do more.

"2) I prefer a carbon tax as a means of "leveling the field" among energy sources."

AGain, the mandate also equals the playing field among energy sources. It applies equally to renewable fuels. The subsidy probably does too. And in both cases, other fuels with more carbon reductions get more credit/subsidy. It's not calibrated quite right (as I said, we tried to change the calibration) but it is not too far off.

"3) I *am* in favor of a tax that captures externalities (30-40c/gal)"

What do yo mean? You mean just carbon externalities? Your number is way higher than most experts advocate. If you mean that is all externalities, the current gasoline tax is around 50 cents per gallon. So then you'd be saying gasoline taxes are too high.

"If there's a tax and someone invents something that delivers energy and avoids the tax, then he can make money on the difference between the cost of the invention and market price. There's no need to capture all the rents (Bill Gates never did), so why is this not a good idea?"

Because it is insufficient and inefficient. Someone could have done this for years in Europe and Japan, but no one ever did. Furthermore, there is a huge market failure you are not addressing.

Bill Gates was able to innovate because his innovation costs were low. (it was him in a garage).

Even without US intervention, there are already trillions of dollars of profits to be made for something that can replace oil. I don't think the lure of market profits is enough.

Talking to CEOs, they say there is too much risk, and the costs are too high (test facilities costs billions).

"much more different from what you advocate "

What I advocate is a long term solution that includes solving the problem of climate change for the whole world (China, India, Africa), as well as geo-political problems as well. And just efficiency alone will never achieve 80% GHG reduction by 2050 or whenever when billions of people start headed toward Western standards of living. We need carbon neutral fuels. And that means innovation. And there are huge market failures in innovation that carbon taxes alone cannot solve.

David Zetland said...

Hi Ben -- I'm not reading anymore, but you may want to check out this post on government picking winners

HoBs said...

that guy doesn't know what he's talking about. yeah, the flex-fuel incentives may be wrong headed, but a relatively cheap intervention (we opposed it, but not strongly given that it is cheap and does have option value). It only costs about $100 per car to make it flex-fuel. Cars have a lifespan of up to 30 years. Therefore, it may indeed be useful to have those cars on the road, if 20 years from now (as projections suggest) there will be cheap and green cellulosic ethanol.

Also, agreed, government is bad at picking winners. I'm not saying it should. but it picks winners all the time in basic research. It's called NSF, and billions of dollars allocated by bureaucrats to pay for among other things, your salary.

Not the most efficient, but no one has come up with a better one.

My argument is just maybe, some of these technologies are like basic research, and need government grants.

We shouldn't be picking winners. But that doesn't mean a tax is the only answer. The policies we proposed apply to all alternative fuels equally proportional to their merit.

"I'm not reading anymore"

I will take that to mean, that I win! ;)