Some figures on the economics of corn ethanol

2008-02-03

in Economics, Geek stuff, The environment

Trustworthy numbers on some climate-related things are virtually impossible to find. Key examples are nuclear power and biofuels.

All the more reason, then, to be thankful that some numbers have been crunched over at R^2. Conclusions:

  1. The corn for one gallon of ethanol costs about US$1.30.
  2. Energy for processing costs another $0.33
  3. Enzymes, yeast, and chemicals are $0.14
  4. Labour and other expenses are $0.23
  5. Capital depreciation costs are estimated at $0.40

He thus concludes that corn ethanol costs about $2.00 per gallon, not including return on investment. This is also after you subtract the revenue from selling the distiller’s dried grains with solubles (DDGS) grown with the corn but not used for ethanol production. Due to the lower energy density of ethanol, this is equivalent to gasoline for $3 a gallon.

While I certainly wouldn’t bet the farm on the accuracy of those figures, I think there is reason to put more stock in them than in estimates from journalists (who lack expertise) or governments (who often have conflicts of interest). Of course, the issue of whether corn ethanol is cheap or expensive doesn’t bear upon some other vital questions: Does it actually reduce fossil fuel usage? Does it produce fewer greenhouse gas emissions on a lifecycle basis? Does making it raise food prices and starve the poor?

Report a typo or inaccuracy

{ 10 comments… read them below or add one }

Bill February 4, 2008 at 6:47 am

I don’t see any reason to put more stock into those numbers.

If the cost to produce a gallon of ethanol, mix it with gasoline to form E85, and ship it across the country to a fueling station were ONLY the cost of production as he lists them, then the retail price including taxes would be significantly higher than they are now, indeed higher than gasoline (G100).

Yet, despite that fact, E85 is generally 20-30% cheaper. This is not using the “equivalent by energy” red-herring either, but raw comparison. If G100 sells for $3/gallon (retail), and E85 sells for 2.15 (retail), AND the tax difference is capped at 54 cents/gallon, there are limited explanations.

1) His numbers are way off – on the high end
2) Everyone in the chain is actually selling at a huge loss

Remember that with retail price you HAVE the factors of transportation and profit.

It reminds me of the Doctor Who line “Yes, well it is impossible, but there it is”. It is also similar to arguments I’ve heard that claim it takes two (or more) gallons of gasoline (equivalent) to make a gallon of ethanol. If that were true, then the costs of producing ethanol would be even higher, say double what R^2 says.

That is one of the major pitfalls with playing with “equivalencies”. They don’t pan out very often. Using real numbers, and real use cases has a far better track record.

For example, R^2 uses an “equivalent” base don energy differences, but fails to account for combustion efficiency differences, the most common failing in this field. Anyone versed in the physics knows that burning alcohol is more efficient than burning gasoline; generally 10-15% using today’s flex-fuel technology (which while not “new” is still immature). This explains why those running E85 tend to not suffer the “full loss” that those stuck on “energy equivalencies” claim MUST exist.

In essence if the fuel has 67% of the energy content but the combustion process is 15% more efficient (it gets 15% more usable energy out) you have a net difference of 77% the energy, which is approximately what we are getting in real world usage.

Ultimately if the costs are as outrages as R^2 claims they are, the market will die horribly next year. The amounts of operational money available to most plants is vanishingly small, not likely to be able to support burning through cash that fast for very long.

As to your end questions, your first is too vague. What emissions are you measuring? Without details there can be no “true” answer. In most of what is a health hazard, the answer is affirmative.

For your second question, it has two problems. Does making it increase food prices? Possibly. Does it starve the poor? No. How can the second part be false if the first is true? Because there are many causes of increasing prices, most of which have a far greater impact.

Two major influences on grain cost are oil and world demand. A major factor on the second aspect there is wheat. When wheat has a bad year, corn will have a bumper year barring environmental conditions precluding it. While it is apparently fun to blame ethanol for corn prices, the facts are that a bad year for wheat (weather) caused a large increase in global demand combined with a sharp rise in the price of crude (oil) played the dominant roles in food price increases. The effect that ethanol did have was in fact a minor player by comparison.

In a situation where ethanol was at best a minor player, you can not claim with honesty that it caused “poor to be starved”. Which, by the way has virtually nothing to do with the price of corn. The fact is we produce more food than we can eat, even in America. Globally more food is thrown away than would be required to feed every person on the planet enough to make then fat.

Starving people are primarily a product of bad infrastructure, terrible government policies, and distribution losses. If somehow every ethanol plant ceased operation and all the corn bound for corn-ethanol plants were now “available”, the “starving poor” would not eat any better.

hmmm ..no comment preview. Hope there aren’t many typoes. ;)
/b

russ February 4, 2008 at 9:26 am

“Anyone versed in the physics knows that burning alcohol is more efficient than burning gasoline; generally 10-15% using today’s flex-fuel technology (which while not “new” is still immature). This explains why those running E85 tend to not suffer the “full loss” that those stuck on “energy equivalencies” claim MUST exist.”

The above assertion is most unusual. If one assumes that the combustion efficiency of ethanol is 100%, 10-15% of the gasoline must be discharged from the exhaust system of vehicles. That might be noticeable.

The measured mileage of gasoline- and ethanol- fuelled is remarkably close to that predicted by their relative heats of combustion.

russ February 4, 2008 at 9:30 am

Not omitted word in para 3: Should read “gasoline- and ethanol-fuelled vehicles”

Anon February 4, 2008 at 9:54 am

It’s interesting to see how many sock puppets are out in the comment threads of blogs defending ethanol. I suppose it’s a relatively cheap way to try to shape public opinion on a very problematic fuel that is highly profitable for some people.

Anon February 4, 2008 at 9:57 am

“For example, R^2 uses an “equivalent” base don energy differences, but fails to account for combustion efficiency differences, the most common failing in this field. Anyone versed in the physics knows that burning alcohol is more efficient than burning gasoline; generally 10-15% using today’s flex-fuel technology (which while not “new” is still immature).”

This just isn’t true. Gasoline has an energy density of 46.9 MJ/kg. For ethanol, it is just 30 MJ/kg.

Burn some and test it for yourself.

. February 4, 2008 at 3:48 pm

But that’s a digression. On the subject of using coal as the source of BTUs for ethanol production, there are two things that stand out. First, the current process of using natural gas to produce ethanol makes little sense, since you can use natural gas directly in a CNG vehicle. You gain little or nothing by turning a BTU of natural gas into a BTU of ethanol (plus some animal feed). However, coal can’t be used directly as automotive fuel, so one can make the argument of upgrading the quality of the energy source by turning some of coal’s BTUs into ethanol.

Tristan February 4, 2008 at 5:31 pm

“Return on investment” is the surplus value of labour – its the fact that when someone makes something, that thing is worth more than the time they put into it – its worth the amount of working time that thing can produce in another worker.

For example, if I grow food, the value of my labour is not the amount of labour I put into growing the food, but the amount of labour the food can produce by feeding workers.

You can say this is “marxist drivel” all you want, the fact is, if labour didn’t add more value to things than the labour cost, we wouldn’t have an economy, and we wouldn’t have any profit that wasn’t just “you managed to sell something for more than it was worth” – which is no more likely than selling something for less than its worth.

Of course, labour doesn’t always add value – but capitalists that don’t manage to hire labourers that add value to their products in general fail as capitalists.

This doesn’t change the analysis of corn oil – it merely says that pretrol made from corn has a surplus labour value (return on investment) equal to the amount per liter that the petrol can reproduce labour (i.e. people must drive to work). If the cost exceeds the value, this will be reflected in the market forces that prohibit the production of corn petrol except for by government subsidy (which of course, distorts the market by adding value arbitrarily in the form of cash infusion).

tristan February 4, 2008 at 5:36 pm

There is one other consideration for corn oil – it is very high in octane, which makes it excellent for high performance cars. It may be profitable for small firms to grow corn to make racing fuel. Engines can produce on average 20 to 30 percent more power on E85 than on regular petrol, because they can be built with higher compression engines.

This is a real consideration in the sense that it means we could build cars with smaller engines, which would have less moving parts, and convert more of the heat energy there is in the petrol into mechanical force. A higher compression engine, ceteris peribus, is always more efficient than a lower compression one. (by the way, ceterus paribus rarely exists in the car world – in the 70s when fuel economy legislation went into force compression ratios went down because at the same time emmisions regulations for NOX went into force – and higher compression engines produce much greater amounts of NOX. However, as soon as companies got their act together building excellent catyletic converters that convert NOX into less scary chemicals, compression ratios have gone back up – and this is another reason why a 3.5 liter Nissan midsize family saloon is considerably faster than a 1987 Ferrari 328, despite weighing more and having the same sized engine.

Raul February 5, 2008 at 2:48 pm

“Everyone in the chain is actually selling at a huge loss”

That’s not necessarily the case. If it costs $2 to make it, and they are currently selling for $2.15, it just means they aren’t making as much as they used to. And we know that’s the case, which is why a number of producers have stopped projects. Think again about what you wrote, and you will see why it is wrong.

“Ultimately if the costs are as outrages as R^2 claims they are, the market will die horribly next year.”

You have seen the stock prices, no? They have been dieing pretty horribly. They won’t die completely, because of the mandates. But it will be tough to make a consistent living at producing ethanol.

“For example, R^2 uses an “equivalent” base don energy differences”

No, he didn’t. He made a comment on that at the end, but his analysis was based purely on the ethanol inputs. It had nothing to do with energy differences. And you really can’t argue with the numbers. It isn’t rocket science. You know how much corn it takes to make a gallon of ethanol. You know how much corn sells for. You know how much DDGS sells for. The rest of the numbers came out of an ethanol producers spreadsheet! The only difference was that he plugged in the current natural gas price.

. September 9, 2009 at 11:25 am

Corn Ethanol

A niche, you say? Aren’t we producing 10 billion gallons of corn ethanol already? True, but I am talking about something that could actually stand on its own in the long run – unsubsidized – and still make a decent net contribution to our energy supplies. In that case, producers might still be able to sell 10-15 billion gallons of ethanol a year and make a profit, but the distribution pattern would be different. In a state with ample rainfall and rich soil, corn ethanol may be able to stand unsubsidized by making and consuming the ethanol locally. Corn ethanol may be a fine solution for Iowa (although E85 is not even cornering the market in Iowa, where it should be in its optimal market). Stretching it beyond a local solution is where the economics start to break down and the scheme only works with subsidies.

Here are some examples of what I am talking about. When corn ethanol is produced far from corn supplies – like in California – the economics became difficult due to the cost of shipping the corn to the plant. I talked about that in 2006, when I warned of the potential problems of Pacific Ethanol’s plans to do just that. They filed for bankruptcy earlier this year.

Another example is when ethanol is produced from a state in which ethanol’s energy balance is poor (e.g., parts of Nebraska, due to corn’s irrigation requirements) and then shipped to California. If you look at the USDA’s most recent paper on corn ethanol’s energy balance (the one in which they used creative accounting), you can see from Table 2 that Nebraska’s energy inputs for growing corn are about 20,000 BTU/bushel above the Midwest average. (By comparison, Iowa’s are 11,000 BTU/bushel under the Midwest average). This has the overall impact of actually causing Nebraska’s net energy from producing ethanol to be negative unless one adds a BTU credit for co-products. With such a marginal energy balance (and I haven’t even mentioned the Ogallala Aquifer) it hardly makes sense to produce ethanol in the drier regions of Nebraska. It makes even less sense to then spend more energy shipping that ethanol far from the point of origin.

Leave a Comment

You can use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Previous post:

Next post: