July 22, 2007
Ethanol and food price volatility
If this is what we get in a good year, what will happen when we have a bad crop?
American consumers are starting to see some of the consequences of our ill -fated ethanol policy in the prices of everything from meat to ice cream. While well-fed Americans may gripe, the implications for those in sub-Saharan Africa are quite alarming.
All of these concerns arise from the higher average price of corn that necessarily results from an increase in the use of the corn crop for ethanol production. But another issue well worth considering is the effect on the volatility of corn prices.
Food prices naturally are quite volatile because unpredictable and uncontrollable variation in weather can produce a bumper crop one year and a big shortfall the next. Usually consumers are able to mitigate somewhat the consequences of the volatility of supply by switching between foods depending on what is most cheap or expensive at the moment. However, whereas the demand for food is relatively price elastic, the demand for gasoline is quite inelastic. If the quantity of ethanol demanded does not fall much when there is a bad crop, the quantity of corn used for all other purposes must make an even bigger proportional adjustment. For example, if 1/2 of our corn crop were devoted to ethanol production and ethanol demand were completely price inelastic, a 10% reduction in corn production would require a 20% reduction in use of corn for other purposes.
A recent analysis by University of Illinois Professors Darrel Good and Scott Irwin notes that over the last half-century, corn-production shortfalls as big as 30% are not that uncommon. Very inelastic demand means that having a stable, reliable source for fuel is a very high priority for consumers. Having the supply for such a commodity depend on something as volatile as U.S. corn production does not seem like such a brilliant idea.
Now this raises a potentially big problem for the U.S. Congress, and I'm not sure how they're going to solve it. Specifically, the question they might want to be mulling over now is, When the next 20% shortfall comes, at whom will they point the finger of blame?
Posted by James Hamilton at July 22, 2007 06:48 AMdigg this | reddit
This year's corn crop will be huge. That's partly because the weather has been pretty good with last week's timely rain, but it's also true because corn has replaced soybean acres (higher tofu prices for my vegan daughter), and even marginal & fallow land has been put under the till. Fencepost to fencepost.
Here's one outlook for corn.
And for soybeans.
But is this level of production sustainable? I kind of doubt it.
Posted by: 2slugbaits at July 22, 2007 12:42 PM
Very nice economics lesson, Professor, about the impact of varying elasticities of demand on one commodity, and the potential implications. Elegant example.
What a shame, too, given how lots of poor folks (e.g., Mexicans) depend upon corn and given the iffy/mixed points about ethanol in regard to air quality.
Posted by: jg at July 22, 2007 12:56 PM
It's a shame. It's not a reliable source and it squeezes out the food. I wonder if on the USDA's web site's pro-ethanol section they still claim ethanol won't affect food prices. They used to.
As for the soy, it's not just tofu. My baby nephew gets fed soy milk more often than not since it doesn't seem to upset his stomach as much. I guess mom and dad will be having a bit less money for college for him if they keep doing that.
Posted by: Allen at July 22, 2007 01:34 PM
Yes, the corn crop and total supply is expected to jump this year according to USDA's July 12 projections. But supply is also projected to grow 1.6% more to 14.2 bill bushels by crop year 2008/09. We produced 4.9 bil gallons of ethanol in 2006 and the forecast is for 12.6 billion in 2008/09(2.75 gal/bu) -- an increase of 157%. At that point, ethanol would use 36% of the corn crop/supply, with projected feed use dropping more than 6.3%. Something will probably have to give here, and I think it will be ethanol production. The great majority of it (75%+) moves in rail tank cars, and there is likely going to be a shortfall - carbuilder order backlog has stretched to about 2years. Beyond that,blender facilities at destinations are not up to handling the volume, at least at this point. BTW the blenders are the guys who get the 51 cents/gal subsidy.
Posted by: wogie at July 22, 2007 03:50 PM
Good to see you roam here as well.
In addition to the transportation constraints, there will also be some severe storage constraints as well. And before that corn can be stored, it has to be dried, which will put even more pressure on natural gas reserves.
As I recall the old "cobweb" model is frequently applied to agricultural commodities. The general rule is that the stability condition requires the demand curve be flatter (more elastic) than the supply curve. Prof. Hamilton's charts don't exactly tell us what the elasticity of supply is for corn, but the extreme volatility suggests that an unstable supply curve may have the same effect on the stability of the cobweb model as a flat supply curve coupled with an inelastic demand curve.
Posted by: 2slugbaits at July 22, 2007 04:34 PM
Ethanol, this creature of high oil prices, the sinister idea of converting food into fuel, is another example of bad investment. But governments are giving subsidies from ordinary tax payers to corn producers, to make the gas cheaper, making the food more expensive. "Bread for today, starvation for tomorrow". It's another example of privatization of earnings and socialization of costs. You know who the decision makers are... The same who make money from Iraq war... And more o less the same that planned the subprime business...
Posted by: jla at July 22, 2007 04:55 PM
It seems like this might happen about once ...after that there would be either a technical or supply solution:
- Technical solution is simply to allow 2 sources of power for cars. Many Hybrids have this allready ...they plug into the wall at night.
- Supply solution is simply to build up reserves in bumper crop years. We have an oil reserve ...why not ethanol?
Conclusion - Having more sources of energy makes it less likely that there will be supply issues. The fact it cuts into the food supply though leaves me wondering how many millions of 3rd world people will starve the first time there is a shortage. :(
Posted by: NoFate at July 22, 2007 05:19 PM
Well, one benefit of the increasing price of corn is that many of my favorite juice beverages are replacing high fructose corn syrup with pure cane sugar. I swore off these drinks years ago because of the hfcs. Sugar tastes so much better, less sweet than hfcs, and is better for you to boot.
Posted by: Thinker at July 22, 2007 05:41 PM
It is nice to see the realization begin that US ethanol policy will affect world food prices, not just US. It may hurt here, but it will be horrible elsewhere.
It will also affect the value of the dollar (strengthen it) and change the balance now in place between oil exporters and food exporters.
There will be some dizzying realignments as this happens.
Posted by: wally at July 22, 2007 06:45 PM
Food uses of corn (corn, cornflour, corn syrup) are trivial in the U.S. with respect to production and stocks. About 3% of U.S. corn production goes to such uses. The same is true of soybeans. The effect of increased corn ethanol production and the natural volatility of agricutural field crop production will primarily have the following impacts: It will raise corn prices (thereby reducing the need for price floor intervention). It will displace soybean and cotton production in the U.S., raising prices for those commodities. It will reduce U.S. export "dumping" of cheap, subsidized corn, soybeans, and cotton on the world market. It will increase production of corn, soybeans, and cotton in the rest of the world, allowing 3rd world farmers to make a living. It will increase feed costs for U.S. livestock producers, causing feed "switching" and increasing the cost of production of meat, dairy, and eggs. Note that ethanol production from corn destroys about 2/3rds of the feed value of the corn (not 100%) thus ethanol displacement of feed uses will not be as large as might be understood from the projected fraction of corn used for ethanol. It will improve the economics of grass-fed beef in the U.S., helping small cow-calf range-fed beef operations at the expense of feedlot operators. It will displace imported oil, thus offsetting trade balance effects of reduced ag exports. It will utilize large quantities of natural gas, increasing the price of that commodity in the U.S., increasing imports from Canada, and increasing the market pressure for LNG imports from overseas. It will reduce the margins of existing corn ethanol operations, due to increased feedstock and fuel costs, and increased ethanol supply, increasing the market pressure to substitute other sources of process heat (corn stovers, etc) for natural gas. It will increase rail volumes of ethanol, leading to upward pressure on bulk transportation costs and increased rail infrastructure expenditure.
Some of these economic effects will create political effects. Better off U.S. commodity farmers and reduced ag subsidy costs in the U.S. may lead to decreased opposition to removal of ethanol import restrictions, given protection by quota of existing ethanol operations. Increased food and natural gas prices/volatility may lead to government creation of effective Strategic grain and natural gas reserves and to federal overrides of local interests blocking LNG gasification facilities.
Posted by: benamery21 at July 22, 2007 07:53 PM
"It will utilize large quantities of natural gas"...has anyone done a detailed look at this? Is the natural gas consumption of ethanol plants significant given the huge total quantity of nat gas used in this country? Will it become significant with the construction of additional ethanol plants, or will we run out of corn before gas consumption reaches levels that will affect the overall market?
Posted by: david foster at July 22, 2007 09:32 PM
Recently we have seen more clearly than ever how congress ignores the majority for the short term benefits of buying votes. We all know that ethanol policy has nothing to do with energy and we also recognize it borders on insane to burn up our food supply while people in the world (Zimbabwe) are bordering on starvation (of course this is another political issue for another thread).
The US is so prosperous that US politicians can thumb their noses at the starving in the world to sit fat and happy in Washington DC dealing out largess to buy farm votes. If I spend too much time thinking about it I get physically ill. And they criticize big business. Give me a break!
Posted by: DickF at July 23, 2007 05:19 AM
"When the next 20% shortfall comes, at whom will they point the finger of blame?" I think the answer is the free market in general with the policy prescription of price controls.
I assume that Brazil does not have a similar issue with regard to ethanol made from sugar cane due to climate?
Posted by: GWG at July 23, 2007 06:13 AM
Here is how the 11.575 billion bushels 2006-07 U.S. corn consumption break down, according to the Good and Irwin paper I cited:
Feed and Residual: 5.85 billion bushels (50.5% of total)
Exports: 2.2 billion bushels (19.0% of total)
Ethanol: 2.125 billion bushels (18.6% of total)
Other: 1.375 billion bushels (11.9% of total)
It's certainly correct that feed is by far the most important single use, but your statement "Food uses of corn (corn, cornflour, corn syrup) are trivial in the U.S." might leave the mistaken impression that the feed component does not matter for the price American consumers pay for food. The examples I give in the opening paragraph are related to meat and dairy costs. Nor should exports be thought of as something that's not showing up in someone's food costs. Decreased acreage of non-corn production also affects food costs.
Good and Irwin calculate that a 1.25 billion bushel increase in corn for ethanol would substitute for 150 million bushels of corn for feeding, a 12% replacement ratio rather than the 33% figure you cite.
Posted by: JDH at July 23, 2007 07:20 AM
Since beef and pork are sometimes referred to as "grain on the hoof" it's clear that higher corn prices translate directly into higher protein food prices.
And I guess if we're listing all the negatives of ethanol subsidies, we might as well note the not insignificant societal costs of American sugar tariff-rate import quotas - a program based on what I once saw Senator Bill Bradley refer to as "The Caribbean Impoverishment Act" . . . . . . .
Posted by: Anarchus at July 23, 2007 08:36 AM
I can't find data on expected production, but Brazilian sugar cane production data can be found here:
There is a small dip in production(smaller than 4 percent) in 2000, which is due to a draught in the south-east region of the country. That draught was so severe that it lead to the energy crisis in 2001.
Posted by: cerqueira at July 23, 2007 10:33 AM
benamery may be right that these higher prices will be good for foreign farmers and even have some positive effects elsewhere through higher incomes for farmers. But there have already been major problems in Mexico due to the substantially higher prices for tortillas, which are the base of the diet for the very poorest people.
Ethanol is not all that good of a deal in terms of net energy, given the natural gas requirements for generating it from maize. It really is unfortunate that Iowa plays such an important role in the US presidential selection process, the highest maize producing state in the US...
Posted by: Barkley Rosser at July 23, 2007 02:16 PM
excellent post prof.
Posted by: sa at July 23, 2007 03:35 PM
Very interesting. I wonder what it would take to get Bill O'Reilly to interview you? He and his core audience need several reality checks. He seems to think that if we dropped tariffs on Brazilian ethanol that all of our oil problems would be solved. I guess he has never thought what the Brazilians would think if we suddenly drove up the price of their alternative? I think Mexico and Brazil might join forces and repel the Yangui invaders.
Posted by: Eric H at July 23, 2007 06:19 PM
Despite the difference in opinion about residual effects between Ben Amery and JDH, I think Ben's post shows just how complex and interwoven is any scenario analysis that apparently starts with a single thread of analysis. It's nearly impossible to determine the overall outcome because of the secondary, tertiary, quaternary and I-don't-know-what-the-next-adjective-is effects.
However, the food vs fuel problem is really an just an outcome of the real problem which is that the EROEI on ethanol is much too close to 1.000 to be any sort of realistic substitute for petroleum based fuels.
Posted by: IntoTheBlack at July 23, 2007 08:03 PM
JDH, et al--I think we are agreed that big changes in U.S. corn ethanol production have the potential to cause significant market dislocation.
However, the ag brief you cite for replacement is attempting to provide guidance on net replacement levels. The 12% replacement rate you cite is after assuming that 1/4 of the byproduct goes to export, which would take us to a 16% replacement rate. My 33% number is based on the actual process outputs, i.e. if you put 1000 bushels of 56lb/bu corn into a distillery you get about 8-9 tons of DDGS (or equivalent amounts of other coproducts, depending on process) out. This DDGS has both a higher energy content by weight and a significantly higher nutritive value (higher protein levels), than the input corn. I do not claim to know what disposal channel all of this byproduct will take. It appears that your source is assuming some other disposal channel for 1/2 of it. I guarantee that they won't be landfilling it with DDGS feed prices over $100/ton, however, there may be more lucrative market options than feed.
Regarding the affect of cheap corn exports on the international food market, I believe that people in largely agricultural poor countries will earn more money to buy higher priced food if they have less subsidized competition from U.S. corn, soybeans and cotton (soybeans and cotton are the chief crops being displaced by increased U.S. corn production). Obviously there are individual winners and losers, but raising international corn prices above the cost of production is a huge net gain for countries that have very little in the way of factors of production besides labor and land. I fully expect SIGNIFICANT increases in world ag production, driven by higher prices. Brazil is doing something similar with sugar: where they have historically dumped massive quantities on the international market, resulting in prices below the cost of production for most other countries, but are now turning some of it into ethanol and letting other sugar producers make some money.
Regarding Mexican tortilla prices, Mexico is a significant net importer of maize, despite the cultural importance of the crop. This is due to abysmal yields, and the need for agricultural modernization. Mexico has ~25% as much land area in corn as the U.S. but produces only ~7-8% as much corn. If the political pressure created by higher corn prices can create the will to modernize (or higher corn prices the capital to do so) than this is all to the good. Soil acidity is probably the biggest factor in low Mexican yields, the fix for that is different strains of corn (which have shown 60% improvements in yields from typical yields on typical soils), and the application of lime or manure (which have shown 200% increases in yields). This is not rocket science. To acheive "maize independence" Mexico needs only about a 33% improvement in yields (which would still put them behind most of the world).
Regarding corn ethanol impacts on natural gas: roughly 2/3 as much natural gas energy is required by the typical U.S. distillation process as is produced as motor fuel. At the estimate of 12.6B gal of ethanol in 2008/9 that is in the neighborhood of 6-7% of U.S. natural gas consumption--definitely not trivial.
Natural gas is not required to make ethanol from corn, just process heat. This is one of the big differences in U.S./Brazilian ethanol production. Brazil does have a superior feedstock in sugarcane but the use of bagasse for process heat is a main factor in substantially raising the net energy value of their ethanol. Burning corn stovers and other ag trash would have the same impact here. Broin (now POET) does this at some plants.
Posted by: Anonymous at July 23, 2007 08:41 PM
Oops, that annonymous post above is mine.
Posted by: benamery21 at July 23, 2007 08:44 PM
benamery21 ---> It's more than 2/3 of the value that gets destroyed. You're sapping the calories out of it which is the main reason for feeding the animals. More so, as ethanol production gets more effecient, that ratio will further decrease.
I also reject that higher prices are "good". We're talking about a commodity. They should be able to offset low prices with volume and effeciences.
Posted by: Allen at July 24, 2007 06:44 AM
I understand the DDGS feed byproduct is only useful feed for ruminants, like cattle. Consequently hogs and, I believe, broilers will not be a market.
Also, don't know where it stands now, but last year the Brazillian government restricted the amount of sugar cane that could be used for ethanol. Presumably, protecting sugar exports.
Posted by: wogie1 at July 24, 2007 09:23 AM
Allen: Basically, the ethanol process converts carbohydrate energy, the protein, fat, minerals, fiber, and vitamins pass thru. Only 8% of feed corn dry matter is protein, typical DDGS is 28%. This is a highly nutritive feed. I am somewhat conversant with dairy nutrition, believe me when I tell you that DDGS is worth more ($ and ME kcal/lb) lb/lb than the input corn, and that the approximate ratio is 2/3rd of input feed calorie value destroyed.
Wogie: Hi!, I had the same opinion regarding nutritive value for non-ruminants (the use I'm familiar with is dairy cattle rations), and 85% is fed to cattle, but I recently saw information that the ME for swine is actually higher than for cattle in 2 separate studies. Didn't make sense and I haven't had time to look into it. I do know that ration fractions are held substantially lower in swine than in cattle.
Posted by: benamery21 at July 24, 2007 05:14 PM
Wogie: My understanding is that the Brazilian gov't controls the world price of sugar (a higher value use of sugarcane than fuel ethanol) by setting quotas (the Saudis of Sugar). They are expanding sugarcane production far more rapidly than the sugar market would allow. They have future cropland available for production of about 6mbd of ethanol.
Posted by: Anonymous at July 24, 2007 05:21 PM
Allen: You are talking about a "commodity" which has become fungible with another commodity. Increased fungibility improves the efficiency of the economy. This understanding may affect your analysis somewhat. Higher prices are a result of higher demand and will necessarily result in negative cost pressure on the 2nd commodity.
What you are doing is taking rents from Saudis and Russian oil producers and multi-national oil refiners and providing them to landowners, farmers, taxpayers(by reducing farm subsidies), and drivers. U.S. food and natural gas prices will see upward pressure, but motor fuel prices will see downward pressure.
Posted by: benamery21 at July 24, 2007 05:35 PM
I've been tracking my gas mileage using ethanol versus regular. Where I live the price differential is 10 cents. I've found that my mileage drops about 3 percent when I run ethanol. I drive a 2005 Honda Civic. Where I live the price differential is such that ethanol is 10 cents a gallon cheaper. So basically the break point is when gas goes above $3 per gallon, it makes more sense to pay the extra 10 cents at the pump and get the additional 3 percent mileage gain. What I don't understand is why the gap between ethanol and regular seems to be fixed at 10 cents irrespective of the base price. The differential was 10 cents back when gas was $2 per gallon and it's still 10 cents with gas well north of $3 dollars.
Posted by: 2slugbaits at July 25, 2007 04:06 PM
Very enlightening writings. I would readily agree that higher ag prices are positive for a majority of the world population.
Posted by: HZ at July 25, 2007 05:58 PM
Hi! When you say "ethanol" what fraction are you talking about? E10, E85, other? I would say that 3% mpg differential with a single vehicle study is well within the margin of error, I see a lot wider variation between tanks of (presumably) identical fuel than that, as I suspect you also do. I would also say that, given the energy per volume content of ethanol, you should expect to see reduced mpg with ethanol. Also, vehicle tuning definitely has something to so with this, flex-fuel vehicles are currently optimized to run on gas, not ethanol.
Posted by: benamery21 at July 25, 2007 11:21 PM
Regarding pricing differential, I presume you have noticed the same thing with regard to premium fuels? Just another indication that retail fuel prices are not set by an efficient market.
Posted by: benamery21 at July 25, 2007 11:23 PM
Speaking as a retired farmer (and as one who has actually grown switchgrass as a commercial cash crop), I just want to note that most discussion of ethanol and especially corn ethanol misses the point of why we have it and where it will go in the future. Corn ethanol was developed as a way to deal with surplus corn after the Nixon administration essentially removed production controls and made us dependent on exports. We had a lot of corn sitting on the ground in piles outside grain elevators. For four centuries, all over the world, commodity prices have been on a steady downhill slide. Today's so-called high prices are lower in constant dollars than they were in the early 50s when families were leaving the farms in droves because they couldn't make a living. Today's demand for ethanol is being driven primarily by the need for an oxygenator to replace MTBE in gasoline. Corn ethanol is a temporary phenomenon. We in farm country understand this. Remember, we're not "corn growers." We're farmers producing cash crops and presently we happen to prefer corn because it has a good cash return. We're not married to corn or to corn ethanol. We control the land and what we chose to grow. We are very interested in what we call the "Third Crop Movement," that is, energy crops in addition to plant and animal crops. When cellulosic liquid fuels become viable, you will discover that you will be begging us to grow corn, since the advantages to farmers of growing perennial biomass crops are so huge, from the perspectives of logistics, capital needs, machinery, cash flow and time, that we won't want to grow corn. Energy crops have the potential of giving us three paychecks: the crop itself, carbon sequestration credits, and wildlife benefits -- all at less risk to us than annual cereal grain crops. We are on the cusp of massive change.
Posted by: Fred Schumacher at August 6, 2007 12:42 PM