This is the concluding post in a series looking at the impact of increased ethanol production on petroleum imports. Previous posts concluded that there has been little measurable impact on our petroleum imports as a result of increased ethanol production. In this post, I provide a spreadsheet to all the data and graphics used, and delve a bit deeper into the issue.
Previous posts in the series were:
The spreadsheet that was used to tabulate all of this information is archived here:
(For some reason the graphs don’t show up in the Google Documents link. However the data and calculations are all there).
One of the most frequently cited reasons for our U.S. ethanol policy is that it will reduce our dependence on foreign oil. Some of the more audacious claims actually suggest that one barrel of ethanol will displace more than one barrel of foreign oil. Here is a sampling of some of the claims. From the Renewable Fuels Association’s (RFA) “Energy Facts”:
FACT: The production and use of 9 billion gallons of ethanol in 2008 displaced the need for 321.4 million barrels of oil. It also saved American consumers and taxpayers $32 billion, an average of more than $87 million a day. This is the equivalent of eliminating oil imports from Venezuela for 10 months, or looked at another way, it would mean that the U.S. would not have to import ANY oil for 33 days.
The RFA’s page on industry statistics shows that ethanol production in 2006 was 9 billion gallons, which is 214 million barrels. Once refined, a barrel of oil will turn into products with an average BTU value of 126,000 BTUs/gal, versus 76,000 BTUs/gal for ethanol; therefore 214 million barrels of ethanol contain the BTU equivalent of 129 million barrels of oil. (Source: ORNL). The claim then is that ethanol with an energy equivalent of 129 million barrels of oil (BOE) displaced more than twice that much oil – 321 million barrels!
The RFA’s source on that was the consulting firm LECG, where director John M. Urbanchuk consults for the Renewable Fuels Association and the National Corn Growers Association. Thus, Urbanchuk is expected to spin a positive ethanol story, but one would hope he could do so without completely sacrificing his credibility. He has also been quoted:
The production of nearly five billion gallons of ethanol means that the U.S. needed to import 206 million fewer barrels of oil in 2006, valued at $11.2 billion. This is money that stayed in the American economy.
Source: Contribution of the Ethanol Industry to the Economy of the United States in 2006 (PDF download)
Even grander claims have been made by the U.S. Government. From DOE Assistant Secretary Alexander Karsner’s keynote address to the RFA’s National Ethanol Conference (link now dead) in Tucson, Arizona:
Last year, we contributed something on the order of a displacing 500 million barrels of oil, oil that we didn’t have to import from regimes that are hostile to our interest or might leverage energy economics over our future.
Over 6 billion gallons of ethanol were produced in the United States last year, and we have an additional 5 billion gallons of refining capacity under construction.
That effort means 500 million fewer barrels of oil that we have to import from the Middle East.
That’s from the U.S. Department of Energy. Those are pretty bold claims. How on earth are people coming up with these numbers? More importantly, can we go to the data and actually see this impact?
Probing the Data
The import situation is complicated by several factors, the biggest of which is the rapid run-up in petroleum prices over the past few years. The increase in prices caused overall demand to fall, which can be seen in Figure 1 below:
It is important to note that “demand” includes all crude oil, natural gas liquids (ethane, propane, butane, etc.), ethanol, fuel gas (offgas from the refinery used as fuel or feedstock), and asphalt. (See the full list of products covered here). This is important to understand, because if ethanol displaces petroleum, it has no impact on overall demand – since it is already included. What you would see in that case is merely a shift between ethanol and gasoline, for instance, with total demand remaining constant (actually it would have to go up a little due to ethanol’s lower BTU content).
The conclusion one draws is also influenced by the time period over which one looks. In the first post in this series, I looked at imports, demand, and ethanol production over the time period 2002 through 2007. The reason for choosing that particular time period was that this was when ethanol was ramping up sharply.
I left off 2008 because of the very sharp drop in demand due to the recession. However, as one reader pointed out, since ethanol is included in the demand number, it doesn’t really matter whether demand went up, down, or stayed constant. If ethanol is displacing imports, we should see that effect even if demand drops sharply. For example, if demand fell by 1 million barrels a day, then all else being equal I would expect imports to fall by 1 million barrels a day. Now add in expanding ethanol production, and I expect imports to fall by more than 1 million barrels a day.
What I observed was that between 2002 and the end of 2007, our petroleum imports do not appear to have been impacted at all by the increase in ethanol production. But that time period is complicated by a couple of things. First, the largest increase in ethanol production took place in 2008. Thus, the largest impact would be expected to show up in 2008 – a year I left off because of the recession effect.
Second, the phase-0ut of methyl-tertiary-butyl-ether (MTBE) took place during this time. I went into detail on how this would have impacted the issue in the second post in this series. The bottom line was that even when MTBE was taken into account, it still did not appear that ethanol production had a measurable impact on petroleum imports.
However, the MTBE phase-out was completed in the first half of 2006. So for the rest of this post, I want to focus on 2007 and 2008. (And as I write this, I don’t know what the answer is; I will work it out as I put the rest of this post together).
During 2007 and 2008, total demand fell by 434 million barrels. Domestic production fell by 74 million barrels. (You can see all of the data in this spreadsheet; there are comments indicating where different data originated). So then all else being equal, I would expect imports to fall by 434 million barrels, but then they also need to make up for the 74 million barrel domestic production deficit. That modifies the expected import change to (-434 million + 74 million) = – 360 million barrels.
Over that two-year time period, net imports actually fell by 466 million barrels. This is the first time period I have looked at over which the import change was less than the demand change, which is what I would expect to see if ethanol was displacing imports. The change certainly isn’t the often exaggerated 200 million or 500 million barrels, but over the course of 2007 and 2008 imports did fall by 106 million more barrels (53 million barrels per year) than would be expected on the basis of demand and domestic production changes. Over the longer time frame of 2002 through 2008, the cumulative increase in imports (+207 million barrels) is very close to what would be expected based on changes in demand and domestic production (-225 million barrels), still implying no measurable impact from ethanol.
How much ethanol was produced over that period of time? Per the RFA’s ethanol statistics, a total of 15.5 billion gallons of ethanol was produced in 2007 and 2008, which amounts to 369 million barrels. On an energy equivalent basis, this is equal to about 215 million barrels of finished petroleum products. Yet the measured fall in imports was less than half that value.
One of the problems here is that we may be looking for a needle in a haystack. By that, I mean that the contribution of ethanol is so small relative to that of overall demand, that any actual displaced imports would be lost in the noise. Figure 2 illustrates:
For this graphic, I have put ethanol production on the same scale as total demand to show the relative contribution. The production for ethanol in 2008 amounted to 0.59 million barrels per day of a total demand of 19.5 million barrels per day. For people who claim that the oil companies are threatened by the ethanol companies, that graphic puts things in perspective.
One could argue that the ethanol impact should show up most strongly in a comparison with gasoline demand. Figure 3 shows that effect:
In fact, gasoline demand* did dip in 2008 by 300,000 bpd. Ethanol may have been part of the reason, but the increase in ethanol production was quite a bit less than the fall in gasoline demand. Corrected for energy content, the ethanol increase was less than half the drop in gasoline demand (which can be mostly explained by higher prices and recession, as shown below).
One thing Figures 2 and 3 show is the dip in demand in 2008, which followed a flattening of demand for a few years prior. Recall that since ethanol is included in the demand number, ethanol can’t be a cause of the drop in demand. Figure 4 shows part of the culprit:
As crude prices began to climb in 2004, crude demand flattened. As the price skyrocketed in 2008, we were also entering a recession. The combination caused a sharp drop in demand. One interesting thing to consider is that since ethanol is mandated in increasing volumes each year, it is not impacted by the drop in demand. While total demand fell by 1.2 million bpd in 2008 relative to 2009, “demand” for ethanol actually increased by nearly 200,000 bpd – because the mandated increase has no allowance for overall drops in demand.
What to conclude from this exercise? The easiest conclusion is that the claims of petroleum import displacement have been at a minimum grossly exaggerated. It may even be that ethanol hasn’t backed any petroleum imports out, or that the impact is so small as to be unnoticeable.
All of these conclusions, however, point toward a common theme: Even our biggest source of alternative fuel is taking very little bite out of our petroleum consumption. Much more effective has been high prices and recession. In fact, I believe it unlikely that any combination of biofuels will ever replace even 50% (net) of our present petroleum consumption. That points toward the need for conservation as a critical component of any major effort to wean off of fossil fuels. Perhaps some combination of conservation, electrification, mass transit, and biofuels can make a significant impact on our fossil fuel consumption. But the graphics above should demonstrate that it isn’t a trivial matter to significantly impact our petroleum consumption.
*Total gasoline demand contains the ethanol contribution. Therefore, Figure 3 shows gasoline after subtracting out the ethanol volumes.
Special thanks to the Energy Information Administration for answering some of my questions about the data.