In an article in today's Washington Post an official of the National Wildlife Federation was quoted linking rejection of the Keystone XL pipeline with breaking our addiction to oil. Even with the administration having delayed its decision on the project until 2013--quite possibly killing it--this point merits further exploration. Just how might we go about breaking that "addiction", and when could we reasonably expect the task to be accomplished? As with everything else to do with energy, the answers to those questions must be based on facts and figures, rather than wishfulness.

The brief quote and its context imply that a decision to forgo additional supplies of oil from Canada or any other source would, by itself, move us significantly closer to breaking our addiction to oil, a rather vague phrase brought into common usage by President Bush's 2006 State of the Union address. Of course if delaying or rejecting the pipeline only results in continued or additional oil imports from other countries, that would be counterproductive from an energy security standpoint, and perhaps even from an environmental perspective. Ending our oil addiction requires more than just a real or artificial supply constraint; it calls for enormous quantities of energy from other sources, mainly for transportation, along with significant improvements in the efficiency with which we use that energy. How soon should we expect such a transformation?

Start with electric vehicles, which are essentially the only pathway by which renewable electricity sources like wind, solar and geothermal power would have any impact on our oil consumption, because less than 1% of US electricity is now generated from oil. Even if EVs turn out to be the long-term solution to our transportation needs, as I suspect, it will be many years before they can displace enough fuel demand to make a dent in our oil addiction. The current goal is to have a million EVs on the road by 2015. As ambitious as that target seems compared to current sales of less expensive hybrid cars, that would constitute just 0.4% of the 238 million cars and light trucks in the US as of 2008. Moreover, even if EVs replaced cars of only average efficiency, one million of them would displace just 31,000 barrels per day of gasoline. In other words, it would take more than 20 million EVs to save the volume of oil that the Keystone Pipeline could have delivered annually.

If we want to kick our oil habit quicker than by waiting for a hundred million EVs to turn up, we'll need an energy source that's compatible with the vast majority of existing cars, and the ones like them that will probably dominate new car sales for some time. Consider ethanol, our largest and most successful alternative energy initiative so far. Through August, ethanol accounted for 9.2% of 2011 US gasoline consumption, nearly four times its contribution in 2005. However, before we could use a lot more ethanol in our cars, in the way Brazil has, we would need to overcome some big hurdles. Raising the proportion of ethanol in gasoline above 10% creates logistical and reliability problems, and the flexible fuel vehicles that can run on nearly pure ethanol are relatively scarce. In addition, we would need to produce most of the incremental ethanol from a feedstock other than corn. With the latest disappointing crop forecast from the US Department of Agriculture, ethanol production will consume about 41% of this year's harvest. Whether or not that's already enough to cause major food vs. fuel concerns, doubling corn use for ethanol would clearly push corn prices up drastically and cause ripple effects throughout the global food economy.

The good news is that biofuels--including better fuels than ethanol--can be produced from a wide variety of non-food crops, along with their efficient production from sugar cane in the tropics. The bad news is that with the exception of cane ethanol, none of these has been demonstrated on anything close to the scale required. Two of the largest cellulosic ethanol projects under construction, POET's Emmetsburg, Iowa project and the Vero Beach, FL facility of INEOS Bio, will together be capable of supplying just 0.02% of US vehicle fuel needs. And until these plants are up and running, their owners won't know whether their economics are sufficiently favorable--even with the current $1.01 per gallon cellulosic tax credit--to provide a basis for building more and larger versions. Although some of the many competing processes for producing biofuels from non-food biomass including wood, waste, dedicated energy crops and algae look very promising, they all face major uncertainties in development and scaling-up, including the scale-up of their supply chains, and none is yet ready for prime time. That might still be the case ten years from now.

Of course there are many other fuels we could put in our cars, after some modifications, including methanol, compressed natural gas (CNG), liquefied natural gas (LNG) or possibly even ammonia. However, the production of all of these, aside from a relatively small amount of landfill gas, is currently based on fossil natural gas, and all would require major investments in infrastructure and/or vehicle fleets. For that matter, 78% of the energy content of corn ethanol comes from natural gas and other fossil fuels--it also consumes enormous quantities of water--and most of the incremental electricity consumed by the first EVs will likely be generated from gas.

Although it appears that we have ample resources of natural gas to expand its use beyond current demand, I'm not sure that's quite what environmentalists have in mind when they talk about breaking our addiction to oil. And so far we've only considered alternatives to gasoline, without factoring in the significant demand for petroleum products for moving goods by truck, train and ship, along with aviation fuels, lubricants and many other products. Together, they account for as much oil as we use in cars, with non-oil alternatives for most of them at an earlier stage than for gasoline. And while energy efficiency measures, including the substantial improvements in vehicle fuel economy that are possible on a technology-neutral basis--including shifting cars to fuel-efficient diesels--can help to reduce the size of the mountain we must climb, they can't turn it into a valley.

Taking all these considerations into account it's not realistic to imagine that we could break our addiction to oil to any great extent for at least another decade. In the interim, we should certainly pursue all options that could alter the feasibility of such a shift in the years ahead, in a manner consistent with the fiscal constraints we face. I'm also not oblivious to what that implies for greenhouse gas emissions and climate change, though I would point out that our use of oil in transportation is neither the worst emissions offender, nor the easiest high-emitting segment of the US energy economy to tackle in that time frame. In the meantime, we are committed by virtue of scale, infrastructure and fleet requirements to burn many billions of barrels of oil over the next few decades, from wherever they may come. In that light, the administration's decision on the Keystone XL pipeline could prove to be a costly misstep, no matter how much political pressure they were under to withhold approval.