This Report
Structure and methodology
Pages 29–126 summarize an independent, transdisciplinary analysis of four ways to displace oil:
- Using oil more efficiently, through smarter technologies that wring
more (and often better) services from less oil (pp. 29–102).
- Substituting for petroleum fuels other liquids made from biomass or
wastes (pp. 103–111).
- Substituting saved natural gas for oil in uses where they're interchangeable, such as furnaces and boilers (pp. 111–122). Note that gas and oil, though sometimes found and thought of together, are utterly different in geology, economics, industry, and culture.
- Replacing oil with hydrogen made from non-oil resources (postponed
to pp. 228–242).
These options will be described first separately and then in integrative combination, because together they can do more than the sum of their parts. Efficiency options are presented for each end-use of oil. Each class of oil-displacing technology is presented as two different portfolios:196
- Conventional Wisdom: Expectations broadly accepted by industry and government; technologies on or soon to enter the market using thoroughly proven methods; surprisingly often already overtaken by the best technologies already on the market.
- State of the Art: Best technologies sufficiently developed by mid-2004, applying established principles and techniques, to be expected confidently as timely and competitive market entrants; well supported by empirical data and industry-standard simulations; require no technological breakthroughs; not all off-the-shelf, but no longer heretical.
The current public-policy approach to implementation might be
described as:
- Gridlock as Usual: Political will, chiefly focused on national policy and
driven by traditional constituency politics, can make modest, incremental advances comparable to those of the past two decades, but cannot execute gamechanging shifts in the status quo.
In contrast, pp. 127–168 present the business case, and pp. 169–226 the policy content, of:
- Coherent Engagement: Advanced technologies are rapidly and widely
adopted via innovative business strategies and a diverse portfolio of innovative policy instruments with trans-ideological appeal, weaving a rich tapestry of experiments by diverse actors.
Readers might expect us to outline at least three197 internally consistent scenarios for America's path beyond oil. But the first is uninteresting:
- Drift: Conventional Wisdom technologies
plus Gridlock as Usual implementation;
We briefly mention it on pp. 180–181 as a recalibration of the base-case forecast described below, just to clarify that some oil savings would occur even without our policy portfolio. We pay slightly more attention to what good policies can do even with incremental technologies:
- Let's Get Started: Conventional Wisdom technologies
plus Coherent Engagement implementation;
But we focus mainly on the "best of both worlds":
- Mobilization: State of the Art technologies
plus Coherent Engagement implementation.
None of these possible futures is a forecast, but we believe all are possible. Their divergent results show the importance of wisely choosing the future we want, then fearlessly creating it.
To ground our possible futures in policymakers' day-to-day reality, each technological option gauges potential oil displacement against the U.S. economic activity and oil consumption projected for 2025 in the Reference Case of the U.S. Energy Information Administration's Annual Energy Outlook 2004.198 EIA uses the National Energy Modeling System (NEMS), which lacks the predictive power, modern technological assumptions, and structural flexibility needed for sound business planning, but is widely used by government. Our analysis uses NEMS's outputs (e.g., forecast oil use by each sector and each class of end-use device in 2025) and inputs (e.g., how briskly new light vehicles sold in 2025 will accelerate, so we can match our assumed vehicle designs to that performance). However, we have chosen not to use the NEMS model itself, nor any other large computer model of the energy system. (Econometric models, though widely used, would be especially misleading because they rely on historic coefficients that our proposed technological and policy changes are meant to transform.) Rather, our calculations have been kept so simple and scrutable that they can all be reproduced on a hand calculator or with a simple spreadsheet. All calculations are shown and documented in the Technical Annex, so they can be checked and so readers who prefer other assumptions can plug in their own. We believe that for modeling the long-term energy system, less is more, that it's better to be approximately right than precisely wrong, and that simplicity and transparency trump complexity and opacity.
As a baseline for energy comparisons, we mainly use the year 2000, both because its statistics are available and stable (those for 2002 and even 2001 are still subject to revision at this writing in mid-2004199) and because 2000 was a broadly typical year for most energy statistics.
Though we remain mindful of market failures and the importance of correcting them, our economic analysis rests on orthodox market principles and economic methods—with one exception. To avoid using a large and opaque model, we haven't performed a general-equilibrium simulation to test how far the strong efficiency improvements we describe would undermine their own viability by reducing the prices of energy or of energy services.200 However, our conclusions are made more robust by multiple countervailing forces, including: About half of the State of the Art end-use efficiency potential using 2004 technologies can compete even at the lowest oil prices (Fig. 3) that might be expected to result even from its full global adoption (its use just in the United States, a fourth of the world market, would cut oil prices by only one-fourth as much); wide global adoption of strong energy efficiency could take about as long as depletion of low- and intermediate-cost oil outside the Gulf, further focusing OPEC's market power; China, India, and others will meanwhile want more oil; once adopted, the efficient vehicles, boilers, insulation, and other oil-using devices are unlikely to be switched back to inefficient ones; and if oil gets too cheap, it could always be taxed. Of course, if we're wrong, a durable oil surplus would be a nice problem to have.
We test cost-competitiveness against EIA's January 2004 projections of real energy costs in 2025, which are 14% above 2002's real costs for world oil, 8% for gasoline, and 19% for retail natural gas. However, EIA's 2025 U.S. Refiner's Acquisition Cost, $26.08/bbl in 2000 $, is well below the ~$35–50/bbl (2004 $) world market prices prevailing in summer 2004. EIA's energy cost projections aren't market price forecasts or expected values; they assume that weather, climate, inventories, regulations, geopolitics, and other background conditions remain "normal." We use them because EIA considers them consistent with its demand projections—our base case.
Our Conventional Wisdom technology options rely heavily on prior studies by authoritative industry, government, and academic teams. Many of these studies are sound, available, documented, and adequate. Where such prior work is unavailable or insufficient, our own State of the Art analysis is documented in the Technical Annex. Our limited analytic resources are focused mainly on the seven biggest terms, each at least 6% of 2025 U.S. oil use, treating only briefly the small terms that make up the last 6%.
We have also focused chiefly on the terms accounting for most of the demand growth.
Our emphasis on empiricism leads us, wherever possible, to rely upon and document actual measurements rather than relying on theoretical projections. (Similarly, where results seemingly contrary to economic theory—such as expanding rather than diminishing returns to investments in energy productivity—have been solidly established by empirical practice, we don't reject them in favor of theory.) And we have followed Aristotle's counsel205 that "it is a mark of educated people, and a proof of their culture, that in solving any problem, they seek only so much precision as its nature permits or its solution requires." Energy analysis, especially looking a quarter-century ahead, is an uncertain art, so we strive to avoid implying spurious precision.
196. We use this term in this report to mean a cluster of internally consistent assumptions that help to understand the nature and implications of choices. This differs from the meaning in scenario planning: Schwartz 1996.
197. We don't explore the off-diagonal matrix element combining State of the Art technologies with Gridlock as Usual implementation because those technologies are unlikely to be commercialized within that policy framework.
198. EIA 2004.
199. For example, EIA's Annual Energy Review showed 2000 petroleum consumption 1.2% higher in its 2002 edition (EIA 2003c) than in its 2000 edition (EIA 2001c). We prefer to wait a few years until the numbers stabilize.
200. Such a calculation would entail many assumptions and great complexity—akin to the inverse of the challenge of calculating macroeconomic effects of increases in oil price, where results vary substantially depending on model structure and assumptions (IEA 2004b, pp. 5–6), especially about exchange rates and monetary policy. Unmodeled effects, such as changes in business and consumer confidence and in gas pricing, can be important (IEA 2004b).
201. Methyl tertiary butyl ether.
202. In addition, vehicular fuel equivalent in 2000 to 0.3% of gasoline was supplied by liquefied petroleum gas (68%),
compressed natural gas (27%), liquefied natural gas (0.02%), methanol (0.003%), and 85% or 95% ethanol (0.02%): EIA 2002.
203. This is because ethanol is due to replace MTBE in 17 states "over the next few years" (EIA 2004, p. 256),
and ethanol is twice as effective an oxygenate per unit volume (EPA 1998).
204. From EERE 2003. The car/light-truck data, which EIA's Annual Energy Outlook doesn't show, were confirmed from the NEMS database. Light trucks include both passenger (Class 1–2) and commercial (Class 3–6).
205. A paraphrase combined from Nicomachean Ethics I:3.24 (Berlin 1094b) and I:7.26 (Berlin 1098a).
(End of excerpt)
