Good news, not-so-great news: The official view on trends in US energy consumption

The Trump Administration is often criticized for doing little (or worse) on energy and issues related to climate change.

Read the most recent Annual Energy Outlook (AEO), however, published on January 29 by the Energy Information Administration, and that critique looks a little off. In fact, the US energy sector is experiencing a world of change—some that the most diehard anti-Trumper could applaud, and others not so much. How these trends will play out in the period to 2050, the theme of the AEO, will affect everything from the balance of trade to geopolitics to the trajectory of greenhouse-gas (GHG) emissions.

The EIA insists that it is not making predictions on what will happen; rather, it is making projections based on a set of assumptions on things like regulation, economic growth, and prices. EIA’s best guesses form the “reference case” scenario. Other assumptions naturally deliver different predictions—sorry, projections.

Now that that’s clear, here are some of the biggest changes the AEO identifies.

The electricity sector: Demand for power grows slowly to 2050—about 1 percent a year—even though the US population and economy will continue to grow faster. The reason: improved energy efficiency and the continued shift toward a more service-oriented, less energy-intensive economy. Also, there will be more and more rooftop solar panels; this privately-generated power, which is tiny at the moment, could account for as much as 3.5% of generation in 2050. Every kilowatt produced on the top of someone’s house is one less watt needed from a power plant.

What if the car and truck fleet went electric? Well, if the US economy does begin to electrify more broadly—something that at least some of my McKinsey colleagues think could happen—that would change things considerably, boosting demand far beyond the reference scenario. But the EIA notes, diplomatically, that right now, there is a “lack of market evidence” that consumers want to buy electric vehicles (EVs). It estimates total sales of only about 2 million battery electric vehicles a year by 2050.

The mix of electricity sources is also changing. Natural gas and ever-cheaper renewables account for almost all new generation capacity to 2050, with renewables becoming more prominent over time. Coal and nuclear decline over the period. Regulation will matter. For instance, if the tax credit for wind is not renewed in the next few years, expect many fewer windmills.  On the other hand, solar production, both big and small, continues briskly to 2050 because the economics work. In broad terms, the EIA projects that all kinds of renewables will account for 38 percent of total generation in 2050, up from 18 percent in 2018.

Oil and gas production: Every year since 1953, the United States has imported more energy than it exported. Until now. Starting this year, thanks to the dual trends of strong domestic production and slow demand, the United States will become a net energy exporter—and  stay that way for a while. 

By 2022, the AEO projects that the United States will produce 14 million barrels of oil a day—almost three times as much as in 2008. Most of this is in the form of “tight oil”— meaning resources that are recovered through fracking, rather than conventional drilling. Tight oil accounts for about 70 percent of US production in the reference-case scenario to 2045, and then begins to decline. That is the projection—albeit this is one that AEO appears less confident of. Producing tight oil is relatively new, dating back only 10 to 15 years, and this short history means the production dynamics of tight-oil formations are not thoroughly understood. The related technology is also new, and keeps improving. So the context is not clear-cut. What can be said, though, is that domestic oil production is at levels that as little as a decade ago would have been considered impossible. Change, in short, can happen fast.

Greenhouse-gas (GHG) emissions: Energy use (per mile) improves for cars, freight, and aircraft by 2050, thanks to newer, cleaner vehicles and better fuel efficiency. And the changes in the power sector described above mean that electricity generation also progresses in terms of emissions intensity. On that basis, the AEO figures that GHG emissions will fall over the next few years—but then rise before 2030, due to higher transport and industrial demand. The cost of renewables will be an important factor in emissions trends. If renewables keep getting cheaper, then emissions will drop; but if costs stagnate, then there is less incentive to displace coal and gas in their favor. All told, the AEO believes that in 2050, US emissions will be 8 percent lower than they are now; that is progress, but considerably less than the US target in the Paris accords.

I have a lot of respect for the boffins at the EIA, and I like that, despite the many uncertainties, the AEO is putting forward a well-researched analysis that we can debate. But as I read through it, it also occurred to me that its projections are positively awash in uncertainty. For example, politics could markedly change the AEO’s assumptions. Bernie Sanders, for example, recently introduced a bill to ban fracking, and Elizabeth Warren is also hostile to the practice, tweeting that she wants to “ban fracking— everywhere.” Maybe EV manufacturers will lick the range problem, and cut costs, and Americans start buying EVs by the millions. Maybe states actually meet some of the aggressive emissions-reduction targets they have set; New York’s goal, for example, is 85 percent by 2050. Maybe there will be some new GHG-eating technology that changes the narrative on climate change. Maybe, maybe, maybe.

Just because we cannot predict (or even project) the future, does not mean that we do not have a responsibility for it. But given the assumptions and uncertainties that are so big a part of the energy sector—and of the Annual Energy Outlook—a degree of caution is in order.  Big changes will happen in energy in the future. Let’s hope they are driven more by technology and consumer choice, and nudged with regulatory incentives rather than forced from the top down.

Graphic credits: Energy Information Administration

All views are mine and not those of McKinsey & Company.