The return of peak oil: Better luck this time?

Remember “peak oil”? This was the prediction, made regularly since the mid-1950s[i], that oil was running out due to inexorable depletion. 1970 was the first such deadline, which was then regularly pushed back. Somehow, although oil kept not running out, the idea endured. In 2002, one of the theory’s most prominent proponents, the late Matthew Simmons, predicted that supplies would peak as soon as 2010: “We are in [the] early stages of a global train wreck when demand outstrips supply and shortages begin. This could easily morph into social chaos and war."

Of course that didn’t happen, just as all similar predictions proved equally off-base.

Now there is a different peak-oil theory—to do with demand, rather than supply. For example, McKinsey’s Global Energy Perspective puts peak demand of about 102 million b/d, “between 2024 and 2027, driven largely by EV uptake—a development that is already underway.”

?Bloomberg’s well-regarded energy columnist, David Fickling, argues that a recession in 2023 will push down demand in the short term, and that long-term trends, in the form of greater fuel efficiency and new technologies, will keep it down. His conclusion: “the world’s appetite for oil is peaking and will soon enter terminal decline.”

Ditto for the International Energy Agency. ?If all current climate pledges are met, it sees peak demand coming in the mid-2020s. If the world goes further, faster, to net zero, then the peak would have already passed, in 2019. And if it’s business-as-usual, think mid-2030s.

In 2020, bp, the European oil major, suggested that the world may never again see 100 million b/d, and that demand might have peaked just before COVID. Its most recent report, however, bp notes that the pandemic had less of an effect and that demand has resumed. With serious climate-mitigation action, it sees consumption peaking in the mid-2020s, and then falling sharply. Under business-as-usual, it’s more like 2030, and then a gradual decline.

On the other hand—and there is always at least one other hand in the oil business—Bloomberg’s Javier Blas takes issue with his colleague, arguing that the IEA itself anticipates demand growing in 2023 by 1.7 million b/d, despite fears of recession, changing commuting patterns, and all the rest. “To achieve the Holy Grail of peak oil demand, first the world needs to show signs that consumption growth is starting to slow down — and quite materially. That’s?yet to happen.”

Vitol Group, an oil trader, puts peak demand no sooner than the 2040s, arguing that even lower demand for road transport would be offset by rises in other areas, such as plastics, chemicals, and aviation.

OPEC doesn’t seem much fussed, either, projecting that demand will continue to grow, albeit slowly, to 110 million b/d by 2045, due to population and economic growth in today’s low- to middle-income countries. (Consumption in the rich world, it argues, has already peaked.)

So, what do I think? Well… while I am always reluctant to make flat commitments about anything to do with oil, here are a few thoughts about the long term.

• The time-frame for peak-oil keeps coming closer, as changes on the ground, in terms of efficiency, renewables, and new technologies, take root.
• ?There is broad agreement on the major trends—that the world is transitioning to less use of fossil fuels, including oil, and that their relative importance in the energy system will decline. The difference is on timing. Early peak oilers see it happening much faster, because there is big and growing pressure to cut emissions, and many of the ways to do so already exist. Skeptics project a more drawn-out process as non-rich countries begin to buy cars, use air conditioning, and so on.
• ?If the world was really serious about getting to net zero by 2050 (or even earlier), then there would have to be truly drastic action on fossil fuels, like shutting down all coal plants now, or banning the sale of internal combustion engine (ICE) vehicles shortly. There may be a handful of very wealthy people or places that do this—but not the billions in developing countries.
• ?Peak oil doesn’t mean no oil. One thing all these predictions have in common is that none of them see oil becoming irrelevant any time soon, just less important. Let’s assume that the most aggressive predictions prove correct—that demand is close to its peak right now (and perhaps even past it), and that it will be about half that in 2050. That’s still a lot of oil, and a lot of energy.

With all that in mind, I think there are two things to consider. First, history shows that the world doesn’t give up on energy sources; it simply adds new ones, and then these displace the old to greater or lesser degrees. (If you have a wood-burning fireplace, you will know what I mean.)

?According to all peak-oil scenarios I know of, oil is going to continue to be an important feature in the global energy landscape for decades to come. It is not separate from, but an important part of, the energy transition. The implication, then, is that its development and use should be as clean and emissions-efficient as possible. Many oil majors already have net-zero plans for their operations in the works. Going further, however, will mean investing in as-yet unknown innovations as well as things like carbon capture and storage —something that has proved too expensive for widespread use to date.

And second, the debate on the date is pointless. Not all the peak-oil predictions can be right. In fact, they could all be wrong.?As Spence Dale, bp’s chief economist, put it, “The range of uncertainty is huge. Small changes in assumptions about the myriad factors determining oil demand, such as GDP growth or the rate of improvement in vehicle efficiency, can generate very different paths.”

For example, the rate of EV take-up has long been consistently over-estimated. Yes, we may hit a tipping point, and then the market could change very fast. But we’re not there yet: close to 98 percent of all vehicles on the road are ICEs. In 2022,?an estimated 13 percent of new cars sold were EVs—a record, but still only about a month’s worth of ICEs. There is a long, long way to go. Similarly, the link between oil and GDP growth is weaker than ever, but it still exists.

So, whether the peak oil demand is last year, 2030, 2040, or some other time, the fact is oil is not going away. Let’s make the best of it.

And, of course, happy new year.

[i] Indeed, the idea goes back even further: the first reference I can find dates to 1919! But the idea really took hold with the famous Hubbert Curve, first described in 1956.

All views are mine and not those of McKinsey & Company or of the Greater Houston Partnership.??