First published at New Scientist, 17 May 2012.

People have fretted about when the world’s oil will start to run out ever since M. King Hubbert came up with the idea of “peak oil” back in the 1950s. The American geologist, who worked for Shell, pointed out that we are destined to reach a moment when oil production stops rising and goes into terminal decline. With it, the era of cheap oil that fuelled the post-war economic boom would end. The idea still provokes great debate, and many forecasters are predicting that global production will peak by the end of this decade as supplies dwindle.

New Scientist 17.5.12

Now there is a different view. A small number of analysts forecast that oil production will start to fall by 2020 – not because we are running out, but because we just won’t need it. They argue that the world will wean itself off oil voluntarily, through major advances in vehicle technology. Peak oil will not be a supply-side phenomenon brought about by shrinking reserves, but by motorists buying electric cars and conventional cars with highly efficient engines. If they are right, this shift will start the long-term transition from oil to electricity as the main transport fuel, reduce economies’ vulnerability to spikes in the oil price, and cap greenhouse emissions from crude oil.

It is a bold prediction. Could it be right? Judging by motor industry investment and the number of new models being launched, the prospects for the electric car are brightening. All the major manufacturers are producing cars with varying degrees of electrification, ranging from hybrids, such as the Volvo V60, that run on petrol and electricity to cars such as the Nissan Leaf that are powered entirely by an electric battery (see “Six degrees of electrification” below). There are now about 130 models in total.

Sales so far have proved disappointing, though. Total car sales in the US last year jumped by a tenth over the previous year. But electric vehicle sales rose just 2.3 per cent, according to research firm WardsAuto. Sales of General Motor’s Chevy Volt missed their target by a fifth, and those of the pioneering Toyota Prius hybrid have been falling since 2007. So can electric vehicles really make a serious dent in global oil demand?

Investment analysts at Deutsche Bank in New York argue in a series of reports that the electric vehicle is a disruptive technology and its short-term potential is widely underappreciated. “Transportation is likely to change more in the next 10 years than over the last 50,” says Dan Galves, the bank’s chief car industry analyst. That’s not because of some imminent technological breakthrough, but because he expects that the relative costs of electric and petrol cars will soon be transformed.

Electric cars are far more expensive to buy than their petrol equivalents, largely because the cost of the lithium-ion battery that powers the vehicle is so high – currently about $12,000. But the fuel costs of electric vehicles are already far lower than for petrol-powered ones. In the US, for example, the petrol for an average car costs about 8 cents per kilometre, compared with less than 2 cents for the electricity to power an electric car. In Europe, where fuel tax is higher, the numbers are 12.5 cents and 2.5 cents, respectively. Either way, that is a huge gap. So for electric vehicles to compete on price, battery costs need only fall far enough to be swallowed by that gap, and Galves believes that it is likely to happen sooner than most people think.

First, he expects the costs of batteries to plummet as mass production ramps up – just as they did for laptops – to less than $7000 by 2015. Second, the gap is likely to widen with most analysts expecting oil prices to keep rising. “On a 10-to-15-year view, it’s almost impossible for electrification not to carve out a decent portion of the market,” says Galves, who expects electric vehicles to be economic within a decade even without the subsidies that many governments currently give.

The effect of falling electric vehicle costs will be reinforced by strengthening fuel efficiency and emissions policies in the world’s most important car markets. The policies of the world’s biggest gas guzzler will soon be among the toughest. In 1975, US president Jimmy Carter passed a law forcing vehicle manufacturers in the US to meet average fuel efficiency standards. For cars, that number has languished at around 27 miles per gallon (11.5 kilometres per litre) since the early 1990s, but recent legislation means average fuel economy must double to 54.5 mpg by 2025. The standard has been rising since 1978, and by 2020 the targets become so demanding, says Galves, that car manufacturers will not be able to meet them without selling a significant number of electric vehicles. Galves expects them to make up a fifth of US car sales in 2020.

The impact will be dramatic. Every day, US vehicles guzzle about 9 million barrels of oil – the biggest single element in our daily global consumption of almost 90 million barrels (see chart, top left). Deutsche Bank oil analysts expect US petrol consumption to plummet, almost halving by 2030. The story is the same in the European Union, which regulates carbon dioxide emissions per kilometre rather than miles per gallon. Cars manufactured there in 2020 must reduce their average emissions by more than a quarter compared with models made in 2015. Such standards will especially encourage electrification because they govern “tailpipe” emissions pumped out in the day-to-day running of car engines and not those emitted while they are being built. By this measure, electric vehicles are zero emission. Deutsche Bank expects them to make up 25 per cent of Europe’s car sales in 2020, accelerating the decline in demand for petrol.

Petrol still rules
So much for the world’s richer nations. In China, where the developing car market is booming, the demand for petrol will continue to rise for at least a decade. Yet the global impact will be limited because the size of China’s car fleet is currently just a fifth of that of the US. The Chinese government too is strongly committed to electric vehicles as one way of tackling appalling air quality in the cities and the country’s dependence on imported oil. Deutsche Bank forecasts that Chinese petrol demand will start to fall from 2025, as electric vehicles become more common.

The net effect is that global petrol demand will peak as early as 2015. “From that point forward,” writes Deutsche Bank’s lead oil analyst Paul Sankey in a research note. “We believe gasoline demand will be on an inexorable and accelerating decline.” And as a result, he argues, global demand for crude oil will go the same way in about 2020.

Others disagree with Deutsche Bank’s analysis. The International Energy Agency has long been dismissive about predictions of an early peak in the global oil supply. It is just as dismissive that demand will decline within the next couple of decades. It forecasts that daily oil demand will rise to 107 million barrels by 2035 on the basis of current government policies. Fatih Birol, the agency’s chief economist, believes that there are simply too many cars in the world – about a billion and rising – for electric vehicles to have a meaningful impact in the short term.

Although most governments have policies to encourage electrification, they are very unlikely to achieve their targets. Even if they do, says Birol, the number of electric vehicles on the road in 2020 will be just 20 million – about 2 per cent of the total fleet. Stefanie Lang, a London-based automobile analyst at investment-research firm Sanford C. Bernstein, agrees that electric vehicles will make only limited progress over the next 10 to 15 years. She argues that they will struggle because they will remain far too expensive and will face fierce competition from the incumbent technology – the internal combustion engine.

Even after a century of development, the internal combustion engine has the capacity to make major improvements in fuel economy, says Lang, rattling off three examples. “Stop/start” mechanisms that kill the engine when the car pauses in traffic can produce average fuel savings of 5 to 9 per cent, and will probably come as standard on all European models by 2015. Fitting cars with smaller engines and turbochargers will use 3 to 6 per cent less fuel to deliver the same performance as conventional engines. Injecting fuel directly into a petrol engine, rather than mixing it first with air in a carburettor, can raise fuel economy by another 3 to 5 per cent. “They aren’t headline grabbing technologies, necessarily,” says Lang, “but they are the low-hanging fruit of fuel efficiency and can reduce fuel consumption across the board.” She forecasts that these and other known technologies will lead to an improvement in efficiency of up to 30 per cent by 2020.

The upshot, according to Lang, is that car manufacturers can meet US and European standards simply by investing in incremental improvements to existing models, rather than struggling to sell more electric vehicles. Such investment could still have a dramatic impact on global oil demand. Although cars would still be fuelled largely by oil, another study shows how the increased efficiency of traditional engines would have much the same effect as electric vehicles. Analysts at engineering consultancy Ricardo in London surveyed the energy efficiency improvements being planned by car manufacturers and plugged them into a global model that includes factors such as government policies, demographics and gross domestic product. They were surprised to find that global oil demand would peak by the end of this decade, and could drop 10 per cent by 2035.

Like others, Ricardo concluded that electric vehicles would make little headway this decade, and that improvements in the efficiency of conventional engines would be the primary factor. Despite an 80 per cent rise in vehicle numbers by 2035, oil demand will fall largely because vehicle efficiency will more than double, claims Peter Hughes, head of Ricardo’s energy practice in London. Other factors lower fuel consumption too: the ageing population in key markets, because older people drive less; working from home; and the oil price, even though the model in Ricardo’s research assumes just $100 per barrel to 2035. The factors working against a growth in demand for oil are increasing in number and intensity, says Hughes. “The world is nearing a paradigm shift in oil demand.”

So what does the motor industry itself think lies ahead? That the internal combustion engine’s days are numbered, for one thing. In a recent survey, consultants KPMG asked 200 top executives of car companies how long they thought the traditional engine would continue to prevail over electric vehicles. Some 70 per cent answered 1 to 10 years, but only 18 per cent thought 10 to 20 years.

One reason for the result could be that electrification is now widely seen as the best way to make major reductions in transport emissions, even taking into account the emissions from generating the electricity in the first place. That is because electric vehicles are far more efficient than petrol cars. Take the Nissan Leaf. It is responsible for just 99 grams of CO2 per kilometre, even when charged on electricity generated by the average mix of coal, natural gas, nuclear and renewables. That makes it 40 per cent cleaner than a typical petrol car in Europe. And as electricity generation becomes cleaner, the emissions of electric vehicles will fall further still – unlike those of cars powered by biofuel or natural gas (see New Scientist, 25 February, p 48).

Lang points out that future improvements to the internal combustion engine will become progressively more expensive and less effective, while legally binding standards get tougher. She reckons the turning point will be 2025, when the US fuel economy standard reaches 54.5 miles per gallon (23 kilometres per litre) and Europe’s upper limit on CO2 emissions for new cars could be as low as 70 grams per kilometre. “It’s going to be very difficult to achieve that with low electrification,” says Lang. Both she and Hughes see electric vehicle sales beginning to take off from around that time.

Rebound effect
In one sense it doesn’t matter when electric vehicles supplant the internal combustion engine. As long as the motor industry delivers the expected efficiency gains somehow, the climate will benefit. But what if both sides of the argument are wrong, and neither technology delivers large cuts in oil demand?

Super-efficient engines may fail to change oil demand if their efficiency gains are eroded by the “rebound effect”, by which rising efficiency stimulates increased consumption. Researchers at the UK Energy Research Centre in London concluded that 10 to 30 per cent of the benefits could be lost because efficiency gains make it cheaper to drive, encouraging people to use their cars more.

Economic growth could hamper progress too: one scenario considered by the International Energy Agency indicates that improvements in fuel economy will be overwhelmed by rising vehicle numbers even if governments rigorously enforce tighter rules on energy efficiency. On the other hand, recession and fiscal austerity could hamper progress if governments start cutting back their financial support for electric vehicles.

If the forecasts of Deutsche Bank, Ricardo and Sanford C. Bernstein are anything to go by, the transition away from oil could be far less painful than many expect. But if technology fails to slake our thirst for oil, then supply will struggle to keep up with demand and peak oil may turn out to be a supply-side phenomenon after all, just as Hubbert predicted all those years ago.

Six degrees of electrification

• A micro hybrid has a conventional internal combustion engine (ICE) with a “stop/start” mechanism that kills the engine whenever it pauses in traffic. This means it needs a more powerful lead-acid battery and starter motor. Advanced versions use this not just to start the engine, but also to drive the car briefly after it restarts, when running an ICE is at its least efficient. Offered as standard on many new cars, it can deliver fuel savings of 5 to 9 per cent. It is not generally considered to be an electric vehicle.

• A mild hybrid is somewhere between a micro and full hybrid. It has regenerative braking, which uses energy that would otherwise be lost as heat during braking to recharge the battery; a traction battery that is used to power the car instead of just the starter motor and peripherals; and an electric motor. But unlike the full hybrid, its electric motor only ever supplements the ICE and never powers the vehicle entirely by itself – so it is not considered an electric vehicle. One version of the Honda Civic is a mild hybrid.

• A hybrid, or full hybrid, such as the Toyota Prius, has an internal combustion engine, an electric motor and a small nickel-metal hydride traction battery. All the electricity is generated on-board by the ICE or regenerative braking. The motors are arranged in parallel, so each can drive the wheels independently. Many combinations are possible, but typically the car will rely on electric power up to about 40 kilometres per hour, when the ICE takes over. The new Prius C can do up to 53 miles per gallon (22.5 kilometres per litre).

• A plug-in hybrid, such as the Volvo V60, has the same configuration as a hybrid, along with a socket to charge the battery from the grid.

• A range-extended electric vehicle, such GM’s Chevy Volt (or Vauxhall Ampera in Europe), is similar to a plug-in hybrid except that the ICE is only there to generate electricity for the battery and electric motor, and never drives the wheels directly. The vehicle travels on grid electricity only for the first 45 kilometres or so, and then switches to electricity from the ICE until the next recharge. The Volt does the equivalent of 40 kilometres per litre.

• A battery electric, such as the Nissan Leaf, has only a battery and electric motor and is entirely dependent on grid electricity and regenerative braking. The Leaf can travel about 160 kilometres on a single charge.


  • Edmund Kurridy

    Isn’t a big part of the ‘bigger picture’ the point that the petroleum-fuelled vehicle business is the core component of the process that converts the heat energy embodied in oil into transferable wealth?

    Therefore, less oil + fewer ICEs will = less wealth for people to buy EVs with. Running-cost comparisons based on today’s situation are also misleading, since electricity prices would rise greatly as oil was withdrawn from the global energy mix.

    Existing coal and gas energy generation capacity is more or less fully committed to maintaining our non-transport infrastructure anyway. Expecting it to ramp up sufficiently to take over oil’s work as well would be a big ask.

    I do expect the balance of vehicle motive power to shift towards EVs in future. But the total number of vehicles of all kinds will surely reduce dramatically if, or rather when, oil all but disappears from export markets in 20 years’ time.

  • Talking of hybrids, this working paper from the IMF is very interesting:

    They appear to be softening their stance towards peak oil. They combine geological supply constraints with the usual demand sider optimists to come up with this conclusion:

    “..our prediction of small further increases in world oil production comes at the expense of a near doubling, permanently, of real oil prices over the coming decade.”

    This means that effeciency gains are unlikely to keep pace with cost increases for running traditional petroleum cars. Whether the electricity cost differential widens even further to drive enormous shifts over to electric & hybrid is debatable, though, as surely electricity demand will surge.

    Either way, transport costs are going up. And up.

  • Mike Grenville

    I didn’t see any mention of electric trucks? While many individual journeys can be avoided, our economy is based on long distance delivery of goods and food by lorry. Unless electric trucks become a significant part of the mix food and other goods costs will rise substantially.

  • To successfully convert 800 million cars to EVs before serious oil decline starts is an untested assumption. Moreover, the energy problem is just moved from oil to other fossil fuels with similar CO2 problems. The 1st priority must be to get rid of coal fired power plants. If we then have still enough power to run electric trains, we’ll be lucky. While there will be car-pooling and less driving because the 1st phase of peak oil (which started in 2005) has already damaged our economy, the real worry is the oil dependence of agriculture and transport of food. That should be our focus for solutions. Cars will be the least of our worries.

  • From a recent CNNMoney article: “During the height of the so-called “SUV craze” in the late 1990s and early 2000s, about one in five vehicles sold in America was an SUV. Today, in an era of near $4 gasoline and heightened environmental awareness, nearly one in three vehicles sold is an SUV.”

  • Derek

    A few more counter-points:

    1.) All advanced batteries rely on rare earth minerals like Lithium that may not be extracted at rates economical enough to satisfy demand for millions of automobiles:

    2.) Many analysts follow an oil supply model that roughly matches the EIA’s methodology, which does not account sufficiently for depletion in older oil fields and declining discovery rates. So, most analysts assume a gradual slope upward and plateauing of oil supply. In reality, we may see production declines of up to 4% per year within the next decade. It’s going to be tough to eliminate 4% of global demand each year when India and China continue to grow at a rate of >2%.

    3.) When you get right down to it, automobiles are the least efficient means of transportation. It’s going to be much easier for us to switch now to denser cities and more mass transit rather than push to keep the automobile dream alive. We need to think in terms of Megajoules per dollar of GDP instead of “how can we keep the cars running?”

  • BS

    US consumers have never demonstrated downsizing anything. So assuming it would take 8-10 years to replace the US vehicle fleet if every new car bought was much more efficient puts us into the 2020s. Does anyone think SUVs or light duty trucks are going away any time soon? So realistically we are looking at perhaps 2030 before the US decreases it’s oil consumption for transportation by a large percentage. That shale oil and tar sands better live up to the hype!

  • BBHY

    Lithium is not a rare earth mineral. It is about as abundant in the Earth’s crust as aluminum. No one is talking about running out of aluminum. Even if the price of lithium doubled, it is not the main driver in the cost of lithium batteries.

    I agree that it is better to move people to cities and use more public transportation. But it is a free country and people are going to do what they want, and a lot of them do not want to live in the city.

    Probably the best way to promote cities is to make them better, so people will want to live there. Electric cars in the cities will reduce the amount of noise and pollution, so they will actually help that process. We should also start a movement to bring back the inner city trolleys.

  • It’s unrealistic to think about this as replacing 800 million cars with EVs – the reason we’ve got 800 million cars today is that we’ve just been through a brief moment in human history (about 100 years) in which oil was cheap and readily available, barring a couple of short-term blips. Since we’re entering a new era in which oil will be neither cheap nor readily available, the very conditions that brought those 800 million cars into being no longer exist. This has massive implications in heavily motorized nations whose economies have grown around the ICE – where economic progress is measured in car sales – but perhaps not such massive implications in other countries.

    China is building 45,000 km of high-speed rail by 2015, while the UK debates whether it can link London with Manchester with a new high-speed line by 2030 – a distance of some 350 km – simply staggering ineptitude.

    An electromobility pathway is not just about private cars: China has 120 million electric bicycles on its roads today, from roughly zero 12 years ago. For people who’ve never owned a car, this is a vastly improved prsonal transportation mode than anything they have been accustomed to.

    Electric trucks: trials are underway to test HGVs with pantographs running from overhead power lines on existing highways.

    Keep in mind the following: the electricity sector MUST and WILL decarbonise over time; electric drive is more than twice as energy efficient as the *theoretical maximum* that mechanical drive can ever attain; the liquid fuel pool gets MORE carbon-intensive over time, not less, as each successive barrel of oil takes more energy to extract.

    An economic paradigm in which advancement is measured by the rate at which we dig stuff out of e ground and set it on fire in order to move things around should be seen for what it is: extraordinarily primitive. Surely humanity can do better than this pile of bollocks.

  • John B.

    The idea of “peak oil” actually goes back to the 19th century. Many failed predictions have been put forth since then, including Hubbert’s failed prediction of peak oil in 1995. A new peak was just set in 2011.

    In reality, world oil reserves are sufficient to last several hundred years. Current peak oil pundits are basically a cult of modern day Luddites “hoping” the world runs out of oil.

    EVs will likely become the dominant form of transportation in the future, because of ever increasing emissions regulations.

  • Boris

    The notion that its not a supply side issue because people will migrate to EV’s before geological constraints “kick in” but then go on to argue this will be encouraged by rising oil prices strikes me a a contradictory if not circular argument.

    Rising oil prices must be in some way be related to geological restraint. Even conspiracy theories based around cartel behaviour such as OPEC have this underlying assumption. The oil has to be geographically limited [hence geologically limited] in location for a cartel to have the power to distort the price. Rarity has to be a issue. Peak demand is in effect no different from peak supply because the alternatives economically driven or not are of less utility.

    Its such a muddled issue it is hard to see what the central point of the piece is?

  • Omri

    If we really want to weather the oil peak, we have to give up rubber wheels on hardtop and get used to turning steel wheens on steel track again.

  • John B.

    Peak conventional oil has pretty much already happened, circa 2005, so Hubbert wasn’t far off if his estimates were made in the 1960s!

    By your definition then maybe you are calling the IMF Luddites for this paper:

    “Our empirical representation of this view models oil supply as a combination of the Hubbert linearization specification of Deffeyes (2005) and a price mechanism whereby higher oil prices increase oil output.”


    “..our prediction of small further increases in world oil production comes at the expense of a near doubling, permanently, of real oil prices over the coming decade.”

    P.S. What estimates of remaining reserves are you referring to?

  • Isn’t a transition that moves us from fossil fuels delivered in one form (unleaded) to another form (electricity) like thinking we can conserve water by using a brass faucet instead of one made from steel?

  • Derek

    BBHY: Lithium is “abundant” meaning a lot of it (in terms of quantity) exists on our planet. However, it doesn’t occur in concentrations great enough to make it economically recoverable for much longer, especially if demand continues to grow.

    It doesn’t matter if the price of Lithium isn’t the main price driver now, it matters whether enough of it can be extracted economically in 5 years.

  • David Lee

    $6,000 reusable battery will last about 8 years. And think about how much money is spend each year on gas (2,000)

Comments are closed.