CAR-BUYERS are getting behind the wheel of an electric vehicle (EV) in ever greater numbers. No wonder, for they are exciting and easy to drive, compared with internal combustion engine (ICE) equivalents. As battery costs tumble, prices are falling. But the shift to EVs means much more than driving pleasure. Transport is responsible for around a quarter of the world’s carbon emissions and road vehicles account for around three-quarters of that share. If there is to be any chance of reaching net-zero by 2050, EVs will need to take over, and soon.

The 6m pioneers who opt for EVs this year will still represent only 8% of all car purchasers. That figure will need to increase to around two-thirds by 2030 and to 100% by 2050 in order to meet net-zero goals. Many an investor is operating on the assumption that this will all happen as smoothly as a Tesla shifts gears. The huge market values of Elon Musk’s company, and of other newcomers such as Rivian with its electric pickup trucks, as well as pricey Chinese EV firms, attest to sky-high confidence. Electric battery-makers, too, are booming and their shares are soaring.

Yet look beyond the glamorous, shiny vehicles stuffed with the latest technology that are the obvious embodiment of the EV revolution, and you can see a merciless bottleneck ready to foul things up. Not even those eyeing an EV purchase are sufficiently aware of it. Governments are only waking up to the problem around now. Put simply: how will all these EVs get charged?

The current number of public chargers—1.3m—cannot begin to satisfy the demands of the world’s rapidly expanding electric fleet. According to an estimate by the International Energy Agency (IEA), a global forecaster, by the end of this decade 40m charging points will be needed, requiring an annual investment of $90bn a year as 2030 approaches. If net-zero goals are to be met, by 2050 the world will need no fewer than 200m public charging points.

It is certainly true that current pledges by governments on the phasing-out of ICE cars and the transition to EVs mean that sales consistent with net-zero look unlikely. Even so, should roads turn electric less speedily, the sums the world needs to spend on charging infrastructure are still stupendous. In a slower scenario envisaged by BloombergNEF (BNEF), a research firm, under which EV penetration continues to rise as battery prices fall, but sales only reach just under a third of all vehicle sales by 2030, roughly $600bn of investment would still be needed by 2040. That would pay for fewer chargers than the IEA foresees—24m public points by 2040, and 309m in total. If net-zero is to be achieved by 2050, BloombergNEF calculates that the cumulative investment required would be a whopping $1.6trn.

To the problem of woefully few public chargers, add the poor operational record of the charging industry thus far. On paper, the number currently exceeds what some authorities reckon is needed. The European Commission, for example, thinks every ten electric vehicles require one public charger. According to the Boston Consulting Group (BCG), a consultancy, there are now five EVs per charging point in the European Union and China and nine in America.

The reality is starkly different. According to a survey of chargers in China by Volkswagen (VW), inoperable or “ICEd” chargers (those blocked either inadvertently or deliberately by fossil-fuel cars) mean that only 30-40% of Chinese public points—there are now 1m—are available at any time. It is safe to assume some inoperability in the EU and America.

Even important drivers deride the experience. Herbert Diess, VW’s chief executive, posted on LinkedIn, a social network, this summer to complain that his holiday had gone less than smoothly because Ionity, a European charging network, provided too few points on the Brenner Pass between Austria and Italy. The firm’s charging facilities in Trento in northern Italy left plenty to be desired. “Anything but a premium charging experience,” he wrote. That vw is a part-owner of Ionity made the criticism sting more.

Drivers can smell trouble ahead. Range anxiety and the availability of public charging is a huge issue (see chart 1). In a recent survey by AlixPartners, a consultancy, in the seven countries that make up 85% of global EV sales, the cars’ high prices came third on the list of main reasons not to switch to battery power; the four others were all anxieties related to charging.

To assess the scale of the challenge, and how it will be met, start with the basics. One big advantage of EVs is that the cars can be charged at home—or at workplaces, if employers install chargers as a perk. In America, 70% of homes have off-street parking where a charger might be installed (the equivalent figure is lower in Europe and China). BCG estimates that the energy demand for home and workplace charging in 2020 accounted for nearly three-quarters of the total in America, around seven-tenths in Europe and three-fifths in China.

Current models of electric cars typically have batteries with ranges of around 250 miles (400km). Some go over 400 miles. The average American drives 30 miles a day, according to Bank of America. Europeans and Chinese drive less than that. That means two types of charger are good enough to top up vehicles, or to give them a bigger boost overnight at home or during the working day. The slowest, delivering up to five miles of range an hour, can do it. So do “level 2” chargers that deliver around 10-20 miles. These chargers are easy on the wallet, too. Using dedicated sockets that cost just a few hundred dollars (they are often subsidised by governments) entitles drivers to the very cheapest electricity tariffs.

Nonetheless, home and workplace charging only gets drivers so far. As EV ownership spreads from wealthier households to people living in flats or dwellings without the ability to plug in at home, a public network becomes more and more vital. In America, Europe and China, demand for public charging as opposed to private is expected to increase (see chart 2). Public chargers come in three varieties. A common kind is kerbside charging, which can be via converted lampposts or other dedicated charging points, where cars might park overnight. Then there is “destination” charging, of the sort that is becoming more widely available in car parks at shopping centres, restaurants, cinemas and other public attractions. For both kinds, which count as level-2 charging, the installation cost is usually between $2,000 and $10,000 per point.

Fast charging, which can typically deliver 60-80 miles of range every 20 minutes, is vital on main roads for drivers making long inter-city trips beyond the range of their vehicles, and in cities for a quick emergency jolt. Commercial vehicles, such as taxis, driving longer distances need fast charging too. But since charging companies are seeking to recoup hefty costs of $100,000 or more per charger, the price is high. To make life easier for its customers, Tesla’s mapping software directs its cars on long journeys and works out the best route weaving through its dedicated, rapid “Supercharger” network. Other new EV models come with similar features for planning long journeys around fast chargers.

In the charging industry’s defence, many in it point out that both EV ownership and charging are in their infancy. Pessimism is unwarranted, they argue, based on just a few years of experience. Only one car in a hundred now on the world’s roads is an EV, after all. And as Pat Romano of ChargePoint, an American firm that is one of the world’s biggest charging companies, notes, this is the start of “a 20 year arc”. But though the charging industry has time to mature, what sounds clear on paper is daunting in reality. The nature of demand for charging at scale is impossible to know as yet, meaning much unpredictability.

Expansion is coming fast, say some. Along with all the momentum from EV-phile governments, the opportunity to make money charging the world’s expanding fleet means that “hyperbolic growth” is on the way, insists James West of Evercore ISI, a bank. But exactly how many public chargers are needed for each EV on the road is “an open question” notes Bank of America. Scott Bishop of Yunex Traffic, a division of Siemens, a German firm that makes charging hardware, notes that there are many different answers to the question of what proportion of slow versus fast chargers will be needed.

Another problem is that the charging industry is made up of many complex layers. Aakash Arora of BCG’s automotive practice calls this the “gnarliest problem of all”. The need to co-ordinate with and get permission from many different parties helps explain the slow roll-out of charging infrastructure. First, there are firms that make the chargers themselves. Then there are the operators. These might own the points, earning money directly from charging. Or they might lease or sell points to site-owners but make money maintaining the chargers and update software when needed. Site-owners, usually businesses, other private landlords or local authorities, provide the locations for chargers and typically charge rent to operators. Service providers allow the charging to happen, with apps or cards that give access to charge points and provide payment mechanisms.

Three kinds of company are coming to rule the EV-charging roost. One is the vertically integrated car giant. Tesla has not revealed what it has spent on its “Supercharger” network, which now numbers 30,000 points worldwide, but it is likely to have been several billion dollars. Other car firms are following, up to a point. BMW, Ford, Hyundai and Mercedes-Benz are partners with VW in Ionity. Its fast-charging network hopes to expand from 1,500 points to 7,000 by 2025. Electrify America, set up by VW as part of its settlement with American regulators over its dieselgate emissions-cheating scandal starting in 2015, now has 2,200 fast chargers in the United States. General Motors says it will spend $750m on charging. Its first move will be to install 40,000 points at dealerships.

Specialist charging firms are also expanding quickly. Several have come to public markets during the past year. None of them are profitable, and their revenues are tiny for now, but their market capitalisations are increasing. The most highly valued (at around $7bn) is ChargePoint, which is based in America with 44% of the public-charging market there; it is also expanding in Europe. EVBox, a Dutch firm, has 300,000 points worldwide including a quarter of Europe’s public level-2 chargers and third of fast-charging points. EVgo has half the fast-changing market in America (excluding Tesla). But as Ryan Fisher of BNEF notes, over the next decade, if governments start to cut subsidies, charging companies will have to find business models that reliably produce profits.

A third category is oil companies. Fearful of losing business at petrol stations, they are developing ambitious schemes. After buying ubitricity, a leading European on-street charging firm, in February, Shell, an Anglo-Dutch oil major, said in August that it planned to roll out 500,000 charging points around the world by 2025, both kerbside and fast charging. BP and Total have also been busy buying charging firms. Utilities are making a push, too. Wallbox, part-owned by Spain’s Iberdrola, sells chargers for homes and workplaces. The Electric Highway Coalition, made up of 17 American power companies including Dominion Energy and Duke Energy, plans to install fast charging along intercity routes.

But grave doubts about the speed of the ramp-up persist nonetheless. Instead of the 40m public chargers the IEA reckons the world will need by 2030 to put the industry on course for net-zero by 2050, BCG forecasts that in America, Europe and China, the world’s main EV markets, there will be only 6.5m. The number of cars per charger will thus rise steeply, it reckons.

Governments will certainly act. America’s infrastructure bill will set aside $7.5bn to enable the installation of 500,000 public points by 2030. Mandates such as that recently announced in Britain requiring new homes, workplaces and retail sites to have charging points, adding 145,000 every year, are likely to become more common.

But the numbers are still small relative to the vast scale of charging networks that the world needs. More money will be needed to update electricity grids to distribute power to the new source of demand. A reason for optimism is that improvements in batteries should continue to offer longer ranges, meaning less need for frequent charging. Newer batteries will be capable of being charged much more quickly and chargers will deliver current more swiftly in future. Drivers must cross their fingers and hope that technology delivers, again.

For more expert analysis of the biggest stories in economics, business and markets, sign up to Money Talks, our weekly newsletter.