JANUARY 17th 2015

S P E C I A L R E P O R T

E N E R G Y A N D T E C H N O L O G Y

Let there be light

20150117-ENERGY.indd 1 09/01/2015 11:08

The Economist January 17th 2015 1

ENERG Y AND TECHNOLOG Y

SPECIAL REPOR T

A list of sources is atEconomist.com/specialreports

An audio interview with the author is atEconomist.com/audiovideo/specialreports

CONTENT S

4 RenewablesWe make our own

6 AfricaA brightening continent

7 New business modelsAll change

9 Energy efficiencyInvisible fuel

1

ACAREFULOBSERVERmightnote the chunkydouble glazingon the ele-gant windows and the heat pump whirring outside the basement en-trance. From the outside the five-storey house in London’s posh NottingHill district looks like any other. Inside, though, it is full of new technol-ogies that aim to make it a net exporter of power. They exemplify manyof the shifts now under way that are making energy cleaner, more plenti-ful, cheaper to store, easier to distribute and capable of being used moreintelligently. The house in Notting Hill is a one-off, paid for by its greenmultimillionaire owner. But the benefits of recent innovations can be

reaped by everybody.That makes a welcome

change from the two issues thathave dominated the debate aboutenergy in the past few decades:scarcity and concerns about theenvironment. Modern life is basedon the ubiquitous use of fossil fu-els, all ofwhich have bigdisadvan-tages. Coal, the cheapest and mostabundant, has been the dirtiest,contributing to rising emissions.Oil supplies have been vulnerableto geopolitical shocks and pricecollusion by producers. Naturalgas has mostly come by pipeline—and often with serious politicalbaggage, as in the case of Europe’sdependence on Russia. Nuclearpower is beset by political trou-bles, heightened by public alarmafter the accident at Japan’s Fuku-shima power station in 2011. Re-newables such as wind and solar—beneficiaries of lavish subsidies—

have so far played a marginal role. The main worries were whetherenough energy would be available forpowergeneration, transport, heat-ing, cooling and industry; and if so, whether it would cookthe planet.

Now new factors are in play. Technological change has broken thepower of the Organisation of the Petroleum Exporting Countries (OPEC)to keep the oil price high. Hydraulic fracturing (“fracking”) and horizontaldrilling have turned America into a big oil producer, with 4m barrels aday coming from sources which used to be deemed “unconventional”.The boom in producing oil and gas from shale has yet to spread to othercountries. America enjoys some big advantages, such as open spaces, ac-commodating laws, a well-developed supply chain and abundant fi-nance for risky projects. So far it has refrained from exporting its crude oilor natural gas, but exports of liquefied natural gas (LNG) will start thisyear. Increased trade in LNG will create a more global gas market andgreater resilience of supply, undermining Russia’s pipeline monopoly inEurope. America is already exporting lightly refined oil.

An increase in supply, a surprising resilience in production in trou-bled places such as Iraq and Libya, and the determination of Saudi Ara-bia and its Gulfallies not to sacrifice market share in the face offalling de-mand have led to a spectacularplunge in the oil price, which has fallen byhalf from its 2014 high. This has dealt a final blow to the notion of “peak

Let there be light

Thanks to better technology and improved efficiency, energy isbecoming cleaner and more plentiful—whatever the price of oil,says Edward Lucas

ACKNOWLEDGMENT S

In addition to the people cited inthis report, the author wouldparticularly like to thank SimonDaniel of Moixa, Edward Osterwaldof CEG Europe, Laura Sandys MP andMagda Sanocka of the IEA; alsoThomas Baker, David Gee and FrankKlose of BCG; Satish Kumar, DaveNichol and Cass Swallow ofSchneider Electric; David Walker, WillGifford and Pe����

aessen of DNV; TimEmrich and Jonathan Gaventa.

2 The Economist January 17th 2015

SPECIAL REPOR TENERG Y AND TECHNOLOG Y

2

1

oil”. There is no shortage of hydrocarbons in the Earth’s crust,and no sign thatmankind isabout to reach “peaktechnology” forextracting them. But the fall has created turmoil in financial mar-kets as energy companies lay offworkers and cut or delay invest-ment projects.

The implications are more complicated than the headlinessuggest. Fora start, lowpricesdo not instantlycause supplycurbsor make investment dry up. Even costly projects do not stoppumping when the oil price falls. Fracking is a small-scale busi-ness. New projects can be halted quickly and restarted when theprice picks up. American frackers are now the world’s swing pro-ducers, reacting to price fluctuations in a way that was once theprerogative ofthe Saudis. On a 15- to 25-year time horizon, today’sslide in the oil price needs to be set against the likely long-termtrend. Futuresmarketsare betting that the oil price will be backto$90 per barrel in the early 2020s.

For now, though, low oil prices put money in consumers’pocketsand give a bitofbreathingspace to governments, makingit easier to cut fossil-fuel subsidies (and perhaps even tax carbonemissions). In 2013 some $550 billion was spent on subsidisingfossil fuels, a policy of extraordinary wrongheadedness that fa-vours the rich, distorts economies and aggravates pollution.

A bigger question on many minds is the effect of rock-bot-tom oil prices on the shift towards low-carbon energy. Solar,wind and other renewables have recently benefited from un-precedented investments: an average of $260 billion a yearworldwide over the past five years. Long, and wrongly, decriedas mere boondoggles, they have begun to show real commercialpromise in places as diverse as India, Hawaii, and parts of Africawhere the climate is favourable, costs are low and other sourcesof power are expensive. Renewables capacity is rising even assubsidies are falling. China, for example, has already installednearly half the 200 gigawatts (GW) of wind power it had been

planning for 2020, so it is sharply cuttingback the subsidies it introduced in 2009.

But the relationship is not alwaysstraightforward. Renewable electricitymainly competes with gas- and coal-firedpower stations, not with oil. In NorthAmerica, low oil prices may, paradoxical-ly, lead to higher natural gas prices. Lessfracking means there will be less of the as-sociated gas that is produced along withshale oil. More broadly, much of the sup-port for renewables has been political,and there is little sign that this is changing.Worries about climate change continue toensure that clean energy enjoys strong po-litical support in many developed coun-tries. Whereas shares in oil companieshave in recent months fallen along withthe price, the S&P Global Clean Energy In-dex, which covers the industry’s 30 big-gest listed companies, has barely budged.

The economics—and particularly thewhopping subsidies of the past decadepaid out in countries such as Germanyand Britain—remain contested. Solar andwind are intermittent, so they are trulyuseful only if the power they produce canbe stored; otherwise they need back-upcapacity, typically from fossil-fuel sources.Dieter Helm, an energy expert at OxfordUniversity, says that subsidies for primi-tive green technology, such as the current

generation of solar panels, have been a “colossal mistake”. Itwould have been much better, he argues, to invest in proventechnologies such as electrical interconnectors (linking Britainand Norway, for example) and support research into new kindsof solar power, such as films that can be applied to any outsidesurface and technologies thatuse a widerchunkofthe spectrum.

Bits of the green-energy world are wilting under the impactof low oil prices. Some biofuels have become less attractive. Thesame is true for electric cars, which currently make up less than1% of America’s light-vehicle fleet. Bloomberg New Energy Fi-nance reckons that with petrol at $3.34 a gallon ($0.87 per litre),that share could rise to 9% by 2020. With petrol at $2.09, it wouldgo up to just 6%. At the same time countries and companiesthinking of switching from oil-fired power generation to renew-ables may reconsider. Saudi Arabia, for example, was planningto invest $110 billion in 41 GW of solar capacity by 2032, but maynow want to thinkagain.

Take the long viewYet the long-term trend is clear. In particular solar electric-

ity, and ways of storing it, are getting ever cheaper and better, asthis special report will show. Sanford C. Bernstein, a researchfirm, sees “global energy deflation” ahead. Most of the invest-ment decisions in the fossil-fuel industry are taken a decade ortwo ahead. The International Energy Agency (IEA), an intergov-ernmental organisation often criticised for its focus on fossil fu-els, says the world will need to stump up about $23 trillion overthe next 20 years to finance continued fossil-fuel extraction, butthe prospect of much cheaper solar power and storage capabiliymay put investors off. The story may be not so much what fallingoil prices mean for clean energy than what the prospect of cleanenergy will mean for the oil price.

Old energy industries are changing too. Gas will become

The Economist January 17th 2015 3

ENERG Y AND TECHNOLOG Y

2

1

more abundant and easier to trade. Even coal, the most widelyused and so farmost pollutingfossil fuel, is not inherently dirty. Itdoes not need to be burned but can be cooked instead to producemethane, which can then be used as a fuel or in petrochemicals.Modern coal-fired plants, though pricey, are far cleaner than thebelching monsters of the past. The heat they produce is used, notwasted as in many traditional power plants. The emissions arescrubbed of the oxides (ofnitrogen and sulphur) that eat away atbodies and buildings. In some projects—albeit for now on a tinyscale—the CO2 is also captured for storage or use. Such improve-ments could make coal as relatively clean as other fossil fuels,though they make commercial sense only if the rules are tilted intheir favour. But if the price of such techniques comes down andthe cost ofpollution goes up, clean coal could be competitive.

Nuclear power, in theory, is a source of cheap, dependable,constant electricity. In practice it is too costly for private investorsto back without government guarantees, and its perceived dan-ger makes it unpopular in some European countries and in Ja-pan. One of several flaws in Germany’s Energiewende—suppos-edly a big shift to green technology—was the hurried abandon-ment of the country’s nuclear capacity. Besides, many of theworld’s existing nuclear power stations will have to close in thecoming two decades. Barring a political shift or a technologicalbreakthrough—perhaps in small, mass-produced nuclearplants—it is hard to see the fortunes ofnuclear energy reviving.

Demand for energy is likely to hold up for some time yet,mainly thanks to rapid economic growth in emerging econo-mies. The IEA predicts that over the next 25 years it will rise by37%. Yet increasing efficiency in energy use and changes in be-haviour have meant that the hitherto well-established link be-tween economic growth and energy use is weakening.

More for lessAmerica’s economy, for example, has grown by around 9%

since 2007, whereas demand for finished petroleum productshas dropped by nearly11%. In Germany household consumptionof electricity is now lower than it was in 1990. Global demandused to rise by 2% a year, but the rate is slowing. Even emissionsin China, the world’s largest and dirtiest energy consumer, maypeak by 2030, thanks to huge investments in new clean-coalpower generation, nuclear and renewable energy and long-dis-tance transmission lines. Simon Daniel, an energy expert, sees

two conflicting trends: on one hand greater efficiency, local pro-duction and storage, on the other increased consumption fromthe billions of new devices that will be hooked up to the “inter-net of things”.

On current form the emissions from oil, gas and coalwould, on most models, make it impossible to keep the rise inglobal temperatures below 2˚C by the year 2100; the most likelyoutcome would be a 4˚C rise, which has prompted calls for mostof the world’s remaining hydrocarbons to be left in the ground.The IEA estimates the investmentneeded for“decarbonising” fu-ture electricity production alone at an astounding $44 trillion.The best hope of avoiding that much warming is a huge increasein energy efficiency.

One bigcomponent ofthat taskwill be to adapt the existingstock of buildings. Amory Lovins, one of the foremost prophetsof energy efficiency and founder of Rocky Mountain Institute, athink-tank and consultancy based in Colorado, believes that thescope for improvement remains huge. He has long preached thatproper building design and energy storage can eliminate theneed for air-conditioning and space heating in most climates,and illustrates this by growing bananas in his own house, on awindswept mountainside in Colorado where winter tempera-tures can drop to –44˚C. Eliminating the heating system for hishouse, he says, saved more money than he spent on insulationand fancy windows. His optimism is slowly winning converts.

Despite all the obstacles, prettymuch all the technology theworld needs for a clean, green future is already available. As A.T.Kearney, a consultancy, notes in a recent report for the World En-ergy Council, a think-tank: “Energy-efficiency potentials com-bined with renewable-energy sources and shale-gas potentialsprovide an abundance of energy that can be made accessiblewith currently available technologies.”

Transmission costs for electricity are plunging, thanks tosolid-state technology, which makes efficient direct-current cir-cuitry safer and more flexible. Power grids which were previous-ly isolated can now be connected: one audacious plan involves a700-mile, £4 billion ($6 billion) linkbetween Britain and Iceland.Such projects are costly up front, but offer big long-term savingsfrom cheaper power, better storage and increased resilience.

More effective management ofsupply and demand also of-fers scope forbigsavings, as this special reportwill show. Sensorscan nowcollectvastamountsofdata aboutenergyuse, and com-

puter power and algorithms can crunchthat information to offer incentives to cus-tomers to curb consumption at peak timesand increase itwhen demand is low. At thesame time business models which canturn a profit from thrifty energy use are de-veloping, and capital markets are wakingup to their potential.

That splendidly energy-efficienthouse in Notting Hill demonstrates justhow much can be done right now, even ifit does not yet come cheap. Its owner, Mi-chael Liebreich, founded a business calledNew Energy Finance, which he sold toBloomberg, a financial-information com-pany, in 2009. He has spent tens of thou-sands of pounds on making his homethrifty, resilient and productive.

The house is no stranger to energyrevolutions. In 1865 its original builders in-stalled a state-of-the-art delivery and stor-age system: a coal hole in the pavement,sealed bya handsome cast-iron hatch. Gas

SPECIAL REPOR T

4 The Economist January 17th 2015

SPECIAL REPOR TENERG Y AND TECHNOLOG Y

2

1

AT FIRST SIGHT the story of renewable energy in the richworld looks like a waste of time and money. Rather than in-

vesting in research, governments have spent hundreds of mil-lions of pounds, euros and dollars on subsidising technologythat does not yet pay its way. Yet for all the blunders, renewablesare on the march. In 2013 global renewable capacity in the powerindustry worldwide was1,560 gigawatts (GW), a year-on-year in-crease of more than 8%. Of that total, hydropower accounted forabout 1,000GW, a 4% rise; other renewables went up by nearly17% to more than 560GW. True, aftereight years ofcontinuous in-crease, the amount invested dropped steeply in 2012 amid uncer-tainty about future subsidies and investment credits. But thanksto increased efficiency, less money still bought more power.

Measuring progress is tricky: the cost of electricity fromnew solar systems can vary from $90 to $300 permegawatt hour(MWh). But it is clearly plummeting. In Japan the cost of powerproduced by residential photovoltaic systems fell by 21% in 2013.As a study for the United Nations Environment Programmenotes, a record 39GW of solar photovoltaic capacity was con-structed in 2013 at a lesser cost than the 2012 total of 31GW. In theEuropean Union (EU), renewables, despite a 44% fall in invest-ment, made up the largest portion (72%) of new electric generat-ing capacity for the sixth year running.

The clearest sign of health in the renewables market issmoke-clogged China, which in 2013 invested over $56 billion,more than all of Europe, as part of a hurried shift towards cleanenergy. China’s investment included 16GW of wind power and

13GW ofsolar. The renewable-power capacity China installed inthat year was bigger than its new fossil-fuel and nuclear capacityput together.

Whetherornot it represents good value formoney in all cir-cumstances at the moment, renewable energy has become a se-rious part of the energy mix. In 2013 Denmark’s wind turbinesprovided a third of the country’s energy supply and Spain’s afifth. Some worries are abating. Though power from solar andwind is intermittent, nature often cancels out the fluctuations:sunny days tend not to be windy, and vice versa.

Both forms ofgeneration have their fans, but solar seems tobe pulling ahead of wind. Wind technology is running upagainst the laws of physics: it is hard to see great new gains in sit-ing, or in the design of bearings and blades. And wind turbinesare widely considered unsightly and noisy. Solar panels, by con-trast, can be surprisingly attractive. Instead of featuring serriedranks ofblackrectangles, the latest designs looklike glittering au-tumn leaves captured in glass.

Solar flareThe main reason for the growth in solar energy, though, is

innovation, not aesthetics. It comes in two forms. The smaller(accounting for around a tenth of existing solar capacity) is ther-mal storage, in which sunlight is concentrated as heat, for exam-ple in molten salt. That can be used to produce steam for powerturbines. After some slackyears this form ofrenewable energy isenjoying a renaissance.

Investment in the second, more widespread form of solarenergy, electricity produced by photovoltaic (PV) cells, fell backin 2013 after ten years when average annual growth was around5����et in the same year total global capacity added in solarelectricity exceeded that in wind for the first time. Solar received53% of the $214 billion invested worldwide in renewable powerthat year. It still provides only a sliver of the world’s energy, andeven by 2020 it will make up just 2% of global electricity supply.But the pace of change is remarkable, with more solar capacityinstalled since 2010 than in the previous four decades.

Along with worries about pollution from other fuels, thebiggest boost to solar—both in the rich and the emergingworld—is its plummeting cost. In a recent report on solar electric-ity the IEA noted that the cost of solar panels had come down bya factoroffive in the past sixyears and the cost offull PV systems,which include other electronics and wiring, by three. The “level-ised cost” (the total cost of installing a renewable-energy systemdivided by its expected energy output over its lifetime) of elec-tricity from decentralised (small-scale) solar PV systems in some

Renewables

We make our own

Renewables are no longer a fad but a fact of life,supercharged by advances in power storage

and then electric lighting, central heatingand hot water came lat-er. But the revolution under its current owner is the biggest yet.Despite the airtight insulation the rooms feel airy. Specially de-signed chimney cowls suck stale, moist air from the house whilea heat exchanger keeps the thermal energy indoors.

The house now requires remarkably little input of energy.Gas and electricity bills for a dwelling of this size would normal-ly run to at least £3,500 ($5,500) a year, but once everything is inplace the owner expects not only to spend nothing but to receivea net payment for the electricity he produces. On the roof is alarge array of solar panels which deliver two kilowatts (kW) ofelectricity on sunny days. Another source of power is a 1.5kWfuel cell in the former coal bunker. It runs on gas, with over 80%efficiency—far more than a conventional power station or boiler.The electricity from these two sources powers the household’s(ultra-frugal) domestic appliances and its low-energy lighting, aswell as a heat pump (a refrigerator in reverse) that provides un-derfloor heating. A water tank stores surplus heat. Spare electric-ity is fed back into the grid.

Mr Liebreich does not claim that his house is easily copied,but he insists that through “thinning mist” the future is visible.“The only things that are inherently costly are the thermody-namic process and resource depletion—for everything else costshave come down, are coming down and will come down in fu-ture,” he says. In short, most of the forces changing the energymarket are pushing in the right direction. 7

The Economist January 17th 2015 5

ENERG Y AND TECHNOLOG Y

2

1

markets is “approaching or falling below the variable portion ofretail electricity prices”, says the report. The IEA expects the costof solar panels to halve in the next 20 years. By 2050, it predicts,solar will provide 16% of the world’s electric power, well up fromthe 11% it forecast in 2010. At times ofpeakdemand in places suchas Hawaii, where electricity would otherwise come from oil-fired power stations, solar electricity produced by unsubsidisedlarge installations is already competitive. Sanford C. Bernstein, aresearch firm, reckons that in the right conditions solar, mea-sured by thermal units produced, is already cheaper than bothoil and Asian LNG, despite the recent dip in the oil price.

Paint me a power stationSuch forecasts are largely based on existing technologies.

Newsolar technology, known as “third generation”, stacks layersof photovoltaic material to capture a much broader section ofthe spectrum, including invisible parts such as infra-red. Suchcells could be printed from graphene (an ultra-light form of car-bon) on a 3D printer. There will no longerbe a need for solarpan-els on rooftops. Instead, any man-made surface could be turnedinto a solar panel with films and paint. In a pilot project in theNetherlands, solar electricity is being generated by a newly builtroad. Dieter Helm, the Oxford-based energy expert cited earlier,believes that solar power will become so cheap that energy willno longer be seen as scarce.

Other forms of “distributed” generation which providepower for flexible local use and storage are also coming up fast.Domestic fuel cells, for instance, are common in energy-hungryJapan. Such fuel cells can run off the gas grid. Its pipes, notes Da-vid Crane, the boss of NRG, an American power company, aresimpler, cheaper and less vulnerable to rough weather than thepoles and wires of the electric grid. Households can turn theirgas into electricity on the spot. That may end up cheaper and

more reliable. Some of that gas could come from

waste products instead of fossil sources.America’s oldest brewery, Yuengling inPennsylvania, has installed a combined-heat-and-power (CHP) plant, fuelled bymethane produced from waste, whichprovides 20% of the brewery’s energyneeds. In Ukraine, which is trying to be-come independent of Russian natural-gassupplies, the European Bank for Recon-struction and Development is financing a2.25MW biogas plant at a sugar refinerynear Kiev. In Britain the first self-poweredsewage works came into operation in Oc-tober 2014, at a saving of £1.3m ($2m) ayear. And biogas now accounts for one-tenth of gas consumption in China, where42m households turn theiranimal and hu-man waste into methane.

Cost apart, the biggest problem withrenewables has always been storing theelectricity they produce. That gave a bigadvantage to incumbent power compa-nies, which could afford large capital in-vestments in generation and storage. Fordomestic consumers, the power producedfrom solar panels on the roof is of limiteduse if they cannot store it, because theystill have to buy from the grid in the eve-ning when they need it most. But if inter-mittent energy can be stored, its econom-

ics are dramatically improved: the cost of installing capacityremains the same but the cost per kilowatt hour shrinks.

The easiest storage is someone else’s. In regimes with “netmetering” rules, common in some green-minded places includ-ing43 American states, the energy utility is obliged to buy renew-able power from small-scale producers at the same price atwhich it sells its own electricity. That is a startlingly good deal forthe producer, less so for the company. But it applies only to smallamounts ofpower and is unlikely to last.

Meanwhile breakthroughs in storage are creating other op-tions. Businesses and households can store cheap, home-gener-ated electricity as thermal energy. An American company calledIce Bear sells a device which makes ice at night with cheap elec-tricity (and in cooler temperatures), then uses it to cool air in thedaytime, saving energy and money.

All these technologies are becoming cheaper and morepractical, and in some countries are boosted by generous subsi-dies. Germany rebates 30% (an average of €3,300, or $4,000) onthe cost of a solar-plus-battery household system, and offerslow-interest credit for the rest. California has legislation in placeunder which a third of its energy must come from renewable re-sources by 2020. The state has told its three large utilities to pro-vide 1.3GW of storage capacity. Around 85MW of this is likely tobe used by small providers with solar panels.

Bloomberg New Energy Finance (BNEF) has done the sumsfor a German household planning to install a 5kW solar systemand a battery with 3kWh of storage capacity at a cost of around€18,000 ($22,000). The solar panels would cut the household’spower consumption by roughly 30%; adding the storage systemcould increase the savingto asmuch as80%. At the currentcost ofthe equipment, and assuming no rise in electricity prices, the re-turn on the investment would be barely 2%. But on more realisticassumptions—a continuing rise in electricity prices of 2% a year

Cost apart,the biggestproblemwithrenewableshas alwaysbeenstoring theelectricitytheyproduce

SPECIAL REPOR T

6 The Economist January 17th 2015

SPECIAL REPOR TENERG Y AND TECHNOLOG Y

2

1

FOR THE WORLD’S1.2 billion poorest people, who are fac-ing a long and perhaps endless wait for a connection to

mains electricity, solar power could be the answer to their pray-ers. A further 2.5 billion are “underelectrified”, in developmentparlance: although connected to the grid, they can get only unre-liable, scanty power. That blights lives too. The whole of sub-Sa-haran Africa, with a population of 910m,consumes only 145 terawatt hours of elec-tricity a year—less than the 4.8m peoplewho live in the state of Alabama. That isthe pitiful equivalent of one incandescentlight bulb per person for three hours a day.

In the absence ofelectricity, the usualfallback is paraffin (kerosene). Lightingand cooking with that costs poor peoplethe world over $23 billion a year, of which$10 billion is spent in Africa. Poor house-holds are buying lighting at the equivalentof $100 per kilowatt hour, more than ahundred times the amount people in richcountries pay. And kerosene is not just ex-pensive; it is dangerous. Stoves and lampscatch fire, maiming and killing. Indoor

fumes cause 600,000 preventable deaths a year in Africa alone.But candles or open fires are even worse—and so is darkness,which hurts productivity and encourages crime.

Africa’s population will nearly double by 2040. The electri-cal revolution nowunderwaythere, and in otherpoorbut sunnyplaces, is coming just in time for all those extra people. It is basedon three big technological changes, all reinforcing each other.

The first is the collapsing cost of solar power. The second isthe fall in the price of light-emitting diodes (LEDs). These turnelectrical power almost wholly into light. Traditional bulbs arefragile and emit mostly heat. The new LED lamps are not onlybright and durable but now also affordable. But lamps are need-ed at night, and solar power is collected in the daytime. So thethird, crucial revolution is in storage.

Fleecing the poorAll in all, the capacity needed to produce a watt of solar

power (enough to run a small light), which in 2008 cost $4, hascome down to $1. The simplest solar-powered lamps cost around$8. That is still a lot for people with very little money, but the sav-ing on kerosene makes it a good investment. Better light enablespeople to study and work in the evening. As well as powering alamp, a slightly larger solar system can charge a mobile phone,for which users in poor countries often pay extortionateamounts. Russell Sturm of the International Finance Corpora-tion (IFC), the part of the World Bank group that works with theprivate sector, cites kiosks in Papua New Guinea where custom-ers pay for each bar of charge shown on the phone’s screen—at acost than can easily reach a stonking $200 per kWh.

Sales of devices approved by the IFC/World Bank’s Light-ing Africa programme are nearly doubling annually, bringing so-lar power to a cumulative total of 28.5m Africans. In 2009 just1%ofunelectrified sub-Saharan Africans used solar lighting. Now itis nearly 5%. The IEA rather cautiously estimates that, thanks tosolar power, 500m people who are currently without electricitywill have at least 200 watts per head by 2030.

But lighting and charging phones are only the first rungs onthe ladder, notes Charlie Miller ofSolarAid, a charity. Radios caneasily run on solar power. Bigger systems can light up a school orclinic; a “solar suitcase” provides the basic equipment needed byhealth workers. A Ugandan company called SolarNow has a$200 low-voltage television set that runs on the direct current(DC) used by solar systems. A British-designed fridge called SureChill needs only a few hours of power a day to maintain a con-stant 4oC. A company in South Africa has just launched solar-powered ATMs for rural areas with intermittent mains power.

Other companies offer bigger systems, for $1,500 and up-wards, which can power “solar kiosks” and other installations

Africa

A brighteningcontinentSolar is giving hundreds of millions of Africans accessto electricity for the first time

and a big fall in the cost of solar capacity and storage—the rate ofreturn could be a juicy10% or more.

The Gigafactory, which will build batteries for the Teslaelectric-car company, aims to cut the cost of battery storage to-wards what many see as a crucial benchmark: $100 per kWhagainst $250 now. That will bring the price ofan electric car closeto parity with that ofa conventional one, maybe even before theend of this decade, hints Elon Musk, the CEO. But better batterieswill have other advantages too. One is that electric cars, whennot being driven (which is 95% of the time, research suggests),can be used for storage. And batteries that are beingdisplaced bymore efficient versions can play a part in domestic storage.

The storage business is booming. Navigant, a consultancy,reckons that in 2014 alone projects amounting to 363MW wereannounced. BNEF estimates thatby2020 some 11.3GW will be in-stalled, 80% of it in America (chiefly California), Germany, Japanand South Korea, and that investment in storage by then will berunning at $5 billion a year.

The biggest advantage of storage is that it dispenses withthe most inefficientpartofthe power industry: the generating ca-pacity that is held in reserve to meet peaks of demand. In thestate ofNew�ork, for example, one-fifth of the generating capac-ity runs for less than 250 hours a year. By some counts, a mega-watt of storage replaces roughly ten megawatts of such generat-ing capacity—an irresistible saving.

In rich countries new forms of storage and generation areeatingaway at the model that has sustained the electricity indus-try since the days of Thomas Edison. In parts of the developingworld where there are no incumbents, they offer the best chancefor whole populations to get any power at all. 7

The Economist January 17th 2015 7

ENERG Y AND TECHNOLOG Y

1

that enable people to start busi-nesses. Beefed up a bit more,these systems can replace dieselgenerators that will powerstores and workshops, millgrain, run an irrigation pump orpurify water. At an even largerscale they become mini-grids. A$500,000 aid-funded project inKisiju Pwani, once one of thepoorest villages in Tanzania,uses 32 photovoltaic solar pan-els and a bank of120 batteries toprovide 12kW of electricity,enough for 20 street lights and68 homes, 15 businesses, a port,the village’s government officesand two mosques.

Three main problems have yet to be resolved. One is quali-ty. Poor consumers mulling a $100 investment need to be surethat their purchase will be robust. The IFC and other aid outfitsare running a scheme to verify manufacturers’ claims. Second,makers of mass-market appliances, used to mains electricity,have been slow to rejig their products to run on the low-voltagedirect current (DC) produced by renewable energy sources andbatteries. Mr Sturm says the industry is waiting for the “holygrail”: a cheap, efficient and reliable DC fan.

The third and biggest constraint is working capital. It typi-cally takes five months from paying the manufacturer to gettingpaid by the customer. Some companies are coming up with inge-nious hire-purchase schemes for bigger systems to spread thecost. Others offer “solar as a service”, where the customer paysmonthly for the power, with maintenance thrown in.

Some experts see solar as a second-best solution. It can im-prove lives but not power an economy. But grid connections inpoor countries are scarce and unreliable, and developing themwould take too long, especially in remote rural areas where thepoorest live. Besides, the power industry’s old business model ofdelivering through the grid over longdistances is in retreat every-where, including in rich countries. 7

THE BASIC MODEL of the electricity industry was to sendhigh voltages over long distances to passive customers.

Powerstationswere bigand costly, builtnext to coal mines, ports,oil refineries or—for hydroelectric generation—reservoirs. Manyof these places were a long way from the industrial and popula-tion centres that used the power. The companies’ main concernwas to supply the juice, and particularly to meet peaks in de-mand. Most countries (and in America, regions) were energy is-lands, with little interconnection to other systems.

That model, though simple and profitable for utilities andgenerators, was costly for consumers (and sometimes taxpay-ers). But it is now changing to a “much more colourful picture”,says Michael Weinhold of Siemens, a big German engineering

company. Not only are renewables playing a far bigger role;thanks to new technology, demand can also be tweaked tomatch supply, not the other way round.

As a result, the power grid is becoming far more complicat-ed. It increasingly involves sending power at low voltages overshort distances, using flexible arrangements: the opposite of thetraditional model. In some ways the change is akin to what hashappened in computing. A 2010 report for BCG, a consultancy,drew a parallel with the switch from mainframes and terminalsto cloud storage and the internet.

Traditional power stations and grids still play a role in thisworld, but not a dominant one. They have to compete with newentrants, and with existing participants doing new things. Oneexample is the thriving business of trading what Mr Lovins ofRocky Mountain Institute has named “negawatts”: unused elec-tricity. The technique is known as “demand response”—adjust-ing consumption to meet supply, not the other way round.

Flattening the peaksThe most expensive electricity in any power system is that

consumed at peak time, so instead of cranking up a costly andprobably dirty power station, the idea is to pay consumers toswitch off instead. For someone running a large cooling, heatingor pumping system, for example, turning the power off for ashort period will not necessarily cause any disruption. But forthe grid operator the spare power gained is very useful.

This has been tried before: in France, after a heatwave in2003 that hit the cooling systems of nuclear power stations andled to power shortages. In response, big energy consumersagreed to cut their power consumption at peak times, in ex-change for generous rebates. The Japanese have installed200,000 home energy-management systems that do somethingsimilaron a domestic scale. Butnewtechnologytakes it to anoth-er level, allowing a lot of small power savings from a large num-ber ofconsumers to be bundled together.

In South Africa companies can sell such spare power them-selves, through a companycalled Comverge. Elsewhere consum-ers earn rebates either from their own power company or from athird-party broker which manages their consumption. In Austin,Texas, for example, 7,000 households have signed up for ascheme in which they get an $85 rebate on an internet-enabledthermostat, such as the Nest, which costs $249. This has otherbenefits for them too, such as allowing them to control theirhome heating and cooling remotely. But it also means that thepower company, Austin Energy, can shave 10MW from its sum-mer peakdemand, typically between 3pm and 7pm.

Nest is selling its programmes all over North America, andmore recently in Britain too. Customers of its “Rush Hour Re-wards” programme can choose between being given notice aday in advance of a two- to four-hour “event” (meaning theirthermostat will be turned down or up automatically) or beingtold ten minutes ahead ofa 30-minute one. This can cut the peakload by as much as 55%. In another scheme customers agree to achange ofa fraction ofa degree over a three-weekperiod.

At an auction in May 2014 at the PJM interconnection,America’s largest wholesale electricity market, 11GW of “nega-watts” were bid and cleared, replacing capacity that would havecome from conventional power stations. In other words, insteadof buying in capacity from power stations that operate only tomeetpeakdemand, itwaspayingcustomersnot to use electricityat that time. In 2013 PJM took $11.8 billion off electricity billsthrough demand response and related efficiency savings. Thefigure for 2014 is likely to be $16 billion.

NRG, America’s biggest independent power company, isalso moving into the market. David Crane, its chiefexecutive, ad-

New business models

All change

The power industry’s main concern has always beensupply. Now it is learning to manage demand

SPECIAL REPOR T

8 The Economist January 17th 2015

SPECIAL REPOR TENERG Y AND TECHNOLOG Y

2

1

mits that some consumers find the idea of saving power “un-American”, but thinks that for companies like his the “mindlesspursuit of megawatts” is a dead end. In 2013 NRG bought a de-mand-response provider, Energy Curtailment Specialists, whichcontrols 2GW of“negawatts”, for an undisclosed sum.

The big question for demand-response companies is theterms on which they compete with traditional generators, whichargue that markets such as PJM are starving the power system ofbadly needed investment. For example, FirstEnergy, a companyin Ohio, suspended modernisation plans at a coal-fired plantwhich failed to win any megawatts in the auction for 2017-18.Such plants are viable only if utilities are paying top dollar forpeak electricity—a cost which is eventually passed on to the con-sumer. Companies like FirstEnergy hope that the Supreme Courtwill overturn a rulingby the Federal EnergyRegulatory Commis-sion that negawatts be treated like megawatts in capacity auc-tions. These worries are already spooking the market. EnerNOC,which bundles together small energy savings from many differ-ent customers to offer negawatts, has seen its share price fall byhalf since May.

Sara Bell, who represents Britain’s demand-response com-panies, notes a market failure: the same companies that generatepower also supply it. She argues that their interest is selling atpeak demand and peak prices—which is the opposite of what acustomer would want.

In any case, the days of the vertically integrated model ofenergy supply are numbered, observes Dieter Helm. Thanks toabundant solarpower, he argues, the energy market increasinglyresembles the economics of the internet, where marginal costsare zero. That “undermines the very idea ofwholesale electricitymarkets”. The future model will be much more fragmented. In-dependent generators, plus new entrants, are already “revolutio-nising the way electricity is sold and used”; new technologieswill make the 21st-century model even more different. “No won-dermanyofthe energygiantsofthe pastare already in such trou-ble,” he says.

No longer so usefulThe combination of distributed and intermittent genera-

tion, ever cheaper storage and increasingly intelligent consump-tion has created a perfect storm for utilities, particularly those inEurope, says Eduard Sal��edruna of IHS, a consultancy. Theyare stuckwith the costsofmaintaining the grid and meeting peakdemand, but without the means to make customers pay for itproperly. Their expensively built generating capacity is over-sized; spare capacity in Europe thiswinter is100GW, or19% oftheconstituent countries’ combined peak loads. Much of that is

mothballed and may have to bewritten off. Yet at the same timenew investment is urgentlyneeded to keep the grid reliable,and especially to make sure itcan cope with new kinds ofpower flow—from “prosumers”back to the grid, for example.

To general surprise, de-mand is declining as power isused more efficiently. Politiciansand regulators are unsympa-thetic, making the utilities payfor electricity generated by oth-er people’s assets, such as roof-top solar, to keep the greens hap-py. At the same time barriers toentry have collapsed. New ener-

gy companies do not need to own lots of infrastructure. Theircompetitive advantage rests on algorithms, sensors, processingpower and good marketing—not usually the strong points of tra-ditional utilities. All the services offered by these new entrants—demand response, supply, storage and energy efficiency—eatinto the utilities’ business model.

For an illustration, look at Hawaii, where solar power hasmade the most inroads. On a typical sunny day, the panels onconsumers’ rooftops produce so much electricity that the griddoesnotneed to buyanypowerfrom the oil-fired generators thathave long supplied the American state. But in the morning andevening those same consumers turn to the grid for extra electric-ity. The result is a demand profile that looks like a duck’s back, ris-ing at the tail and neckand dipping in the middle.

The problem for the state’s electricity utilities is that theystill have to provide a reliable supply when the sun is not shining(it happens, even in Hawaii). But consumers, thanks to “net me-tering”, may have an electricity bill of zero. That means the util-ities’ revenues suffer, and consumers without solar power (gen-erally the less well-off) cross-subsidise those with it.

Rows about this are flaring across America. The HawaiianElectric Power Company, the state’s biggest utility, is trying to re-strict the further expansion of solar power, telling new consum-ers that theyno longerhave an automatic right to feed home-gen-erated electricity into the grid. Many utilities are asking reg-ulators to impose a fixed monthly charge on consumers, ratherthan just let them pay variable tariffs. Since going completely off-grid still involves buying a large amount of expensive storage,the betting is that consumers will be willing to pay a monthly feeso they can fall backon the utilities when they need to.

Consumers, understandably, are resisting such efforts. InArizona the utilities wanted a $50 fixed monthly charge; the reg-ulator allowed $5. In Wisconsin they asked for $25 and got $19.Even these more modest sums may help the utilities a bit. But thebigger threat is that larger consumers (and small ones willing tojoin forces) can go their own way and combine generation, stor-age and demand response to run their own energy systems, of-ten called “microgrids”. They may maintain a single high-capaci-ty gas or electricity connection to the outside world for safety’ssake, but still run everything downstream from that themselves.

The Economist January 17th 2015 9

ENERG Y AND TECHNOLOG Y

2

1

THE CHEAPEST AND cleanest energy choice of all is not towaste it. Progress on this has been striking yet the potential

is still vast. Improvements in energy efficiency since the 1970s in11 IEA member countries that keep the right kind of statistics(America, Australia, Britain, Denmark, Finland, France, Ger-many, Italy, Japan, the Netherlandsand Sweden) saved the equiv-alent of 1.4 billion tonnes of oil in 2011, worth $743 billion. Thissaving amounted to more than their total final consumption inthat year from gas, coal or any other single fuel. And lots ofmon-ey is being invested in doing even better: an estimated $310 bil-lion-360 billion was put into energy efficiency measures world-wide in 2012, more than the supply-side investment inrenewables or in generation from fossil fuels.

The “fifth fuel”, as energy efficiency is sometimes called, isthe cheapest of all. A report by ACEEE, an American energy-effi-ciency group, reckons that the average cost of saving a kilowatthour is 2.8 cents; the typical retail cost of one in America is 10cents. In the electricity-using sector, saving a kilowatt hour cancostas little asone-sixth ofa cent, saysMrLovinsof RockyMoun-tain Institute, so paybackcan be measured in months, not years.

The largest single chunk of final energy consumption, 31%,is in buildings, chiefly heating and cooling. Much of that is wast-ed, not least because in the past architects have paid little atten-tion to details such as the design ofpipework(long, narrow pipeswith lots of right angles are far more wasteful than short, fat andstraight ones). Energy efficiency has been nobody’s priority: ittakes time and money that architects, builders, landlords andtenants would rather spend on other things.

In countries with no tradition of thrifty energy use, theskills needed are in short supply, too. Even the wealthy, know-ledgeable and determined Mr Liebreich had trouble getting thebuilders who worked on his energy-saving house to take his in-structions seriously. Painstakingly taping the joins in insulatingboards, and the gaps around them, seems unnecessary unlessyou understand the physics behind it: it is plugging the last fewleaks that brings the biggest benefits. Builders are trained to wor-ry about adequate ventilation, but not many know about themarvels ofheat exchangers set in chimney stacks.

Snug as a bug in a rugFor new buildings, though, energy efficiency is becoming

an important factor. The 99-storey Pertamina Energy Tower be-ingbuilt in Jakarta, forexample, will be so thrifty that wind, solarand geothermal energy can meet all its power needs. “Energy-plus” buildings can even harvest energy from their environmentand inhabitants and export it. Like cars, new buildings are typi-cally much more efficient than the ones they replace.

But old cars are scrapped more often than old houses. Thebiggest problem in energy efficiency is adapting existing build-ings. Circle Housing, a large British housing association, has65,000 dwellings, with tenants whose incomes are typically be-low £20,000 ($32,000) a year, low by British standards. Annualenergybills in the worstproperties can be a whopping£2,000, soCircle is knocking some of the houses down. Their replacementsbring the bills down to about £450. In new “passive houses”,

Energy efficiency

Invisible fuel

The biggest innovation in energy is to go without

Some organisations, such as military bases, may have spe-cific reasons to want to be independent of outside suppliers, butformost ofthem the main motive is to save money. The Universi-tyofCalifornia, San Diego (UCSD), forexample, which until 2001had a gas plant mainly used for heating, changed to a combined-heat-and-power (CHP) plant which heats and cools 450 build-ingsand provideshotwater for the 45,000 people who use them.The system generates 92% of the campus’s electricity and saves$8m a year. As well as 30MW from the CHP plant, the universityhas also installed more than 3MW in solar power and a further3MW from a gas-powered fuel cell. When demand is low, thespare electricity cools 4m gallons (15m litres) of water for use inthe air-conditioning—the biggest load on the system—or heats itto 40˚ to boost the hot-water system. Universities are ideal forsuch experiments. As autonomous public institutions they areexempt from fiddly local rules and from oversight by the utilitiesregulator. And they are interested in new ideas.

Places like UCSD not only save money with their micro-gridsbutadvance research aswell. Aserveranalyses84,000 datastreams every second. A company called ZBB Energy has in-stalled innovative zinc-bromide batteries; another company istrying out a 28kW supercapacitor—a storage device far faster andmore powerful than any chemical battery. NRG has installed arapid charger forelectric vehicles, whose past-their-prime batter-ies are used to provide cheap extra storage. And the universityhas just bought 2.5MW-worth of recyclable lithium-ion iron-phosphate battery storage from BYD, the world’s largest batterymanufacturer, to flatten peaks in demand and supply further.

In one sense, UCSD isnota good customerfor the local utili-ty, San Diego Gas & Electric. The microgrid imports only 8% of itspower from the utility. But it can help out when demand else-where is tight, cutting its own consumption by turning down air-conditioners and other power-thirsty devices and sending thespare electricity to the grid.

UCSD is one of scores of such microgrids pioneering newways of using electricity efficiently and cheaply through betterdesign, data-processing technology and changes in behaviour.The IEA reckons that this approach could cut peak demand forpower in industrialised countries by 20%. That would be goodfor both consumers and the planet. 7

SPECIAL REPOR T

built from mass-produced pre-fabricated energy-efficientcomponents at roughly thesame cost as ordinary ones,bills fall to £350. The inhabit-ants have to get used to notopening windows, whichwastes heat and upsets the ven-tilation system. But Europe al-ready has 30,000 such build-ings, and more are on the way.

For Circle’s existing stock,with bills averaging £1,240 ayear, “energy champions”—ten-ants who are trained to helpothers with similar housingand lifestyles—offer simple tips(switching off appliances, turn-

ing down thermostats) that save an average of £250 a year. Help-ing tenants shop around for good deals on gas and electricitysavesanother£150. Butafter that it getsmuch more expensive. Re-fitting a house with double glazing, cavity-wall and loft insula-tion, a heat pump and an energy-efficient boilermay save anoth-er £150, but requires an investment of £3,000 to £8,500. In mosthouses and offices, saving £3 a week is not worth a lot of hassle.Tenants do not want to invest in properties they do not own, andlandlords do not really care how much their tenants pay for theirenergy. Besides, better insulation may simply mean that peoplewear lighter clothes indoors rather than turn the heating down.

One answer to this market failure is to bring in mandatorystandards for landlords and those sellingproperties. Another involves energy-ser-vice companies, known as ESCOs, whichguarantee lower bills in exchange formodernisation. The company can devel-op economies of scale and tap financialmarkets for the upfront costs. The savingsare shared with owners and occupiers. ES-COsare alreadya $6.5 billion-a-year indus-try in America and a $12 billion one in Chi-na. Both are dwarfed by Europe, with €41billion ($56 billion) last year. Navigant Re-search, the consultancy, expects this todouble by 2023.

That highlights one ofthe biggest rea-sons for optimism about the future of en-ergy. Capital markets, frozen into cautionafter the financial crash of 2008, are nowdoing again what they are supposed to do:financing investments on the basis of fu-ture revenues. The growth ofa bond market to pay for energy-ef-ficiency projects was an encouraging sign in 2014, when $30 bil-lion-40 billion were issued; this year’s total is likely to be $100billion.

Solar energy is now a predictable income stream drawingin serious money. A rooftop lease can finance an investment of$15,000-20,000 with monthly payments that are lower than thecustomer’s current utility bill. SolarCity, an American company,has financed $5 billion in new solar capacity, raising money ini-tially from institutional investors, including Goldman Sachs andGoogle, butnowfrom individual private investors—who also be-come what the company calls “brand ambassadors”, encourag-ing friends and colleagues to install solar panels too.

The model is simple: SolarCity pays for the installation,then bundles the revenues and sells a bond based on the expect-

ed future income stream. Matu-rities range from one to sevenyears. The upshot is that the costof capital for the solar industryis 200-300 basis points lowerthan that for utilities.

A virtuous circle is emerg-ing which is confounding thedoomsters. It rests on five ele-ments. The first is abundant en-ergy, above all from new solartechnology: a sliver now, butalso a dagger in the heart of thefossil-fuel industry. The grid par-ity which Hawaiian rooftops of-fer today will be possible inmany more locations in future—and not just on rooftops in directsunlight, but from any surface indaytime. That shapes the futureinvestment climate. The price offossil fuels will always fluctuate.Solar is bound to get cheaper.

The second part of the cir-cle is storage. Batteries are get-ting cheaper, more powerfuland more prevalent, for exam-ple in electric cars. So, too, areother ways of storing energy,such as warm water and ice.That deals with the biggest dis-advantage of solar power, its in-termittent nature. Some of this may be achieved through big in-terconnectors that can shift power to countries with the rightgeography for hydro-electric generation. But even more impor-tant may be the aggregation of lots ofsmall-scale storage.

That reflects the third element: distribution. Consumers arenow in a position to be small producers and storers of energy.Thatcreates resilience in the network, alongwith greater efficien-cy and more innovation. Perhaps fuel cells will become smallerand cheaper, makingup a networkofmicropowerstations wher-ever the gas pipelines run. Perhaps they will remain toys for therich. But whereas innovation in the power network of the past—

big, centralised and regulated—was slow, in the new, decentral-ised grid of the future it will move ever faster.

The fourth part of the circle is intelligence. The internet hasmade it possible for its users to generate, store and manage dataefficiently. Now processing power and algorithms will do thesame for electricity. Whether that comes from smart meterswhich manage consumption in the home or from individualsmart devices programmed to maximise theirefficiency remainsto be seen. Given the riskofcyber-attacks, security will need seri-ous thought. But overall the grid is getting smarter, not dumber.

The fifth and final part is finance. Business models for newenergy systems are now proven, both in the rich world and inemerging economies. A wave of money is breaking over the oldmodel, sweeping away incumbents. If they and their friends ingovernment try to hold it back, everyone will suffer.7

10 The Economist January 17th 2015

SPECIAL REPOR TENERG Y AND TECHNOLOG Y

Universities February 14th America’s Hispanics March 14thFamily companies April 4th

The price of fossil fuels will always fluctuate. Solar isbound to get cheaper