The Carbon Capture Business

September 2021

If you want to become the next Elon Musk, success in one infrastructure business will guarantee your untold riches: Carbon Capture.  The only catch?  Still being alive when it’s time to cash in.

At the rate efforts to stem carbon emissions are going, and observing how changed climate just from GHG emissions to date is affecting hundreds of millions across the globe, many parts of planet earth will become more and more unpleasant to live in.  The latest IPCC report sounded its loudest alarm bells to date, and yet there are few political signs that getting gas-guzzling cars and trucks off the road, or shutting any significant fraction of the world’s coal-fired generating plants, is going to happen anytime soon.  In the best of political will and consensus cases, we’ll still have way too much carbon in the air for the planet’s climate to be as we would want it – soon, and for centuries.

Enter Carbon Capture.

The idea of sucking some of that excess carbon out of the air — negative emissions — in order to reduce global warming has been around for a couple of decades now.  Working prototypes first started appearing around 2015, and the biggest carbon capture plant to date was turned on just this week, in Iceland.  It can actually work.  At a very small scale, and a very high cost – for now.  In today’s post, we’ll have a look at where this business is, and where it might be going.

Carbon capture comes in two basic flavors: point-specific, and general.  Publicity and early investment have been more focused on the former, notably carbon capture facilities being tied to coal-fired generation – often referred to as CCS, or Carbon Capture and Storage plants.  There are currently some 20 CCS plants in operation worldwide, with a combined capacity of about 40MT of CO2 per annum.  The first large-scale facility was installed in 2014 at the Boundary Dam coal plant in Saskatchewan.  The track record of these early efforts has been… mixed.  While technical performance has generally been in line with expectations (an efficient CCS facility should be able to remove some 90% of an associated plant’s CO2 emissions), the overall economics have been marginal.  A highly publicized effort to incorporate CCS onto coal-fired generation by Duke Energy was abandoned, and the PetraNova plant in Texas, the largest in the world when launched in 2017, was shut down earlier this year for losing money. 

PetraNova CCS facility

The International Energy Association (IEA), among others, remains bullish on CCS.  While public attention has been primarily on CCS in conjunction with power generation plants, CCS can be applied to many kinds of industrial processes with GHG emissions.  Facilities tied to plants which produce more concentrated CO2 streams – such as ethanol or natural gas – require less energy to separate out carbon for subsequent storage, and thus have significantly lower costs than those tied to power generation.  The IEA notes that after a time of declining investment pipelines, plans for more than 30 new integrated CCS facilities have been announced in the last three years, with a combined CO2 capture capacity of around 90 Mt per year.  That, however, is still less than 0.1% of estimated global CO2 emissions.  The IEA stated earlier in 2021 that carbon-neutrality by 2050 would require the capacity to remove 1 billion tons of CO2 from the atmosphere every year.  A recently issued report by the National Academies of Science puts that number 10 times higher, at 10 Gigatons per annum.

This week public attention has been on the other “flavor” of carbon capture, non-point-specific carbon capture, often referred to as Direct Air Capture (DAC).  Instead of being located alongside a specific source of emissions, DAC facilities can be sited anywhere – often near potential carbon storage sites, and extract carbon from the atmosphere itself.  According to Bloomberg New Energy Finance, the current global capacity of DAC is 6,415 tons of annual CO2 capture, to which the September 8th launch of the Orca Climeworks plant in Iceland will add another 4,000 tons: a big jump, yet still a tiny fraction of the capacity of point-specific CCS technologies, which themselves are a tiny fraction of the excess CO2 in the atmosphere.

Climework’s Orca DAC plant — Iceland

There are plans for DAC capacity to get bigger, much bigger, and we’d say the force is with DAC.  There are three main players as of now – Climeworks, Carbon Engineering, and Global Thermostat.  Bloomberg NEF says their capacity will “increase 150-Fold by 2024.”   That would be the year in which Carbon Engineering plans to open a 1 million TPA facility.  While this would still lag the impact and capacity of point-specific CCS technologies, we think that DAC has a couple of decisive advantages over CCS in the medium to longer term.  The first advantage is location: a DAC plant (which essentially looks like a shipping container) can be placed anywhere, and extract CO2 that entered the atmosphere from any number of sources.  At the far larger scales on the road to removing 1 billion tons per year, or 10 times that, from the atmosphere, not being tied to power generation or other industrial processes will become a big advantage.  The second big advantage will be political attractiveness.  In the next 1-2 decades, as political pressure to take action grows in tandem with global warming and its impacts on populations everywhere, and extracting carbon from the atmosphere gets seen more and more as one essential part of the solution, DAC will be a far more attractive place to channel global investment and subsidies.  It is not that DAC doesn’t have a moral hazard problem – it’s just that DAC’s moral hazard is and looks far, far smaller than that of point-specific Carbon Capture.  Some degree of public subsidy and support has flowed to CCS – the IEA says $2.8 billion in public grants have accompanied the $15 billion in private investment in already commissioned CCS plants, and places like Wyoming and North Dakota are lobbying hard for more.  But environmental opposition to such support is already fierce – as it “rewards” large polluters directly, and scaling up such public support is going to be extremely difficult politically, if not impossible.  In contrast, DAC doesn’t reward any polluters, and its own scale won’t be significant early enough to really impact the political debates around taxing, decommissioning or prohibiting fossil fuel usage.  Yet the political demand for removing carbon from the atmosphere, just to make the problem less bad in the coming decades and beyond, will become huge – and DAC is far more likely to be the beneficiary of tax monies channeled to attacking that problem.

Issues

Several obstacles stand between today’s small-scale DAC and an eventual large-scale flood of public support and investment.  Leaving aside the combination of moral hazard and the creation of public subsidies, chief among the obstacles are (i) costs, (ii) energy intensity, and (iii) what to do about the carbon.

DAC costs and business model.  DAC is for now the most expensive of carbon capture technologies.  According to Christopher Gebald, co-founder of Climeworks, current costs run at $600-800 per ton of CO2 extracted from the atmosphere; CCS costs related to power generation might run at 10-15% of this, between $60-120 per ton, while CCS costs related to more “pure” forms of CO2 emissions such as ethanol production can be as low as $20/ton (still leaving costs related to carbon separation and storage).  Revenues are far lower, with the largest source presently from selling to oil companies for enhanced oil recovery – but even that isn’t high enough to keep some CCS plants in operation, and it is hard to envisage the petroleum industry paying much more.  Some CO2 is recycled to industrial producers which use it as feedstock.  Global Thermostat sells its DAC-derived carbon dioxide to soft-drink producers, but it is also hard to envisage a large-scale business model around this.  For now, the economics of decommissioning coal fired plants is more attractive.  The business model of DAC, however, may become more positive.  Economics of scale will be a big factor: the Orca plant in Iceland was made entirely by hand; as demand and volumes increase by orders of magnitude, it should become far cheaper to build DAC plants, as we’ve seen happen in other technologies such as battery storage.  Gebald of Climeworks projects costs at around $200-300/ton by 2030, and $100-150/ton by 2040.  Steve Oldham, CEO of Carbon Engineering, claims that his company is already capable of building plants with costs closer to $100/ton.  On the revenue side, larger, commercial-scale plants will allow DAC players to sell offsets to firms looking to reduce their emissions.  But public sector payments are still likely to be the big revenue source for DAC.  The first steps in this direction in the US were taken this Spring, with the passage of the 45Q rule, which provides a tax credit of $50/ton for captured and sequestered CO2.

Energy Intensity.  Direct Air Carbon Capture is highly energy intensive – mostly a function of the fact that CO2 is a small fraction of the air DAC absorbs, meaning that energy needs for separation of the carbon are high.  How energy intensive?  Research firm Carbon Brief claims DAC could account for as much as 25% of global energy consumption by the end of the century (direct CO2 capture machines could use a quarter of global energy in 2020).  DAC technology, however, is still far up the learning curve.  Liquid solvents for example require a far lower temperature to run separation processes, in which case waste heat – at near-zero marginal cost – could become a replacement energy source.

What to do with the carbon?  Storing the extracted carbon presents challenges.  The issue is less the existence of potential underground storage with appropriate geology (less prone to re-release of the carbon than, say, forests which might get consumed by fires), and more with the need to get the carbon from where it is extracted to where it is stored.  This is especially an issue for CCS, whose location is determined by emission sources such as power or industrial plants.  A big part of costs and land issues for CCS, if this should scale up substantially, will be pipeline networks to transport the extracted carbon.  Such pipelines exist today (piping carbon to oil fields for enhanced recovery), and studies are underway in several places for this kind of infrastructure at a larger scale.  This should be less of an issue for DAC, which can be sited closer to storage areas, requiring far less transport infrastructure and investment.  DAC is also less land-intensive than solar or wind farms, and unlikely to compete with other priority land uses, like agriculture. 

The impact that carbon capture technology will have on reducing carbon in the atmosphere, over the next two decades, is likely to be… minuscule.  Which, paradoxically, may help make it an even bigger business than it would otherwise be… maybe by 2050.  By 2050 (and certainly way before), the political pressure to remove large amounts of GHGs from the atmosphere will have become enormous – unlike the tepid interest which the idea now attracts.   By then DAC technology should have sufficiently matured to be more economic to operate, and be able to grow faster with fewer subsidies.  As heat waves, mega-fires and super-storms occur in more places and more frequently than we’d want to imagine at this point – along with whatever other climate problems may be in store as global temperatures rise – there could literally be no end to the appetite for building more and more carbon capture machines.  In the second half of this century this could be, no joke, the largest infrastructure business in the world. 

A business that won’t be mature or profitable for maybe 25 years sounds like it’s too early for a good investment.  Or maybe not.

Index to Previous Climate Adaptation Posts

German Floods and Performance Bonds

August 2021

In mid-July, some 250 people were killed in Germany and Belgium as rain-swollen rivers flooded towns over a wide area.  More than 10 inches of rain fell in 48 hours in some spots; Cologne received 6 inches in 24 hours.  It was the deadliest natural disaster to hit Germany in over 50 years.  Economic losses are estimated at over $3 billion, with the total likely to rise much higher.  Germany was not alone in experiencing extreme rainfall in July.  One Sunday, Londoners were hit with a month’s worth of rain within a few hours.  In Central China, rain amount records were set, over a million people were affected, and the subway in Zhengzhou – a city of 5 million – flooded while passengers were trapped in trains.  This a year after several million were displaced by flooding in the Yangtze River Basin.  And in the Berkshires of Massachusetts, July 2021 became the rainiest year since records were first kept – in 1891.

Floods in Germany (Reuters)

In our previous column, Infrastructure Ideas wrote about rising water levels along coasts, and the infrastructure implications of plans to build seawalls to defend many cities.  As last month has shown, once again, weather-related flood events are increasing far from the seas as well.  Floods are both damaging existing infrastructure, creating repair and restoration needs, and triggering plans for new infrastructure investments to help cities adapt to rising flood risks.

Too much water in many places, and not enough in others.  July’s extreme weather events were not limited to flooding: in the Western US, in Turkey, in Greece and in Sardinia, wildfires also set records and damaged widespread areas.  Some of these wildfires are expected to burn on into the Fall.  Much of the Western US also saw unprecedented heat waves in July, setting the stage for the fires – as did Moscow, among other places.  Last year it was Australia.  In an era of climate change, extreme weather events are becoming more common, and the IPCC — the Intergovernmental Panel on Climate Change – tells us that the frequency of these extreme events will increase as global temperatures rise.  As a headline from the New York Times says “No One is Safe.” 

From the standpoint of infrastructure, these floods, and the wildfires, share one important thing in common.  They result from extreme weather events which are unpredictable.

General trends are clear: more floods in some places, and more heat and fires in others.  Sea level rises are increasingly observable, and “predictable” in the short term.  But the timing and scale of downpours is – generally speaking – not predictable, and neither are the location or breadth of wildfires.  With Climate Change, we already observe that extreme events occur on shorter notice, with both more intensity and severity than before – and, as July has demonstrated, outside of any forecast range.

This lack of predictability, in an age of adaptation to climate change, has significant implications for infrastructure.  The big implication is that related infrastructure investments — being made with a short (or no) planning period, and subject to a large range of uncertainties as to how soon they are needed, how frequently they’ll be used, and the magnitude of the problem they seek to solve — will tend to have some of the least desirable characteristics of infrastructure projects.  Notably, these investments can expect to be characterized by (a) frequent design changes, (b) significant delay risks, and (c) large cost overruns.  Frequent design changes will almost inevitably stem from the uncertainties involved, and from the politics surrounding how best to respond.  Risks of delays and overruns go hand-in-hand with frequent design changes in all construction projects.

In normal times, public authorities asking for infrastructure projects, and lenders supporting the projects, always look to lay this kind of risk off to sponsors and construction companies.  Completion guarantees from sponsors and performance bonds from construction companies are the primary instruments to shift these risks.  A consequence of climate change, and the rapid rise in adaptation-related infrastructure investments, is that it will become more difficult for these risks to be shifted in the way public authorities and lenders typically require.  The culprit will be unpredictability.  With the higher risks of delays and overruns coming from that unpredictability, the size of adaptation-related infrastructure performance bonds will strain the balance sheets of many construction companies.  Where sponsors themselves are also construction companies, required completion guarantees will make the problem worse.  And the construction companies will note, often correctly, that weather-related sources of cost overruns – as well as overruns stemming from political disagreements on how best to respond to extreme weather events – are outside of their control, making them even more unwilling to take on these risks.  We can therefore expect to see that many infrastructure investments intended to help cities and other areas adapt to more extreme weather events – urgent investments when the need for them becomes clear – will get at best delayed and at worse stuck due to the unwillingness of parties to bear the risks stemming from higher unpredictability.

Keeping infrastructure investments flowing as the need to adapt to extreme weather events grows may therefore require something new.  For developing countries, funding for these higher risk investments may simply get swept up into their general need for additional finance related to climate change: yet one more problem to solve.  For wealthier and middle-income countries, the solution may wind up in the domain of insurance.  The likely best way to manage the risks from unpredictability will be diversification of that risk across a very large pool of geographies and projects.   One model may be the World Bank’s Disaster Risk Financing and Insurance program, developed in the mid-2010s, which was created to pool weather-related risks for low-income countries. 

Floods in Germany, fires in the Mediterranean, these are disasters whose occurrence, timing and scope are increasingly unpredictable.  Yet that such events will occur more frequently is itself predictable.  Infrastructure investments may in at least some cases mitigate the damages and deaths from further extreme weather events, and will in many cases be needed to repair damages.  These adaptation-related investments will present different problems than traditional infrastructure, due to the unpredictability of specific severe weather events.  The biggest problem is likely to revolve around Performance Bonds, and the ability of construction companies to absorb unpredictability risk.  Let’s hope insurance can provide a solution.

Index to Previous Columns on Climate Adaptation and Infrastructure

Seawalls and Emerging Markets

July 2021

Built on beautiful Biscayne Bay, money has flowed from the sea to Miami – especially to its real estate developers — for centuries.  It is starting to flow back to the sea.

Miami flooding — from the Miami Herald

Last month, the US Corps of Engineers released a draft study for how best to protect the city of Miami from rising seas and recurring flooding.  The Engineers’ recommendation: a $6 billion, 6-mile long, and up to 20-foot-high seawall.  City and state politics are now mired in a high-profile back-and-forth on whether to proceed (see “A 20-Foot Sea Wall? Miami Faces the Hard Choices of Climate Change”).  Similar plans to build large and expensive seawalls are being debated in other American cities: Houston, San Francisco, Charleston, and Honolulu for a few, with New York City looking at the most grandiose plans of all, costing well over $100 billion.  A 2019 report noted that the cost of building the seawalls under debate in the US could run to $416 billion – the same cost as the build-out of the entire national interstate highway system.  Across the Atlantic Europe already has seawalls in a number of places: Venice, London, St Petersburg, the Netherlands.  A gargantuan project – nearly 400 miles long – is under discussion to protect European coastlines along the North Sea – at a preliminary cost estimate of half a trillion dollars.  Along the Pacific Singapore and Shanghai are among (the few and wealthy) Asian cities with seawalls.

Rotterdam’s Seawall

There is still novelty around the idea.  Until the last decade, one would have been hard pressed to find “seawall” in anyone’s definition of infrastructure.  Ports have built jetties in many places to protect harbors, but these have been much smaller endeavors.  Yet the future where one can plausibly project seawalls becoming one of the 3 or 4 largest categories of infrastructure spending around the world, capturing hundreds of billions of dollars, has come quickly.  A future where seawalls will be the single largest ticket item in the budget of many coastal cities, at times dwarfing their combined spending on all other infrastructure combined.  This is another example of how disruptions have upended the once stable and fairly predictable world of infrastructure, whether disruptions from technology – such as wind turbines or batteries – or from other sources, like climate change.

Fear of rising sea levels from the melting of glaciers is galvanizing the newfound interest in seawall building.  Hundreds of millions of people live in coastal cities with low elevations and many, like those in Miami, are already seeing the increased flooding that will worsen in coming years.  As the World Economic Forum states, “Even if we collectively manage to keep global temperatures from rising to 2°C, by 2050 at least 570 million cities and some 800 million people will be exposed to rising seas and storm surges. And it is not just people and real estate that are at risk, but roads, railways, ports, underwater internet cables, farmland, sanitation and drinking water pipelines and reservoirs, and even mass transit systems.”  Estimates of the sea level rise itself, which may sound small or slow, tend to understate the problem.  Only about 1/3 of future coastal flooding risk is from rising sea levels that would permanently submerge low-lying areas, while 2/3 of the risk comes rather from the likely increase in extreme high tides, storm surges and breaking waves.  Cities are looking at a variety of ways to protect themselves, looking to better absorb and drain water faster, but attempting to keep water away is on nearly every wish list. 

New research (see “A Space Laser Shows How Catastrophic Sea Level Rise Will Be”) shows that for several of these coastal cities, the issues of rising seas and more severe storms will be made worse by yet another problem: sinking.  As populations in many of these urban areas have grown rapidly, over-extraction of ground water is causing the ground to subside.  Cities built on river deltas usually sit on several layers of clay, deposited over time as sediments by the river, with underlying aquifers.  When the aquifers get drained to provide water to the city’s population, the clay collapses into the space which had held water.  The more an urban center grows, the more people it needs to hydrate, which increases the rate and severity of subsidence.  Djakarta is the prime example of this effect, with subsidence having been a key factor in last year’s decision by the Indonesian government to move the capital to a different location (see Capital Punishment (or So Long, Djakarta ?)), but it is far from the only one.

The surge in interest in seawalls as the centerpiece of the solution for many cities will keep engineers occupied and planners preoccupied.  It is still very early days in the growth of what will likely be one of future infrastructure’s largest areas.  Today we’ll look at just a couple implications of this coming boom, especially as regards developing countries.

We’ll start with one safe assumption about this new type of infrastructure: if the seawalls get built, they’ll cost a lot more than the amounts now projected – even the $400+ billion estimated for the US.  Seawalls will fit squarely into the type of infrastructure prone to frequent and large cost overruns (think of tunneling projects, like Boston’s infamous “Big Dig,” or of large hydroelectric dams, with average overruns approaching 50%).  They will be highly politicized investments, with continued debate about every detail (whose property is disturbed, whose views are affected, which houses are outside the protection zone, what is the timeline – and especially, who pays), and debate about just how high the tidal or storm surges they’re built to prevent will be and how soon.  This means the construction of these barriers will be subject to frequent change orders, the perfect recipe for more cost overruns.  And they may become obsolete fairly quickly, depending on the pace of climate change and glacier melt in the coming decades.  It would not be a big stretch to see the US spend over $1 trillion on seawalls in the coming 20 years, nor would it be a big stretch to see global spending on such projects well over $5 trillion.  That’s a lot of infrastructure spending

A second safe assumption about seawalls?  You won’t find many in Emerging Markets any time soon. 

And that will become a big deal.

Cities in lower-income countries stand to be disproportionately affected by rising seas.  While all coastal cities will be affected by sea-level rises, some will be hit much harder than others. Asian cities will be particularly badly affected. About 4 out of every 5 people impacted by sea-level rise by 2050 will live in East or South East Asia – several hundreds of millions of people.  Africa is also highly threatened, due to rapid urbanization in coastal cities and the crowding of poor populations in informal settlements along the coast.  The list of most affected cities includes Mumbai, Kolkata, Dhaka, Guangzhou, Rangoon, Ho Chi Minh City, Manila, Dakar, Alexandria, Lagos, Abidjan, among many others.  Leaving aside China, most of these Emerging Markets cities and their national governments have one thing in common when looking at seawalls as part of their adaptation plans: a lack of capital. 

The list of Emerging Markets countries with cities affected by rising seas looks an awful lot like the list of Emerging Markets countries with large infrastructure deficits – already.  The capital requirements for building seawalls to protect their coastal cities from increased flooding will absorb a large share of their capital that is already needed for deficient infrastructure: for some smaller countries, the cost of seawalls may approach the size of their entire current infrastructure budgets.  It is no surprise, therefore, that a list of cities actively considering seawalls is 90%+ in developed markets (including China).  Djakarta – banking on financial support from the Netherlands – is the only city in a lower-income country with an advanced plan. 

While it is not surprising that attention to seawalls is almost entirely concentrated in more developed countries, the absence of such attention in Emerging Markets has some important implications worth noting.

1.         Flooding increases in coastal cities and the inability of those in low-income countries to engineer solutions (or at least what may appear to be solutions) to offset sea-level rise will lead to much larger-scale relocation of populations in the Emerging Markets than what we will see in the US, Europe and the richer Asian countries.  Some of that relocation may be organized, at least to an extent, along the model of Indonesia’s announced move of the country’s capital, and much of it is likely to be dis-organized, in the form of migration – in country where inland options may be available, and cross-border where those options are not available.  As the World Economic Forum states it, “The coming decades will be marked by the rise of ex-cities and climate migrants.”  To date much of this climate migration has been relatively “invisible,” contained within countries.  Don’t expect this to continue.  The cry we have seen in early 2021 for better equity in the distribution of COVID-19 vaccines may presage a louder cry in years to come for better equity in the building of seawalls.

2.         Given that the wealthy countries that dominate the Boards of International Financial Institutions will want to see as little large-scale cross-border migration as possible, and will have to devote plenty of capital to their own climate adaptation plans, we will undoubtedly see a big push for the IFIs to engage in helping Emerging Markets fund seawalls.  With the scale of the financing challenge, this will be the domain of the large global and regional multilateral development banks, and will stretch their balance sheets. Should a large-scale Climate Adaptation Fund emerge, as has been discussed for many years, and could safely assume that a large share of its capital would wind up going into this area.

3.         There will even greater interest in “innovative financial solutions” than there is for traditional forms of infrastructure.  Don’t be surprised to see mechanisms through which the local private sector in coastal cities (especially companies serving consumers in these cities, such as retail, telecommunications, and producers of consumer goods) “help finance” some kind of Public-Private Partnerships (it will sound better than to say they are being taxed) in order to preserve their own revenues.  And don’t be surprised to see some mechanism emerge whereby wealthy countries contribute to some kind of “Fund” to help finance seawalls in lower-income countries.  It would be the same kind of general principle which has been discussed now for decades for Climate Change adjustment funds, but would have the clear advantage, relative to current discussion, of going to concrete (pun intended) objectives.  In the US, we have seen the building of a wall to limit immigration generate considerable political momentum: one can imagine building of walls further away, with the same idea of limiting immigration in mind, will also generate plenty of political momentum in the future.

Seawalls: coming soon for infrastructure budgets – ready or not.

Index to previous Infrastructure Ideas columns about Climate Adaptation

Asia’s Energy Transformation: Vietnam

June 2021

As the climate keeps warming, many in the United States and Europe are taking a long list of actions and arguing for more.  How hot the earth gets, however, more than anyplace else, hinges on the actions taken – or not taken – in Asia.  Asia has the world’s largest population, the world’s fastest growing economy, and – for climate, more important than anything else – close to 80% of the world’s coal-fired generation.  The path Asia takes – and takes in this decade – will do more to determine the path of climate change the rest of the century.  The path Asia takes, in turn, depends on the path that its own large economies take.  Infrastructure Ideas has previously examined the dynamics of the energy transformation, especially whether countries will or will not add yet more coal-burning electricity capacity, in India, Pakistan, Bangladesh, and Indonesia.  Today we’ll look at another of the region’s critical economies: Vietnam.

Vietnam’s population of 97 million ranks 15th in the world, and its energy consumption growth of over 10% a year the last several decades has been one of the 5 highest in the world.  As population and incomes continue to rise, the demand for electricity in the country is expected to more than double by 2030.  Generation capacity is expected likewise to more than double, from the current 55 gigawatts to 130 GW, at an estimated cost of US$150 billion – and then to more than double again by 2045, to 277 GW.  Coal-fired generation is the largest source of power in Vietnam, accounting for about 53% of demand. Aside from coal, hydropower accounts for about ¼ of capacity, according to the IEA.  Natural gas makes up some 16% of demand, and non-hydro renewables about 7%.

Coal in Vietnam is not only the largest source of power in the country, it has also been the fastest growing, with capacity having increasing by nearly 15 times since 2005, to about 25 GW in 2019.  As the ability to build more large dams along the Mekong River basin has become very constrained, the government increasingly has turned to new coal plants instead of hydropower.  With the expected strong growth in future electricity demand, Vietnam’s earlier power sector plans called for building more than another 45 GW of new coal-fired generation capacity by 2030, which would nearly triple the country’s existing coal fleet.  According to Bloomberg New Energy Finance, Vietnam’s coal-fired pipeline is the 4th largest in the world today, with some 17 GW under construction and another 29 GW in advanced planning stages.  This comes to about 15% of the total planned new coal capacity worldwide, excluding China, and if built, these plants would contribute to adding annual emissions of some 500 metric tons a year of CO2.  Enough to make the world significantly hotter.

Mong Duong coal plant, Vietnam

Energy policy in Vietnam, fortunately, is in transition.  The country continues to envisage rapid further growth in electricity consumption as it develops, but where that added electricity is to come from is changing fast.  In the past two years, Vietnam has gone from almost entirely fossil-fuel and hydropower-based to a solar and wind powerhouse.  With a different sequence than most of the world, Vietnam moved first to aggressively adopt solar generation, especially rooftop solar.  From less than 2 GW of capacity in 2016, solar generation capacity now exceeds 11 GW – 5 GW of which was installed just in 2020.  Vietnam even showed the third-biggest growth in rooftop solar installations globally in 2020.  Yet the biggest energy headlines for Vietnam are now elsewhere – in offshore wind.  Onshore wind plants in Vietnam have begun to appear, but sites are constrained by the lack of available land.  The country has turned its eyes offshore, as the offshore wind sector has begun to mature worldwide (see Infrastructure IdeasOffshore Wind – the Next Big Thing).  In 2021, Vietnam is forecast to install 1 GW of wind capacity, triple its existing capacity and surpassing Thailand—at present Southeast Asia’s front-runner in installed wind capacity.  And in July 2020, the Vietnamese government approved the assessment of the area off the cape of Kê Gà in south Vietnam to build the world’s largest offshore wind farm with a capacity of 3,400 MW – larger than any existing generating facility in the country.

With – at last – renewables coming to Vietnam, the country’s planners are rethinking Vietnam’s large-scale plans for future coal-fired generation.  Several factors are coming into play: (a) the government has seen that investors and banks will finance new wind and solar generation, and that this source of power is cheaper than it had expected; (b) internal demand is geographically uneven, with both demand and growth highest in the south of the country – where offshore wind potential is the greatest; (c) the communist government is also ill-at-ease with both recent demonstrations against coal-fired power station projects, and with the risk of electricity shortages – with fossil-fueled capacity taking much longer to bring online than wind and solar; (d) sources of external capital to finance new coal plants are getting harder to come by; and (e) Vietnam itself stands to be heavily impacted by sea-level rise, with its extensive low-lying urban and agricultural areas along the Mekong Delta.

The government’s evolving thinking has begun to take shape in the draft form of “PDP-8,” its eighth multi-year Power Development Plan.  Released in February 2021, the draft calls for both wind and solar generation capacity to rise to about 20 GW each by 2030, with their share of generation jumping from about 7% today to 30% by 2045.  Coal, as a share of the country’s generation mix, is projected to be cut in half, to about 27%.    The National Steering Committee for Power Development has recommended eliminating about 15 gigawatts of planned new coal plants by 2025, according to the state-controlled news website VietnamPlus.  The draft PDP-8 proposes no new coal-fired power plants except those already under construction or planned for completion by 2025 or sooner.  This would still, however, leave almost 20 GW of new coal capacity to come online this decade.  And the battle for how to meet yet another doubling of demand in the following decade has not been joined.

PDP 8 — IHS Markit

As the planners deliberate, the environment around Vietnam keeps changing as well.  For one, financing for coal plants continues to get more complicated.  Japan has been a big financier of the sector in Vietnam, but Mitsubishi – one of Japan’s largest players in coal — announced in February it would no longer support one 2 GW and $2B flagship coal project, Vinh Tan 3.  Conversely, financiers are eager to finance renewable generation: two wind power plants, Phu Lac 2 and Loi Hai 2, just this month closed a financial package from the IFC.  For another, Vietnam has not really seen yet how cheap wind and solar power have become around the world.  A late-comer to renewables procurement, Vietnam still offers a feed-in tariff mechanism to project developers, at 8.5 cents per kilowatt-hour – more than triple what it costs to procure new wind power capacity in the United States.  As it moves this year to more efficient auction mechanisms for new capacity, and assuming it improves its PPA framework, Vietnam should start seeing renewable prices far lower than what it has been paying to date.  And thirdly, Vietnam has yet to dip its toes into energy storage.  As costs continue to plunge and availability expand, battery storage could help Vietnam meet its growing electricity demand with significantly less future expansion of new generation capacity.

Vietnam completed its five-year general elections for the National Assembly in May.  By the end of June, the government is expected to release the final version of PDP-8.  In a largely state-controlled economy such as Vietnam’s, formal government plans rule the roost, and PDP-8 will determine whether Vietnam sticks to earlier plans to move full steam ahead with building large-volume and high-emission new coal generation, or whether it will continue to cut back on new coal plans and switch even more strongly in the direction of renewable energy.  A great deal – of emissions and climate change – hinges on the decision, and on Vietnam’s continued energy transition. 

Previous Infrastructure Ideas Posts on Energy: Index

Ransomware and the Pipeline

May 2021

In the last week, one of the largest fuel pipelines in the United States has been shut down as it deals with a ransomware attack.  This is the highest-profile infrastructure cyber-attack on the energy system in the US, and a reminder that this “new” problem is getting much worse – and will continue to do so.  Today we’ll take a look at some implications of this latest attack, and of cyber-risk trends for infrastructure.

The Colonial Pipeline System

Infrastructure Ideas has been writing about infrastructure cyber-risks for some time, and one of our Ten Infrastructure Predictions for 2021 was that these risks would grow for utilities.  Unfortunately, we were right.  The attack on the Colonial Pipeline, which operates the largest fuel pipeline between Texas and New York, has disrupted availability of gasoline and jet fuel for a week – with long lines at gas stations in some areas, and a state of emergency declared by the Governor of Virginia.  The 5,500 mile pipeline carries nearly one-half of the motor and aviation fuels consumed in the Northeast and much of the South (see “What We Know about the Colonial Pipeline Attack,” from the New York Times).  Colonial, the pipeline operator, reported that hackers had infiltrated corporate data, not control of the pipeline itself, but that Colonial had shut down operation of the pipeline to prevent further damage and contain risks.  The FBI has attributed the hack to a Russia-based criminal group known as “Darkside,” which specializes in ransomware attacks against English-language targets.  As of this writing pipeline operations have yet to return to normal.

A customer help pumping gas at Costco, as other wait in line, on Tuesday, May 11, 2021, in Charlotte, N.C. Colonial Pipeline, which delivers about 45% of the fuel consumed on the East Coast, halted operations last week after revealing a cyberattack that it said had affected some of its systems. (AP Photo/Chris Carlson)

The Colonial Pipeline ransomware attack is far from the only headline regarding cyber-attacks on infrastructure in the first months of 2021.  A report in February from the industrial cybersecurity firm Dragos named four separate hacker groups with ties to Russian intelligence services as having targeted industrial control systems in the United States.  One group, named “Kamacite,” reportedly works in cooperation with the GRU, Russia’s largest foreign intelligence agency and has targeted US electricity and oil and gas firms, and is said to have gained network access to firms on several occasions.  Another February report, this one from IBM, found the energy sector to be the third most frequently targeted in 2020 (after finance and manufacturing), up six places from 2019.  Aside from energy, other attacks have targeted the water sector.  An as-yet-unknown hacker gained access to the controls of a water treatment facility in Oldmar, Florida, and attempted (unsuccessfully) to introduce large amounts of lye into the city’s water.  In February, an ex-employee of a water company near Little Rock, Arkansas, was indicted for accessing and attempting to disrupt the company’s systems after being let go.  In 2020, a likely Iranian hacker was found offering to sell network access to a water treatment plant in Florida over the messaging app Telegram.  A recent study profiled in Wired (Water Supply Hacks Are a Serious Threat – and Only Getting Worse) found dozens of hacking incidents at US water installations, with a continued rise over the last decade.  Water utilities turn out to be far more vulnerable to cyber-risks, in spite of the focus of most headlines on electric utilities, as so many water utilities are small and lack the administrative capacity and resources to protect themselves against rapidly evolving attack risks.

The underlying dynamics indicate that infrastructure cyber-risks are, unfortunately, getting much worse.  For one, the growing use of digital controls to manage electricity and other energy installations opens new entry points for hackers to exploit.  Second,  the sheer number of actors involved or with the potential to be involved cyber-attacks is growing rapidly: barriers to entry are low, and the trend towards ransomware attracts criminal groups across the  board.  As one cyber-expert cited in the Dragos report puts it, “A lot of groups are appearing, and there are not a lot going away.”  One element of this week’s Colonial Pipeline attack highlights the issue: the group apparently responsible, dubbed “Darkside”, operates on a business model whereby it develops hacking tools and then sells, rents or leases them to other parties.  It does not require much imagination to see how this will accelerate the availability of hacking tools.  Third, with the multiplication of actors comes a multiplication of targets.  One group Dragos has dubbed “Stibnite” has targeted Azerbaijani electric utilities and wind farms using phishing websites and malicious email attachments: if firms in Azerbaijan are becoming targets, firms in places such as Jordan, Indonesia, Mexico and elsewhere cannot be far behind.  Utilities in lower-income countries, lacking in managerial and financial resources to adequately defend themselves, utilities in areas of internal or external conflict, attractive targets for political or ideological reasons, and utilities in high-crime countries with already diversified and sophisticated criminal groups, are all going to be at particularly high risk in coming years.  Fourth, the types of infrastructure cyber-risks are also expanding.  Ransomware attacks are the flavor of the day, and with the proliferation of hacking tools among criminal networks will doubtlessly expand.  These are expensive and disruptive, but the damage to date from these attacks has been limited in scope and in time.  Yet more aggressive and destructive attacks are unlikely to be far away.  As an alarming new book by Nicole Perloth, This is How They Tell Me the World Ends: the Cyberweapons Arms Race (for a short version, see the excellent review by Sue Halpern in the New York Review of Books, “Weaponizing the Web”) points out, an important feature of cyber-weapons is that they are very cheap compared to traditional “hard” weaponry.  Perloth tells the story of seeing a young Iranian at a hacking conference in Miami demonstrate how to break into the power grid in five seconds: “With his access to the grid, he told us, he could do just about anything he wanted: sabotage data, turn off the lights, blow up a pipeline or chemical plant by manipulating its pressure and temperature gauges.  He casually described each step as if he were telling us how to install a spare tire, instead of a world-ending cyberkinetic attack that officials feared imminent.”  Hacking tools can give intruders access to even critical infrastructure such as nuclear facilities, the power grid, and air traffic control.  But they are relatively cheap compared with other weapons of mass destruction, and for sale in a market that is robust, largely out of sight, and welcoming to anyone with piles of cash at their disposal, whatever their motivation. 

Disruptive technologies continue to change the face of infrastructure.  In many cases, this is bringing lower costs, better services, more convenience and reduced emissions.  Technology, though, is agnostic: the Colonial Pipeline cyberattack is a reminder that disruption can be negative as well as positive.  For infrastructure operators and investors today, there is a clear message from these attacks.  Cyber-risks are not going away, and are going to get worse.  Investments in cyber-security (the FBI, after the Colonial Pipeline breach, has issued a useful “tip sheet” to key US infrastructure providers), insurance, and the ability to re-launch systems after an attack are all going to be increasingly important.  The worst situation will be to be unprepared. 

Previous Infrastructure Ideas columns on Disruptive technologies

Not Just Another $2.5 Trillion Infrastructure Plan

April 2021

Earlier this month, the Biden administration announced a much-anticipated infrastructure plan for the United States.  Named “the American Jobs Plan,” the proposal would invest $2.5 trillion (or $2.3T, $3T, or $4T, depending on who is doing the math).  Outside of China, this is the largest national infrastructure plan ever put forward.  The bill is … complicated: we won’t attempt to summarize it, and instead will highlight some of its more interesting aspects.

1) Size.  The first and most obvious feature of the Biden proposal is its estimated cost.  $2.5 trillion is… a lot of money.  It’s pretty close to estimates of annual global spending on infrastructure altogether – which ranges from about $2T to $4T, depending who is estimating and what is being counted as infrastructure.  In comparison, what is usually thought of as the US’ previously largest infrastructure plan, Eisenhower’s development of the interstate highway system in the late 1950s, had a price tag of… $25 billion, or 1% of the price tag of Biden’s plan (yes, we’ll leave inflation aside for now, just making a general comparison).  The next biggest infrastructure proposal that’s been out there is the Modi plan for India from 2019, which clocked in at $1.3T.  But let’s look at the size of this proposal a bit more closely. 

Now, the Biden proposal looks very large compared to global infrastructure spending, and some – for example those focused on trying to find capital for infrastructure in developing countries – have expressed concern that very high US infrastructure spending might “crowd out” access to capital elsewhere.  As we noted, $2.5T is a lot of money.  The Biden plan, however, talks not about spending this in 2021, or even 2022, but over 15 years.  That’s smart, of course; it’s good to match financing with actual expenditures, and many infrastructure projects take long years to properly design and build.  It also changes the way one looks at the size of the plan.  For those doing their math, $2.5T divided evenly over 15 years comes to $167 billion a year – which looks a bit less impressive.  $167B a year is, well, less than half what the US spends annually on infrastructure today, and comes to less than 1% the size of the US economy.  China’s estimated annual spend on infrastructure?  $2.5 trillion. 

So it is a bit hard at this stage to tell how impressive the scale of this plan.  Ambitious by recent US standards, definitely.  Ambitious by global standards, and enough to “fix” the country’s infrastructure problems?  This we’ll only be able to see as details, timing, and implementation of the Biden infrastructure plan become clearer.

2) Is it infrastructure?  Immediate Republican criticism of the Biden plan claims that the proposal is not infrastructure, and rather a mislabeled left-leaning social welfare push (Senator Roy Blunt claimed on Fox News “even if you stretch the definition of infrastructure some, it’s about 30 percent of the $2.25 trillion they’re talking about spending”).  It can be difficult to tell substance from rhetoric in most pronouncements from today’s Republican party, yet the Biden plan does rely on an extremely broad definition of infrastructure.  Child care, retrofitting of buildings, and school modernization are rarely seen in most economists’ definitions of infrastructure.  This is not necessarily good or bad: infrastructure is notoriously difficult to define precisely; it does however mean that the bill’s proposed spending on “hard infrastructure,” or what fits in narrow definitions of infrastructure, will be less than the headline numbers might suggest.  Then again technology has so transformed the economies of today, and continues to do so, that maybe it calls for a wholly different approach to how countries think about infrastructure.

3) Politics are unavoidableInfrastructure Ideas has written often about how politics can impede infrastructure investments.  The US is a clear example of this; though the Republican party now seems disposed to support at least some scale-up of federal infrastructure investment, the narrow Democratic legislative majorities and Republican criticism render uncertain the passage of Biden’s bill.  It is interesting to note two big planks of the administration’s strategy on this.  First, note that the infrastructure plan is not named “the infrastructure plan.”  Rather, it is called the “American jobs plan,” a label which makes attacking the bill politically trickier, and is aimed squarely at the block-collar electorate which has been switching away from democrats.  Second, the bill contains several provisions aimed at the “left behind” segments of the electorate which have become increasingly Republican, most notably $100 billion targeted at high-speed broadband networks throughout the country, aiming to make broadband access universal and to drive down the costs for internet.  The coming months will show whether the strategy proves successful.

4) A changing view on transport.  American infrastructure plans have typically talked about roads and bridges.  The Biden proposal does allocate money – some $115B — for roads, but this is accompanied by large provisions for promoting electric vehicles, mass transit, and to a lesser extent rail transport.  $174 billion, or about 28% of the transportation portion of the bill, would go for electric vehicles alone. That includes a network of 500,000 electric vehicle stations, using electric vehicles in bus fleets, and replacing the federal government’s fleet of diesel transit vehicles with electric vehicles. It would also offer tax incentives and rebates for electric cars.  The plan would also allocate $85B to public transit, and some $80B to railroads, for a growing backlog of Amtrak repairs as well as improvements and route expansion.

5) Is this the Biden climate plan?  A good column in The Atlantic calls this “the one thing to understand about Biden’s infrastructure plan.”  The piece claims “there is only one serious vehicle to pass climate policy through Congress during the Biden administration—and it’s this infrastructure plan. If recent history is any guide, the bill is the country’s one shot to pass meaningful climate legislation in the next few years, if not in the next few decades.  In short: If you want the United States to act at a national level to fight climate change, this bill is it. This is the climate bill.”  This is a perceptive analysis: the proposed bill has several climate-related provisions, notably funds for decarbonization technologies in several industries, for the creation of a “civilian climate corps”, modeled on the depression-era Civilian Conservation Corps which contributed greatly to infrastructure building in the 1930s and 1940s, and for some $16 million to plug oil and gas wells and reclaim abandoned mines.  Yet this past week’s White House Climate Summit indicates something much more ambitious is on the way from the administration on climate change.  The politics on such a separate climate bill, though it would be surely needed to meet the administration’s stated climate ambitions, will be orders of magnitude more complicated than for the infrastructure bill.

6) A big role for states and cities.  Most national infrastructure plans create a detailed agenda of projects to be implemented by national agencies, and discussions about US infrastructure in the press in recent years have taken the same tone – often harkening back to the development of the US interstate highway system.  But in the US, a majority of infrastructure investment takes place at the sub-sovereign level, through states and cities.  The Biden plan usefully focuses on using federal spending to leverage and incentivize, not replace, local infrastructure investment.  Thanks for this surely go at least in part to Transportation Secretary Pete Buttigieg, a former a mayor.

7) Know-how, not just money.  It is often difficult to find the funding for infrastructure, but it is always easy to mis-spend money when it’s found.  Mis-spending money comes from many directions: from selecting projects to build which have much less of an impact than others might have (the famous and regrettably plentiful “bridge to nowhere” tales); from poor project design or procurement; from corruption and fraud; and just from plain re-invention of the wheel, which is always more expensive.  Understanding what has worked well and has not worked in other places is one of the most effective tools against mis-spending.  Project development agencies have regularly been highlighted by the World Bank Group and the G-7 as one of the most useful, and cost-effective, means of getting infrastructure plans implemented in emerging markets.  Virginia’s Office of Public-Private Partnerships is a good example of such an agency in the US.   The Biden proposal at least moves in this direction, though details remain to be spelled out.  The proposal says it will “focus on knowledge, best practices, to modernize faster, speed adoption of new technologies, and increase efficiency of sub-national infra planning and effectiveness.”  The White House Fact Sheet goes on to say that “When Congress enacts the American Jobs Plan, the President will bring the best practices from the Recovery Act and models from around the world to break down barriers and drive implementation of infrastructure investments across all levels of government to realize the President’s vision of safe, reliable, and resilient infrastructure. Critically, in order to achieve the best outcomes on cost and performance for the American people, the Administration will support the state, local, and tribal governments delivering these projects through world-class training, technical assistance, and procurement best practices. In addition, the President’s plan will use smart, coordinated infrastructure permitting to expedite federal decisions while prioritizing stakeholder engagement, community consultation, and maximizing equity, health, and environmental benefits.”  A big step forward from previous plans.

8) Some notable absences.  There are a few things which might have been expected to be part of the Biden infrastructure plan which aren’t there.  For one, there is no list of projects, as one finds in so many national plans.  That’s probably good, since as noted above states and cities in the US have a better sense of what individual projects are needed locally, but it has implications, on which we’ll expand below.  There is also no “infrastructure bank.  Infrastructure banks have become an increasingly popular tool in many countries, and more recently in some US states and cities, like the Washington DC Green Bank.  A typical role envisaged for these banks is to help projects leverage public money with private capital.  A recent column in Forbes made a big plug in this direction.  Infrastructure Ideas believes these types of entities can be useful, though less for their fund-raising role, and more for their expertise-sharing capability.  Given the acknowledgement of the importance of know-how outlined above, and the absence of specifics on how this would be done, we may yet see some version of an “infrastructure bank” emerge.  And there is less than what one would have expected for cyber-security, a growing concern for infrastructure of many kinds.

9) The big unknown: how much will happen?  The Biden administration’s plan could make a major difference in the quality of US infrastructure.  But a plan needs to become passed legislation, then become concrete investment projects, and then those investments need to be executed.  Will it happen ?  How much and how fast?  These remain hard to answer questions.  As noted above, the headline ticket cost is very large relative to recent US and global infrastructure spending, yet if spread over 10-15 years the amounts are less extraordinary.  Relative to GDP, many countries publish larger infrastructure plans – though not many of these come to pass as announced: the Biden plan is also accompanied by a fairly clear funding strategy, which is often not the case.  Spreading expenditures over time does increase political risk, as elections may change control of different government branches, and at the moment no one would describe US politics as stable. 

It may also take quite some time to translate the plan into a list of concrete projects.  Demand for funding is likely to vastly exceed funds available, and there will be a lot of competition for the attention of administration members who are seen as making decisions on allocating funds.  This is normal in infrastructure, but it does not always go well.  Diseconomies of scale exist, and the more decisions need to be made by more people, the more chances there are for things to go awry and/or get bogged down.  Having an entity with a mandate to both ensure know-how is passed along (as noted in #7 above), and to ensure that the program as a whole move forward (possibly within Transportation or other departments), would go a long way to turning more of the plan into reality.

One thing we can say safely: these are exciting times for infrastructure in the US.  And no one has been able to say that in a long time.

Where did all the Chinese money go?

March 2021

It seems like just yesterday that Chinese financing was the future of emerging market infrastructure.  Chinese banks, awash with liquidity, and state-owned companies were everywhere, buying assets and announcing new multi-million-dollar energy and transport projects.  By about 2011 Chinese infrastructure financing exceeded infrastructure financing from the World Bank and other multilateral institutions, and in 2013 new President Xi Jing, in two major policy speeches, announced the largest infrastructure financing ever – the Belt and Road Initiative.  The future appeared to be one of endlessly increasing financial flows from Beijing, and a Chinese-enabled jump in infrastructure investment across the developing world.

Instead, it turns out that the 2013 announcement of the Belt and Road Initiative (BRI) coincided with the high-water mark in Chinese infrastructure financing.  From some $40+ billion per year at its peak, Chinese financing to emerging markets has fallen to less than a quarter of this level.  According to data collected by Boston University, Chinese energy-related financing, accounting for roughly one-half of these flows over the past two decades, fell to below $5 billion in 2020 – and it was not just COVID: the trend has been steadily downwards:

Source: Boston University

In a report earlier this month, the Financial Times said that Chinese overseas energy finance collapsed to its lowest level since 2008 in 2020.  More than half of the low figure of overseas energy lending was accounted for by a single project (the Ajaokuta-Kaduna-Kano gas pipeline in Nigeria, funded by China Exim).  One widely-watched and reported area, Chinese bank lending to energy in Latin America, in 2020 was… zero. 

Where did it all go?

Before taking a stab at this question, let’s have a quick look at where the BRI has been.  Total funding has been on the order of $50-100 billion per year, 2/3 of which going to energy and transport.  Most of the loans have come on terms that are more generous than developing countries can get from private investors, but much more costly than funds from Western donors or the concessional windows of the multilateral development banks.  In energy along, according to the Boston University database, Chinese banks have provided some $245 billion since 2000 – roughly half of this for power generation.  $127 billion, slightly more than half, has gone to coal or oil-related projects.  Funds have been spread widely: some $76B to Europe and Central Asia, some $68B to Asia, some $46B to Latin America, and $53B to Africa.

The aggregate numbers have been very large – far larger than the capital flows coming to emerging market infrastructure from the international development community over this period.  All this during a period of recurrent calls for prioritizing funds to help developing countries improve their infrastructure, as a high-priority element of poverty reduction.  Yet in spite of bringing large numbers to an area of need, the BRI has come in for plenty of criticism, centered on a few areas:

  • Lack of attention to environmental and social effects
  • Financing of investment projects that are low priority and/or that won’t deliver
  • Driving up the debt burden of countries to unsustainable levels

1)         At a project level, Chinese funding’s lesser bureaucracy and requirements, relative to those of international donor financing, has been to some degree an attractive feature for recipient countries.  At a strategic level, Chinese appetite for financing coal-fired power generation has also been attractive to countries with big energy deficits and undeveloped indigenous coal reserves, like Pakistan.  Over $50 billion of Chinese overseas financing has gone to support coal.  Both these advantages have drawn criticism, and this criticism is increasingly in conflict with President Xi Jiping’s stated objective of China’s being seen as a leader on environment and climate issues (see Infrastructure Ideas’ previous comment on this conflict, as well as that from this month’s World Economic Forum). 

2)         There have been repeated headlines about “white elephants” being funded by the BRI, ahead of infrastructure projects with arguably higher development impact.  One poster child of this type has been the Hambatota Port project in Sri Lanka, held up as something that suits Chinese interests far more than those of the host country.  The Overseas Development Institute tracks BRI projects which run into trouble, and points to some 15 in difficulties, worth over $2 billion.  More recently, there have been headlines around the cancellation of several planned coal power investments in the BRI pipeline, including the $8-10 billion Hamrawein plant in Egypt (which would have been the second largest coal-fired plant in the world).  Both Bangladesh and Pakistan, traditionally short of electricity, are seeing growing concerns about potential overcapacity, which has analysts pointing to poor sector planning related to the BRI-backed projects.  And one major study of over 2,500 Chinese-backed projects, led by Aid Data, concluded that BRI “economic benefits accrue disproportionately to politically-privileged regions:” 164% more projects took place in a political leader’s home province when in power.  

3)         Concerns about a “debt trap” brewing for countries who borrow heavily from Chinese institutions have grown significantly with the COVID-19 induced recession across many developing countries.  These concerns dovetail with those about poor project selection.  In the case of Sri Lanka, Foreign Policy drew a link between the large borrowing for Hambantota and the fact that the port was in the then-President’s home district.

The impact of the BRI lies somewhere between the positives of large flows to infrastructure and the criticisms of the initiative.  Certainly from a climate perspective, extensive Chinese backing for coal-fired generation is a big problem, and it may increasingly be a problem at a sector level for some recipient countries.  The data on project-level environmental issues is more inconclusive.  The massive Aid Data study already cited found no clear links to poor environmental outcomes.  And the same study, comparing BRI and World Bank Group projects, concluded that the outcomes of the Chinese-backed projects are not inferior in terms of impact on growth; data from the OECD DAC produces similar results. The ODI numbers of 15 projects in trouble, worth about $2.4 billion, translates to around 1-3% of projects supported by China.  While difficulty knowing when projects underperform imply these numbers are likely understated, problem project rates of 1-3% would be warmly welcomed by any infrastructure investors or lenders.  Meanwhile another study, by the Center for Global Development, was muted in its views of the impact of the BRI on debt sustainability in recipient countries. 

It is also of interest to note that the bulk of the reduced Chinese capital for infrastructure in 2020 flowed, actually, to the region that needs help the most – Sub-Saharan Africa – even though the BRI geography is not obvious.  Aside from Nigeria, Chinese lending in 2020 also funded hydropower projects in Ivory Coast (US$286 million) and Rwanda (US$214 million), as well as a solar project in Lesotho (US$70 million). 

So why the big decline since 2015?

For the driving forces, look to macroeconomics, credit concerns, and a change of instruments.

1)         For the Chinese government, the past decade has been a favorable period for lending to developing countries globally, with low global interest rates and excess domestic reserves in China.  The combination of repeated calls by the donor community for more infrastructure financing and the failure of donors, the US, or international agencies to respond with substantial increases in financing also opened an important strategic door for the BRI.  As growth in emerging markets levelled off – well before the COVID recession – this led both to credit risks and to reputational issues.  From a policy perspective, large-scale flows to emerging market infrastructure is aimed at producing impressive economic results, for which China can then take credit.  The lack of emerging market growth, after two decades of ramped up lending and seven years of the BRI, dims the attraction of the strategy.

2)         As the Center for Global Development’s Scott Morris has pointed out, the massive scale up in lending by China always carried risks, and – concerns about debt repayments have grown.  China’s largest overseas exposure is to Venezuela, a country unlikely to make good on its debts.  Growing calls for debt relief surely are a significant factor in the major retrenchment in China’s overseas lending.  The same sovereign credit risks are likely to have an impact on the balance sheets of the Chinese government’s large external lenders. Large scale write offs may become inevitable, limiting the lending capacity of these institutions. 

3)         The large numbers in flows from China over the past two decades have come almost entirely from loans.  Loans have the advantage of creating large, publicity-friendly headline numbers.  Loans have the disadvantage that at times they fail to adequately recognize and price risks.  Arguably this dynamic has become more visible for China with the BRI, and China’s state-owned banks have gone through a similar cycle with domestic lending.   One important development which has been not so visible in recent years has been the turn towards the use of Chinese-backed equity financing for overseas infrastructure.  In Latin America, for example, where 2020 Chinese lending fell to zero, Chinese-led equity funds had a busy year, highlighted by China Three Gorges’ acquisition of Peruvian utility Luz del Sur, backed by the Chinese-Latin America Industrial Cooperation Investment Fund (CLAI).  CLAI, along with the China-LAC Cooperation Fund, are examples of emerging new approaches to infrastructure by the Government of China.  They will generate fewer headlines, smaller numbers, but may yet achieve more strategic objectives going forward than large-scale BRI lending.  Time will tell.

Chinese financing has become one of the big stories in emerging markets infrastructure.  The eye-grabbing numbers of capital flows of 2010-2015 may become a thing of the past, and it may be that this financing does not take emerging markets infrastructure to a whole new level, as many thought and some hoped it would.  The BRI will not be the magic wand that solves this development problem once and for all.  But, Chinese financing is adapting, as the greater recognition of credit risks and turn to equity show.  More adaptations will likely follow. 

Undoubtedly, the Chinese National People’s Congress taking place this week, the vehicle for setting five-year national strategies in many areas, will have this on the agenda.  Look for the BRI to be reaffirmed, for headline capital flow number objectives to be tempered, for a “greening” of the BRI, and for China to continue to be a large factor in this area for years to come.

Lessons from the Texas Grid Failure

February 2021

Everything is big in Texas – even the crises.

Over the past weekend, a blast of winter cold and snow spread from Canada, bringing sub-freezing temperatures and heavy snow to the center of the continent as far south as Mexico.  Northern states managed; Texas did not.  As Texans huddled inside and turned up thermostats, the grid failed.  Over four million Texans lost heat and power, more than forty have died, and tens of millions are under boil water advisory.  Long lines of people assembled in the cold, seeking propane, firewood, or help.  Five days later, many still have no power.

What happened, and what can we learn?

Stripping out the political rhetoric (to which we’ll return), there were four interlocking causes to Texas’ grid failure: high demand, supply failures, Texas’ peculiar regulatory environment, and lack of preparation.  The last of these is arguably least the fault of Texas’ utilities, yet also arguably the most important factor from which to draw lessons.  Cutting across all four of these causes is something larger – climate change.  Let’s take each of these in their turn.

High demand.  When the unexpected cold arrived over the weekend of February 13-14, demand for power and heating surged.  It’s not that Texans turned up the thermostats, but it was taking a lot more electricity – or natural gas, depending on the household – to keep the inside temperature warm.  Texans consume an average of roughly 50 gigawatts (GW) of electricity at a time, and the state has a generating capacity of about 85GW.  By February 15, demand had hit 71 GW.  In the best of cases, this would have caused generators and grid operator to scramble to meet demand, without much room to maneuver.  It was not the best of cases. 

Supply failures.  This is where most of the headlines have landed.  The state’s grid operator said that on February 15 about 34 gigawatts of power were offline.  That’s almost half of the peak power demand which hit the grid.  Wind turbines were the focus of the blame from Texas’ fossil-fuel fraternity, including Governor Greg Abbott.  In practice, inoperative wind turbines accounted for… 4 GW of lost supply.  30 GW, the vast bulk of lost supply, came from natural gas, and to a lesser extent coal and nuclear.  Frozen gauges and instruments at natural gas, coal and nuclear plants cut into operations.  Natural gas-fired plants also had to deal with low gas pressure in their supply lines, as demand for gas from pipelines feeding home heating also surged while gas producers dealt with frozen pipes and difficulties at cold well-heads.  Freezing also forced one of the state’s nuclear power plants offline. The Texas grid, with demand exceeding 70 GW, could only muster 51 GW of supply.  Grid operators, faced with the choice between a “managed” failure of rolling and hopefully temporary blackouts and an unmanaged grid meltdown, went with the blackouts. 

Texas’ peculiar regulatory environment.  Every grid has its own distinct regulatory environment, whether across the different parts of the United States, Europe, or across 200-plus of the rest of the world’s countries.  But some are more distinctive than others, and this is the case of Texas.  The two key characteristics of the Texas system are deregulation and independence – neither unique, but pushed farther in Texas than in most places.  In terms of deregulation, the Texas market encourages constant competition between providers of energy.  This is a good thing in terms of creating incentives for utilities to provide power to customers as cheaply as possible, and a great many other energy markets now do the same thing.  Yet the Texas market is unusual in lacking an accompanying regulatory feature which is common in most other competitive power markets – coupling payments for electricity delivered with capacity payments – payments that create incentives for generators to keep their plants available to deliver power, and not avoid the costs of maintenance (or, say, winterizing).  Most grids have kept this – Texas has not.  It is not just opponents of renewables who are taking the opportunity to politicize the current crisis: proponents of old-styled fully regulated and vertically-integrated monopoly utilities now claim – incorrectly – that any kind of deregulation and competition is bad.  What one should aim for is both cheap (from competition) and reliable (from capacity reserves) electricity.  Texan “independence” has also been a factor.  Texas has far fewer cross-state border electrical connections relative to size than other states, partly because this serves to make the state less subject to federal rules than would otherwise be the case, and partly because Texas is Texas.  This week it has meant that a tool available to many other grid operators – buying surplus power from neighboring grids – was not available to deal with Texas’ crisis. 

Lack of preparation.  This is in some ways the most interesting factor in the crisis.  After all, weekend temperatures were far worse elsewhere than in Texas.  The Southwest Power Pool (SPP), a 14-state system stretching from North Texas to the Canadian border, had limited and rapidly solved problems.  Same with the Midcontinent Independent System Operator (MISO), a 15-state system going from Louisiana to Manitoba.  Yet it was in “relatively warm” Texas that the grid failed from the cold.  ERCOT, the main Texan grid, did to some extent plan for winter troubles. But not enough: ERCOT predicted winter storm-driven demand would peak at about 67 GW (not 71), and that outages from coal and natural gas plants would be limited to about 14 GW (not 39).  Texas had most importantly not asked generators to prepare for cold – insulation and heaters helped keep SPP and MISO electricity coming into the system when Texan turbine blades and gas-fired boilers just froze.  Not exactly rocket science from a technical point of view – though costs that have to be paid for, in a regulatory environment that does not compensate maintaining availability.

Here are four key lessons we can take away from Texas’ tragic week.

  1. Climate change will further stress our power grids
  2. Grids need more investment.
  3. The regulatory environment for electricity transmission needs to catch up with climate change.
  4. Politics is a big part of the infrastructure problem

Lesson # 1.       Climate change will further stress power grids

Texas’ crisis started when it got cold – very cold, and unusually cold.  “Climate change” is often mis-translated into “Global Warming.”  The world is definitely warming – a lot.  But climate change effects include extreme weather events of all sorts, even more cold snaps like the one the southern US is experiencing.  The likely culprit here is warming in the Arctic, which has loosened the mechanism that usually keeps the jet stream far north, locking Arctic air away.  This week the jet stream has bent far south, bringing cold air with it.  The huge winter storms that plunged large parts of the central and southern United States into an energy crisis this week, sounded an alarm for power systems throughout the country.  Electricity systems are designed to handle spikes in demand, but the wild and unpredictable weather linked to global warming is making those spikes much bigger.   As climate change accelerates, many electric grids will face extreme weather events that go far beyond the historical conditions those systems were designed for (see “A Glimpse of America’s Future: Climate Change Means Trouble for Power Grids”). As the NYT also points out, “blackouts in Texas and California have revealed that power plants can be strained and knocked offline by the kind of extreme cold and hot weather that climate scientists have said will become more common as greenhouse gases build up in the atmosphere.”  The common theme in the two states is that many power plants are much more sensitive to temperature changes than the utility industry has acknowledged, said Jay Apt, co-director of the Carnegie Mellon Electricity Industry Center: “coal plants and gas plants have problems in both heat and cold.” Last August, several power plants fired by natural gas stopped generating electricity as Californians were cranking up air-conditioners because equipment at the plants malfunctioned in the hot weather.   Pedro Pizarro, CEO of Edison International, the parent company of California’s second-largest investor-owned utility, said no utility in Texas or California had anticipated the kinds of extreme weather that hit the two states: “both the Texas event and the California event are really good examples that we are all living with climate change.  Electric grid systems need to be able to deal with the new normal.”

In turn, the process of adjustment itself to climate change will create still yet more stress on power grids.  As ever-increasing amounts of intermittent power generation are built, accounting for greater share of generation, and more energy storage is connected both in homes (back of the meter storage) and grids (front of the meter storage) – both good things for emission reduction and for cheap electricity – yet further investments will be needed in grids to adjust to this, too.

Lesson # 2.       Grids need more investment.

Even without the effects of extreme climate events, investment in electricity transmission has lagged behind needs for many years.  Investments in new generating facilities – particularly in renewables – have captured the vast bulk of the sector’s financing in recent decades.  Clean and cheap power is good, but it does require transmission facilities to keep up.  The intermittent nature of wind and solar power can cause grid instability: ten years ago it was thought that intermittent generation of as little as 15-20% could cause major instability; now Denmark, other European countries and grid operators have shown that with sufficient investments in technical resiliency, systems can tolerate having at least 50% of their capacity from intermittent sources.  But a great many middle and lower-income countries have yet to make such investments in technical resiliency.  Wind and solar power, while increasingly the cheapest alternative for new electricity, also changes the geography of generation, requiring new and enlarged transmission lines to take advantage of the cheap power.  More transmission lines would help cheap wind power from the Great Plains, including north Texas, come to markets in the west and east of the US – or, as we saw this week, help an overloaded grid more quickly resolve its problems.  This issue has bedeviled systems across the world, and will only do so more as the energy transition continues.  Unfortunately, growing cyber-security risks to utilities also now require more investment for hardening controls, as Infrastructure Ideas has previously commented.  This week’s crisis was accompanied by at least one faked news story purporting to be from a Texas utility – it could have been much worse if hackers had been targeting the state.  Intelligence agencies warn that the United States’ power grids are increasingly vulnerable to attacks from hackers sponsored by foreign adversaries. Hardening the nation’s electrical supply against cyberwarfare is also needed.

Infrastructure Ideas had predicted that 2020 would be the year that policy-makers woke up to the problem and began catching up – then the pandemic intervened.  2021 may be the year.

Note that if you’re an emerging market, the events in Texas should be even more of a wake-up call.  Antiquated, underinvested electricity transmission grids are a problem that’s going to get a lot worse.  Use available public funding – and World Bank and other external funding where available – to strengthen transmission.  Don’t use those scarce public funds – especially after the COVID pandemic and its fiscal impacts — for Chinese-offered coal-fired generating plants: the debt which may be offered to fund them will have to be paid back before long, their electricity is more expensive than renewables, and their raw material supply will become increasingly problematic.  And don’t use scarce public funds for new wind and solar generating capacity – the world is awash in private capital willing to finance that. 

Lesson # 3.       The regulatory environment for electricity transmission needs to catch up with climate change.

As noted above, generating plants failed in Texas in part because they were not hardened to withstand the sort of severe weather that struck the state this week. This was avoidable. Texas had a similar freeze in 2011, after which the need for system upgrades and better planning was obvious.  Yet none of this happened.  Again as summarized by the NYT, grids can be engineered to handle a wide range of severe conditions.   Wind turbines can be equipped with heaters and other devices so that they can operate in icy conditions — as is often done in the upper Midwest, where cold weather is more common. Gas plants can be built to store oil on-site and switch over to burning the fuel if needed, as is often done in the Northeast, where natural gas shortages are common. Grid regulators can design markets that pay extra to keep a larger fleet of backup power plants in reserve in case of emergencies, as is done in the Mid-Atlantic.  Energy storage can make a big difference, especially as longer-term storage technologies start to emerge.  But these solutions all cost money, and grid operators are often wary of forcing consumers to pay extra for safeguards.  “Building in resilience often comes at a cost, and there’s a risk of both underpaying but also of overpaying,” said Daniel Cohan, an associate professor of civil and environmental engineering at Rice University. “It’s a difficult balancing act.” 

As extreme weather events become more common – without becoming more predictable – a regulatory environment which recognizes these risks, and forces and/or creates incentives for utilities to make these kinds of precautionary investments becomes more important.  Cross-border regulations and cooperation, which encourage investment in transmission lines that can both transfer cheap energy when it is generated and when one system has a failure, also take on a new importance.

Lesson # 4.       Politics is a big part of our infrastructure problem

The Texas crisis shows us that money and change are both needed to improve the infrastructure needed to keep millions from freezing unnecessarily.  Yet infrastructure is easily politicized – and that’s a big problem. 

Building long-term infrastructure is a complex and risky undertaking.  Countries, states and cities that have done so successfully – at least for a time – have benefitted from a good degree of popular consensus as to needs and (at least in general) solutions.  Without that consensus, it is much harder for the public sector to build long-term infrastructure across electoral cycles, much harder to do things differently than before, and much harder to complement public finances with private capital.  This week in Texas illustrates how easy it is to undermine popular consensus.  Under a Republican Governor, Rick Perry – also a former Republican Secretary of Energy, no less – fossil-fueled Texas had achieved a remarkable popular consensus on the value of wind power.  Drive across the north of the state and it seems like a forest of turbines, stretching for dozens of miles: even hotel rooms feature photographs of turbines.  Yet the first thing current Republican Governor Greg Abbott blamed for the crisis was wind energy.  It was surely much easier than blaming one’s own lack of preparedness.  And it illustrates how much harder the next phase of the energy transition will be in Texas.

Modernizing the electrical grid to make it more resilient, more efficient and more secure is the worst kind of challenge: complex, expensive and easy to ignore.  Local companies see overall demand for electricity leveling out, thanks to more efficient homes and businesses, which means a future without growth for their bottom lines.  They lack motivation and capital to build a stronger grid on their own.  Solving the problem will take political will, yet as the past has shown us, once heat and light are restored and the press has stopped covering the crisis in Texas, elected officials and policymakers will tend to quickly move on as well.  

Let’s see how Texas, and the new Biden administration, rise to the challenge.  Communicating clearly to voters what the problem really is and what is needed – and countering false claims about what the problem is not – will be an important part of the challenge.

In closing, this paragraph from today’s Washington Post captures the issue:

The grid is responsible for modern civilization, yet its failure can imperil life itself. It’s a 20th-century antique that, when foiled by 21st-century factors, can send us back to the 19th century, when Thomas Edison’s company built coal-fired dynamos on Pearl Street to illuminate Lower Manhattan.

The Demise of Loon

February 2021

Disruptive technologies have been reshaping the face of infrastructure as never before.  From wind turbines, to solar cells, smart meters, distributed generation, and now battery storage, new technologies have almost completely displaced traditional fossil-fuel sources in electricity supply.  Electric vehicles, automation, and mass-data-enabled business models like Uber and Lyft are similarly turning transportation infrastructure upside down – with Tesla now worth far more than General Motors. Hardly anyone remembers rotary phones, let alone land lines.  Yet not all promising technology breakthroughs survive to become the “next normal,” even with the deepest pockets behind them.

Meet Loon.

It was only a year ago that Loon balloons were being touted as the next big thing in telecommunications, especially in remote or disaster-hit areas.  Owned by no less than Google itself (through Google’s parent company, Alphabet), Loon’s high-flying balloons were, well, high fliers.  Enabled by artificial intelligence, Loon’s helium-filled polyethylene balloons were able to hover in one place for extended periods, serving as air-borne cellular towers in areas where such towers either didn’t exist or had been knocked out.  AI helped predict wind speed and direction at various heights, then use that information to raise and lower the balloon accordingly.  Loon’s balloons work by beaming Internet connectivity from ground stations to a balloon 20 kilometers overhead. The signal is then sent across multiple balloons, creating a network of floating cell sites that deliver connectivity directly to a user’s phone or computer router.

A Loon Balloon in action

Each Internet-enabled balloon covers a large area—roughly 30 times greater than a ground-based system—Loon can provide service to traditionally hard-to-reach or underserved regions.  Loon’s earliest markets were, not surprisingly, exactly these kinds of places: outlying areas of Brazil, Indonesia, New Zealand, Peru, Puerto Rico, and Sri Lanka, and disaster zones.  The first big commercial contract came in Kenya, where a 2018 agreement with Telkom Kenya to support internet access for the 70% of Kenyans who did not have it was followed by emergency grants of balloons to help the Kenyan government deliver health care messaging during the early months of the COVID pandemic.  For Google, all this was great public relations, as well as part of a strategic approach to diversify into other cutting-edge technologies.  The company touted Loon’s ability to bring the internet to the “bottom of the pyramid,” or “the bottom billion.”  A $125 million investment from partner Softbank in 2019 made the future seem promising.

Then in late January, 2021, came the news from Google headquarters: it was pulling the plug on Loon.  In a blog post, Google said “the road to commercial viability has proven much longer and riskier than hoped. So we’ve made the difficult decision to close down Loon.”  For all its promise and disruptive technology, Loon ran out of money. 

What happened?  A mix of good and bad news.  On the positive side, Loon’s technology worked.  The underserved and disaster-hit were able to get cellular and internet service.  At the same time, on the positive side for the under-served “bottom of the pyramid,” but bad news for Loon, other existing technologies were at work narrowing the innovation’s market window.  When Loon was started a decade ago, the share of the world’s population without internet availability was 25% — nearly 2 billion people.  Today it is about 7%, a bit more than half a billion, only a quarter of the market opportunity it was before.  The bad news for Loon is that is that its model is more expensive than more traditional existing technologies, and that the remaining population without service is the most expensive to reach, and the least likely to be able to afford the prices that would keep Loon airborne – not to mention the problem of administering payment of bills.  Potential commercial success had gone from “around the bend” to out of sight.

Google and Alphabet will go on.  New technologies will continue to appear.  But as always, spreading the benefits to the least well-off is the hardest part.  For now, the remaining poor unserved will continue to fall behind the rest of the world.  Yet Loon’s failure is a failure to be applauded.  Perhaps a foundation will partner with international funding agencies to continue stationing floating cellular towers to reach the poor – on a fully subsidized basis – if connectivity is deemed an important enough developmental goal.  A lot of rooftop solar has been built this way in low-income countries, and product development costs – possibly as much as half a billion dollars – have already been absorbed by one of the few companies which could afford it.  Time will tell.

And in the meantime, this tale illustrates that technological disruption in infrastructure continues to broaden – and that the early winners become the likeliest to have the money to kickstart more innovation.  Loon’s balloons may or may not live another life; that there will be more tales such as Loon’s is certain.

Previous Posts on Disruptive Technologies

Ten Infrastructure Predictions for 2021

January 2021

As for the past several years, we start the new year by looking into our crystal ball and seeing what these twelve months are likely to bring for infrastructure operators, investors and policy-makers (see here for Infrastructure Ideas2018, 2019 and 2020 predictions, and here for how well the predictions tracked for 2018, 2019 and 2020).  Here are ten infrastructure predictions for 2021.

  1. A post-COVID boom for new renewable capacity.  The ongoing COVID pandemic and its ensuing disruptions was the obvious big infrastructure story for 2020, but there were a few segments of outperformance.  Renewable energy managed to hold its own, and now after a few years of generally flat levels of activity globally, is poised to return to significant growth.  Global investment in new renewable energy capacity inched up 2% in 2020 to $304B, according to Bloomberg New Energy Finance, but this level has been essentially flat since 2015.  Underneath the aggregate numbers, patterns are more positive than they’ve been in some time for growth: 2020 was underpinned by new renewables investments in Europe, which is likely to continue to be the case under the EC’s “green recovery” plans; meanwhile investment fell in the two largest individual markets, China and the US, to $84B and $49B respectively.  In both China and the US, we can expect the combination of a return of demand growth (China’s economic growth rate is forecast to be the highest in years), the cost advantages of renewables, and the return of pro-renewable policy under the Biden administration, to underpin a jump in new wind and solar investments in both these markets.  In China in particular, we expect prices of new solar capacity to drop significantly, as the country continues its transition away from its older Feed-in-Tariff procurement mechanism for domestic solar generation towards competitive auctions.  Look for a record-breaking year in total investment and in Gigawatts of new solar capacity added worldwide, and another record for renewables as a share of net new generating capacity added worldwide, at over 70%.
  2. The energy storage market gets back on track.  Prices of energy storage have been tumbling, while the size of utility-installed batteries has been soaring.  The cost of a four-hour storage addition to new generation capacity has fallen from over $80/megawatt-hour in 2010 to less than $10 today.  Nonetheless new installed energy storage capacity has fallen by 15-20% each of the last two years, down to $3.6 billion in 2020, largely due to regulatory uncertainties.  2021 will see a completely different story.  With solar-plus-storage costs for new generation capacity beginning to match the costs of new gas-fired plants, more and more utilities are switching new projects from gas to renewables plus storage.  And with the Biden administration in the US focused on getting a favorable regulatory environment in place, we can expect a surge in new capacity additions in the US.  The last few months have already seen this emerging: according to the Energy Storage Association, fourth-quarter 2020 deployments of energy storage in the US more than doubled those of any previous quarter on record.  The EIA expects a record 4.3 GW of new battery power to be added worldwide in 2021, and we agree – that should imply about $5 billion in investment — and we also expect grid-scale capacity to exceed not only 2 GW for the first time, but to reach between 2.5 and 3 GW.
  3. More airline bankruptcies.  The Fallout of COVID for all sorts of transport infrastructure for moving people has already been horrendous – whether airlines, mass transit, or taxis.  The flow of red ink is far from over.  Infrastructure Ideas reviewed the situation of airlines back in mid-2020 (The Airline Shakeout Starts Up), and by year-end over 40 carriers had declared bankruptcy.  Today many others have fragile Balance Sheets from hemorrhaging cash all year, and there is little sign of any turn around in air traffic demand in the next few months.  IATA says airlines lost over $80 billion in 2020, and projects the industry to lose $5-6 billion a month in the coming year.  Watch for more carriers to fall by the wayside in 2021 (for more see Over 40 airlines have failed in 2020 so far and more are set to come).
  4. Rethinking mass transit.  COVID has also been a disaster for mass transit infrastructure everywhere.  Ridership across US metropolitan systems fell by 65-90%.  Revenue shortfalls have forced transit authorities to cut routes and frequencies, and delay expansion and maintenance.  These measures will unfortunately create a negative feedback loop: transit systems which run fewer, slower routes, less reliably, will attract fewer riders, even when pandemic concerns eventually recede.  The $20 billion for mass transit in the Biden Administration’s “Rescue America” plan will reduce the damage to an extent, but we can expect the financial wreckage to last several years.  Infrastructure Ideas expects several consequences: (a) further reductions and delays in planned expansions of mass transit systems worldwide; (b) a sharp falloff in interest in new subway plans, including across Emerging Markets, and their replacement by cheaper Bus Rapid-Transit plan; (c) new partnerships between municipal mass transit systems and “shared mobility” players (bicycle, scooter, and car-sharing companies). 
  5. Cyber risks grow for Utilities.  Regrettably, this has become a “safe” annual prediction.  2020 saw a worldwide increase in the frequency and scope of cyberattacks on a wide range of targets, including infrastructure.  Aside from the much-publicized Solar Winds hack which, along with breaching several parts of the US government, exposed several infrastructure systems in the US, 2020 also saw several other known, and no doubt more unknown, attacks.  In February, a US natural gas compressor unit was closed for two days after dealing with one incident.  In April, a pair of cyberattacks were reported on electric utilities in Brazil.  In June, Ethiopia reported it had thwarted a cyberattack from an Egyptian group aimed at creating pressure against the filling of the Grand Ethiopian Renaissance Dam on the Nile.  Concerns run across geographies, including Africa.  A recent McKinsey analysis found three characteristics that make the energy sector especially vulnerable to contemporary cyberthreats:  an increased number of threats and actors targeting utilities, including nation-state actors and cybercriminals; utilities’ expansive and increasing attack surface; the electric-power and gas sector’s unique interdependencies between physical and cyber infrastructure.  Look for the headlines to get worse in 2021.
  6. Joint action on climate… finally.  With the exit of the Trump administration, the stage is reset for multilateral progress on climate change.  2020 was either tied or in second place for the hottest year on record, and that’s with the pandemic-induced slowdown in economic activity and emissions.  Most analysis now show the world on track for at least 3 degrees if not significantly more of warming (see McKinsey’s analysis of the world being on a 3.5 degree track), and the damage from heat waves, storms and flooding continues to increase.  Joe Biden already on his first day in office committed the US to return to the Paris Climate Accords, and China has become more aggressive in its emission reduction undertakings.  Look for new substance at the November climate summit in Glasgow, COP26.  In particular, look for (a) announcements of further reductions beyond those undertaken by countries in the Paris Accords, and (b) the emergence of a clearer tracking system to “grade” countries on how their actions are matching their commitments – a key missing element in the global frameworks to date.  The first baby steps beyond “voluntary” action.
  7. Cash for clunkers makes headway.  Coal-fired plant closures have brought constant positive emissions-related headlines over the past few years.  Last week came the announcement of the upcoming closure of a large Florida coal plant – 18 years early.  As a good piece by Justin Guay in Green Tech Media put it “Nearly every day, articles appear announcing new record lows in coal generationcoal retirements and the generalized economic train wreck that is the coal industry.”  Yet these headlines are not yet enough to bring the world back to a 2-degree warming scenario – and probably not enough to keep it even to a 3-degree scenario.  To be on track to meet the Paris Agreement goals, every coal fired-plant in the OECD would have to be offline by 2030, and every coal-fired plant in the rest of the world would have to be by 2040.  In OECD countries, almost half the existing coal-fired generation plants are not earmarked for retirement before 2030, so a lot of work will be needed there.  The biggest rich-country coal users – Japan, the US, Germany and Australia – are in the best of cases a decade off schedule.  Yet this dwarfs the complications of the rest of the world, especially Asia.  China has 1,000 gigawatts of young coal plants – almost half the world’s total coal-generation capacity, and is still building new ones.  India and the rest of Asia have about 400 gigawatts of coal-fired generation, need much more electricity, and are still locked in internal debates as to how much of their future energy needs are to be with coal (see Infrastructure Ideas’ series on Asia’s Energy Transformation: Pakistan, Bangladesh, India, and Indonesia).  The technical lives of many of these plants will stretch long past when they would need to be shuttered to meet the Paris accords, and many of them are insulated from the declining economics of coal by quasi-monopolies and/or long-term contracts.  According to Carbon Tracker, the US and the EU will, by next year, be paying coal plants over $5 billion to stay in operation, through contracted capacity payments.  It would be much better to use these funds to buy the plants out and close them, in effect a “cash-for-clunkers” program as Justin Guay labels it.  As an earlier Infrastructure Ideas piece puts it, “Money is Coming for Coal.”  The funds would be needed to buy out legacy operators, and to support affected workers and communities.  This will be controversial, and complex to design and implement.  But with emissions likely rebounding again and a more favorable political environment, look for paying coal to go away to get on the table in 2021.
  8. The US gets its trillion-dollar infrastructure plan.  There were plenty of promises in 2016 about a trillion-dollar infrastructure plan for the US to fix many of its problems, but this never materialized.  Now with a new administration, and democratic control of both houses of Congress, there will surely be such a plan put in place in 2021, with roll-out getting underway.  The nomination of Pete Buttigieg as Secretary of Transport indicates that urban infrastructure will be a priority, and that municipal authorities will get much more say going forward on how funding helps address cities’ infrastructure needs.  Buttigieg had his own trillion-dollar plan as a candidate (see “Inside Buttigieg’s $1 Trillion Infrastructure Plan”) in the primaries, and stated “as a former mayor, I know that priority-based budgets made locally are better than budget-based priorities set in Washington.”  This will be in sharp contrast to the previous four years, when whatever federal funding trickled out was aimed almost entirely at the rural areas which were the base of Donald Trump’s support.  Climate adaptation and road rebuilding were high on both Buttigieg and Joe Biden’s campaign pronouncements, so look for major spending in these areas in 2021.  President Biden apparently also plans to re-create a version of the depression-era Civilian Conservation Corps to work on climate adaptation projects.
  9. The BRI gets a facelift.  In September, Chinese President Xi Jinping pledged to make China carbon neutral by 2060, and to “bring forward” an earlier pledge to start reducing GHG emissions by 2030.  The announcement was widely welcomed, but it will be a hard slog to turn into reality: with economic pressures, 2020 saw a sharp increase in the number of permits for new coal-fired plants issued in China.  China’s emissions progress will likely stay in the limelight as international climate discussions get more serious in 2021, thanks to the re-engagement of the US (see above).  At the same time, China’s flagship international initiative, the Belt and Road Initiative (BRI), is seeing increased criticism of its environmental and climate impacts.  Coal-fired generation plants have been big recipients of support under the BRI, particularly in South Asia.  Announcing some sort of “greening” of the BRI going forward would be low hanging fruit for Xi Jinping to avoid focus on the BRI’s environmental negatives at a time China wants to be seen as a leader of the international agenda.  Look for this to come to pass later in 2021.
  10. This is (not) the time for the Emerging Markets infrastructure boom.  There is one coming – really!  For years policy-makers, analysts and investors have looked at Emerging Markets as the great future of infrastructure.  Large infrastructure deficits, growing wealth and demand for services among the population, higher returns than in wealthy markets, coupled with a “wall of money” from institutional investors looking to get some yield on their excess liquidity.  In 2021 it … will not happen.  The demand pull will stay largely theoretical.  Of the ten or so larger economies that make up 80% of collective GDP of Emerging Markets, four of the biggest – Brazil, Mexico, South Africa and Turkey – will at best remain hamstrung from a combination of COVID and internal politics, and at worse turn their back on private investment.  The “push” from investors will be going elsewhere.  Between a big push for new infrastructure in the US, and the European “Green Recovery” plan, investors and infrastructure companies will be looking for their opportunities in developed markets.  And between the Trump tax cuts and forthcoming public spending increases, look for interest rates to start inching up, further reducing the push from institutional investors.  At some point continued internal pressures, and limited public spending options, will lead to a wave of Emerging Market reforms.  Just don’t look for it in 2021.