Category Archives: greenhouse gas emissions

The case for additional actions in sectors covered by the EUETS is now even stronger

Recently agreed reforms to the EUETS mean that excess allowances in the MSR will be cancelled.  This further strengthens the case for actions such as phase-out of coal plant, increasing energy efficiency and deploying more renewables.

About a year ago I looked at whether additional actions to reduce emissions in sectors covered by the EUETS do in practice lead to net emissions reductions over time [i].

It is sometimes claimed that total emissions are always equal to the fixed cap, and by implication additional actions do not reduce total emissions.  This is sometimes called the “waterbed hypothesis” by analogy – if you squeeze in one place there is an equal size bulge elsewhere.

Although often repeated, this claim is untrue.  Under the EU ETS at present the vast majority of emissions reductions from additional actions will be permanently retained, reflecting the continuing surplus of allowances and the operation of the MSR.  Furthermore, over the long term the cap is not fixed, but can respond to circumstances.  For example, tighter caps can be set by policy makers once emissions reductions have been demonstrated as feasible.

When I last looked at this issue, the fate of additional allowances in the MSR remained necessarily speculative.  It was clear that additional excess allowances would at least not return to the market for decades.  It also seemed likely that they would be cancelled.  However, no cancellation mechanism was then defined.

This has now changed with the trilogue conclusions reached last week, which include a limit on the size of the MSR from 2023.  The limit is equal to the previous year’s auction volume, and is likely, given the size of the current surplus, to lead to large numbers of allowances being cancelled in the 2020s.

With this limit in place there is a very clear pathway by which allowances freed up by additional actions, such as reduced coal burn or increased renewables, will add to the surplus, be transferred to the MSR then cancelled (see diagram).  Total emissions under the EUETS will be correspondingly lower.

There is now a clear mechanism by which additional actions reduce total emissions

Modelling confirms that with the limit on the size of the MSR in place a large majority of reductions from non-ETS actions are retained, because additional allowances freed up almost all go into the MSR, and are then cancelled.  This is shown in the chart below for an illustrative case of additional actions which reduce emissions by 100 million tonnes in 2020.  Not all of the allowances freed up by additional actions are cancelled.  First there is a small rebound in emissions due to price changes (see references for more on this effect).  Then, even over a decade, the MSR does not remove them all from circulation.  This is because it takes a percentage of the remainder each year, so the remainder successively decreases, but does not reach zero.  If the period were extended beyond 2030 a larger proportion would be cancelled, assuming a continuing surplus.  Nevertheless over 80% of allowances freed up by additional actions are cancelled by 2030.

The benefit of additional actions is thus strongly confirmed.

The large majority of allowances freed up by additional actions are eventually cancelled

Source: Sandbag

When the market eventually returns to scarcity the effect of additional actions becomes more complex.  However additional actions are still likely to reduce future emissions, for example by enabling lower caps in future.

Policy makers should pursue ambitious programmes of additional action in sectors covered by the EUETS, confident of their effectiveness in the light of these conclusions.  Some of the largest and lowest cost gains are likely to be from the phase out of coal and lignite for electricity generation, which still accounts for almost 40% of emissions under the EUETS.  Continuing efforts to deploy renewables and increase energy efficiency are also likely to be highly beneficial.

Adam Whitmore – 15th November 2017

[i] See  For further detail see .  A study by the Danish Council on Climate Change reached similar conclusions, extending the analysis to the particular case of renewables policy.  See Subsidies to renewable energy and the european emissions trading system: is there really a waterbed effect? By Frederik Silbye, Danish Council on Climate Change Peter Birch Sørensen, Department of Economics, University of Copenhagen and Danish Council on Climate Change, March 2017.

New long term targets for emissions reduction are needed.

The UK and other jurisdictions need to set target dates for reaching net zero greenhouse gas emissions.  These need to be reinforced by new targets for 2060 that are at least close to zero, and by reaffirmed or strengthened targets for 2050.

Ten years ago setting emissions reduction targets for 2050 was a major step forward

2018 sees the tenth anniversary of the UK’s Climate Change Act[i].  This remarkable piece of legislation established a legally binding obligation for the UK to reduce its greenhouse gas emissions by 80% from 1990 levels by 2050, with obligations along the way in the form of five year carbon budgets.  So far progress has been remarkably good, though significant challenges remain.

Other jurisdictions also adopted 2050 targets at around the same time.  In 2005 California also set a target of an 80% reduction from 1990 levels[ii].  In October 2009 the EU established a long term EU goal for reducing emissions by 80-95% from 1990 levels by 2050[iii].

At the time these targets were path breaking.  However, ten years on there are good reasons for reviewing and extending them.

But now the world has moved on …

  • When the targets were established, the period to 2050 seemed long enough to give appropriate strategic guidance to policy makers and investors. However, future dates are now ten years closer.  A 2060 target now gives about the same time horizon for planning as the 2050 targets did when they were established.
  • The Paris Agreement sets targets to limit temperature rises which imply stringent limits on cumulative emissions. It also sets a goal of net zero global emissions in the second half of the century.
  • A fifth or more of the world’s carbon budget that remained in 2008 has since been used up[iv], increasing the urgency of emissions reductions.

Extending targets to reflect these changes would have some clear benefits … 

Together these changes imply a strong case for setting new targets now.

The most compelling target would be a date by which emissions must fall to net zero.  Such a target would make it clear to all sectors that they need to completely decarbonise by a specified date.  At the moment emissions of up to 20% of 1990 levels are allowed even in 2050.  This allows each of those sectors where decarbonisation is more difficult – for example parts of industry, agriculture or residential heating – to largely continue in a belief that there will still be plenty of room for them within the 2050 emissions limit, even though this cannot be true for most sectors.  This in turn allows them to continue to believe they can carry on indefinitely without taking the steps needed to decarbonise.  A date for reaching zero makes it clear this can’t happen.

Setting stringent target for 2060 – at or close to zero – would also give investors in low carbon infrastructure greater confidence, and deter investment in higher carbon alternatives. In the case of the UK and California, a simple extrapolation of their current targets would suggest a 2060 target of a 93% reduction from 1990 by 2050, reaching zero by 2065.

As part of the process of setting these longer term goals the existing 2050 targets need to be at least reaffirmed and preferably tightened.  If this is not done there is the risk that policy makers will simply see the problem as having become more distant, and delay action.  This is the last thing that the climate needs.

2050 targets may also need to be revised …

As a first step, the EU’s target of 80-95% cuts clearly needs to be made more precise.  The current uncertainty of a factor of four in the level of emissions allowed in 2050 is too wide for sensible policy planning.

However the events of the last ten years also raise the question of whether the stringency of the 2050 targets need to be increased, with implications for later periods.  The UK Government’s former Chief Scientific Adviser Sir David King and others have suggested that there is a strong case for the UK seeking to reach net zero emissions by 2050[v].  The difference in cumulative emissions in declining linearly to net zero by 2050 instead of by 2065 is substantial, at a little over 3 billion tonnes – equivalent to about 8 years of current UK emissions.

The goal of reaching zero emissions by 2050 is clearly desirable in many ways.  However there is a risk that it may have unwanted side effects.  The government’s advisory body, the Committee on Climate Change has pointed out that policies are not in yet place even to meet current goals for the fifth carbon budget in around 2030[1].  The route to net zero emissions in 2050 – just over 30 years from now – looks even less clear.  Indeed reaching that goal even by 2065 remains challenging.  If even tighter targets are introduced they may come to be regarded as unrealistic, which may in turn risk weakening commitment to them.  A somewhat slower emissions reduction track may prove a relatively acceptable price to pay for retaining the credibility and integrity of the targets.

Whatever the judgement on this, the need for longer term targets is clear.  Governments need to set dates for reaching net zero emissions.  These need to be supported by targets for 2060 that specify continued rapid reductions in emissions after 2050, and by reaffirmation of 2050 targets, tightening them as necessary.  These new targets will in turn help stimulate the additional actions to rapidly reduce emissions that are ever more urgently needed.

Adam Whitmore – 6th November 2017





[iii] measures_data/docs/pressdata/en/ec/110889.pdf

[iv] The calculation is based on data in the IPCC Fifth Assessment Report, Synthesis Report.  This quotes a  cumulative budget of 3700 billion tonnes of CO2 for a two thirds probability of staying below 2 degrees.  Of this 1800 billion tonnes had been used by 2011.  Assuming CO2 emissions of roughly 40 billion tonnes p.a. including land use gives a remaining budget in 2008 of 1920 billion tonnes.  Over the subsequent ten years about 400 million tonnes CO2, which is just over a fifth of 1920 billion tonnes, have been emitted.


Climate change: how did we get here, and why is it so hard to fix? (Part 1)

Activities that cause emissions are ubiquitous, diverse and deeply embedded in modern life.  The world’s energy system is huge and long-lived.  This makes emissions tough to deal with. 

This post is the first of two stepping back a little from the specific topics I usually cover to take a very high level look at why the climate change problem is so hard to fix.  This first post looks at how we got here and (at a very high level) the physical and engineering challenges of addressing the climate change problem.  The next post will consider some of the political and psychological barriers to greater action.

The consequence of industrialisation

The world’s climate was remarkably stable from before the birth of agriculture, some 8-10,000 years ago, until very recent times[1].  Human civilisation grew up in a stable climate, and knew nothing else, despite the calamities caused on occasions by storms, floods, drought, and so forth.

Industrialisation changed this.  There is no single year that definitively marks the beginning of industrialisation, but 1776 probably as good a reference point as any.  It was an eventful year, with the US Declaration of Independence giving history one of its most famous dates, while elsewhere the first edition of Adam Smith’s Wealth of Nations was published and the Bolshoi Theatre opened its first season.  But in the long view of history perhaps more important than any of these was that James Watt’s steam engines began to power industrial production[2].  This, more than any other event, marks the beginning of the industrial era.

In the nearly two and a half centuries since 1776, world population has grown by almost a factor of about 10.  Economic output per person has also grown by a factor of about 10.  Taking these two changes together, the world’s economic activity has increased by a factor of about 100.  This has put huge stresses on a range of natural systems, including the atmosphere[3],[4].

The increase in the use of fossil fuels has been even greater than the increase in industrial activity.  Around 12 million tonnes of fossil fuels, almost entirely coal, were burnt each year before 1776[5].  Today the world burns about 12 billion tonnes of fossil fuels each year, an increase of a factor of 1000[6].

This huge increase in the burning of fossil fuels is now – together with deforestation, agriculture and a few other activities – changing the make-up of the atmosphere on a large scale.  This in turn, is changing the world’s climate.   Within a single human lifetime – just one percent or so of the time since the birth of agriculture – changes to the climate are likely to be much greater than human civilisation has ever before experienced.  The consequences of these changes are likely to be largely harmful, because manmade and natural systems are largely adapted to the world we have, not the one we are making.

The characteristics of the systems that have led to these changes also make the problems hard to address.

The scale of emissions is huge …

The scale of CO2 emitted from the energy system is vast, around 36 billion tonnes p.a.  If this were frozen into solid form as “dry ice” it would cover the whole of Manhattan Island to the depth of about two thirds of the Empire State building.

The system that generates these emissions is correspondingly huge.  The world’s energy system cost tens of trillions of dollars to build, and is correspondingly immensely expensive to replace.

The diversity and dispersion of emissions makes the problem more challenging …

The problem is worse even than its scale alone suggests.  It would be simpler to deal with emissions if they were all in one place, whether Manhattan or elsewhere, and in solid form.  Instead emissions are dispersed across billions of individual sources around the world.  And they come from many different types of activity, from transporting food and powering electronics to heating and cooling homes and offices.  There is no single technology doing one thing to be replaced, but a wide diversity of technologies and applications.

And once emissions get into the atmosphere the greenhouse gases are very dilute.  Carbon dioxide makes up only 400 parts per million (0.04%) of the atmosphere.  Among other things this makes capture of CO2 once it has got into the atmosphere difficult and expensive.

And assets producing emissions are very long lived …

Energy infrastructure often lasts many decades, so changing infrastructure tends to be a long term process, with premature replacement expensive.  And on the whole the existing system does its job remarkably well.  Some political considerations aside, there would be little need for very rapid changes to the system if it were not for climate change and other forms of pollution.

Energy is central to modern life …

Finally it’s not possible to simply switch off the world’s energy system because it is essential to modern life.  Hurricane Sandy disrupted much of New York’s energy system, and the consequences of that gave an indication of how quickly modern life collapses without critical energy infrastructure.

These physical characteristics of the problem are compounded by the political and psychological obstacles to change at the necessary scale.  I will return to these in my next post.

Adam Whitmore – 22nd May 2017


[1] This climatically stable period since the end of the last ice age between 11,000 to 12,000 years ago is referred to as the Holocene.  Agriculture started not long after the ice sheets retreated and the world warmed.  Human activity has now led to a new period, referred to as the Anthropocene.

[2]  The first use of the Watt engine to provide the rotary power, which was crucial for factories, was a little later in 1782 at the Soho manufactory near Birmingham.



[5]Reliable data is obviously hard to come by that far back, but See Energy for a Sustainable World: From the Oil Age to a Sun-Powered Future By Vincenzo Balzani, Nicola Armaroli .  They estimate 10 million tonnes in 1700 and 16 million tonnes by 1815.  The majority of the increase would have been in the later part of this period.  See also Socioecological Transitions and Global Change, edited by Marina Fischer-Kowalski, Helmut Haberl, who quote estimates of 3 million tonnes p.a. in 1700 in the UK, a large proportion of the world total at the time, with little increase to 1776.  This consumption included a few primitive, inefficient steam engines, used mainly for pumping water from coal mines themselves.  The Newcomen steam engine required such large quantities of coal that it was rarely economic to site it away from coal mines.  The Watt engine was more than twice as efficient.

[6] My estimate of the total mass of coal, oil and gas, based on data in BP statistical review of World Energy.