Why Have Climate Negotiations Proved So Disappointing?

Document technical information

Format pdf
Size 171.2 kB
First found May 22, 2018

Document content analysis

Category Also themed
not defined
no text concepts found


Carl Perkins
Carl Perkins

wikipedia, lookup




Sustainable Humanity, Sustainable Nature: Our Responsibility
Pontifical Academy of Sciences, Extra Series 41, Vatican City 2014
Pontifical Academy of Social Sciences, Acta 19, Vatican City 2014
Why Have Climate Negotiations
Proved So Disappointing?
I’m grateful to the organizers for proposing this question for my title, because it’s important. People often complain that the climate negotiations have
been disappointing, only to wring their hands and say that we must do better.
But unless we know the reasons why the negotiations have been disappointing,
we won’t know how to do better. Using a medical metaphor, if our diagnosis
of the illness is wrong, our recommended treatment is unlikely to heal the
patient. Indeed, the wrong treatment may only make the patient sicker.
One of the striking things about the climate negotiations is that the negotiators have admitted that they have failed to meet their own goal.
In the Framework Convention on Climate Change, adopted in 1992,
parties agreed that atmospheric concentrations of greenhouse gases should
be stabilized “at a level that would prevent dangerous anthropogenic interference with the climate system”. Later, in the non-binding Copenhagen
Accord adopted in 2009, countries recognized “the scientific view that the
increase in global temperature should be below 2 degrees Celsius”. Finally,
in Cancun in 2010, the parties to the Framework Convention reaffirmed
this goal, but added that it may need to be strengthened, limiting temperature rise to 1.5°C.
After Copenhagen, countries submitted pledges for reducing their emissions. However, analysis by Rogelj et al. (2010) shows that even an optimistic
reading of these pledges implies that mean global temperature will surpass
the 2°C temperature change target.
The negotiators agree with this assessment. In Durban in 2011, they noted
“with grave concern the significant gap between the aggregate effect of Parties’
mitigation pledges in terms of global annual emissions of greenhouse gases
by 2020 and aggregate emission pathways consistent with having a likely
chance of holding the increase in global average temperature below” the
agreed threshold. Rogelj et al. (2010: 1128) describe this behaviour as being
“equivalent to racing towards a cliff and hoping to stop just before it”.
Note that the Copenhagen pledges are voluntary. It’s possible that they’ll
be exceeded. As bad as things look now, they could turn out to be worse.
The problem isn’t disagreement about what should be done. Support
for the 2°C goal is universal.The problem is that this is a global goal. Every-
Sustainable Humanity, Sustainable Nature: Our Responsibility
one is responsible for meeting it, meaning that no country is responsible
for meeting it. Limiting climate change requires very broad cooperation. It
requires collective action.
The reason collective action has eluded us so far is that reducing emissions is a prisoners’ dilemma game. Each country is better off when all
countries reduce their emissions substantially. But each country has only a
small incentive to reduce its own emissions.
The Kyoto Protocol asks some countries to reduce their emissions beyond “business as usual”, and so confronts the prisoners’ dilemma head on.
But Kyoto has failed in its mission to limit emissions.1 The Copenhagen
Accord, by focusing on the need to avoid “dangerous” climate change, has
tried to reframe the problem. Will Copenhagen succeed? Reframing is the
right strategy. But we need a different framing. Although my main purpose
is to explain why the negotiations have been disappointing, I also want to
suggest how an understanding of this failure can provide insights into how
we might do better. Countries are currently trying to negotiate a new kind
of climate agreement for adoption in 2015. My paper ends by suggesting
how a different framing of the climate collective action problem could turn
the negotiations around.The climate negotiations needn’t be as disappointing as they have been so far.
The “dangerous” climate change game
Here is a way to think about the “dangerous” climate change game. Let’s
say that there exists a red line for “danger”, and that countries know what
this red line is. For example, it might be the 2°C goal. Let’s also say that the
impact of crossing this threshold is expected to be so severe relative to the
costs of staying clear of it that all countries, collectively, prefer to stay clear of
it. Then it’s obvious what countries should do if they act collectively: they
should limit concentrations of greenhouse gases to avoid crossing the red line.
The problem, of course, is that countries don’t act collectively. They’re
sovereign. They act independently. So, we should ask: What incentives do
countries have to stay within the good side of the red line?
Under reasonable assumptions, the game I have just described is not a
prisoners’ dilemma. It’s a “coordination game” with two Nash equilibria
(Barrett 2013).2 In one, countries stay just within the “safe” zone. In the
See, especially, research by Aichele and Felbermayr (2011), discussed later in this
paper, which takes into account the effect of Kyoto on trade “leakage”.
In a coordination game, people want to do what others are doing. A car may be
driven on the left or right side of the road. So, on which side of the road should you
Sustainable Humanity, Sustainable Nature: Our Responsibility
other, they breeze past the tipping point, making catastrophe inevitable. In
general, game theory has trouble predicting how countries will behave in
a coordination game. However, since the “bad” Nash equilibrium is so obviously bad, staying within the red line is focal (Schelling 1960). Moreover,
with the help of a treaty, countries can virtually guarantee that they will
coordinate around the “good” Nash equilibrium.
Here is how the treaty should be written. It should assign to every country an emission limit, with each country’s limit chosen to ensure that, when
all the limits are added up, concentrations stay within the “safe” zone. The
limits should also be chosen to ensure that every country is better off staying
within its assigned limit, given that all the other countries stay within their
assigned limits. Finally, the agreement should only enter into force if ratified
by every country.
The beauty of such a treaty is that it makes every country pivotal. If every
other country behaves as required, each country has an incentive to behave as
required.The reason: even the slightest slip up guarantees catastrophe.
This treaty, like all good treaties, transforms the game. In the treaty participation game, every country has a dominant strategy to participate.Why?
Every country has nothing to lose by joining. If the agreement fails to enter
into force, each country would be free to act as it pleased. If the agreement
were to enter into force, however, then it would be binding on all parties,
and catastrophe would be avoided – the outcome every country prefers.
Every country is thus always better off participating.
How would the emission limits for individual countries be chosen? This
is the bargaining problem. If countries were “symmetric”, bargaining would
be simple. Technically, a wide variety of allocations would satisfy the requirements I described above, but an equal allocation would be focal, and
for that reason is to be expected. Asymmetry makes bargaining more complex, but so long as collective action promises all countries an aggregate
gain, there will exist an allocation of responsibility that will be acceptable
to every country (this allocation won’t be unique and may require side payments, but these are relatively minor points compared with the imperative
to avoid catastrophe).
drive? The answer depends on where everyone else is driving. In Italy, it’s obvious that
you should drive on the right. In the UK, it’s obvious that you should drive on the left.
These different outcomes (driving on the left and driving on the right) are each a “Nash
equilibrium”. Given that others are driving on the left (right), each driver chooses to
drive on the left (right).
Sustainable Humanity, Sustainable Nature: Our Responsibility
The threat of catastrophe simplifies the negotiation problem. It makes
each county’s promise to stay within its agreed limits credible.
The vulnerability in this game isn’t the behavior of the countries. It is
the credibility of the science – in particular, the science of locating the critical
tipping point.
In the dangerous climate change game, it would be irrational for any
country to exceed its assigned amount of emissions when doing so would
cause atmospheric concentrations to cross the catastrophic tipping point.
In this case, free rider deterrence depends on the credibility of Nature’s
threat to tip a critical geophysical system. Yes, in the dangerous climate
change game, Nature is an important player.
The importance of scientific uncertainty
What would countries need to do collectively to prevent dangerous climate change?
The Copenhagen Accord says:
To achieve the ultimate objective of the Convention to stabilize
greenhouse gas concentration in the atmosphere at a level that would
prevent dangerous anthropogenic interference with the climate system, we shall, recognizing the scientific view that the increase in global
temperature should be below 2 degrees Celsius [emphasis added], on the
basis of equity and in the context of sustainable development, enhance our long-term cooperative action to combat climate change.3
Why defer to the “scientific view”? The reason is that it simplifies the negotiations. It allows negotiators to bargain over individual country shares.
However, while scientists warned of “climate disaster” before the Framework Convention was adopted (see, in particular, Mercer 1978), I think the
Framework Convention caused scientists to focus on this question at least
as much as previous scientific research caused negotiators to focus on it.
Reference to “the scientific view” [emphasis added] implies that there is
strong agreement among scientists about the threshold.There isn’t.The only
“scientific view” that I detect in the literature is that thresholds are likely
to exist.4
The Copenhagen Accord was written somewhat hastily, and this temperature target is
identified without reference to a base level of temperature. In Cancun the following year,
negotiators clarified that the temperature reference target was the pre-industrial level.
Rapid changes in temperature have been observed in the paleoclimatic record, an
example being the Younger Dryas; see Broecker (1997).
Sustainable Humanity, Sustainable Nature: Our Responsibility
Although temperature thresholds are uncertain, the uncertainties involved in choosing a target are even greater than this.The Framework Convention specifies the target in terms of concentrations, not temperature, and
converting temperature to concentrations introduces an additional layer of
uncertainty – something known as “climate sensitivity” (Roe and Baker
2007). Moreover, we don’t know the quantity of global emissions (expressed, perhaps, as a cumulative sum) needed to meet any particular concentration target, due to uncertainty in the carbon cycle. For example, there
is uncertainty about how much of the CO2 emitted will be taken up by
soils and the oceans. The fifth assessment report by the Intergovernmental
Panel on Climate Change calculates a global “budget” in terms of cumulative emissions that will keep temperature change within 2˚C, but all of these
values are probabilistic. Even if we knew for certain that 2˚C were the true
red line for climate change (and we don’t know this), countries would have
to decide how to balance the cost of reducing emissions with reductions in
the risk of crossing the red line.
The dangerous climate change game with uncertain thresholds
Uncertainty about the threshold for “danger” changes the climate change
game fundamentally.
Consider a very simple game – no treaty. In stage 1, countries choose
their emission levels independently. In stage 2, Nature chooses the tipping
point. When making their choices in stage 1, the players know the probability density function for the tipping point.What they don’t know is which
value under this function will be chosen by Nature – the “true” value for
the tipping point.
An example will make this clear. Rockström et al. (2009) argue that atmospheric CO2 concentrations should be constrained “to ensure [emphasis
added] the continued existence of the large polar ice sheets”. They note
that the paleoclimatic record implies “that there is a critical threshold between 350 and 550 ppmv”, and interpret this as saying that if concentrations
are limited to 350 ppmv, then the ice sheets will be preserved, whereas if
concentrations rise to 550 ppmv, then the ice sheets will be lost. In between
these values there is a chance that the ice sheets will disappear, with the
probability increasing with the concentration level. (For reference, last
month’s reading from Mona Loa was about 399.65 ppmv; at the start of the
industrial revolution, concentrations were about 280 ppmv; when the
Framework Convention was adopted they were 356 ppmv).
Sustainable Humanity, Sustainable Nature: Our Responsibility
Assume that the probability density function is uniform over the range
(350, 550).5 Assume also that the expected aggregate benefit of reducing the
threat of catastrophe exceeds the cost. What should countries do? Under
reasonable assumptions (Barrett 2013), the collective-best outcome is to limit
concentrations to 350 ppmv.This implements the “precautionary principle”.
Acting independently, however, countries have incentives to reduce emissions
only up to the point where their expected individual marginal benefit equals
marginal cost – under reasonable assumptions, a substantially smaller value.
Indeed, very simple calculations show that it will probably pay individual
countries to abate so little that will they blow right past the critical threshold.
They will cross the 550 ppmv line, guaranteeing catastrophe.
You might think this is just theory and that country representatives
wouldn’t be this dumb. I think there are good reasons to take the prediction
Astrid Dannenberg and I have tested these predictions in the experimental lab with real people playing for real money (Barrett and Dannenberg
2012). Putting people into groups of 10, we find that when the threshold
is certain, 18 out of 20 groups avoid catastrophe.6 By contrast, when the
threshold is uncertain, catastrophe occurs with probability 100% for 16 out
of 20 groups and with probability no less than 80% for the rest.7
How sensitive are these results? Intuitively, there should exist a critical
amount of uncertainty such that if uncertainty were greater than this
amount, catastrophe would be bound to occur, whereas if uncertainty were
less than this, catastrophe would be avoided.This is exactly what the theory
predicts (Barrett 2013), and in further experiments (Barrett and Dannenberg
2014a), Astrid Dannenberg and I have shown that this result is also robust.
To the left of a critical “dividing line” for threshold uncertainty, we find
that catastrophe is avoided with high probability almost all the time. Just to
the right of the dividing line, by contrast, catastrophe occurs with proba-
This means that the probability that the threshold lies between 350 and 400 ppmv
is the same as the probability that the threshold lies between 400 and 450 ppmv, between
450 and 500 ppmv, and between 500 and 550 ppmv. It also means that the probability
that the threshold lies below 350 or above 550 ppmv is zero.
In each of the two failing groups, just one individual, a bad apple, caused the trouble,
pledging to contribute his or her fair share and then choosing to contribute nothing.
Interestingly, theory predicts that uncertainty about the impact of crossing a critical
threshold should make no difference to collective action (Barrett 2013), another
prediction confirmed in the experimental lab (Barrett and Dannenberg 2012). It is only
uncertainty about the tipping point that matters, and this is a purely scientific matter.
Sustainable Humanity, Sustainable Nature: Our Responsibility
bility 100%. This research suggests that negotiators can’t rely on science to
solve their collective action problem. Even the new science of “early warning signals” won’t be able to shrink uncertainty by enough to transform
the behavior of nation states.
This research is helpful, I think, because it is completely consistent with
the behavior we are observing in the real world. As noted before, countries
have agreed to limit temperature change to 2˚C, but they have pledged
emission reductions that virtually guarantee overshooting of this target.
If the science of climate change were much more certain, the prospect
of catastrophe would give countries the discipline they needed to act in
their collective-best interest. It would make the dangerous climate change
game a coordination game. Scientific uncertainty makes the emission reductions game a prisoners’ dilemma. The most important thing about a
prisoners’ dilemma is that the collective-best outcome cannot be sustained
by non-cooperative behavior. It requires enforcement, something that the
international system is very bad at doing.
A climate Doomsday Machine
It is tempting to consider an analogy to nuclear arms control. Herman
Kahn (1961: 107) proposed construction of a Doomsday Machine,
a device whose function is to destroy the world. This device is protected from enemy action (perhaps by being situated thousands of
feet underground) and then connected to a computer, in turn connected to thousands of sensory devices all over the United States.The
computer would be programmed so that if, say, five nuclear bombs
exploded over the United States, the device would be triggered and
the world destroyed. Barring such problems as coding errors (an important technical consideration), this machine would seem to be the
‘ideal’ [deterrent]. If Khrushchev ordered an attack, both Khrushchev
and the Soviet population would be automatically and efficiently annihilated.
A Climate Doomsday Machine would connect all the world’s nuclear
bombs to a computer, which in turn would be linked to a sensor at the top
of Mona Loa in Hawaii. This is where readings are taken of atmospheric
concentrations of greenhouse gases. Today, as noted before, the concentration level is about 400 ppmv. The computer could be programmed to destroy the world should this level top, say, 500 ppmv. With the trigger for
catastrophe being certain, theory and experimental evidence strongly suggest that this device would give the world all the encouragement needed
to stay within 500.
Sustainable Humanity, Sustainable Nature: Our Responsibility
Of course, I’m not seriously proposing this. The proposal is
unacceptable.8 However, the idea behind it is worth thinking about. The
Doomsday Machine is a purely strategic device. Its sole purpose is to change
the incentives countries have to rein in their emissions and save the world
from dangerous climate change. It works by transforming the prisoners’
dilemma into a coordination game. Thinking about it begs the question:
Are there acceptable strategic approaches that could have a similar effect? I
shall return to this point later in the paper.
Framing reconsidered
Has a focus on “dangerous” climate change really made no difference?
Under certain conditions, theory suggests that uncertainty about the threshold could mean that behavior won’t change at all, even though the consequences of failing to act will be much worse because of the threat of
catastrophic climate change (Barrett 2012).
But is this result to be believed? Many predictions of analytical game
theory are disproved in the experimental lab. For example, cooperation in
a prisoners’ dilemma typically exceeds the Nash equilibrium prediction
(though the level of cooperation declines rapidly as the players learn that
their efforts to cooperate are not reciprocated). In a one-shot test of the
theory, Astrid Dannenberg and I found that cooperation was higher for the
prisoners’ dilemma with an uncertain threshold for “catastrophe” compared
to a prisoners’ dilemma without any risk of “catastrophe” (Barrett and Dannenberg 2014b).This suggests that, given the risk from “dangerous” climate
change, the wording of Article II of the Framework Convention has probably helped (though it is the real risk rather than the wording that would
affect behavior). Unfortunately, our experiment also showed that the additional cooperation wasn’t enough to prevent catastrophe from occurring.
Strategies of reciprocity
My description of the dangerous climate change game left out the role
of an international agreement. I explained before that a treaty could change
the incentives in the game with a certain threshold, ensuring coordination.
Could a treaty help overcome the incentive to free ride when the threshold
Nor did Kahn recommend the Doomsday Machine: “If one were presenting a
military briefing advocating some special weapon system as a deterrent … the Doomsday
Machine might seem better than any alternative system; nevertheless, it is unacceptable”.
(Kahn 1961: 104-105).
Sustainable Humanity, Sustainable Nature: Our Responsibility
is very uncertain? Theory suggests that an agreement would help very little
(Barrett 2013). The reason is the difficulty of enforcing an agreement to
limit emissions.
Atmospheric concentrations of CO2 are determined by the aggregate
behavior of all countries (as mediated by the carbon cycle). Strategies of
reciprocity work very well in two-player games. They work less well when
the number of players is large.
In the climate change game, how many players really matter? Some
countries are bigger emitters than others. However, the top ten emitters account for only about two-thirds of total emissions, and stabilizing concentrations requires driving global net emissions to zero, necessitating the
engagement of nearly all countries.9
The temptation to free ride is further aggravated by the high marginal
cost of reducing emissions substantially. It is sometimes argued that reducing
emissions is cheap. If this were true, however, collective action would be easy.
Another problem is the lack of correlation between a country’s contribution to emissions and its vulnerability to climate change.To illustrate,William
Nordhaus (2011) has calculated that the “social cost of carbon” is more than
twice as large for Africa, a continent of more than 50 states, as it is for the
United States. Moreover, this gap is growing.Yet, Africa’s emissions are tiny
when compared to those of the United States. Africa is both more vulnerable
to climate change and less able to prevent it from occurring.
Finally, globalization amplifies the incentives to free ride. Abatement by
a single country or coalition of countries will tend to shift emissions towards
the countries that fail to act – a phenomenon known as “leakage”.
It is well known that infinitely repeated play of the prisoners’ dilemma
can allow the full cooperative outcome to be sustained as a (subgame perfect) Nash equilibrium, provided discount rates are sufficiently low.The reason is that, should a country “cheat” on an agreement to limit emissions,
the others can reciprocate. This suggests that cooperation should be easy.
The flaw in this perspective is that it considers only the interests of individual countries. It ignores these countries’ collective interests.
Imagine that all the world’s countries come together and negotiate an
agreement that maximizes their collective interests. Later, one country announces that it will withdraw.This withdrawal would harm the other states,
and they would like to punish this country (or, better yet, threaten to punish
it, hoping to deter its withdrawal). In the context of a treaty, they would
Unless, that is, substantial amounts of carbon are removed from the atmosphere.
Sustainable Humanity, Sustainable Nature: Our Responsibility
naturally cooperate to punish the deviant country. To deter a deviation, their
punishment must be big enough that the deviant would be better off remaining in the agreement than withdrawing and facing the punishment.
But the punishment must also be credible. Given that this country has withdrawn, the remaining N – 1 countries must be better off when they impose
the punishment than when they do not impose it (or when they impose a
weaker punishment). Because so many countries remain in the agreement,
it will only pay these countries to cut their abatement a little. A larger punishment wouldn’t be credible. A small punishment, however, would be too
little to deter a defection. Continuing in this way, it is easy to see that an
agreement to limit emissions is only self-enforcing if the number of countries participating is very small. For then, should one country withdraw, the
remaining countries would have an incentive to drop their abatement significantly. However, once participation shrinks to such a low level, the treaty
achieves very little.
It may be possible for countries to sustain a high level of participation, but
this would only be true if the gains to cooperation were small. It may also be
possible to sustain a high level of participation if the ambition of the treaty
were set very low.What isn’t possible is for countries to sustain a high degree
of cooperation when the gains from cooperation are very large.10
I have so far focused on participation. What about compliance? A flaw
in the approaches to enforcement taken previously is that they either ignore
participation (as in Chayes and Chayes 1995) or fail to distinguish between
participation and compliance (as in Downs, Rocke, and Barsoom 1996).
Under the rules of international law, countries are free to choose whether
or not to participate in a treaty. However, the countries that choose to participate are legally obligated to comply with it (pacta sunt servanda, meaning
“agreements must be kept”). The easiest way to avoid needing to comply
is therefore not to participate in the first place – or to withdraw after becoming a party. From the perspective of game theory, the problem is coming
up with a credible punishment that is large enough to deter non-participation. Once this is done, deterring non-compliance is easy. Remember, larger
deviations can only be deterred by larger punishments, and larger punishments are less credible. What’s the biggest harm a party could ever do? Behaving as it would were it not a party to the agreement (any bigger harm
would not be credible). So, if the parties can deter non-participation, they
can easily deter a smaller deviation of non-compliance. From both perspec-
All of these points are developed in detail in Barrett (2003).
Sustainable Humanity, Sustainable Nature: Our Responsibility
tives, then, deterring non-participation is the binding constraint on enforcement (Barrett 2003).
This is theory. Is the reasoning compelling? I know of no example from
international cooperation that challenges this perspective.The World Trade
Organization might appear to be an exception, but trade isn’t a global public
good. Trade is a bilateral activity, and strategies of reciprocity are very effective in sustaining cooperation amongst pairs of players. The Montreal
Protocol on protecting the ozone layer might appear to be another exception, but as I shall explain later, this treaty works very differently.
Enforcement of the Kyoto Protocol
The Kyoto Protocol looks at climate change as a prisoners’ dilemma,
demands that certain countries cooperate, and then does nothing about enforcement.
How do we know this? The United States participated in the Kyoto negotiations. President Clinton signed the treaty. However, the United States
never ratified Kyoto. One reason for this is that there were no consequences
to the United States for not ratifying the agreement. Non-participation by
the United States was not deterred.
Canada ratified Kyoto, but failed to adopt the domestic legislation
needed to implement its obligations. As a consequence, Canada’s emissions
exceeded the limit set by Kyoto limit. Once in this situation, Canada had
three options. It could buy permits or offsets to stay in compliance; it could
stay in the agreement and be in non-compliance; or it could withdraw from
the agreement. In contrast to the first option, withdrawal would be costless.
In contrast to the second option, withdrawal would not violate international
law. Not surprisingly, Canada decided to withdraw. The Kyoto Protocol
could not deter Canada from withdrawing.
Compliance with the Kyoto Protocol by other parties is uneven (Haita
2012). However, there are ways to get around compliance. For example,
Japan is maneuvering to achieve compliance partly by purchasing “assigned
amount units” from Ukraine, when Ukraine’s emissions are well below its
“assigned amounts”. In other words, in buying these units from Ukraine,
Japan can comply without emissions being reduced anywhere. This may
seem crazy but the treaty was written to allow this trading in “hot air”.
Finally, countries like China and India are not subject to limits on their
emissions.They participate in Kyoto.They comply with it. But Kyoto does
not require that these countries do anything.
Overall, did Kyoto contribute to meeting the objectives of the Framework Convention? Did it reduce global emissions? Econometric analysis
Sustainable Humanity, Sustainable Nature: Our Responsibility
of the Kyoto Protocol by Rahel Aichele and Gabriel Felbermayr (2012:
351) shows that Kyoto did reduce the emissions of participating countries.
However, its effect on global carbon emissions “has been statistically indistinguishable from zero”.
The most important indicator of whether the objectives of the Framework
Convention are being met is whether the growth in atmospheric concentrations is slowing. It isn’t. If anything, the rate of increase has gone up.11 We are
no closer now to addressing this great problem than we were more than
twenty years ago when the Framework Convention was adopted.
Pivot: From Kyoto to Paris
The Copenhagen talks were supposed to provide a successor agreement
to the Kyoto Protocol. They failed.
A new agreement is now being negotiated under the “Durban Platform”.
It is supposed to be ready for adoption by 2015, when the parties to the Framework Convention meet in Paris. It is supposed to be implemented by 2020.
Since Kyoto’s emission limits ended in 2012, this leaves a gap of eight
years.To fill the gap, Kyoto was given an extension in the form of the Doha
Amendment (which has yet to enter into force). However, it is a further
sign of Kyoto’s failings that Japan, New Zealand, and Russia declared their
intention not to participate this time around.12
The new agreement being negotiated now won’t repeat all of Kyoto’s
mistakes, but there is no indication yet that it will improve much on what
countries would have done in the absence of cooperation. Kyoto referred
to its emission limits as “commitments”. However, countries were never
truly committed to meeting these limits; they couldn’t be committed to
meeting these limits so long as Kyoto lacked the means to compel parties
to do more than they were willing to do unilaterally. It is a sign of where
the current round of negotiations are going that in Durban countries agreed
to negotiate a new “protocol, legal instrument or agreed outcome with
legal force”, and that in the recent Warsaw talks they agreed to negotiate
“contributions” rather than “commitments”.
The world clearly needs a new model for cooperation on climate change.
The data can be found at http://www.esrl.noaa.gov/gmd/ccgg/trends/
The European Union will participate, because this agreement only requires that
Europe meet the target it declared it would meet unilaterally. The new government in
Australia has introduced legislation to repeal the previous government’s climate
legislation, a sign that Australia may not ratify the Doha Amendment.
Sustainable Humanity, Sustainable Nature: Our Responsibility
Why the Montreal Protocol succeeded
Kyoto lacks a strategic design.The emissions targets and timetables were
chosen in the expectation that they would be met. No consideration was
given to whether the treaty created incentives for them to be met.
The Montreal Protocol, negotiated to protect the stratospheric ozone
layer, was designed very differently. Remarkably, while the Montreal Protocol was not intended to reduce greenhouse gases, it has been much more
successful at doing this than the Kyoto Protocol. It turns out that ozone in
the stratosphere is a greenhouse gas (protecting the ozone layer will thus
add to climate change), as are the chemicals that deplete stratospheric ozone
(reducing these emissions will thus help mitigate climate change) and many
of their substitutes (use of these will thus add to climate change). Calculations by Velders et al. (2007) show that, by phasing out the ozone-depleting
substances that double as greenhouse gases, the Montreal Protocol has done
four times as much to limit atmospheric concentrations as the Kyoto Protocol aimed to do.
Why did Montreal succeed where Kyoto failed? A key reason for its
success is the threat to restrict trade – in particular, a ban on trade in controlled substances between parties and non-parties (Barrett 2003).The most
important motive for the trade restriction was to enforce participation in
the agreement (Benedick 1998: 91). If participation could be enforced, then
trade leakage would be eliminated; moreover, compliance could also be enforced (Barrett 2003). Crucially, the trade restrictions in the Montreal Protocol are a strategic device.Their purpose was not to be used; their purpose
was to change behavior.
How do the trade restrictions work? Imagine that very few countries
are parties to an agreement to limit emissions, and your country is contemplating whether or not to join. If you join, you will have to reduce your
emissions.Your country will pay the cost, and the benefits will be diffused;
the incentives to free ride will not be blunted. If, in addition, you are now
also prohibited from trading with non-parties – the vast majority of countries – than you will be doubly harmed. By joining, you not only forfeit
the benefits of free riding; you also lose the gains from trade.
Now imagine that almost every other country is a party to the same
agreement. If you join, you still lose the benefits of free riding. But now
you are able to trade with the vast majority of countries. If the gains from
trade exceed the loss from free riding, your country will be better off joining. Put differently, if every country is a party to such an agreement, none
will wish to withdraw. The agreement will sustain full participation by
means of a self-enforcing mechanism – the trade restriction. Most remark-
Sustainable Humanity, Sustainable Nature: Our Responsibility
ably, in equilibrium, trade will never be restricted, just as the Doomsday
Device would never be detonated. It is the credible threat to restrict trade
(detonate the device) that disciplines behavior.
Notice that a key feature of this strategy is that “enough” countries participate in the agreement. If too few participate, none will want to participate.
If enough participate, everyone will want to participate. Somewhere in between there exists a “tipping point” for participation. Once this tipping point
has been identified, the treaty only needs to coordinate participation – something treaties can do very easily. As I said before, treaties need to ensure that
countries are steered towards the desired outcome. In the Montreal Protocol,
this was achieved by the minimum participation clause (Barrett 2003).
The trade restrictions are an acceptable alternative to the Doomsday
Machine. Like the Doomsday Machine, trade restrictions transform the prisoners’ dilemma into a coordination game.
Coordination in climate treaties
The Montreal Protocol could be amended to achieve more for the climate. Hydrofluorocarbons (HFCs) do not deplete the ozone layer, and so
are not currently regulated by the Montreal Protocol. However, HFCs are
a very potent greenhouse gas – one of the six gases controlled by the Kyoto
Protocol. Kyoto has done very little to limit HFCs. In May 2011, the United
States, Canada, and Mexico proposed amending the Montreal Protocol to
control HFCs. If adopted, such an amendment would represent a significant
departure from the approach taken so far to address climate change. It would
mean addressing one piece of the problem, rather than all of it in a comprehensive way. And it would likely involve using trade restrictions for purposes of enforcement. I want to underline that the application of trade
restrictions is purely strategic. Other proposals for trade restrictions in climate policy are very different; their purpose is to be used, not to alter behavior strategically (Barrett 2011).
There are other opportunities for transforming the climate change game
from a prisoners’ dilemma into a coordination game, especially the application of technology standards (Barrett 2003, 2006).
Let me give one example. Another greenhouse gas known as perfluorocarbons or PFCs is emitted in the process of manufacturing aluminum. Apparently, these emissions can be eliminated if the anodes being used now
are replaced with inert anodes. According to the United States Environmental Protection Agency’s webpage, “This technology is being pursued
aggressively through a joint R&D program that has been established between the aluminum industry and the U.S. Department of Energy in its
Sustainable Humanity, Sustainable Nature: Our Responsibility
Industrial Technology Program”.13 The new anode is expected to be available in 10-15 years.
Here, then, is a suggestion. A new agreement should be negotiated requiring that producers adopt the new technology, and that all parties agree
to import aluminum only from countries that participate in the agreement.
This approach creates a “tipping” phenomenon. Provided enough countries
join the agreement, all will want to join it. Why? To be outside the agreement when most countries are inside means not being able to trade in aluminum with most of the world.
This proposal for aluminum would be different from Montreal.The trade
restriction would be based on a process standard, not a product standard.
However, I believe it would still be effective. I also believe it would be compatible with the WTO, partly because of the exemptions allowed under Article XX, but also because it would be adopted by a multilateral agreement.
A final point. Another reason Montreal works is that it includes side
payments to address related equity issues. Side payments could also be included in an agreement establishing a aluminum production standard. As
in Montreal, any side payments should be based on the “incremental costs”
of adopting the standard.Transfers should be small, and the countries giving
the money should know what they are getting for their money.
Again, this is just one example of how the negotiations could be made
more effective.14 My aim here is not to develop a comprehensive approach
to future climate negotiations but to suggest a new direction. If the negotiators understood their job as needing to achieve coordination, and to think
strategically, they would achieve more – and the climate negotiations
wouldn’t prove so disappointing.
Aichele, Rahel and Gabriel Felbermayr
(2012).“Kyoto and the Carbon Footprint
of Nations”, Journal of Environmental Economics and Management 63: 336-354.
Barrett, Scott (2003), Environment and Statecraft:The Strategy of Environmental TreatyMaking, Oxford: Oxford University Press.
Barrett, S. (2006). “Climate Treaties and
‘Breakthrough’Technologies”. American
Economic Review (Papers and Proceedings)
96(2): 22-25.
Barrett, Scott (2011).“Rethinking Climate
Change Governance and Its Relationship
to the World Trading System”, The World
Economy 34(11): 1863-1882.
Barrett, Scott (2013).“Climate Treaties and
For other examples of sectoral approaches to reducing emissions, see Barrett
Sustainable Humanity, Sustainable Nature: Our Responsibility
Approaching Catastrophes”, Journal of
Environmental Economics and Management,
66(2): 235-250.
Barrett, Scott and Astrid Dannenberg
(2012). “Climate Negotiations Under
Scientific Uncertainty”, Proceedings of the
National Academy of Sciences, 109(43):
Barrett, Scott and Astrid Dannenberg
(2014a),“Sensitivity of Collective Action
to Uncertainty about Climate Tipping
Points”, Nature Climate Change 4: 36-39.
Barrett, Scott and Astrid Dannenberg
(2014b),“Negotiating to Avoid ‘Gradual’
versus ‘Dangerous’ Climate Change: An
Experimental Test of Two Prisoners’
Dilemmas”, in Todd L. Cherry, Jon Hovi,
and Dave McEvoy (eds.), Toward a New
Climate Agreement: Conflict, Resolution, and
Governance, London: Routledge.
Benedick, R.E. (1998). Ozone Diplomacy:
New Directions in Safeguarding the Planet,
Enlarged Edition, Cambridge, MA: Harvard University Press.
Broecker,W.S. (1997).“Thermohaline Circulation, the Achilles Heel of Our Climate System:Will Man-Made CO2 Upset the Current Balance?” Science
Chayes, A. and A.H. Chayes (1995), The
New Sovereignty. Cambridge, MA: Harvard
University Press.
Downs, G.W, D.M. Rocke, and P.N. Barsoon (1996), “Is the Good News About
Compliance Good News About Cooperation?” International Organization 50:
Environmental Protection Agency (2006).
Global Mitigation of Non-CO2 Greenhouse
Gases, Washington, DC: EPA.
Haita, Corina (2012). “The State of Compliance in the Kyoto Protocol”, International Center for Climate Governance;
Kahn, Herman (1961), “The Arms Race
and Some of Its Hazards”, in Donald G.
Brennan (ed.), Arms Control, Disarmament,
and National Security, New York: George
Braziller, pp. 89-121.
Mercer, J.H. (1978). “West Antarctic Ice
Sheet and CO2 Greenhouse Effect: A
Threat of Disaster”, Nature 271: 321-325.
Nordhaus, William (2011). “Estimates of
the Social Cost of Carbon: Background
and Results from the RICE-2011 Model”. Cowles Foundation Discussion Paper
No. 1826,Yale University.
Rockström, J., W. Steffen, K. Noone, Å.
Persson, S. Chapin III, E.F. Lambin,T.M.
Lenton, M. Scheffer, C. Folke, H.J.
Schellnhuber, B. Nykvist, C.A. de Wit,
T. Hughes, S. van der Leeuw, H. Rodhe,
S. Sörlin, P.K. Snyder, R. Costanza, U.
Svedin, M. Falkenmark, L. Karlberg, R.W.
Corell,V.J. Fabry, J. Hansen, B.Walker, D.
Liverman, K. Richardson, P. Crutzen, and
J.A. Foley (2009).“A Safe Operating Safe
for Humanity”, Nature 461: 472-475.
Roe, G.H. and M.B. Baker (2007). “Why
is Climate Sensitivity So Unpredictable?”
Science 318: 629-632.
Rogelj, J., J. Nabel, C. Chen, W. Hare, K.
Markmann, M. Meinshausen, M. Schaeffer, K. Macey, and N. Höhne (2010).
“Copenhagen Accord Pledges are Paltry”,
Nature 464: 1126-1128.
Schelling,T.C. (1960). The Strategy of Conflict.
Cambridge, MA: Harvard University
Velders, G.J.M., S.O. Anderson, J.S. Daniel,
D.W. Fahey, and M. McFarland (2007),
‘The Importance of the Montreal Protocol in Protecting Climate’, Proceedings
of the National Academy of Sciences 104
(12): 4814-4819.
Sustainable Humanity, Sustainable Nature: Our Responsibility

Similar documents


Report this document