Full Terms & Conditions of access and use can be found at
Ethics, Policy & Environment
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/cepe21
Consumption-Based Emissions Accounting and
To cite this article: Olle Torpman (2022): Consumption-Based Emissions Accounting and
Historical Emissions, Ethics, Policy & Environment, DOI: 10.1080/21550085.2022.2058296
To link to this article: https://doi.org/10.1080/21550085.2022.2058296
© 2022 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Published online: 01 Apr 2022.
Submit your article to this journal
Article views: 203
View related articles
View Crossmark data
Consumption-Based Emissions Accounting and Historical
Swedish University of Agricultural Sciences, Uppsala, Sweden; Department of Philosophy, Stockholm
University, Stockholm, Sweden; Institute for Futures Studies, Stockholm, Sweden
This paper argues that, unlike the production-based emissions
accounting (on which emissions are attributed to producers of
goods and services), the consumption-based emissions account-
ing (on which emissions are attributed to consumers of these
goods and services) can solve the problem of historical emis-
sions. This problem concerns the question of how to assign
remedial responsibility for emissions that were made by people
who are now dead. Since historical emissions are embedded in
the goods consumed by present consumers, and since present
consumers can (unlike past producers) do something about
their emissions, a consumption-based accounting can contribute
to solving the climate crisis.
accounting; climate ethics;
Human induced climate change stems from humans’ emissions of greenhouse gases.
Such emissions are made when we produce and use goods and services, such as
electricity, housing, food, and transportation. One ethically interesting question is on
whose books these emissions should be put. Should the responsibility for emissions
be attributed to those whose productions give rise to the emissions, or to those who
consume the products? This question is relevant for several reasons. First, it has
implications for how to calculate individuals’ as well as nations’ carbon footprints.
Second, it has implications for how the costs of climate change mitigation and
adaptation should be divided between people. Third, it inuences what is a just
distribution of emissions permits, since those who have emitted more in the past
should plausibly receive fewer permits to emit in the future.
On the one hand, it might seem appropriate to assign responsibility for emissions to the
producers of those goods and services whose productions generate the emissions – i.e. to
those who burn the coal to produce electricity, those who farm the animals for meat
production, those who produce the cement for bridge and house construction, those
who cut the forests for land use and furniture production, and so on. This corresponds to
an accounting method that is called the production-based emissions accounting. What speaks
CONTACT Olle Torpman olle.torpman@iﬀs.se Institute for Futures Studies, Hollandargatan 13, 111 36
ETHICS, POLICY & ENVIRONMENT
© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License
(http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any med-
ium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
in favour of this approach is that the producers appear to be causally responsible for the fact
that the emissions take place. Without production of any goods or services, no emissions
would be generated.
On the other hand, it might seem just as appropriate to assign responsibility for
emissions to the consumers of all those goods and services whose productions generate
the emissions – i.e. to those who use the electricity to heat and illuminate their homes,
those who eat the meat, those who use the bridges, those who buy the furniture, and so
on. This corresponds to an accounting method that is called the consumption-based
emissions accounting. What speaks in favour of this approach is that it is unlikely that
producers would give rise to any emissions at all, if there were no consumer demand in
the rst place. In that sense, consumers may also be regarded as causally contributing to
the related emissions.
In climate ethics, there is a debate over which of these accounting methods is most
Besides the issue of causal responsibility, questions regarding eectiveness,
political feasibility, technological feasibility, and justice are also considered relevant as to
which emissions accounting method is most appropriate.
Although the production-
based accounting has been widely used by the international community – including
the IPCC – many climate ethicists argue that the consumption-based accounting is
morally preferable. For instance, it is argued to be fairer since it does not credit people
in poor countries for the emissions that are due to products that are in the end consumed
by people in rich countries. Moreover, the consumption-based accounting is argued to be
advantageous for the reason that it does not allow (rich) nations to escape responsibility
for emissions by oshoring production to other (poorer) nations.
My aim in this paper is to oer an additional argument in favour of consumption-based
emissions accounting, by showing that it can provide an intuitive answer to a separate
problem discussed in the climate ethics literature: The problem of historical emissions.
This problem concerns the question of how to assign remedial responsibility for the
emissions that have already been made by people in the past.
While the production-
based accounting attributes remedial responsibility for all historical emissions to the past
people who produced these emissions, I argue that the consumption-based accounting
attributes remedial responsibility for historical emissions to present consumers of the
goods and services that stem from these past productions.
I will start by making some preliminaries that are crucial to the argument. First, it should
be noted that emissions-accounting methods can be considered in dierent ways. For
instance, they could be considered as merely conceptual accounts, thus saying something
only about who counts as an ‘emitter’ or ‘polluter’. As such, the emissions-accounting
methods may be used for bookkeeping purposes. However, they could alternatively be
understood as normative accounts, thus saying something also about who bears respon-
sibility for emissions. Since we are here interested in responsibility for emissions, I shall
understand the emissions-accounting methods as normative.
There are dierent ways in which ‘responsibility’ could be understood. For instance, it
could be understood either (i) in causal terms, where an agent is responsible for an
outcome if and only if the outcome would not have been realized without the agent’s
intervention; (ii) in moral terms, where an agent is responsible for an outcome if and only
if she is blameworthy for that outcome; or (iii) in remedial terms, where an agent is
responsible for an outcome if and only if she is liable to solve, or pay the costs for, the
problems that are due to this outcome. In this paper, I will focus on the notion of remedial
responsibility – about who should pay the ‘social costs’ of historical emissions.
Indeed, the problem of historical emissions is a problem exactly about remedial
responsibility. It is not merely an empirical problem about who emitted how much or
when – it is rather a normative problem about the forward-looking duty to do something
about the historical emissions. I stipulate that ‘historical emissions’ are emissions made by
people who are now dead. I qualify these as emissions made between year 1750, marking
the beginning of the industrial era, and 1990. I choose 1990 since it marks the year of the
rst IPCC report, after which emitters could no longer be considered excusably ignorant,
while keeping in mind that the problem of allocating responsibility to the dead is
a separate problem from that of excusable ignorance.
For an answer to the problem of historical emissions to be intuitively plausible,
I assume that it will have to meet the following conditions (the rst being theoretical,
the second being practical):
(i) It provides a theoretically backed-up identication of the relevant duty-bearers,
in the sense that it explains why the identied agents are the relevant duty-
(ii) It provides a practicable recommendation, in the sense that the identied duty-
bearers could do something about the problem at issue.
As this means, I regard these conditions as jointly sucient for an answer to the
problem of historical emissions to be intuitively plausible. In other words, I take it that
a ‘solution’ to the problem of historical emissions would be a solution only if it satises
these conditions. As I will argue below, the answer given by the consumption-based
accounting meets both of these conditions, while the answer given by the production-
based accounting does not.
However, I do not think that the emissions-accounting methods as such provide
sucient conditions for remedial responsibility for emissions. Reasonably, some addi-
tional conditions apply. On the plausible assumption that ‘ought’ implies ‘can’, an avoid-
ability condition requires that an agent must in some sense be capable of avoiding an
outcome in order to be remedially responsible for that outcome. Moreover, a foreseeability
condition requires that the agent must also in some sense be capable of foreseeing an
outcome in order to be remedially responsible for that outcome. In any case, I will assume
that an emissions-accounting method is a necessary component of a complete account of
remedial responsibility for emissions.
One might think that since no present consumer can foresee or avoid emissions that
have already been made in the past, no present consumer could fulll the conditions for
being remedially responsible for historical emissions. Hence, the argument I am present-
ing below – that the consumption-based emissions accounting can provide an intuitive
answer to the problem of historical emissions – would make no sense. In the present
context, however, the conditions for remedial responsibility are supposed to apply
directly to present people’s consumption of things rather than to their ancestors’
ETHICS, POLICY & ENVIRONMENT 3
production of these things. And, it is neither unavoidable for present people to consume
these things, nor unforeseeable to them that these things embody historical emissions.
Hence, the conditions at issue do not rule out the argument I will be giving.
3. The Argument
The argument consists of two premises and one conclusion. The rst premise says that the
things that are consumed at present were to a signicant extent produced by people who
are no longer around. The second premise says that the past production of these things
yielded emissions. The conclusion, following from these premises, is that many of the
things that are consumed in the present embody emissions made in the past. With the
consumption-based emissions accounting plugged in to this conclusion, the remedial
responsibility for historical emissions is assigned to present people in proportion to the
extent they consume goods and services that are due to emitting productions in the past.
Regarding the rst premise, it might seem as if most of the things we consume today
are made by our contemporaries. But that is not generally the case. For instance, the
current infrastructures of modern societies are to a large extent due to past activities of
the industrialization. When people nowadays use roads, bridges, railways, hospitals,
schools, factories, and so on, they use facilities that were produced by others before
them. It is not uncommon that the houses we live in, for instance, were built before we
Even modern technology stems from the productions conducted by past people, since
without the past industrialization this technology and knowledge would not exist today.
Although modern service providers – such as Spotify, Uber, Facebook, Google – are
created by our contemporaries, these services would not have been possible if our
ancestors had not previously taken certain steps of the industrial revolution. Whatever
goods or services we have ourselves produced recently, we did not start from scratch.
Even if some of the nal products (e.g. buildings) we consume are made by present
people, the components (e.g. building blocks) were made by our predecessors. Thus, the
things that we now consume are (to a greater or lesser extent) dependent on productions
of people in the past.
Only non-articial (i.e. natural) resources – e.g. the water in the
oceans, or the oxygen in the air – can be consumed without any prior production.
Of course, there is a question of how much of present goods and services are attribu-
table to productions in the past. This, however, is not a problem in principle, but only in
practice. And, it is practically possible to calculate and determine at least some rough
numbers regarding these proportions. I will get back to this in the next section.
Regarding the second premise, it is quite obvious that the production of things in the
past caused emissions. Around half of all emissions made since 1750, marking the
beginning of the industrial era, were made before 1990.
Indeed, it is the industrial
revolution that has given rise to human induced climate change, through the use of fossil
fuels. Given that yesterday’s production methods were less energy ecient than today’s
production methods, it is moreover clear that the production of things in the past yielded
more emissions of greenhouse gases than similar production methods yield today.
From these two premises – (i) that many of the goods and services that present
people consume were produced in the past, and (ii) that the past production of
these things caused emissions – it follows that the things that are consumed by
present people embody historical emissions. On the consumption-based emissions
accounting, the remedial responsibility for these emissions are (at least to some
extent) attributed to present consumers. Since this answer provides a theoretically
backed-up identication of some relevant duty-bearers (i.e. it explains why those
identied are remedially responsible), it meets condition (i) from the previous sec-
tion. Since the identied duty-bearers are alive and could do something about these
emissions, this answer meets condition (ii) as well. Consequently, the consumption-
based emissions accounting gives an intuitive answer to the problem of historical
It is interesting to note that the consumption-based accounting implies that later
generations will be responsible for the emissions we now make, in proportion to the
extent they consume the products that embody these emissions. Hence, the consump-
tion-based accounting incentivizes people of the present generation to produce long-
lasting goods, since doing so will spread out the remedial responsibility for our
production-related emissions to the future people who could continue using these
things after us.
This section brings up, and responds to, some objections that can be raised against the
argument presented above, or against the consumption-based emissions accounting
when applied to the case of historical emissions.
4.1. Objection #1: The Consumption-Based Accounting is Impracticable in the
Case of Historical Emissions
As hinted at in the previous section, one might worry that the consumption-based
emissions accounting is impracticable in the case of historical emissions. First, there are
questions regarding the exact extent to which the things that present people now
consume were produced in the past, and questions regarding how much emissions
were caused by these productions in the past and how much of these past emissions
can be attributed to present consumers. Second, one might question whether present
people can really make corrections for emissions that were made in the past.
What concerns the rst issue, consider the following example for clarication.
Suppose that a particular apartment house was built in year 1850. Suppose further-
more that the total amount of emissions that were caused by the initial construction
of that apartment house was 2000 tons of greenhouse gases. Let us assume that,
without further maintenance, it is habitable for a total of 200 years, meaning that
each year corresponds to 10 tons of greenhouse gas emissions. If we assume that the
apartment house can host 100 persons living there simultaneously, we get that the
associated emissions would be .1 ton per person and year for living in that apart-
By doing the same sort of calculation for all goods and services that people use,
we will get the total annual amounts of consumption-based emissions for every
person. Just as with the apartment house, the total annual amount of emissions
per person will involve some amounts of historical emissions. By subtracting all
ETHICS, POLICY & ENVIRONMENT 5
present emissions – say, those made after 1990 – we would get the amount of
historical emissions that the consumption-based accounting attributes to each indi-
vidual. The same sorts of calculations can of course be made for collectives of
people, such as nations or rms.
Sure, if we do not know exactly for how long various products will last, or how
much emissions were made when producing them, or exactly how many people will
use them, then we would not be capable of answering exactly how to divide the
historical emissions embodied in those products. The consumption-based accounting
would thus be indeterminate with respect to historical emissions. But as the example
above suggests, it is possible to make at least rough estimates regarding the relevant
numbers – and so to an extent that lets us say something substantially about how to
attribute remedial responsibility to present people for historical emissions on the
basis of a consumption-based emissions accounting.
What concerns the second issue (about whether present people could really do
something about the historical emissions for which they are assigned remedial
responsibility), it should be acknowledged that present people cannot have histor-
ical emission undone, of course. However, remedying historical emissions does not
require taking these emissions back, or have them undone. It only requires paying
a proportionate part of the costs for the problems that the past emissions have
caused. There are several ways to do this. We could either (i) pay for measures
aimed at mitigating climate change (e.g. through investment in development of
renewable energy sources, or in activities that absorbs emissions already made), (ii)
pay for measures aimed at helping people to adapt to unmitigated climate change,
or (iii) pay compensation to those who are nevertheless, and unjustly, aected by
unmitigated climate change. It is therefore not the case that the consumption-
based emissions accounting is impracticable in any sense that makes it inappropri-
ate as an account for how to attribute remedial responsibility for historical
Still, it should be acknowledged that the amount of historical emissions that the
consumption-based accounting covers is limited. For one reason, consumption-based
accounting will only cover the emissions of production in the past leading up to present
consumption, which means that it will not cover the past emissions for the products that
have not yet been consumed. Also, the consumption-based accounting will not cover
emissions related to consumption in the past. In other words, it will ascribe remedial
responsibility for a substantial amount of historical emissions to people who are now
dead, since these people actually consumed a substantial amount of what was pro-
duced in the past. Take, for instance, the greenhouse gases emitted by people driving
cars in 1950.
Even if it is true that the consumption-based accounting cannot deal with all
historical emissions, it attributes a non-negligible amount of the remedial responsi-
bility for these emissions to present people, since these emissions can be tied to
goods and services that are consumed by present people. Moreover, as compared to
the production-based accounting, it sidesteps the question of whether (or to what
extent) the past producers were justiably ignorant about the emissions caused by
their productions. Even if they did not back then know about their contribution to
climate change, we know about it now when we use their products. At the very least,
therefore, the consumption-based accounting is capable of dealing with a larger
amount of historical emissions than the production-based accounting.
4.2. Objection #2: The Argument is Based on Faulty Assumptions Regarding the
Notion of Consumption
Another objection to the consumption-based emissions accounting in the case of
historical emissions is that there seems to be a dierence between rst-hand
and second-hand consumption, as it were. One might argue that when I buy
a new smartphone, for instance, the consumption-based emissions accounting attri-
butes all the emissions that are due to the production of that smartphone to me.
Hence, there will be no historical emissions to inherit for someone who buys that
smartphone second-hand from me, since my rst-hand consumption has ‘washed it
free’ from emissions, so to speak. The same would be true of many other products,
such as the apartment house in the case above. As this implies, no historical
emissions will be attributed to present people for using things that have already
been used by others before them.
However, this understanding of the consumption-based accounting is implausible. The
reason is that it equates ‘consumption’ with ‘purchase’. But just because I have purchased
a smartphone on Monday does not mean that it is already consumed before Tuesday.
Instead, ‘consumption’ should be understood as ‘use’, where a product is fully consumed
once it cannot be (or is not) used anymore. This corresponds to the distinction in
economics between ‘costs’ and ‘payment’, where the costs for a good or service does
not arise immediately with the payment for that good or service, but in proportion to the
utilization of that good or service. While purchasing or buying a product is (typically)
a one-time event, consuming that same product is (typically) a gradual process. Thus,
a more plausible understanding of ‘consumption’ implies that the consumption-based
emissions accounting does not attribute all emissions of a product to the rst-hand
consumer of that product.
There are at least two questions to raise here. First, what if a rst-hand consumer of
a product internalizes all the external (i.e. social) costs related to the emissions of that
product, e.g. via emissions osetting? Then, of course, there would be no emissions left to
inherit for any second-hand consumer of that same product. However, since it is typically
not the case that rst-hand consumers completely oset the emissions embodied in the
things they consume, it is typically not the case that there are no historical emissions left
for second-hand consumers of those products to inherit. Still, this is a contingent assump-
tion. If a rst-hand consumer actually osets all emissions embodied in the products she
consumes, she then can ask for compensation for the osetting costs if selling these
products further to second-hand users. In such a case, the second-hand user does not
incur responsibility for historical emissions, while the production emissions are still
covered by the rst-hand user.
The second question is: What if someone buys a smartphone and then never uses it?
Would such a person avoid being credited with the emissions related to that smartphone?
No. Just because using is not equal to buying does not mean that buying is not an
instance of using (or, in eect, consumption). Indeed, buying is one type of using. If you
ETHICS, POLICY & ENVIRONMENT 7
buy something and then never use it before you through it away, then you are the one
and only consumer of that thing. Hence the consumption-based accounting would
attribute all emissions for that thing to you.
Here it is important to note that two types of emissions can be distinguished: the
emissions of producing a product, and the emissions of further usage of that product. If
you buy a car, for instance, you are responsible for the emissions that were necessary to
produce the car, plus the emission that are a result of your mileage. As this moreover
means, you are responsible for the emissions it took to produce the car even if you would
never drive it.
So, the objection at issue does not undermine the argument that the consumption-
based emissions accounting can answer the problem of historical emissions.
4.3. Objection #3: The Production-Based Accounting Can Also Answer the Problem
of Historical Emissions
A third objection to my argument would be that the production-based emissions
accounting is also capable of answering the question of historical emissions, and that
there is therefore nothing virtuous about the consumption-based accounting in this
regard. There are at least two ways in which this could be argued.
First, the production-based emissions accounting implies that it is the past producers that
are remedially responsible for the historical emissions. And this identies (correctly or not)
some relevant duty-bearers, meaning that it meets condition (i) as presented in section 2.
However, this answer fails to meet condition (ii), for the reason that the past people are not
alive today and hence cannot make any remedies for their historical emissions. Consequently,
the production-based emissions accounting fails to provide an intuitive answer to the problem
of historical emissions. The consumption-based emissions accounting, on the other hand, does
not fail in this regard, since it assigns a signicant amount of the remedial responsibility for
these emissions to people who are alive today and can do something about it.
Second, however, one might argue that the most appropriate understanding of the
production-based accounting is a territory-based understanding, on which emissions are
attributed to the particular nation states on whose respective territories the production of
emissions-generating goods and services take place.
Since nation states are existent over
longer periods of times than individual people, this implies that the production-based
accounting answers the problem of historical emissions by attributing remedial responsibility
for those emissions to the present people of those nation states.
But, this solution to the problem of historical emissions is not due to any intrinsic feature of
the production-based accounting. Rather, it is due to a collectivist reading of remedial respon-
sibility that it employs. On a collectivist reading of remedial responsibility, any emissions
accounting method would be capable of solving the problem of historical emissions – since
collectives are existing over longer time periods, stretching through several generations. This
suggests that the problem of historical emissions would not even arise on a collectivist reading
of moral responsibility. Nevertheless, the collectivist reading of remedial responsibility is
problematic. For one reason, it rejects the plausible assumption that justice is owed to
particular people, as it implies that some individuals (e.g. present people of a certain nation)
are to be disadvantaged for the harmful actions of other individuals (e.g. past people of that
The consumption-based accounting, on the other hand, does not rely on any collecti-
vist reading of remedial responsibility. Instead, it solves the problem of historical emis-
sions while retaining an individualist notion of remedial responsibility. Therefore, the
production-based emissions accounting is inferior to the consumption-based emissions
accounting in the case of historical emissions.
4.4. Objection #4: Other Principles are Already Doing the Job of the Consumption-
In relation to the previous argument, one might argue that we do not need the
consumption-based accounting to solve the problem of historical emissions, since
other climate-relevant principles are already in place to do that job. One such principle,
with similar implications as those of the consumption-based accounting, is the
Beneciary Pays Principle (BPP). According to this principle, an agent is remedially
responsible for historical emissions to the extent she has benetted from the activities
that caused these emissions. Another such principle is the Ability to Pay Principle (APP),
according to which an agent is remedially responsible for emissions in proportion to her
ability to remedy these emissions. On a consumption-based interpretation of ‘polluter’,
a third principle with similar implications would be the Polluter Pays Principle (PPP),
according to which an agent is remedially responsible for emissions in proportion to
the extent she gave rise to them. As this means, all these principles answer the problem
of historical emissions.
I have two replies here, as to why we still need the consumption-based emissions
accounting. Take BPP and APP rst. Although these principles are capable of attributing
responsibility for the past emissions to people who could do something about them (i.e.
the present rich), it is not clear why these people are the relevant duty bearers. Both these
principles have a hard time explaining this. Although capability is a necessary condition
for remedial responsibility (‘ought’ implies ‘can’), it is certainly not sucient. For that
reason, the answers to the problem of historical emissions, as provided by these princi-
ples, would in any case fail to meet condition (i) from section 2.
When it comes to PPP, it seems capable of meeting condition (i), since it is quite clear
why polluters should pay. However, it is capable of attributing responsibility for the past
emissions to people who could do something about them, only given a consumption-
based interpretation of ‘polluter’. On a production-based interpretation, PPP would
attribute responsibility for historical emissions to people who are now dead. This indicates
that the debate over emissions accounting methods is relevant after all, and that there is
a role for the consumption-based emissions accounting to play. Interestingly, the case for
consumption-based accounting would thus provide an indirect argument in favour of
PPP, since PPP’s inability to account for historical emission has been one of the main
objections against it.
More importantly, the problem of historical emissions consists of two parts: one
corrective question about how to divide the costs for the climate eects of unmitigated
emissions, and one distributive question about how to divide the remaining carbon
budget in view of the emissions already made.
And principles such as the BPP, APP,
and PPP can answer only the corrective question about how the costs for dealing with the
eects of climate change should be divided. They cannot provide an answer to the
ETHICS, POLICY & ENVIRONMENT 9
distributive question about how to fairly divide the atmosphere’s capacity to absorb
In answering the distributive question, we need to know how to
attribute emissions (historical as well as contemporary). And this knowledge can be
provided only with the help of emissions accounting methods. Hence, the corrective
principles – i.e. BPP, APP, and PPP – cannot replace the consumption-based emissions
4.5. Objection #5: The Consumption-Based Accounting Implies Too
Another worry regarding the consumption-based accounting, unveiled by the case of
historical emissions, is that it might yield counterintuitive recommendations – or
even recommendations that are counterproductive with respect to climate change
For instance, I claimed in section 3 that the consumption-based accounting gives us
incentives to produce long-lasting goods, since doing so will spread out the remedial
responsibility for our production-related emissions to future people who would continue
to use these things after us. At a closer look, however, it seems that this will in eect give
us a disincentive to clean up production methods – precisely because the responsibility
will be dispersed over a longer time if we can increase the lifespan of products. As it
seems, increasing the lifespan of products is cheaper than cleaning up production
methods. For example, some problems reducing the lifespan of technological equipment
(e.g. microwaves, washing machines) have to do with software issues or parts that wear
out. Of course, we should solve these issues, but it would be bad to continue producing
the raw materials needed (e.g. steel) with coal-powered energy. And we do not want to
saddle future generations with responsibility for the emissions resulting from our dirty
Even though it is true that, by producing long-lasting consumer goods, the consump-
tion-based emissions accounting would give us less incentive to improve production
methods, it does not follow that it incentivizes us to do nothing about these production
methods. In fact, to produce long-lasting goods and to improve production methods
would decrease our emissions even further. This would not only provide us a safer climate,
but also a larger carbon budget to spend on more important things than dirty production
methods. The right response to the objection is, therefore, to insist that consumption-
based accounting gives us reason to do both.
Still, the implication that the responsibility will be dispersed over a longer time is
problematic for another reason: It disregards the fact that it matters when emissions are
emitted. The earlier greenhouse gases are emitted, the more damage they can do to the
climate system. This implies that the sooner we can prevent them from entering the
climate system, the less damage they can do. Since consumption-based emissions
accounting spreads responsibility over time, it gives us no incentive to postpone the
emissions (or prepone the preventions of them).
In response to this, it should be noted that the aim of the emissions-accounting
methods is simply to provide an answer to the question about on whose books
emissions (historical and contemporary) should be put. As this moreover implies,
they do not aim to tell us exactly how we should remedy the emissions which are
10 O. TORPMAN
put on our books, nor when we should remedy those emissions. These are separate
questions to be answered with reference to other considerations (or principles), with
which the emissions-accounting methods will be compatible. With that said, the fact
that the consumption-based emissions accounting is silent on this issue does not
mean that it is irrelevant.
A separate objection, which also points out that the consumption-based accounting will
sometimes yield counterintuitive recommendations, is that it cannot ascertain that its recom-
mendations will be fair. For one reason, it will assign remedial responsibility for historical
emissions to people to who are poor. For example, it will require a poor single mother to pay
for the emissions of the car (production of car and usage) which she needs to drive her kids to
school. As this unveils, the consumption-based emission accounting seems insensitive to the
fact that people have dierent needs and capabilities.
I agree that the division of climate burdens should be sensitive to such dierences between
people. However, it is hard to see how a production-based emissions accounting would fare
better in this respect. Indeed, many producers are also poor, and often poorer than consumers.
More importantly, as I mentioned in section 2, no emissions accounting method provides
a sucient condition for remedial responsibility, since further conditions – such as foresee-
ability and avoidability, etc. – apply. The consumption-based emissions accounting will thus
work in conjunction with such conditions. In terms of the avoidability condition (as implied by
the principle that ‘ought’ implies ‘can’), this means that consumers who are poor, and typically
have fewer options and less capabilities, will actually be less (remedially) responsible for their
(historical) emissions – even on a consumption-based emissions accounting.
The climatic changes that we experience today stem partly from the emissions of our
ancestors. Since these people are not alive today, they cannot themselves make remedies
for their emissions. Hence, it might seem that there is no one left responsible for dealing with
these historical emissions. In this paper, I have argued that the consumption-based emissions
accounting can, while the production-based emissions accounting cannot, provide an intuitive
solution to the problem of historical emissions. It does so by attributing remedial responsibility
for the historical emissions to present consumers of things in proportion to the extent they
stem from the historical productions that caused these emissions. Of course, the consumption-
based accounting cannot deal with all historical emissions, since some of these emissions are
tied to goods and services that were consumed in the past. However, it is superior to the
production-based accounting, which covers a far less amount of historical emissions.
Moreover, the consumption-based accounting provides an advantageous interpretation to
the traditional interpretation of the Polluter Pays Principle.
1. See Broome (2012, p. 70). Of course, this inuence is contingent. One might think that
emissions permits should be divided on the basis of needs or capacities, which would not
require any information about people’s past emissions.
2. See Roser and Tomlinson (2014, p. 238).
ETHICS, POLICY & ENVIRONMENT 11
3. Note that other accounting methods are possible. For instance, a mix-based emissions
accounting would split the attribution of emissions between producers and consumers.
See Lenzen et al. (2007) and Steininger et al. (2016). In this paper, I will put this mixed
approach to one side, since it does not aect the moral ordering between the production-
based and the consumption-based accountings. Were we to compare the production-based
approach to the mix-based approach, then the argument that I present below will speak in
favour of the mix-based approach.
4. Peters (2008) and Steininger et al. (2016) provides interesting discussions on these issues.
5. See Duus-Otterström and Hjorthen (2018), Chancel and Piketty (2015), Davis and Caldeira
(2010, p. 5691), Peters and Hertwich (2008), and Mittiga (2018). In this paper, I will not take
a stand on whether these arguments are sound.
6. For discussions about this problem, see García-Portela (2019), Caney (2012), Gardiner et al.
(2010), and Duus-Otterström (2014).
7. Note that even under a merely conceptual understanding the accounting methods could be
used to answer who is responsible for emissions, given that these accountings were com-
bined with some normative principle. I will sidestep this possibility here.
8. For a discussion about these separate issues, see Caney (2010, pp. 208–11).
9. Some might want to add a third condition requiring that the recommendation will
also contribute to solving the climate problem. Since the fullment of such a condition
would more or less follow from the fullment of condition (ii), and since it would not
make any dierence with respect to the comparison between the production-based
and the consumption-based accountings, I choose to omit it here. What concerns
condition (ii), I take “do something about” to mean “remedy”. As will be argued in
section 4, remedying historical emissions could be done in a variety of ways.
10. I follow the existing literature when I say that things “embody” emissions, by which it is
meant that things come with, involve, or are attached to emissions. See for instance
Peters and Hertwich (2006), and Duus-Otterström and Hjorthen (2018).
11. It is not entirely unproblematic to explain how a certain outcome is “dependent on”
a previous activity, or how that dependency is relevant for responsibility. However, this is
not a problem only for the consumption-based emissions accounting, but also for the so-
called Beneciary Pays Principle, according to which the beneciaries of a certain past activity
should pay the social costs related to this activity.
12. See data from Carbon Dioxide Information Analysis Center(CDIAC): https://cdiac.ess-dive.lbl.
Users/olto0208/Dropbox/Filoso/ARTIKLAR FOR PUBLICERING/Consumption-Based and Hist
13. Steininger et al. (2016).
14. C.f., Caney (2009, pp. 135–137), and Mittiga (2018, pp. 164–5).
15. For the distinction between corrective and distributive principles, see Vallentyne (2007,
16. For this separate debate, see, e.g. Caney (2012) and Torpman (2019).
17. I wish to thank an anonymous reviewer for the objections brought up in this section.
18. An earlier version of this paper was presented at the department of philosophy, Stockholm
University, and at the Institute for Futures Studies, in Stockholm, Sweden. I am thankful to the
audiences for helpful comments. Special thanks to Orri Stefánsson, Lisa Hecht, Robert J.
Hartman, Göran Duus-Otterström and two anonymous referees. I gratefully acknowledge the
nancial support fromRiksbankens Jubileumsfond (grant number grant number M17-0372:1).
No potential conict of interest was reported by the author(s).
12 O. TORPMAN
Olle Torpman http://orcid.org/0000-0003-4541-0970
Aonis, S., Sakai, M., Scott, K., Barrett, J., & Gouldson, A. (2017). Consumption-based carbon accounting:
Does it have a future? WIREs Climate Change, 8(1), e438. https://doi.org/10.1002/wcc.438
Broome, J. (2012). Climate matters – Ethics in a warming world. W.W. Norton & Co.
Caney, S. (2009). Justice and the distribution of greenhouse gas emissions. Journal of Global Ethics, 5
(2), 125–146. https://doi.org/10.1080/17449620903110300
Caney, S. (2010). Climate change and the duties of the advantaged. Critical Review of International
Social and Political Philosophy, 13(1), 203–228. https://doi.org/10.1080/13698230903326331
Caney, S. (2012). Just emissions. Philosophy & Public Aairs, 40(4), 255–300. https://doi.org/10.1111/
Chancel, L., & Piketty, T. (2015). Carbon and inequality: From Kyoto to Paris. Trends in the global
inequality of carbon emissions (1998-2013) & prospects for an equitable adaptation fund. Paris
School of Economics.
Davis, S., & Caldeira, K. (2010). Consumption-Based accounting of CO
emissions. PNAS, 107(12),
Duus-Otterström, G. (2014). The problem of past emissions and intergenerational debts. Critical
Review of International Social and Political Philosophy, 17(4), 448–469. https://doi.org/10.1080/
Duus-Otterström, G., & Hjorthen, F. D. (2018). Consumption-based emissions accounting: The
normative debate. Environmental Politics, 28(5),866–885.
García-Portela, L. (2019). Individual compensatory duties for historical emissions and the
dead-polluters objection. Journal of Agricultural & Environmental Ethics, 32(4), 591–609. https://
Lenzen, M., Murray, J., Sack, F., & Wiedmann, T. (2007). Shared producer and consumer responsi-
bility — Theory and practice. Ecological Economics, 61(1), 27–42. https://doi.org/10.1016/j.ecole
Liu, L. (2015). A critical examination of the consumption-based accounting approach. WIREs Climate
Change, 6(1), 1–8. https://doi.org/10.1002/wcc.325
Mittiga, R. (2018). Allocating the burdens of climate action: Consumption-based carbon accounting
and the polluter-pays principle. In B. Edmondson & S. Levy (Eds.), Transformative climates and
accountable governance. Palgrave Studies in Environmental Transformation 157–194.
Peters, G. P. (2008). From production-based to consumption-based national emission inventories.
Ecological Economics, 65(1), 13–23. https://doi.org/10.1016/j.ecolecon.2007.10.014
Peters, G. P., & Hertwich, E. G. (2006). Pollution embodied in trade: The Norwegian case. Global
Environmental Change, 16(4), 379–387. https://doi.org/10.1016/j.gloenvcha.2006.03.001
Peters, G. P., & Hertwich, E. G. (2008). Post-Kyoto greenhouse gas inventories: Production versus
consumption. Climate Change, 86(1–2), 51–66. https://doi.org/10.1007/s10584-007-9280-1
Roser, D., & Tomlinson, L. (2014). Trade policies and climate change: Border carbon adjustments as
a tool for a just global carbon regime. Ancilla Luris, 75, 223–245.
S. M. Gardiner, S. Caney, D. Jamieson, & H. Shue (Eds.). (2010). Climate ethics – Essential readings.
Oxford University Press.
Steininger, K. W., Lininger, C., Meyer, L. H., Muñoz, P., & Schinko, T. (2016, January). Multiple carbon
accounting to support just and eective climate policies. Nature Climate Change, 6(1), 35–41.
Torpman, O. (2019). The case for emissions egalitarianism. Ethical Theory and Moral Practice, 22(3),
Vallentyne, P. (2007). Distributive Justice. In R. E. Goodin, P. Pettit, & T. W. Pogge (Eds.), A companion
to contemporary political philosophy, 2nd Volume Set (2nd ed.). Wiley Blackwell 548–562.
ETHICS, POLICY & ENVIRONMENT 13