Description
This paper analyses the development of carbon markets: markets in permits to emit greenhouse gases or in credits
earned by not emitting them. It describes briefly how such markets have come into being, and discusses in more detail
two aspects of the efforts to ‘make things the same’ in carbon markets: how different gases are made commensurable,
and how accountants have struggled to find a standard treatment of ‘emission rights’. The paper concludes by discussing
the attitude that should be taken to carbon markets (for example by environmentalists) and the possibility of developing a
‘politics of market design’ oriented to making such markets more effective tools of abatement.
Making things the same: Gases, emission rights and the
politics of carbon markets
Donald MacKenzie
School of Social and Political Studies, University of Edinburgh, Adam Ferguson Building, Edinburgh EH8 9LL, Scotland, United Kingdom
Abstract
This paper analyses the development of carbon markets: markets in permits to emit greenhouse gases or in credits
earned by not emitting them. It describes brie?y how such markets have come into being, and discusses in more detail
two aspects of the e?orts to ‘make things the same’ in carbon markets: how di?erent gases are made commensurable,
and how accountants have struggled to ?nd a standard treatment of ‘emission rights’. The paper concludes by discussing
the attitude that should be taken to carbon markets (for example by environmentalists) and the possibility of developing a
‘politics of market design’ oriented to making such markets more e?ective tools of abatement.
Ó 2008 Elsevier Ltd. All rights reserved.
Introduction
Around the world, markets in permits to emit
greenhouse gases or in credits earned by not emit-
ting them are emerging. Some already exist; others
are in construction.
1
This article describes brie?y
the route – at the level of ‘policy’ – that has led to
their emergence. It then delves a little deeper into
the conditions of possibility of these markets, by
examining two examples of what it takes to make
the entities traded in these markets ‘the same’. The
examples are how the destruction of one gas in
one place is made commensurate with emissions of
a di?erent gas in a di?erent place, and how accoun-
tants have sought (so far with only limited success)
to make ‘emission rights’ equivalent. Finally, the
article discusses the issue of politics: the question
of the attitude that should to be taken to carbon
markets (for example by environmentalists, espe-
cially those who conceive of themselves as oppo-
nents of ‘capitalism’), and the tightly-related issue
of the process of market design viewed, as it has
to be, as politics.
Although the article draws upon the ‘?nitist’ per-
spective sketched brie?y below (see Barnes, Bloor,
& Henry, 1996; Hatherly, Leung, & MacKenzie,
submitted for publication), its approach is
prompted by the view of economic life suggested
by the ‘actor-network’ theory of Michel Callon
and Bruno Latour (for which see, for example,
Latour, 2005). In Callon’s and Latour’s view, the
characteristics of an ‘actor’ – a term which, following
1570-8705/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2008.02.004
E-mail address: [email protected]
1
This paper was originally presented to the workshop ‘Carbon
Markets in Social-Science Perspective’, Durham University, 7
November 2007. The research was supported by a UK Economic
and Social Research Council Professorial Fellowship, RES-051-
27-0062, and I am deeply grateful to the Institute of Advanced
Study, Durham University for supporting the workshop and for a
Fellowship which enabled me to complete the paper.
Available online at www.sciencedirect.com
Accounting, Organizations and Society 34 (2009) 440–455
www.elsevier.com/locate/aos
semiotics (especially Greimas, 1987), they view as
encompassing more than just human beings – are
not intrinsic, but are the result of the networks of
which the actor is made up and forms part. What
we call ‘capitalism’, for example, is not an entity
with ?xed characteristics. ‘Que faire contre le capi-
talisme?’, they write: ‘D’abord e´videmment ne pas
y croire’ (Callon & Latour, 1997, p. 67). What is
to be done against capitalism? First of all, of course
do not believe in it.
In Callon’s and Latour’s view, economic life is
‘performed’ – framed and formatted – by ‘econom-
ics at large’, a term that encompasses not just the
academic discipline but also economic practices
such as accounting and marketing (Callon, 1998,
2007). The characteristics of economic actors and
of markets arise from, amongst other things, the
‘dispositifs de calcul’ (Callon & Muniesa, 2003) –
the calculative mechanisms – of which they are
made up.
If the characteristics of ‘capitalism’ are not inher-
ent, they can be changed by changing the calculative
mechanisms that constitute it. The markets in green-
house-gas emissions that are being constructed
globally are a set of experiments (Muniesa & Cal-
lon, 2007) in the validity of this prediction. Hith-
erto, greenhouse-gas emissions have been, in
economists’ familiar terminology, an ‘externality’:
from the viewpoint of the emitter, they bore no cost,
and so did not ?gure in emitters’ economic calcula-
tions. The goal of a carbon market is to bring emis-
sions within the frame of economic calculation by
giving them a price. In such a market, emissions
bear a cost: either a direct cost (because allowances
to emit greenhouses gases need to be purchased), or
an opportunity cost (because allowances that are
not used to cover emissions can be sold, or because
credits can be earned if emissions are reduced below
‘business as usual’). A carbon market is thus an
attempt to change the construction of capitalism’s
central economic metric: pro?t and loss, the ‘bottom
line’.
The experiments in carbon-market construction
have scarcely begun, so the validity of the prediction
that capitalism can be ‘civilized’ (Latour, forthcom-
ing) by changing calculative mechanisms remains
undecided. We do not yet know whether the bottom
line will be changed to any substantial extent, in
particular to an extent su?cient to keep global
warming below the threshold (uncertain and ?ercely
contested, but often taken to be 2 °C) beyond which
the risk of severe impacts rises sharply (Schellnhu-
ber, 2006).
In consequence, this paper is necessarily preli-
minary. The empirical material on which I am draw-
ing is limited. It consists primarily of a set of 24
interviews conducted with people involved with car-
bon markets (particularly with the European Union
Emissions Trading Scheme) as market designers, as
carbon traders and brokers, or as members of
NGOs seeking to in?uence the evolution of carbon
markets. This interview material is supplemented
by analysis of relevant documents such as monitor-
ing reports and contributions to the debate in
accountancy touched on below.
The article’s main aim is simply to help broaden
social-science research on carbon markets, both in
terms of its disciplinary base (though their origins
lie in economics, carbon markets cannot be under-
stood by the conventional tools of that discipline
alone) and in terms of its empirical focus. In that
latter respect, I hope to show that it is productive
to investigate not just overall questions such as the
reasons why policy-makers might choose carbon
markets rather than other tools to combat global
warming, but also the speci?cs of how carbon mar-
kets are constructed. Whether or not carbon mar-
kets are environmentally and economically
e?ective depends on such speci?cs, and the issues
involved are various and demand inter-disciplinary
treatment. One of the two topics examined below
– how di?erent gases are made commensurable –
is a natural question for the social studies of science
and technology; the other – how to ?nd a standard
treatment of ‘emissions rights’ – is a question obvi-
ously suitable for researchers in accounting.
Although for reasons of space I do not discuss them
here, questions for other disciplines can also easily
be identi?ed: for example, vastly more needs known
about how emission reduction projects in develop-
ing countries actually work in practice, a question
that raises issues ranging from how veri?cation is
conducted to the impact of projects on local com-
munities and local environments. Investigating such
issues in genuine depth requires the skills of,
amongst others, anthropologists and other area
specialists.
Because the speci?cs of market design matter, I
make no apology for the fact that this article
touches upon matters of apparent detail. The com-
mensurability of gases and the accounting treatment
of emission rights are inevitably ‘technical’ ques-
tions, and those technicalities cannot altogether be
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 441
avoided: they matter to overall outcomes. The com-
mensurability of gases, for example, is crucial to
how the world’s two main existing carbon markets
– the European Union Emissions Trading Scheme
and the Kyoto Protocol Clean Development Mech-
anism – interrelate, while there is at least tentative
evidence that the accounting treatment of emissions
rights a?ects ?rms’ behaviour in carbon markets. It
is precisely issues of this detailed kind that an e?ec-
tive, inter-disciplinary analysis of carbon markets
will need to address.
Carbon markets
Carbon markets come in two main species: ‘cap
and trade’ and ‘project-based’.
2
Let me begin with
the former. It involves a government or other
authority setting a ‘cap’ – a maximum allowable
aggregate total quantity of emissions – and selling
or giving the corresponding number of allowances
to emitters. The authority then monitors emissions
and ?nes anyone who emits without the requisite
allowances. If the monitoring and penalties are
stringent enough, aggregate emissions are thus kept
down to the level of the cap. The ‘trade’ aspect of
cap and trade arises because those for whom reduc-
tions are expensive will want to buy allowances
rather than incurring disproportionate costs. The
requisite supply of allowances is created by the
?nancial incentive thereby provided to those who
can make big cuts in emissions relatively cheaply.
They can save money by not having to buy allow-
ances, or (if allowances are distributed free) can
earn money by selling allowances they do not need.
So, as already noted, emissions, which previously
had no monetary cost, now have one.
The origins of the idea of controlling emissions
via a cap and trade scheme can be traced to the
work of Nobel Laureate Ronald Coase (1960), but
a more proximate source is the University of Tor-
onto economist Dales (1968a, 1968b), who ?rst
put forward the idea in something like full-?edged
form.
3
Emissions markets were implemented in rel-
atively minor and sometimes ham-?sted ways in the
1970s and 1980s, mainly in the United States (see,
e.g., Hahn, 1989). It was only in the 1990s that the
idea became mainstream.
The crucial development was the start of sulphur-
dioxide trading in the US in 1995 (for which see,
especially, Ellerman, Schmalensee, Bailey, Joskow,
& Montero, 2000; Burtraw, Evans, Krupnick, Pal-
mer, & Toth, 2005). It had been known for twenty
years or more that damage to the environment and
to human health was being caused by sulphur-diox-
ide emissions, notably from coal-?red power sta-
tions, which react in the atmosphere to produce
‘acid rain’ and other acid depositions. Numerous
bills were presented to Congress in the 1980s to
address the problem, but all failed in the face of
opposition from the Reagan administration and
from Democrats who represented states that might
su?er economically from controls on sulphur-diox-
ide, such as the areas of Appalachia and the mid west
in which coal deposits are high in sulphur.
Sulphur trading broke the impasse. It combined a
simple, clear goal that environmentalists could
embrace (reducing annual sulphur-dioxide emissions
from power stations in the US by 10 million tonnes
from their 1980 level, a cut of around a half) with
a market mechanism attractive to at least some
Republicans. The cut was achieved in practice far
more cheaply than almost anyone had imagined.
Industry lobbyists had claimed it would cost $10 bil-
lion a year, while the actual cost was around $1 bil-
lion. Allowance prices of $400 a tonne were
predicted, but in fact prices averaged around $150
or less in the early years of the scheme. The ?exibility
that trading gave to utilities helped to reduce costs
(by around a half compared to having to meet a stan-
dard that imposed a uniform maximum emission
rate: see Burtraw et al., 2005; Ellerman et al., 2000)
but other factors were equally important. ‘Scrub-
bers’ to remove sulphur from smokestacks turned
out to be cheaper to install and to run than had been
anticipated, and rail-freight deregulation sharply
reduced the cost of transportation from Wyoming’s
Powder River Basin, the main source of low-sulphur
coal in the United States (Ellerman et al., 2000).
That the sulphur-dioxide market was, broadly, a
success shaped how the Clinton Administration
approached the negotiations that led to the 1997
Kyoto Protocol. In the Protocol, the industrialized
nations undertook that by Kyoto’s 2008–2012 ‘com-
mitment period’ they would have limited their
greenhouse-gas emissions to agreed proportions of
2
This article concentrates on regulatory markets, largely
setting aside the ‘voluntary’ market, in which, for example, ?rms
choose to ‘o?set’ their emissions, even though they are under no
compulsion to do so: see, for example, Bumpus and Liverman
(forthcoming).
3
The history of emissions trading will be treated in more detail
in MacKenzie (submitted for publication). The brief account
given here draws upon that in MacKenzie (2007).
442 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
their 1990 levels: 93% for the US, 92% for the Euro-
pean Community overall (with varying levels for its
member states), and so on.
At the insistence of the US, Kyoto gave its signa-
tories sulphur-like ?exibility in how to meet their
commitments. The Protocol contains provision for
a cap and trade market between nation-states.
States with caps they will exceed can pay others that
in the commitment period are emitting less than
their caps for their unneeded ‘Kyoto units’ (quanti-
ties of carbon dioxide or their equivalents in other
gases: see below). Just how much trading of such
units between nation-states there will be remains
to be seen: it is possible it will be quite limited. More
signi?cant so far have been two other Kyoto mech-
anisms – ‘Joint Implementation’ and, especially, the
‘Clean Development Mechanism’ (CDM) – which
are project-based schemes, not cap and trade.
Let me concentrate on the CDM (for which see,
for example, Lecocq & Ambrosi, 2007), which is
the more important of the two. It is a crucial – per-
haps the crucial – aspect of the Kyoto Protocol
(Grubb, 1999), crystallizing the political compro-
mise at the Protocol’s heart, between the refusal of
developing countries to take on emissions caps
and the Clinton Administration’s conviction that
global emissions could be restrained far more
cheaply if the developing world were part of the
Kyoto regime. The CDM allows the creation of
Kyoto units from projects in developing countries
approved by the Executive Board of the CDM (a
body established under the United Nations Frame-
work Convention on Climate Change).
To gain approval, it must be shown that a project
is ‘additional’ (that it would not take place without
CDM funding) and that it will reduce emissions
below the ‘baseline’ level they would have been at
without the project. A developing world entity, or
industrialized-world government, corporation, bank
or hedge fund can then earn the di?erence between
emissions with and without the project in the form
of a speci?c type of Kyoto units: ‘Certi?ed Emission
Reductions’ (CERs). A CER is a credit, not a permit
or allowance: it does not directly convey any right to
emit. However, some governments are purchasing
CERs as a way of meeting their Kyoto caps, and cru-
cially CERs also have monetary value because the
European Union permits its member states to issue
allowances in the most important cap and trade mar-
ket, the European Union Emissions Trading Scheme
(ETS), in exchange for the surrender of CERs (Euro-
pean Parliament, Council, 2004). A credit earned in,
for example, China or India can thus be transformed
into a permit to emit in Europe.
As with the CDM, the ETS, launched in January
2005, was shaped by political exigencies.
4
What
pushed Europe towards trading rather than its initial
preference, harmonized carbon taxes, was in good
part an idiosyncratic feature of the political proce-
dures of the European Union. Tax measures require
unanimity: a single dissenting country can block
them. Emissions trading, in contrast, counts as an
environmental, not a tax matter. That takes it into
the terrain of ‘quali?ed majority voting’. No single
country can stop such a scheme: doing so takes a
coalition of countries su?ciently populous (since
voting weights roughly follow population) to form
a ‘blocking minority’. A plan for a Europe-wide car-
bon tax had foundered in the early 1990s in the face
of vehement opposition from industry and from par-
ticular member states (notably the UK), and its
advocates knew that if they tried to revive it the una-
nimity rule meant they were unlikely to succeed. ‘We
learned our lesson’, one of them told me in interview.
Hence the shift in allegiance to trading.
In terms of volume of transactions, the ETS is the
largest greenhouse-gas market. The scheme has had
its di?culties – the over-allocation, violent price
?uctuations and ‘windfall pro?ts’ discussed below
– but it saw trades worth $24 billion in 2006 (World
Bank, 2007, p. 3). The prospect of ‘monetizing’
CERs via the ETS is the main driver of investment
in Clean Development Mechanism projects, which
generated CERs worth $5 billion in 2006 (World
Bank, 2007, p. 3). Between them, the ETS and
CDM form the core of the world’s carbon markets,
and it is on them that this paper focuses.
Making things the same: Gases
The political decision to create a carbon market
such as the CDM or ETS is not the same as con-
structing such a market. A new commodity – an
emission allowance or emission credit – needs
brought into being: de?ned legally and technically,
allocated to market participants, made transferable
and tradable, and so on. To give a ?avour of what is
involved, let me concentrate on one issue: the heter-
ogeneity of the means by which the ‘sameness’ – the
fungibility of allowances and credits – necessary for
4
On the emergence of the ETS, see, e.g., Zapfel and Vainio
(2002), Christiansen and Wettestad (2003), Damro and Me´ndez
(2003), Wettestad (2005) and Cass (2005).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 443
a carbon market is brought into being.
5
Consider,
for example, two very di?erent sites: the central-area
combined heat and power plant of Edinburgh Uni-
versity, situated a couple of hundred meters from
my o?ce there, and the refrigerant plant operated
by Zhejiang Juhua Co., 6.5 km south of Quzhou
City in China’s Zhejiang province. How is the activ-
ity at one made commensurable with that at the
other, so that both can form part of the same mar-
ket, and CERs at Zhejiang Juhua’s plant can be
equivalent to the ETS allowances that a European
emitter such as Edinburgh University needs?
As its name indicates, the combined heat and
power plant in Edinburgh generates electricity (by
burning natural gas in a device that resembles a
giant car engine), and uses what would otherwise
be waste heat to warm up nearby buildings. Because
its thermal input capacity is slightly greater than the
20 MW threshold of the European Union Emissions
Trading Scheme, this plant became part of the ETS
in January 2008. (Most such installations have been
part of the scheme since its launch in 2005, but
Edinburgh University was exempted from the ?rst
phase because of its involvement in an earlier, vol-
untary UK emissions trading scheme.) CO
2
emis-
sions from the combined heat and power plant are
measured using a gas corrector meter (the interface
of which is shown in Fig. 1) on the large pipe that
takes gas from the national gas grid into the plant.
It is called a ‘corrector meter’ because it samples
temperature and pressure, and can thus convert vol-
umes into masses of gas input, which are in turn
converted to estimates of CO
2
output using stan-
dard multiplication factors.
Zhejiang Juhua Co. is involved in something
quite di?erent, the manufacture of HCFC-22 (chlo-
rodi?ouromethane), which is used mainly as a
refrigerant (especially in air conditioners), though
also as a foam blower and as a chemical feedstock.
The standard process used to produce chlorodi?ou-
romethane involves combining hydrogen ?uoride
and chloroform, using antimony pentachloride as
a catalyst, and even when optimized the process
leads to a degree of ‘over?uoridation’: tri?uorome-
thane, HFC-23, is produced as well.
6
HFC-23 is,
unfortunately, long-lived in the atmosphere and an
e?cient absorber of infrared radiation; the combi-
nation makes it a very potent greenhouse gas.
Fig. 1. The interface of the gas corrector meter in the input pipe to Edinburgh University’s central area combined heat and power plant.
Photograph courtesy David Somervell, Estates and Buildings, University of Edinburgh.
5
On commensuration in the SO
2
market, see Levin and
Espeland (2002).
6
‘HFC-23’ and ‘HCFC-22’ are not standard chemical formu-
lae, but instances of a code, widely used in the refrigerant
business, for identifying haloalkanes. The standard formula for
tri?uoromethane is CHF
3
, but ‘HFC-23’ is how it is referred to in
carbon markets.
444 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
Until recently, the HFC-23 was discharged into the
atmosphere. Now, the Zhejiang Juhua plant’s waste
gases are fed into a specialised incineration furnace
imported from Japan, in which they are mixed with
hydrogen, compressed air and steam, burned at
1200 °Cusing a high-intensity vortex burner, and thus
converted to hydrogen ?uoride, carbon dioxide and
hydrogen chloride. These products pass through a
quencher (in which they are cooled rapidly to mini-
mize the formation of dioxins), and the resultant acid
solution is either sold or disposed of via a facility for
treating ?uoric waste (CDM Executive Board, 2007).
As already noted, to gain approval it must be
shown that a Clean Development Mechanism project
reduces emissions below the ‘baseline’ level they would
have had in the absence of the project, which in many
cases is a tricky exercise in establishing a credible coun-
terfactual (Lohmann, 2005): for an introduction to the
issues involved, see Michaelowa (2005). In the case of
HFC-23 decomposition, however, a straightforward
argument has su?ced: without the decomposition pro-
cess, the HFC-23 would, as already noted, simply have
been discharged into the atmosphere (CDM Executive
Board, 2007). The amount actually decomposed then
needs measured, but in such a way that a connection
is kept to the baseline of the HFC-23 that would have
been emitted in the absence of the decomposition
incinerator. (The quantity of HFC-23 generated is
a?ected by the precise parameters of the HCFC-22
production process, and hence there is a need to reduce
the incentive to operate the process in an unoptimized
way and generate unnecessary HFC-23 in order to
earn credits by destroying it.) So to standard equip-
ment such as ?ow meters and a gas chromatograph
is added a regulation: for each tonne of HCFC-22 pro-
duced, there is a maximum mass of HFC-23 whose
decomposition can earn credits.
7
Crucially, the allowable mass of HFC-23 that the
measurement devices reveal has been decomposed is
then multiplied by 11,700.
8
By decomposing a tonne
of HFC-23 in China, one can – via the link between
the CDM and ETS – earn allowances to emit 11,700
tonnes of CO
2
in Europe. Certi?ed Emission Reduc-
tions are now a major income stream for China’s
refrigerant plants, and for the Chinese government
(which imposes a 65% tax on them, hypothecated
for environmental purposes). Indeed, HFC-23
decomposition is the biggest single sector of the
Clean Development Mechanism, accounting for
67% of the CERs generated in 2005 and 34% of those
generated in 2006 (World Bank, 2007, p. 27). Since
the price of CERs is likely to be a chief determinant
of the European carbon price – and thus, for exam-
ple, a major input into electricity prices – the e?ects
of the commensuration are considerable.
The crucial ?gure, 11,700, is the product of a cal-
culation of the ‘global warming potential’ (GWP) of
HFC-23 published by the Intergovernmental Panel
on Climate Change. Set up in 1988 by the World
Meteorological Organization and United Nations
Environmental Programme, the IPCC has as its
remit the establishment of authoritative scienti?c
knowledge about climate change (see Agrawala,
1998a, 1998b). As the IPCC put it in 1990, GWP
is ‘[a]n index... which allows the climate e?ects of
the emissions of greenhouse gases to be compared.
The GWP depends on the position and strength of
the absorption bands of the gas, its lifetime in the
atmosphere, its molecular weight and the time per-
iod over which the climate e?ects are of concern’
(Houghton, Jenkins, & Ephraums, 1990, p. 45).
Although very similar notions are to be found in
the scienti?c literature of the time (see, e.g., Lashof
& Ahuja, 1990), it was the IPCC itself that gave
‘global warming potential’ its canonical de?nition
GWP ¼
R
TP
O
a
x
½xðtÞ?dt
R
TP
O
a
r
½rðtÞ?dt
x designates the gas in question (e.g., HFC-23). a
x
is
an estimate of the e?ect on the radiation balance at
the tropopause (the boundary of the upper and lower
atmosphere) of an increase in the amount of gas in
the atmosphere, an e?ect measured in watts per
square meter per kilogram. x(t) is the mass of the
7
The mass of HCFC-22 produced (which is determined by
weighing shipping containers and storage tanks) determines the
‘eligible quantity’ of HFC-23: the quantity for the incineration of
which credits can be earned. For each tonne of HCFC-22
produced by the standard antimony pentachloride process, the
eligible quantity of HFC-23 is 0.0137 tonnes, corresponding to
the lowest recorded emission level from a process optimized to
minimize HFC-23 production (see McCulloch, 2005, p. 11). The
mass of gas fed into the incinerator is determined from the
readings of a ?ow meter, and the concentration of HFC-23 in it is
determined by gas chromatography of periodic samples. (A
correction for leakage is also applied.) The product of mass of gas
(in tonnes) and HFC-23 concentration, up to the maximum given
by the eligible quantity, is, as noted in the text, then multiplied by
11,700 to give the quantity of Certi?ed Emission Reductions
earned (SGS United Kingdom Ltd., 2007).
8
I am grateful to Thomas Grammig and to members of the
audience of a talk I gave at the University of Oxford for sparking
my interest in how gas equivalents are brought into being.
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 445
gas that will remain in the atmosphere at time t from
l kg released at time zero. TP is the overall time per-
iod in question: in the calculation in the HFC-23
commensuration, it is 100 years. The denominator
is the equivalent integral for the reference gas, CO
2
.
The expressions in this equation inscribe complex
processes. r(t), for example, is not (and obviously
could not be) determined by releasing a kilogram
of carbon dioxide and measuring what happens over
a century: it is a mathematical function generated
from a standard model (the Bern model: see, e.g.,
Siegenthaler & Joos, 1992) of the exchange of car-
bon between the atmosphere, the oceans and the ter-
restrial biosphere. a
x
and a
r
, likewise, are in part the
products of sophisticated spectroscopic studies,
recorded largely in a database managed by the Har-
vard–Smithsonian Center for Astrophysics. (The
database was originally a military project, designed
to enhance understanding of absorption of infrared
radiation with a view to improving the detection of
heat sources: see Taubes, 2004.) But a
x
and a
r
also
assume a scenario that is believed to be helpful in
predicting the climatic impact of a gas. In this sce-
nario, temperatures in the stratosphere, which are
understood as adjusting relatively quickly to such
perturbations, have done so, while temperatures in
the lower atmosphere and at the earth’s surface
(which adjust only slowly) have not.
9
Again, the sce-
nario cannot be observed empirically, so modelling
as well as spectroscopy is involved in the determina-
tion of a
x
and a
r
.
In 1990, the IPCCfelt able to o?er estimates of the
GWPs of only 19 gases, not including HFC-23, and it
labelled the ?gures ‘preliminary only’ (Houghton
et al., 1990, p. 59 & Table 2.8, p. 60). By 1995–1996,
the list had expanded to 26, and included HFC-23,
the GWP of which was estimated as 11,700 (Hough-
ton et al., 1996, table 2.9, p. 121). Both the notion of
‘global warming potential’ and the IPCC’s mid-1990s
estimates of GWPs were then inscribed into the
Kyoto Protocol, which laid down that they should
be used to translate emissions of other greenhouse
gases into their equivalents in CO
2
and that the
IPCC’s mid-1990s estimates should be used until
the end of the 2008–2012 commitment period.
10
The ‘exchange rate’ of 11,700 used to translate
HFC-23 into CO
2
is thus an example of ‘black-box-
ing’ in the sense of Callon and Latour (1981) and
MacKenzie (1990, p. 26). GWPs could be contested
in at least two senses. First, whether GWPs really
give the best estimates of the climatic e?ects of dif-
ferent gases could be and has been challenged (see
Shackley & Wynne, 1997, and also Shine, Fug-
lestvedt, Hailemariam, & Stuber, 2005, and the liter-
ature cited in the latter): for example, the choice of a
100-year time period is in a sense arbitrary, and very
di?erent GWPs can be generated if, for example, 25,
50 or 500 years is used.
11
Second, GWP estimates
were acknowledged to be subject to signi?cant
uncertainties, of the order of ±35% (Houghton
et al., 1996, p. 73, 119). By 2007, for example, the
consensus estimate of the global warming potential
of HFC-23 had increased from 11,700 to 14,800
(Intergovernmental Panel on Climate Change,
2007, p. 212). Neither of these two forms of chal-
lenge, however, has spilled over into the carbon
market. GWPs, with their apparent simplicity and
the black-box ‘possibility of use by policy-makers
with little further input from scientists’ (Shine
et al., 2005, p. 297) remain the way in which di?erent
gases are made commensurable, and the inscription
of the mid-1990s’ estimates of GWPs into the Kyoto
Protocol means that uncertainties and the changing
estimates of GWPs remain inside the black-box: a
matter for technical specialists, not carbon traders.
This black-boxing is crucial to allowing carbon
markets to encompass greenhouse gases other than
CO
2
: liquidity in such markets would be greatly
reduced if the relevant ‘exchange rate’ between
gases had to be negotiated ad hoc for each
9
‘The long-term forcing is. . .more accurately represented by
that acting after the stratosphere has returned to a state of global-
mean radiative equilibrium. Studies with simple models show that
the climate response, that is, the surface temperature change, is
proportional to the radiative forcing when the radiative forcing is
de?ned in this way. . . Importantly, the proportionality constant
is found to be the same for a wide range of forcing mechanisms’
(Pinnock, Hurley, Shine, Wallington, & Smyth, 1995, p. 23227).
10
See article 5, paragraph 3 of the Kyoto Protocol, the text of
which is available athttp://unfccc.int/resource/docs/convkp/
kpeng.html. Accessed 24.03.06.
11
Amongst other criticisms is ‘the fact that, despite its name, the
global warming potential does not purport to represent the
impact of gas emissions on temperature. The GWP uses the time-
integrated radiative forcing and this does not give a unique
indication of the e?ect of pulse emissions on temperature, because
of large di?erences in the time constants of the various
greenhouse gases. Although a strong greenhouse gas with a short
lifetime could have the same GWP as a weaker greenhouse gas
with a longer lifetime, identical (in mass terms) pulse emissions of
the two gases could cause a di?erent temperature change at a
given time. Economists have also criticised the GWP concept for
not being based on an analysis of damages caused by the
emissions’ (Shine et al., 2005, p. 282).
446 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
transaction. Note that the black-boxing rests upon
a ‘social’ factor: the authority of the Intergovern-
mental Panel on Climate Change. Although that
authority has been challenged by climate change
‘sceptics’ and ‘deniers’, public controversy has
focused on the reality, extent of and evidence for
anthropogenic climate change, not on matters of
‘detail’ such as GWPs, debate over which has
taken place only in much more limited circles.
The IPCC’s authority in such detailed matters is
thus an essential part of ‘making things the same’
in carbon markets, by keeping the ‘exchange rates’
between gases inside the black-box and separate
from political and economic disputes.
It is perfectly possible, however, that this black-
boxing may become harder in the future. At the
time of the Kyoto Protocol, it is unlikely that any-
one imagined that the ?gure of 11,700 for the global
warming potential of HFC-23 would determine a
?ow of funds of the order of $3.5 billion (the likely
total value of credits from HFC-23 decomposition
up to 2012: see Wara, 2007). As negotiations begin
over a successor to Kyoto, however, the ?nancial
consequences of such ?gures can now be seen. It is
possible that GWPs will remain in practice unchal-
lenged – it would be very hard, given the diversity
of economic interests involved, to get agreement
on a measure other than GWPs, or on anything
other than the IPCC’s estimates of them (which
are a ‘focal point’ in game-theoretic terms), so no
party to the negotiations may attempt to do so –
but it is not a foregone conclusion.
Making things the same: ‘Emission rights’
Gases are thus made the same by a combination
of measurement devices, complex natural science,
and the capacity (at least so far) of the Intergovern-
mental Panel on Climate Change to keep the estima-
tion of global warming potentials bracketed o?
from carbon-market politics. But practices of many
other kinds are also needed to make ‘carbon’ fungi-
ble, and amongst these accounting is of particular
importance.
12
The European allowances that Edin-
burgh University needs to emit carbon dioxide
and the CERs generated by Zhejiang Juhua Co.
are items that Europe’s (or indeed China’s) accoun-
tants have not previously encountered. What kind
of items are they? What accounting treatment
should they receive? These questions are signi?cant
for the operation of carbon markets, since account-
ing makes economic items visible, and whether and
how it does so is consequential.
13
Hatherly et al. (submitted for publication) argue
that a ‘?nitist’ perspective is useful for the analysis
of accounting, especially of accounting classi?ca-
tion, and it is particularly appropriate here. In this
perspective, how to classify an item (not just an
accounting item, but an item of any kind) is always
implicitly a choice. Past classi?cations – which are
always ?nite in number, hence ‘?nitism’ – in?uence
present classi?cations by analogy (‘this item is like
previous items we classi?ed as X, so this should be
classi?ed as an X’), but do not determine them.
Of course, classi?cation often does not feel like a
choice. Classi?ers – bookkeepers, accountants, orni-
thologists, botanists, and so on – often, probably
normally, come across items that seem familiar
and simply ‘see’ them as an X (‘this is an X’, not
‘I am classifying this as an X’). Items that seem to
classi?ers to be unfamiliar are thus of particular
analytical interest, because they make implicit
choice explicit. Instead of relying on habit and rou-
tine, those involved have consciously and explicitly
to decide what classi?cation is appropriate, and
the debate that is often sparked can reveal the con-
tingencies that a?ect classi?cation.
In the run-up to the launch of the European
Union Emissions Trading Scheme, the International
Financial Reporting Interpretations Committee
(IFRIC), a subsidiary body of the International
Accounting Standards Board, discussed how to
apply accounting standards to the new items, which
it called ‘emission rights’, which were about to come
into being. What kind of items were they? For
example, were they indeed ‘rights’? The IFRIC con-
cluded that they were not: ‘an allowance itself does
12
I am deeply grateful to Allan Cook, who served as Co-
ordinator for the International Financial Reporting Interpreta-
tions Committee at the end of the period in question for his help
in the research underpinning this section. Cook (forthcoming) is
his own account of these events. For broader legal debate over
the nature of carbon credits and allowances, see, Wemaere and
Streck (2005).
13
The issue of devising appropriate frameworks for making
carbon emissions ‘visible’, for example in corporate accounts, has
received considerable attention: see, for example, the work of
Fred Wellington and his colleagues at the World Resources
Institute (such as Lash & Wellington, 2007) and The Prince’s
Charities (2007). How to account for emissions allowances,
however, has received much less attention: see Cook (forthcom-
ing) and Casamento (2005).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 447
not confer a right to emit. Rather it is the instru-
ment that must be delivered in order to settle the
obligation that arises from emissions’ (IFRIC,
2004, p. 19).
14
An allowance was, however, in the IFRIC’s view
clearly an asset. But what was its nature? Was it an
‘intangible asset’ – ‘An identi?able non-monetary
asset without physical substance’ (IASB, 2005, p.
2227) – and thus within the scope of International
Accounting Standard (IAS) 38? Or was it a ‘?nan-
cial instrument’ – a ‘contract that gives rise to both
a ?nancial asset of one entity and a ?nancial liability
or equity instrument of another entity’ (IASB, 2005,
p. 2219) – and thus within the scope of the standard
governing such instruments, IAS 39? Some of those
who commented on the IFRIC’s initial draft argued
that an allowance was indeed a ?nancial instrument,
but the IFRIC disagreed: though allowances ‘have
some features that are more commonly found in
?nancial assets than in intangible assets’ – such as
being ‘traded in a ready market’ – they were not
?nancial instruments (IFRIC, 2004, p. 21).
An allowance was thus, in the IFRIC’s view, an
intangible asset, and therefore governed by IAS
38. If governments issued allowances at less than
their market value (most have issued them free-of-
charge) the di?erence was, IFRIC decided, a ‘gov-
ernment grant’, and its accounting treatment should
therefore follow the relevant standard, IAS 20.
Emissions themselves – as noted, previously outside
an economic or accounting frame – now had to
come within it. The emissions of those governed
by cap-and-trade schemes should, said the IFRIC,
be treated as giving rise to liabilities that were ‘pro-
visions’ whose treatment should follow IAS 37
(IFRIC, 2004, p. 7).
The IFRIC’s conclusions – crystallised in IFRIC
Interpretation 3: Emission Rights, issued in Decem-
ber 2004, on the eve of the start of the European
Union Emissions Trading Scheme – thus made
‘emission rights’ the same by laying down a homo-
geneous approach to accounting for them, in which,
for example, an allowance received free by an indus-
trial company or bought by an investment bank
were both treated in the same way as intangible
assets. However, IFRIC 3 encountered strong oppo-
sition, with critics arguing that the relationship of
IFRIC 3 to the three relevant standards – IAS 20,
37 and 38 – would create accounting mismatches,
especially in the light of anticipated changes to
IAS 20, which if made will mean that non-repayable
government grants have to be recognized when they
are received (see Cook, forthcoming). For example,
the fair value of the allowances that a company
received free would have to be recognized immedi-
ately as income, while the costs of the corresponding
emissions would be recognized only gradually as
they accumulated.
Re?ecting the criticism of IFRIC 3, the Euro-
pean Financial Reporting Advisory Group told
the European Commission in June 2005 that the
interpretation ‘will not always result in economic
reality being re?ected’, and recommended that the
Commission not endorse it.
15
The following month,
the International Accounting Standards Board,
while defending IFRIC 3 as ‘an appropriate inter-
pretation’ of existing accounting standards,
acknowledged that it ‘creates unsatisfactory mea-
surement and reporting mismatches’ and withdrew
it.
16
There was, of course, a ‘bottom line’ issue under-
pinning the controversy surrounding IFRIC 3. Cor-
porations generally fear earnings volatility: there is
a widespread conviction that investors prefer earn-
ings that rise smoothly to those that ?uctuate, even
around the same underlying trend. IFRIC 3 threa-
tened to produce volatility that, in its critics’ eyes,
would be arti?cial. For example, the advantage,
for corporations, of classifying an ‘emission right’
14
‘It therefore follows that a participant in a cap and trade
scheme does not consume the economic bene?ts of an allowance
as a result of its emissions. Rather a participant realises the
bene?ts of that allowance by surrendering it to settle the
obligation that arises from producing emissions (or by selling it
to another entity). Therefore, the IFRIC observed that amorti-
sation, which is the systematic allocation of the cost of an asset to
re?ect the consumption of the economic bene?ts of that asset
over its useful life, is incompatible with the way the bene?ts of the
allowances are realised. Although the IFRIC agreed that this
observation pointed to precluding amortisation, it agreed with
those respondents who highlighted that in some cases such a
requirement could be inconsistent with the requirements of IAS
[International Accounting Standard] 38. The IFRIC therefore
decided not to proceed with its proposal... that allowances should
not be amortised. Nonetheless, for most allowances traded in an
active market, no amortisation will be required, because the
residual value will be the same as cost and hence the depreciable
amount will be zero.’ (IFRIC, 2004, pp. 22–23).
15
Letter from Stig Enevoldsen to Alexander Schaub, 6 May
2005. Available fromhttp://www.iasplus.com/interps/
ifric003.htm. Accessed 11.07.07.
16
International Accounting Standards Board, ‘IASB withdraws
IFRIC Interpretation on Emission Rights’, available from http://
www.iasplus.com/interps/ifric003.htm. Accessed 11.07.07.
448 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
as a ?nancial instrument would have been that it
would make available the ‘hedge accounting’ treat-
ment permitted under IAS 39. If allowances could
‘be treated as the hedging instrument of a forecast
transaction (ie future emissions)’ (IFRIC, 2004, p.
20), then allowances and the corresponding emis-
sions would o?set each other. If a company received
N free allowances, forecast emissions of N tonnes of
carbon dioxide, and emitted N tonnes, then its earn-
ings would at no point be a?ected. ‘Carbon’ would
thus remain invisible.
The withdrawal of IFRIC 3 means that it
remains permissible to treat carbon in this way: as
inside an economic frame, but in a sense invisibly
so, since no accounting recognition is needed if the
above conditions are met. A survey by Deloitte
(2007) found that some market participants were
doing just that. Others were in e?ect following
IFRIC 3, while others again were doing so partially,
treating the provision for the liability created by
emissions in a di?erent way.
17
The attempt to make
‘emission rights’ the same has, in this sense, so far
failed.
The partial invisibility of carbon also means that
the incorporation of the carbon price into the mar-
ket’s ‘calculative mechanisms’ (Callon & Muniesa,
2003) is only partial. Although it is impossible to
be certain, there is tentative evidence from my inter-
view data of e?ects of both the accounting visibility
of carbon in some ?rms and its invisibility in others.
Consider, for example, the e?ect of the European
Union Emissions Trading Scheme on electricity
prices. If allowances are distributed free, one might
na? ¨vely think that they should have no e?ect on the
price of electricity. If a generator is given enough
allowances to cover its emissions (most generators
have actually had to buy some allowances, but let
me set that aside), what it charges customers surely
shouldn’t change? An economist will quickly tell
you what’s wrong with that argument. As already
noted, there’s an opportunity cost involved. In a
‘perfect market’, a pro?t-maximizing ?rm will pro-
duce electricity only if the price it receives is greater
than what it can earn by not generating electricity
and selling its stocks of the required inputs: its coal,
its gas, and now its carbon allowances (Point Car-
bon, 2007, pp. 24–25). If its allowances can com-
mand a non-zero price, the price of electricity
must rise correspondingly.
According to an interviewee in the electricity
market, however, it has required accountants to give
force to this economists’ reasoning. The ‘na? ¨ve’ view
prevailed in the industry until explicit valuations of
allowances started to be made. The price e?ect
‘should’ have been manifest in forward contracts
covering supply from January 2005 (the start of
the ETS) onwards, but apparently it initially was
not.
18
The e?ect began in the UK only once January
2005 was reached, and analysis by the consultancy
Point Carbon (2007) suggests it was even slower to
appear on the Continent. (Once the e?ect began,
the result in the UK was, for example, an increase
in domestic electricity prices in 2005 of around
7%
19
– for example, about £20 on a £300 annual bill
– and it is increases of this kind that are the source
of the much-criticized ‘windfall pro?ts’ that electric-
ity generators have made from the Emissions Trad-
ing Scheme.)
Carbon has thus been ‘visible’ for some time in
the electricity sector. When, in contrast, carbon is
kept invisible in accounting terms e?ects of three
17
Deloitte (2007) does not estimate the relative prevalence of the
three forms of accounting treatment. Those in the third category
‘recognise a provision on the following bases: To the extent that
the entity holds a su?cient number of allowances, the provision
should be recognized based on the carrying value of those
allowances (i.e., the cost to the entity of extinguishing their
obligation). To the extent that the entity does not hold a su?cient
number of allowances, the provision should be recognized based
on the market value of emission rights required to cover the
shortfall; and the penalty that the entity will incur if it is unable to
obtain allowances to meet their obligations under the scheme,
and it is anticipated that the penalty will be incurred (note that
the obligation to deliver allowances must still be ful?lled).’
(Deloitte, 2007, p. 3)
18
This is an interviewee’s assertion. Unfortunately, I do not
have access to the price data needed to test it quantitatively.
19
Calculation by Karsten Neuho?, quoted on BBC Radio 4,
‘File on Four’, 5 June 2007. Controversy is growing across
Europe about these ‘windfall pro?ts’. In the UK, for example, the
energy regulator Ofgem has called for the windfall pro?ts of the
UK’s electricity generators in the 2008–2012 phase of the
Emissions Trading Scheme – which Ofgem estimates at £9 billion
– to be used to help customers in fuel poverty (Crooks, 2007). In
Germany, the Bundeskartellamt (Federal cartel o?ce) charged
electricity generator RWE with behaving illegally by incorporat-
ing in the price it charged industrial consumers the market value
of permits it had received free. The case was settled out of court in
September 2007, with RWE continuing to defend its pricing but
agreeing that in 2009–2012 it would hold annual auctions of
quantities of power almost equivalent to its annual sales to
German industry (46,000 GWh in total over the 4 years) and
transfer to the purchasers, free of charge, the corresponding
carbon allowances if it had received these at no cost (RWE AG,
2007).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 449
kinds can be anticipated. The ?rst, which is hypo-
thetical (I have no direct evidence on the point),
would be to undermine a major desired e?ect of
a carbon market: incentivizing even those compa-
nies which have ‘enough’ allowances to cut their
emissions so as to generate income by selling
allowances. For this e?ect to be realized, allow-
ances need to be seen as assets with potential
monetary value, not simply as means of complying
with regulatory requirements. The second, related
e?ect (of which there is some tentative evidence)
is to delay the sale of allowances by those who,
even without abatement, have more allowances
than they need. The sale of allowances – and also
lending allowances for short sale – means that
they can no longer be kept invisible. They must
be recognized in accounting terms, and, for exam-
ple, a tax liability may be crystallized. This disin-
centive may reinforce other reasons for not
selling, such as the fact that emission levels will
in general be known in advance only approxi-
mately and the lack of a culture of proprietary,
risk-taking trading in many industrial companies
(in contrast to electricity suppliers, which are
active traders) that would permit the sale of allow-
ances that probably – but not certainly – will not
be needed.
My interview data do not permit me to judge
the relative importance of the various reasons for
postponing the sale of allowances that are likely
to be surplus to requirements, but those intervie-
wees with whom I explored the topic all believed
delayed sale to be a real phenomenon. It has been
consequential because the complex process of set-
ting national allocations for the ?rst phase (Janu-
ary 2005–December 2007) of the European scheme
led to over-allocation of allowances. The extent of
over-allocation was, however, not clear initially,
and the failure of those who were ‘long’ allow-
ances to bring them to market led to a constric-
tion of supply, which helped market prices to
rise to €31/tonne (see Fig. 2). Curiously, when
the extent of over-allocation became clear in the
spring of 2006, prices – though plunging dramati-
cally – did not initially fully re?ect the fact that
allowances no longer had any signi?cant economic
value. It took several months for the market price
of a phase-one European allowance to fall close to
zero (only in 2007 did prices become in e?ect zero,
with allowances towards the end of the year cost-
ing less than €0.10/tonne). Interviewees suggested
that delayed sale by those who were ‘long’ allow-
ances accounts for this paradoxical behaviour of
the carbon price. Even though it was clear that
allowances were intrinsically close to worthless
(because, in aggregate, there were more of them
than would be needed), they still commanded a
price of several euros, because not enough were
brought to market.
20
The third – again hypothetical – e?ect of the
accounting invisibility of carbon may be to
strengthen the hand of managers whose interests
lie in protecting market share by not passing on to
customers the opportunity cost of allowances that
have been allocated free, even when passing on the
cost is pro?t-maximizing for their ?rms. The extent
to which ?rms pass on the opportunity cost is cru-
cial to the environmental e?ects of a cap-and-trade
market – if they pass it on, there is likely to be car-
bon ‘leakage’ from the scheme, as imports from out-
side its boundaries become more attractive – and
there is ?erce controversy over likely behaviour in
this respect. Economists tend to predict pro?t-max-
imization, cost pass-through and thus leakage, while
?rms themselves tend to argue that market share
will be protected and costs will not be passed
through, at least in full. Unfortunately, empirical
analysis of the Emissions Trading Scheme so far is
too limited to be con?dent how ?rms outside the
electricity sector have behaved in this respect: see
Carbon Trust (2008).
Fig. 2. Price history of allowances, phase I of European Union
Emissions Trading Scheme. Courtesy Point Carbon.
20
One interviewee, at a hedge fund, reported making a
considerable amount of money by taking a short position in
allowances in this period.
450 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
The politics of carbon markets
One could go deeper into the issue of fungibility,
of making things the same. A trade, for example, is
a legal transaction requiring documentation, and
with three bodies (the International Swaps and
Derivatives Association, the European Federation
of Energy Traders, and the International Emissions
Trading Association) competing in this sphere,
interviewees reported that it has taken orchestrated
action to reduce the di?erences to a level at which a
trade documented in one format can be regarded as
similar enough to one documented in another, for
example for one to be used to hedge the other.
There has also, for instance, been sharp criticism
from competitors of the e?orts by Barclays Capital,
a leading player in the carbon market, to standard-
ise CERs via its SCERFA (Standard CER Forward
Agreement). The competitors regard a SCERFA as
speci?c to Barclays, not as a ‘standard’ entity.
Instead, however, let me consider the question of
the attitude to be taken to carbon markets. There is
a great deal of suspicion of them, ranging from right-
wing distaste for emissions caps to leftwing hostility to
an extension of market relations. The e?orts at market
construction so far have led to some environmental
bene?ts – for example, because of HFC-23’s potency,
curbing emissions of it is very valuable – but also sig-
ni?cant problems. There has, for example, been only
modest abatement by Europe’s electricity producers
(the sharp rise in gas prices in 2005–2006 swamped
any carbon-price incentive to switch from coal to
gas), while the mechanism discussed above led them,
as noted, to make substantial windfall pro?ts.
Similarly, the large sums that can be earned by
decomposing HFC-23 also create substantial pro?ts,
because the costs of decomposition are modest. A
specialized incinerator of the kind needed costs
around $4–5 million to install and $20,000 a month
to run (McCulloch, 2005, p. 12). Even with China’s
65% tax, a large HCFC-22 plant can recoup the
installation cost in a few months and go on to earn
revenues of well over a million dollars a month.
There is debate over just how much the subsidy
increases HCFC-22 production: McCulloch (2005)
argues that because the cost of HCFC-22 is only a
small proportion of the costs of the products in
which it is used,
21
a reduction in the price of
HCFC-22 will not expand the market for it very
much. However, the de facto subsidy may slow the
replacement of HCFC-22 by more environmentally
friendly refrigerants. (HCFC-22 is an ozone deplet-
er, the use of which as a refrigerant will eventually
be phased out under the Montreal Protocol govern-
ing such substances, and it is also a greenhouse
agent, though not as potent as HFC-23.) Because
of fears of this kind, ‘new’ HCFC-22 production
(i.e., over and above 2000–2004 levels) is currently
not eligible for CDM credits, but the consequence
is that there is no economic incentive not simply to
discharge HFC-23 from such new production into
the atmosphere rather than decomposing it.
In the light of issues such as these, it is tempting
to conclude that carbon markets are inherently
?awed means of achieving abatement. As Callon
(1998) points out, constructing a market requires
an enormous degree of ‘cooling’: of knowledge, of
metrologies, of actors, of identities, of interests. In
a perceptive article, Lohmann (2005, pp. 211 and
229) applies Callon’s analysis to the carbon market
and essentially concludes that market construction
will indeed fail: ‘conditions are not cool enough
for the spadework for commercial relations’, and
‘an unstoppable fount of complexity’ has been
uncorked.
Indeed, much of what I have described is consis-
tent with a bleak, essentialized view of capitalism, as
inherently irresponsible and environmentally dam-
aging, rather than Callon and Latour’s more opti-
mistic perspective. Yet the conclusion that carbon
markets are inherently ?awed carries a risk. Aban-
donment of such markets might well mean no seri-
ous international abatement e?orts, rather than
abatement by other means. If the Emissions Trading
Scheme were abandoned, could the European
Union ?nd a viable alternative, and how long would
it take? The political viability of a harmonized car-
bon tax, the obvious other route, remains question-
able, because of the unanimity required.
Similarly, political constraints mean that if inter-
national agreement on a replacement for the Kyoto
Protocol can be reached, it is likely to include some-
thing similar to the Clean Development Mechanism.
The CDM is, as noted, a result of the need to secure
developing-country participation in abatement
e?orts in a context in which the developing world
was and is unwilling to take on caps: even caps post-
poned to a later date, given the risk that by then
many of the cheaper opportunities for abatement
might be exhausted. The reluctance is understandable,
21
An air conditioning unit retailing at $500–1000 needs less than
a kilogram of HCFC-22, which costs around $1–$2 (McCulloch,
2005, p. 7).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 451
given the desire not to allow a problem caused by
the industrialized countries to serve as a brake upon
development, and it is likely to persist – even in a
context in which China, in particular, no longer ?ts
the traditional template of a developing country.
Abatement e?orts in the developing world are thus
likely to continue to require funding from the devel-
oped world. Of course, such funding could be
achieved by direct government aid – Wara (2007)
points out that HFC-23 decomposition could have
been achieved far more cheaply via this route than
via the CDM – but that again raises the question
of whether governments would in practice make
the requisite large transfers of resources.
22
To conclude that carbon markets must fail may
also be unduly pessimistic, in that it would miss
the extent to which carbon markets hitherto have
been experimental, in the case of phase 1 of the
European Union Emissions Trading Scheme, quite
explicitly so: interviewees involved in establishing
it reported the many compromises that had to be
made to get it up and running, such as the fact that
it was possible to challenge only the most egre-
giously over-generous national allocations of allow-
ances. While existing carbon markets
unquestionably have major ?aws, those ?aws are
increasingly becoming manifest, and ways of reme-
dying them are available. Thus, windfall pro?ts
within the European scheme could be eliminated
by moving from free allocation to full auctioning
(Dales’s original proposal), and there is now a real
possibility that this will happen from 2013 on, at
least in the electricity sector.
If carbon markets are here to stay, can they be
improved? One example of a successful intervention
is of particular interest from the viewpoint of this
paper, because it involves making things not the
same. NGOs, especially the World Wildlife Fund,
have sought to create a separate category of ‘gold
standard’ CERs, covering only renewable energy
and energy conservation projects, and excluding
industrial gas projects such as HFC-23 decomposi-
tion.
23
The gold standard is a form of cooling in
Callon’s sense (as with the CDM as a whole, there
is a formal methodology, automated tools, a role
for auditors, and so on), and there are ‘bottom-line’
e?ects. Although from the viewpoint of the Kyoto
Protocol or of monetizability via the European
Emissions Trading Scheme, an ordinary and a gold
standard CER are identical, my interviewees
reported that the market price of the latter is now
around 10–20% higher. (They suggest that the cause
of the higher price is that those who are buying
CERs not just for compliance but to achieve ‘car-
bon neutrality’ or other forms of o?setting fear rep-
utational risk if it is discovered that ‘neutrality’ is
being achieved via industrial gas projects such as
HFC-23.) ‘Multiple monies’ have emerged in the
carbon market, as a result of intervention by
activists.
The intervention by the World Wildlife Fund and
other NGOs was informal: it did not alter the for-
mal procedures of the CDM. However, NGOs are
also seeking to practise a politics of market design
in a more formal sense, seeking to alter rules and
procedures. That, indeed, is precisely the course of
action that Callon and Latour’s perspective implies.
If markets are plural – Callon’s best-known work is
titled The Laws of the Markets (Callon, 1998) – and
‘capitalism’ has no unalterable essence, then this
may indeed be productive.
Such e?orts are too recent and too limited to
know whether they will be successful. However, it
is worth noting that changes in market design of a
kind that seem potentially achievable could be con-
sequential. Take the underlying issue of a carbon
market versus a carbon tax. Many environmental
activists prefer the latter, as do some economists
such as Nordhaus (2007). Nordhaus argues that
the classic analysis by Weitzman (1974) of the con-
ditions that in?uence the relative e?ciency of ‘quan-
tity-based’ instruments (such as a cap-and-trade
scheme) and ‘price-based’ instruments (such as a
carbon tax) suggests, given the speci?c cost-bene?t
features of combating global warming, the superior
e?ciency of a carbon tax.
Yet carbon markets seem politically feasible,
even in the US; carbon taxes may not be, even in
Europe. Intriguingly, however, a cap-and-trade
market, with full auctioning rather than free alloca-
tion, can be equivalent to an optimally set tax. In
both, polluters pay, either by having to buy permits
or by paying the carbon tax. Indeed, under admit-
tedly ‘idealized conditions’ (Hepburn, 2006, p.
229) they pay the same amounts, and the environ-
mental outcomes are the same. Thus, if the relation-
ship between emission levels and the carbon price is
known with certainty, either a cap-and-trade market
22
For an intriguing suggestion of a means of achieving north–
south transfers at a su?cient level to make a signi?cant impact on
developing countries’ needs to adapt to climate change, see
Mu¨ ller and Hepburn (2006).
23
Seehttp://www.cdmgoldstandard.org. Accessed 17.01.08.
452 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
or a correctly set tax can achieve a required level of
abatement, and the necessary tax rate will be the
same as the allowance price. Of course, the relation-
ship between emission levels and the carbon price is
not known with certainty, and for that and other
reasons the full equivalence between tradable per-
mits and a tax does not pertain in the real world.
However, economists’ analyses suggest ways of
designing a carbon market that might make it and
a tax more closely equivalent in practice. These
include rules facilitating the ‘banking’ of permits
for future use and the ‘borrowing’ of permits from
future years, regulated perhaps by an adjustable
requirement for ?rms to hold a certain amount of
permits in reserve, analogous to the adjustable
reserves that banks are required to hold (Newell,
Pizer, & Zhang, 2005).
Precisely because of the similarity of auctioning
to a carbon tax, emissions markets seem almost
always initially to involve free allocation, because
this reduces lobbying against them and political
opposition. However, once markets are well-estab-
lished, as the European Union Emissions Trading
Scheme now is, shifting to auctioning may become
easier (especially now the ‘economic experiment’
of Phase I of the ETS has made publicly visible
the problems that free allocation leads to). For
example, in October 2007 Sweden announced that
it was ending free allocation of allowances to its
electricity and heat sectors.
24
Indeed, as noted
above, it seems increasingly likely that in the third
phase of the ETS, from 2013 onwards, auctioning
may be much more heavily employed, at least for
sectors such as electricity that cannot in practice
easily move production outside of the European
Union.
The e?ort to shift the ETS to auctioning is ‘pol-
itics’ of a classic, recognizable kind, involving gov-
ernments, the policy-makers of a supranational
body, nation-state representatives, ?erce industry
lobbying against auctioning, and so on. Not all
the politics of carbon markets, however, ?ts that
recognizable template. Neither the IPCC nor the
International Accounting Standards Board see
themselves as political bodies, and indeed it is of
particular importance that the former not be seen
as political, despite the e?orts of its critics to paint
it as such. Yet they are arguably locales of ‘sub-
politics’ in Beck’s sense: politics ‘outside and
beyond the representative institutions of the polit-
ical system of nation-states’ (Beck, 1996, p. 18; see
Holzer & Sørensen, 2003). For example, the
IFRIC and now the International Accounting
Standards Board (which is turning its attention
to emission rights) have to contend with pressure
that has had the e?ect of blocking e?orts to ‘make
things the same’ in carbon markets. In the case of
the IPCC, the key ‘subpolitical’ matter is, para-
doxically, preserving the boundary between ‘sci-
ence’ and ‘politics’, since that boundary is
precisely what is needed to facilitate political
action, because it matters that action can be seen
as based upon ‘sound science’.
The subpolitics of carbon markets may seem
esoteric, and it is certainly not simple, but it is
important. Clearly, such markets are only one tool
for combating global warming, and other tools are
also important: direct regulation, carbon taxes
(where these are feasible), greatly increased public
expenditure on research and development and on
necessary infrastructure (for example, the electric-
ity grid changes needed to make increased renew-
ables production more attractive economically),
the removal of the many subsidies for fossil-fuel
extraction and use, and so on (see, for example,
Lohmann, 2006; Prins & Rayner, 2007). Neverthe-
less, making carbon markets more e?ective is cru-
cial, and the esoteric nature of their subpolitics
means that researchers have a particularly salient
role to play in bringing to light matters of appar-
ent detail that in fact play critical roles in this
respect.
It is this author’s hope that this paper will
encourage the work of this kind that is so badly
needed. The existing and planned experiments in
changing capitalism’s bottom line are heteroge-
neous, widely di?used worldwide, and involve many
aspects – scienti?c, technological, political, account-
ing, sociological, anthropological, geographical –
beyond economics as narrowly conceived. The
experiments need ‘witnesses’ (Shapin & Scha?er,
1985), and those witnesses must be multiple: lay as
well as professional, from many countries, and if
they are academics from many disciplines.
25
Carbon
markets need to become part of a process of ‘social
learning’ (qv Williams, Stewart, & Slack, 2005), in
which institutions to mitigate climate change are
24
Announcement of Environment Minister Anders Calgren,
reported by news service Point Carbon (www.pointcarbon.com).
11.10.2007.
25
I owe this way of formulating the matter to Andrew Barry.
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 453
created, evaluated and reshaped.
26
Such multiple
witnessing and social learning needs to concern
not just the overall features of carbon markets,
but the crucial ‘nuts and bolts’ of their construction,
questions such as how di?erent carbon sources and
sinks are commensurated, how allowances are trea-
ted in accounting terms, and many other such mat-
ters that I have been unable to discuss for space
reasons. If this modest paper recruits others to take
part in this multiple witnessing and social learning,
then it will have achieved its goal.
References
Agrawala, S. (1998a). Context and early origins of the intergov-
ernmental panel on climate change. Climatic Change, 39,
605–620.
Agrawala, S. (1998b). Structural and process history of the
intergovernmental panel on climate change. Climatic Change,
39, 621–642.
Barnes, B., Bloor, D., & Henry, J. (1996). Scienti?c knowledge: A
sociological analysis. Athlone and Chicago, London: Chicago
University Press.
Beck, U. (1996). World risk society as cosmopolitan society?
Ecological questions in a framework of manufactured uncer-
tainties. Theory, Culture and Society, 13(4), 1–32.
Bumpus, A. G., & Liverman, D. M. (submitted for publication).
Accumulation by decarbonisation and the governance of
carbon o?sets. Economic Geography.
Burtraw, D., Evans, D. A., Krupnick, A., Palmer, K., & Toth, R.
(2005). Economics of pollution trading for SO
2
and NO
x
.
Annual Review of Environmental Resources, 30, 253–289.
Callon, M. (Ed.). (1998). The laws of the markets. Oxford:
Blackwell.
Callon, M. (2007). What does it mean to say that economics is
performative? In D. MacKenzie, F. Muniesa, & L. Siu (Eds.),
Do economists make markets? On the performativity of
economics (pp. 311–357). Princeton, NJ: Princeton University
Press.
Callon, M., & Latour, B. (1981). Unscrewing the big Leviathan:
How actors macro-structure reality and how sociologists help
them to do so. In K. K. Cetina & A. V. Cicourel (Eds.),
Advances in social theory and methodology: Toward an
integration of micro- and macro-sociologies (pp. 277–303).
Boston: Routledge and Kegan Paul.
Callon, M., & Latour, B. (1997). ‘Tu ne calculeras pas!’ Ou
comment syme´triser le don et le capital. Revue du MAUSS, 9/
1, 45–70.
Callon, M., & Muniesa, F. (2003). Les marche´s e´conomiques
comme dispositifs collectifs de calcul. Re´ seaux, 21(122),
189–233.
Carbon Trust (2008). EU ETS impacts on pro?tability and trade:
A sector by sector analysis. London: Carbon Trust.
Casamento, R. (2005). Accounting for and taxation of emission
allowances and credits. In D. Freestone & C. Streck (Eds.),
Legal aspects of implementing the Kyoto Protocol mechanisms:
Making Kyoto work (pp. 55–70). Oxford: Oxford University
Press.
Cass, L. (2005). Norm entrapment and preference change:
The evolution of the European Union position on
international emissions trading. Global Environmental Pol-
itics, 5(2), 38–60.
CDM Executive Board. (2007). Project 0868: No. 2 HFC-23
Decomposition Project of Zhejiang Juhua Co., Ltd., PR
China. . Accessed 11.06.07.
Christiansen, A. C., & Wettestad, J. (2003). The EU as a
frontrunner on greenhouse gas emissions trading: How did it
happen and will the EU succeed?. Climate Policy 3, 3–18.
Coase, R. H. (1960). The problem of social cost. Journal of Law
and Economics, 3, 1–44.
Cook, A. (forthcoming). Emission rights: From costless activity
to market operations. Accounting, Organizations and Society.
Crooks, (Ed.). (2007). Watchdog wants £9bn windfall electricity
pro?ts clawed back. Financial Times 17 January, p. 1.
Dales, J. H. (1968a). Land, water, and ownership. Canadian
Journal of Economics, 1, 791–804.
Dales, J. H. (1968b). Pollution, property and prices: An essay in
policy-making and economics. Toronto: University of Toronto
Press.
Damro, C., & Me´ndez, P. L. (2003). Emissions trading at Kyoto:
From EU resistance to Union innovation. Environmental
Politics, 12(2), 71–94.
Deloitte. (2007). Accounting for emission rights. . Accessed 31.07.07.
Ellerman, A. D., Schmalensee, R., Bailey, E. M., Joskow, P. L.,
& Montero, J.-P. (2000). Markets for clean air: The US Acid
Rain Program. Cambridge: Cambridge University Press.
European Parliament, Council. (2004). Directive 2004/101/EC of
the European Parliament and of the Council. O?cial Journal
of the European Union, 338, 13 November, 18–23.
Greimas, A. J. (1987). On meaning: Selected writings in semiotic
theory. London: Pinter.
Grubb, M. (1999). The Kyoto Protocol: A guide and assessment.
London: Royal Institute of International A?airs.
Hahn, R. W. (1989). Economic prescription for environmental
problems: How the patient followed the doctor’s orders.
Journal of Economic Perspectives, 3(2), 95–114.
Hatherly, D., Leung, D., & MacKenzie, D. (submitted for
publication). The ?nitist accountant: Classi?cations, rules and
the construction of pro?ts. In T. Pinch & R. Swedberg (Eds.),
Living in a material world: On the mutual constitution of
technology, economy, and society. Cambridge, MA: MIT
Press.
Hepburn, C. (2006). Regulation by prices, quantities, or both: A
review of instrument choice. Oxford Review of Economic
Policy, 22, 226–247.
Holzer, B., & Sørensen, M. P. (2003). Rethinking subpolitics:
Beyond the ‘Iron Cage’ of modern politics. Theory, Culture
and Society, 20(2), 79–102.
Houghton, J. T., Jenkins, G. J., & Ephraums, J. J. (Eds.). (1990).
Climate change: The IPCC scienti?c assessment. Cambridge:
Cambridge University Press.
26
There is of course a trade-o? between the need to evaluate and
improve a market’s design and the need for rules that are stable
over reasonably long time periods. The European Union’s trade-
o? seems reasonable – an explicitly experimental three-year initial
phase, then a ?ve-year second phase, followed by a third phase
that is likely to last eight years (2013-20).
454 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
Houghton, J. T., Meira Filho, L. G., Callander, B. A., Harris, N.,
Kattenberg, A., & Maskell, K. (Eds.). (1996). Climate change,
1995: The science of climate change. Cambridge: Cambridge
University Press.
IASB (International Accounting Standards Board) (2005). Inter-
national Financial Reporting Standards (IFRSs) 2005. Lon-
don: IASB.
IFRIC (International Financial Reporting Interpretations Com-
mittee) (2004). IFRIC interpretation 3: Emission rights.
London: International Accounting Standards Board.
Intergovernmental Panel on Climate Change. (2007). Climate
change 2007: The physical science basis. . Accessed 31.08.07.
Lash, J., & Wellington, F. (2007). Competitive advantage on a
warming planet. Harvard Business Review, 85(3), 95–102.
Lashof, D. A., & Ahuja, D. R. (1990). Relative contributions of
greenhouse gas emissions to global warming. Nature, 344(5),
529–531.
Latour, B. (2005). Reassembling the social: An introduction to
actor-network theory. Oxford: Oxford University Press.
Latour, B. (submitted for publication). Re´sume´ d’une enqueˆte sur
les modes d’existence, ou Bref e´loge de la civilisation qui vient.
Unpublished typescript.
Lecocq, F., & Ambrosi, P. (2007). The clean development
mechanism: History, status, and prospects. Review of Envi-
ronmental Economics and Policy, 1, 134–151.
Levin, P., & Espeland, W. N. (2002). Pollution futures: Commen-
suration, commodi?cation, and the market for air. In A. J.
Ho?man & M. J. Ventresca (Eds.), Organizations, policy, and
the natural environment: Institutional and strategic perspectives
(pp. 119–147). Stanford, CA: Stanford University Press.
Lohmann, L. (2005). Marketing and making carbon dumps:
Commodi?cation, calculation and counterfactuals in climate
change mitigation. Science as Culture, 14, 203–235.
Lohmann, L. (2006). Carbon trading: A critical conversation on
climate change, privatisation and power. Development Dia-
logue, 48(September), 4–359.
MacKenzie, D. (1990). Inventing accuracy: A historical sociology
of nuclear missile guidance. Cambridge, MA: MIT Press.
MacKenzie, D. (2007). The political economy of carbon trading.
London Review of Books (pp. 29–31). UK: Nicholas Spice.
MacKenzie, D. (submitted for publication). Material markets:
Facts, technologies, politics. Oxford: Oxford University Press.
McCulloch, A. (2005). Incineration of HFC-23 waste streams for
abatement of emisisons from HCFC-22 production: A review
of scienti?c, technical and economic aspects. Comberbach,
England: Marbury Technical Consulting. . Accessed
11.10.07.
Michaelowa, A. (2005). Determination of baselines and addi-
tionality for the CDM: A crucial element of credibility of the
climate regime. In F. Yamin (Ed.), Climate change and carbon
markets: A handbook of emission reduction mechanisms
(pp. 289–304). London: Earthscan.
Mu¨ ller, B., & Hepburn, C. (2006). IATAL: An outline proposal for
an international air travel adaptation levy. Oxford: Oxford
Institute for Energy Studies.
Muniesa, F., & Callon, M. (2007). Economic experiments and the
construction of markets. In D. MacKenzie, F. Muniesa, & L.
Siu (Eds.), Do economists make markets? On the performativity
of economics (pp. 163–189). Princeton, NJ: Princeton Univer-
sity Press.
Newell, R., Pizer, W., & Zhang, J. (2005). Managing permit
markets to stabilize prices. Environmental and Resource
Economics, 31, 133–157.
Nordhaus, W. D. (2007). To tax or not to tax: Alternative
approaches to slowing global warming. Review of Environ-
mental Economics and Policy, 1, 26–44.
Pinnock, S., Hurley, M. D., Shine, K. P., Wallington, T. J., &
Smyth, T. J. (1995). Radiative forcing of climate by hydro-
chloro?uorocarbons and hydro?uorocarbons. Journal of
Geophysical Research, 100, 23227–23238.
Point Carbon (2007). Carbon 2007: A new climate for carbon
trading. Oslo: Point Carbon.
Prins, G., & Rayner, S. (2007). Time to ditch Kyoto. Nature,
449(25), 973–975.
RWE AG. (2007). RWE erzielt Einigung mit Bundeskartellamt in
CO
2
-Verfahren. Essen: RWE AG, 27 September. . Accessed 28.09.07.
Schellnhuber, H. J. (Ed.). (2006). Avoiding dangerous climate
change. Cambridge: Cambridge University Press.
SGS United Kingdom Ltd. (2007). Veri?cation and certi?cation
report for HFC23 decomposition project of Zhejiang Juhua
Co., Ltd., PR China. . Accessed 11.05.07.
Shackley, S., & Wynne, B. (1997). Global warming potentials:
Ambiguity or precision as an aid to policy? Climate Research,
8(2), 89–106.
Shapin, S., & Scha?er, S. (1985). Leviathan and the air-pump:
Hobbes, Boyle, and the experimental life. Princeton, NJ:
Princeton University Press.
Shine, K. P., Fuglestvedt, J. S., Hailemariam, K., & Stuber, N.
(2005). Alternatives to the global warming potential for
comparing climate impacts of emissions of greenhouse gases.
Climatic Change, 68, 281–302.
Siegenthaler, U., & Joos, F. (1992). Use of a simple model for
studying oceanic tracer distributions and the global carbon
cycle. Tellus, 44B, 186–207.
Taubes, G. (2004). An interview with Dr. Laurence Rothman,
.
Accessed 31.05.07.
The Prince’s Charities (2007). Accounting for sustainability.
London: The Prince’s Charities.
Wara, M. (2007). Is the global carbon market working?. Nature
445(8), 595–596.
Weitzman, M. L. (1974). Prices vs. quantities. Review of
Economic Studies, 41, 477–491.
Wemaere, M., & Streck, C. (2005). Legal ownership and nature
of Kyoto units and EU allowances. In D. Freestone & C.
Streck (Eds.), Legal aspects of implementing the Kyoto
Protocol mechanisms: Making Kyoto work (pp. 35–53).
Oxford: Oxford University Press.
Wettestad, J. (2005). The making of the 2003 EU emissions
directive: An ultra-quick process due to entrepreneurial
pro?ciency?. Global Environmental Politics 5, 1–23.
Williams, R., Stewart, J., & Slack, R. (2005). Social learning in
technological innovation: Experimenting with information and
communication technologies. Elgar: Cheltenham.
World Bank (2007). State and trends of the carbon market, 2007.
Washington, DC: World Bank.
Zapfel, P., & Vainio, M. (2002). Pathways to European green-
house gas emissions trading history and misconceptions.
. Accessed 19.09.06.
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 455
doc_341600365.pdf
This paper analyses the development of carbon markets: markets in permits to emit greenhouse gases or in credits
earned by not emitting them. It describes briefly how such markets have come into being, and discusses in more detail
two aspects of the efforts to ‘make things the same’ in carbon markets: how different gases are made commensurable,
and how accountants have struggled to find a standard treatment of ‘emission rights’. The paper concludes by discussing
the attitude that should be taken to carbon markets (for example by environmentalists) and the possibility of developing a
‘politics of market design’ oriented to making such markets more effective tools of abatement.
Making things the same: Gases, emission rights and the
politics of carbon markets
Donald MacKenzie
School of Social and Political Studies, University of Edinburgh, Adam Ferguson Building, Edinburgh EH8 9LL, Scotland, United Kingdom
Abstract
This paper analyses the development of carbon markets: markets in permits to emit greenhouse gases or in credits
earned by not emitting them. It describes brie?y how such markets have come into being, and discusses in more detail
two aspects of the e?orts to ‘make things the same’ in carbon markets: how di?erent gases are made commensurable,
and how accountants have struggled to ?nd a standard treatment of ‘emission rights’. The paper concludes by discussing
the attitude that should be taken to carbon markets (for example by environmentalists) and the possibility of developing a
‘politics of market design’ oriented to making such markets more e?ective tools of abatement.
Ó 2008 Elsevier Ltd. All rights reserved.
Introduction
Around the world, markets in permits to emit
greenhouse gases or in credits earned by not emit-
ting them are emerging. Some already exist; others
are in construction.
1
This article describes brie?y
the route – at the level of ‘policy’ – that has led to
their emergence. It then delves a little deeper into
the conditions of possibility of these markets, by
examining two examples of what it takes to make
the entities traded in these markets ‘the same’. The
examples are how the destruction of one gas in
one place is made commensurate with emissions of
a di?erent gas in a di?erent place, and how accoun-
tants have sought (so far with only limited success)
to make ‘emission rights’ equivalent. Finally, the
article discusses the issue of politics: the question
of the attitude that should to be taken to carbon
markets (for example by environmentalists, espe-
cially those who conceive of themselves as oppo-
nents of ‘capitalism’), and the tightly-related issue
of the process of market design viewed, as it has
to be, as politics.
Although the article draws upon the ‘?nitist’ per-
spective sketched brie?y below (see Barnes, Bloor,
& Henry, 1996; Hatherly, Leung, & MacKenzie,
submitted for publication), its approach is
prompted by the view of economic life suggested
by the ‘actor-network’ theory of Michel Callon
and Bruno Latour (for which see, for example,
Latour, 2005). In Callon’s and Latour’s view, the
characteristics of an ‘actor’ – a term which, following
1570-8705/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2008.02.004
E-mail address: [email protected]
1
This paper was originally presented to the workshop ‘Carbon
Markets in Social-Science Perspective’, Durham University, 7
November 2007. The research was supported by a UK Economic
and Social Research Council Professorial Fellowship, RES-051-
27-0062, and I am deeply grateful to the Institute of Advanced
Study, Durham University for supporting the workshop and for a
Fellowship which enabled me to complete the paper.
Available online at www.sciencedirect.com
Accounting, Organizations and Society 34 (2009) 440–455
www.elsevier.com/locate/aos
semiotics (especially Greimas, 1987), they view as
encompassing more than just human beings – are
not intrinsic, but are the result of the networks of
which the actor is made up and forms part. What
we call ‘capitalism’, for example, is not an entity
with ?xed characteristics. ‘Que faire contre le capi-
talisme?’, they write: ‘D’abord e´videmment ne pas
y croire’ (Callon & Latour, 1997, p. 67). What is
to be done against capitalism? First of all, of course
do not believe in it.
In Callon’s and Latour’s view, economic life is
‘performed’ – framed and formatted – by ‘econom-
ics at large’, a term that encompasses not just the
academic discipline but also economic practices
such as accounting and marketing (Callon, 1998,
2007). The characteristics of economic actors and
of markets arise from, amongst other things, the
‘dispositifs de calcul’ (Callon & Muniesa, 2003) –
the calculative mechanisms – of which they are
made up.
If the characteristics of ‘capitalism’ are not inher-
ent, they can be changed by changing the calculative
mechanisms that constitute it. The markets in green-
house-gas emissions that are being constructed
globally are a set of experiments (Muniesa & Cal-
lon, 2007) in the validity of this prediction. Hith-
erto, greenhouse-gas emissions have been, in
economists’ familiar terminology, an ‘externality’:
from the viewpoint of the emitter, they bore no cost,
and so did not ?gure in emitters’ economic calcula-
tions. The goal of a carbon market is to bring emis-
sions within the frame of economic calculation by
giving them a price. In such a market, emissions
bear a cost: either a direct cost (because allowances
to emit greenhouses gases need to be purchased), or
an opportunity cost (because allowances that are
not used to cover emissions can be sold, or because
credits can be earned if emissions are reduced below
‘business as usual’). A carbon market is thus an
attempt to change the construction of capitalism’s
central economic metric: pro?t and loss, the ‘bottom
line’.
The experiments in carbon-market construction
have scarcely begun, so the validity of the prediction
that capitalism can be ‘civilized’ (Latour, forthcom-
ing) by changing calculative mechanisms remains
undecided. We do not yet know whether the bottom
line will be changed to any substantial extent, in
particular to an extent su?cient to keep global
warming below the threshold (uncertain and ?ercely
contested, but often taken to be 2 °C) beyond which
the risk of severe impacts rises sharply (Schellnhu-
ber, 2006).
In consequence, this paper is necessarily preli-
minary. The empirical material on which I am draw-
ing is limited. It consists primarily of a set of 24
interviews conducted with people involved with car-
bon markets (particularly with the European Union
Emissions Trading Scheme) as market designers, as
carbon traders and brokers, or as members of
NGOs seeking to in?uence the evolution of carbon
markets. This interview material is supplemented
by analysis of relevant documents such as monitor-
ing reports and contributions to the debate in
accountancy touched on below.
The article’s main aim is simply to help broaden
social-science research on carbon markets, both in
terms of its disciplinary base (though their origins
lie in economics, carbon markets cannot be under-
stood by the conventional tools of that discipline
alone) and in terms of its empirical focus. In that
latter respect, I hope to show that it is productive
to investigate not just overall questions such as the
reasons why policy-makers might choose carbon
markets rather than other tools to combat global
warming, but also the speci?cs of how carbon mar-
kets are constructed. Whether or not carbon mar-
kets are environmentally and economically
e?ective depends on such speci?cs, and the issues
involved are various and demand inter-disciplinary
treatment. One of the two topics examined below
– how di?erent gases are made commensurable –
is a natural question for the social studies of science
and technology; the other – how to ?nd a standard
treatment of ‘emissions rights’ – is a question obvi-
ously suitable for researchers in accounting.
Although for reasons of space I do not discuss them
here, questions for other disciplines can also easily
be identi?ed: for example, vastly more needs known
about how emission reduction projects in develop-
ing countries actually work in practice, a question
that raises issues ranging from how veri?cation is
conducted to the impact of projects on local com-
munities and local environments. Investigating such
issues in genuine depth requires the skills of,
amongst others, anthropologists and other area
specialists.
Because the speci?cs of market design matter, I
make no apology for the fact that this article
touches upon matters of apparent detail. The com-
mensurability of gases and the accounting treatment
of emission rights are inevitably ‘technical’ ques-
tions, and those technicalities cannot altogether be
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 441
avoided: they matter to overall outcomes. The com-
mensurability of gases, for example, is crucial to
how the world’s two main existing carbon markets
– the European Union Emissions Trading Scheme
and the Kyoto Protocol Clean Development Mech-
anism – interrelate, while there is at least tentative
evidence that the accounting treatment of emissions
rights a?ects ?rms’ behaviour in carbon markets. It
is precisely issues of this detailed kind that an e?ec-
tive, inter-disciplinary analysis of carbon markets
will need to address.
Carbon markets
Carbon markets come in two main species: ‘cap
and trade’ and ‘project-based’.
2
Let me begin with
the former. It involves a government or other
authority setting a ‘cap’ – a maximum allowable
aggregate total quantity of emissions – and selling
or giving the corresponding number of allowances
to emitters. The authority then monitors emissions
and ?nes anyone who emits without the requisite
allowances. If the monitoring and penalties are
stringent enough, aggregate emissions are thus kept
down to the level of the cap. The ‘trade’ aspect of
cap and trade arises because those for whom reduc-
tions are expensive will want to buy allowances
rather than incurring disproportionate costs. The
requisite supply of allowances is created by the
?nancial incentive thereby provided to those who
can make big cuts in emissions relatively cheaply.
They can save money by not having to buy allow-
ances, or (if allowances are distributed free) can
earn money by selling allowances they do not need.
So, as already noted, emissions, which previously
had no monetary cost, now have one.
The origins of the idea of controlling emissions
via a cap and trade scheme can be traced to the
work of Nobel Laureate Ronald Coase (1960), but
a more proximate source is the University of Tor-
onto economist Dales (1968a, 1968b), who ?rst
put forward the idea in something like full-?edged
form.
3
Emissions markets were implemented in rel-
atively minor and sometimes ham-?sted ways in the
1970s and 1980s, mainly in the United States (see,
e.g., Hahn, 1989). It was only in the 1990s that the
idea became mainstream.
The crucial development was the start of sulphur-
dioxide trading in the US in 1995 (for which see,
especially, Ellerman, Schmalensee, Bailey, Joskow,
& Montero, 2000; Burtraw, Evans, Krupnick, Pal-
mer, & Toth, 2005). It had been known for twenty
years or more that damage to the environment and
to human health was being caused by sulphur-diox-
ide emissions, notably from coal-?red power sta-
tions, which react in the atmosphere to produce
‘acid rain’ and other acid depositions. Numerous
bills were presented to Congress in the 1980s to
address the problem, but all failed in the face of
opposition from the Reagan administration and
from Democrats who represented states that might
su?er economically from controls on sulphur-diox-
ide, such as the areas of Appalachia and the mid west
in which coal deposits are high in sulphur.
Sulphur trading broke the impasse. It combined a
simple, clear goal that environmentalists could
embrace (reducing annual sulphur-dioxide emissions
from power stations in the US by 10 million tonnes
from their 1980 level, a cut of around a half) with
a market mechanism attractive to at least some
Republicans. The cut was achieved in practice far
more cheaply than almost anyone had imagined.
Industry lobbyists had claimed it would cost $10 bil-
lion a year, while the actual cost was around $1 bil-
lion. Allowance prices of $400 a tonne were
predicted, but in fact prices averaged around $150
or less in the early years of the scheme. The ?exibility
that trading gave to utilities helped to reduce costs
(by around a half compared to having to meet a stan-
dard that imposed a uniform maximum emission
rate: see Burtraw et al., 2005; Ellerman et al., 2000)
but other factors were equally important. ‘Scrub-
bers’ to remove sulphur from smokestacks turned
out to be cheaper to install and to run than had been
anticipated, and rail-freight deregulation sharply
reduced the cost of transportation from Wyoming’s
Powder River Basin, the main source of low-sulphur
coal in the United States (Ellerman et al., 2000).
That the sulphur-dioxide market was, broadly, a
success shaped how the Clinton Administration
approached the negotiations that led to the 1997
Kyoto Protocol. In the Protocol, the industrialized
nations undertook that by Kyoto’s 2008–2012 ‘com-
mitment period’ they would have limited their
greenhouse-gas emissions to agreed proportions of
2
This article concentrates on regulatory markets, largely
setting aside the ‘voluntary’ market, in which, for example, ?rms
choose to ‘o?set’ their emissions, even though they are under no
compulsion to do so: see, for example, Bumpus and Liverman
(forthcoming).
3
The history of emissions trading will be treated in more detail
in MacKenzie (submitted for publication). The brief account
given here draws upon that in MacKenzie (2007).
442 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
their 1990 levels: 93% for the US, 92% for the Euro-
pean Community overall (with varying levels for its
member states), and so on.
At the insistence of the US, Kyoto gave its signa-
tories sulphur-like ?exibility in how to meet their
commitments. The Protocol contains provision for
a cap and trade market between nation-states.
States with caps they will exceed can pay others that
in the commitment period are emitting less than
their caps for their unneeded ‘Kyoto units’ (quanti-
ties of carbon dioxide or their equivalents in other
gases: see below). Just how much trading of such
units between nation-states there will be remains
to be seen: it is possible it will be quite limited. More
signi?cant so far have been two other Kyoto mech-
anisms – ‘Joint Implementation’ and, especially, the
‘Clean Development Mechanism’ (CDM) – which
are project-based schemes, not cap and trade.
Let me concentrate on the CDM (for which see,
for example, Lecocq & Ambrosi, 2007), which is
the more important of the two. It is a crucial – per-
haps the crucial – aspect of the Kyoto Protocol
(Grubb, 1999), crystallizing the political compro-
mise at the Protocol’s heart, between the refusal of
developing countries to take on emissions caps
and the Clinton Administration’s conviction that
global emissions could be restrained far more
cheaply if the developing world were part of the
Kyoto regime. The CDM allows the creation of
Kyoto units from projects in developing countries
approved by the Executive Board of the CDM (a
body established under the United Nations Frame-
work Convention on Climate Change).
To gain approval, it must be shown that a project
is ‘additional’ (that it would not take place without
CDM funding) and that it will reduce emissions
below the ‘baseline’ level they would have been at
without the project. A developing world entity, or
industrialized-world government, corporation, bank
or hedge fund can then earn the di?erence between
emissions with and without the project in the form
of a speci?c type of Kyoto units: ‘Certi?ed Emission
Reductions’ (CERs). A CER is a credit, not a permit
or allowance: it does not directly convey any right to
emit. However, some governments are purchasing
CERs as a way of meeting their Kyoto caps, and cru-
cially CERs also have monetary value because the
European Union permits its member states to issue
allowances in the most important cap and trade mar-
ket, the European Union Emissions Trading Scheme
(ETS), in exchange for the surrender of CERs (Euro-
pean Parliament, Council, 2004). A credit earned in,
for example, China or India can thus be transformed
into a permit to emit in Europe.
As with the CDM, the ETS, launched in January
2005, was shaped by political exigencies.
4
What
pushed Europe towards trading rather than its initial
preference, harmonized carbon taxes, was in good
part an idiosyncratic feature of the political proce-
dures of the European Union. Tax measures require
unanimity: a single dissenting country can block
them. Emissions trading, in contrast, counts as an
environmental, not a tax matter. That takes it into
the terrain of ‘quali?ed majority voting’. No single
country can stop such a scheme: doing so takes a
coalition of countries su?ciently populous (since
voting weights roughly follow population) to form
a ‘blocking minority’. A plan for a Europe-wide car-
bon tax had foundered in the early 1990s in the face
of vehement opposition from industry and from par-
ticular member states (notably the UK), and its
advocates knew that if they tried to revive it the una-
nimity rule meant they were unlikely to succeed. ‘We
learned our lesson’, one of them told me in interview.
Hence the shift in allegiance to trading.
In terms of volume of transactions, the ETS is the
largest greenhouse-gas market. The scheme has had
its di?culties – the over-allocation, violent price
?uctuations and ‘windfall pro?ts’ discussed below
– but it saw trades worth $24 billion in 2006 (World
Bank, 2007, p. 3). The prospect of ‘monetizing’
CERs via the ETS is the main driver of investment
in Clean Development Mechanism projects, which
generated CERs worth $5 billion in 2006 (World
Bank, 2007, p. 3). Between them, the ETS and
CDM form the core of the world’s carbon markets,
and it is on them that this paper focuses.
Making things the same: Gases
The political decision to create a carbon market
such as the CDM or ETS is not the same as con-
structing such a market. A new commodity – an
emission allowance or emission credit – needs
brought into being: de?ned legally and technically,
allocated to market participants, made transferable
and tradable, and so on. To give a ?avour of what is
involved, let me concentrate on one issue: the heter-
ogeneity of the means by which the ‘sameness’ – the
fungibility of allowances and credits – necessary for
4
On the emergence of the ETS, see, e.g., Zapfel and Vainio
(2002), Christiansen and Wettestad (2003), Damro and Me´ndez
(2003), Wettestad (2005) and Cass (2005).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 443
a carbon market is brought into being.
5
Consider,
for example, two very di?erent sites: the central-area
combined heat and power plant of Edinburgh Uni-
versity, situated a couple of hundred meters from
my o?ce there, and the refrigerant plant operated
by Zhejiang Juhua Co., 6.5 km south of Quzhou
City in China’s Zhejiang province. How is the activ-
ity at one made commensurable with that at the
other, so that both can form part of the same mar-
ket, and CERs at Zhejiang Juhua’s plant can be
equivalent to the ETS allowances that a European
emitter such as Edinburgh University needs?
As its name indicates, the combined heat and
power plant in Edinburgh generates electricity (by
burning natural gas in a device that resembles a
giant car engine), and uses what would otherwise
be waste heat to warm up nearby buildings. Because
its thermal input capacity is slightly greater than the
20 MW threshold of the European Union Emissions
Trading Scheme, this plant became part of the ETS
in January 2008. (Most such installations have been
part of the scheme since its launch in 2005, but
Edinburgh University was exempted from the ?rst
phase because of its involvement in an earlier, vol-
untary UK emissions trading scheme.) CO
2
emis-
sions from the combined heat and power plant are
measured using a gas corrector meter (the interface
of which is shown in Fig. 1) on the large pipe that
takes gas from the national gas grid into the plant.
It is called a ‘corrector meter’ because it samples
temperature and pressure, and can thus convert vol-
umes into masses of gas input, which are in turn
converted to estimates of CO
2
output using stan-
dard multiplication factors.
Zhejiang Juhua Co. is involved in something
quite di?erent, the manufacture of HCFC-22 (chlo-
rodi?ouromethane), which is used mainly as a
refrigerant (especially in air conditioners), though
also as a foam blower and as a chemical feedstock.
The standard process used to produce chlorodi?ou-
romethane involves combining hydrogen ?uoride
and chloroform, using antimony pentachloride as
a catalyst, and even when optimized the process
leads to a degree of ‘over?uoridation’: tri?uorome-
thane, HFC-23, is produced as well.
6
HFC-23 is,
unfortunately, long-lived in the atmosphere and an
e?cient absorber of infrared radiation; the combi-
nation makes it a very potent greenhouse gas.
Fig. 1. The interface of the gas corrector meter in the input pipe to Edinburgh University’s central area combined heat and power plant.
Photograph courtesy David Somervell, Estates and Buildings, University of Edinburgh.
5
On commensuration in the SO
2
market, see Levin and
Espeland (2002).
6
‘HFC-23’ and ‘HCFC-22’ are not standard chemical formu-
lae, but instances of a code, widely used in the refrigerant
business, for identifying haloalkanes. The standard formula for
tri?uoromethane is CHF
3
, but ‘HFC-23’ is how it is referred to in
carbon markets.
444 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
Until recently, the HFC-23 was discharged into the
atmosphere. Now, the Zhejiang Juhua plant’s waste
gases are fed into a specialised incineration furnace
imported from Japan, in which they are mixed with
hydrogen, compressed air and steam, burned at
1200 °Cusing a high-intensity vortex burner, and thus
converted to hydrogen ?uoride, carbon dioxide and
hydrogen chloride. These products pass through a
quencher (in which they are cooled rapidly to mini-
mize the formation of dioxins), and the resultant acid
solution is either sold or disposed of via a facility for
treating ?uoric waste (CDM Executive Board, 2007).
As already noted, to gain approval it must be
shown that a Clean Development Mechanism project
reduces emissions below the ‘baseline’ level they would
have had in the absence of the project, which in many
cases is a tricky exercise in establishing a credible coun-
terfactual (Lohmann, 2005): for an introduction to the
issues involved, see Michaelowa (2005). In the case of
HFC-23 decomposition, however, a straightforward
argument has su?ced: without the decomposition pro-
cess, the HFC-23 would, as already noted, simply have
been discharged into the atmosphere (CDM Executive
Board, 2007). The amount actually decomposed then
needs measured, but in such a way that a connection
is kept to the baseline of the HFC-23 that would have
been emitted in the absence of the decomposition
incinerator. (The quantity of HFC-23 generated is
a?ected by the precise parameters of the HCFC-22
production process, and hence there is a need to reduce
the incentive to operate the process in an unoptimized
way and generate unnecessary HFC-23 in order to
earn credits by destroying it.) So to standard equip-
ment such as ?ow meters and a gas chromatograph
is added a regulation: for each tonne of HCFC-22 pro-
duced, there is a maximum mass of HFC-23 whose
decomposition can earn credits.
7
Crucially, the allowable mass of HFC-23 that the
measurement devices reveal has been decomposed is
then multiplied by 11,700.
8
By decomposing a tonne
of HFC-23 in China, one can – via the link between
the CDM and ETS – earn allowances to emit 11,700
tonnes of CO
2
in Europe. Certi?ed Emission Reduc-
tions are now a major income stream for China’s
refrigerant plants, and for the Chinese government
(which imposes a 65% tax on them, hypothecated
for environmental purposes). Indeed, HFC-23
decomposition is the biggest single sector of the
Clean Development Mechanism, accounting for
67% of the CERs generated in 2005 and 34% of those
generated in 2006 (World Bank, 2007, p. 27). Since
the price of CERs is likely to be a chief determinant
of the European carbon price – and thus, for exam-
ple, a major input into electricity prices – the e?ects
of the commensuration are considerable.
The crucial ?gure, 11,700, is the product of a cal-
culation of the ‘global warming potential’ (GWP) of
HFC-23 published by the Intergovernmental Panel
on Climate Change. Set up in 1988 by the World
Meteorological Organization and United Nations
Environmental Programme, the IPCC has as its
remit the establishment of authoritative scienti?c
knowledge about climate change (see Agrawala,
1998a, 1998b). As the IPCC put it in 1990, GWP
is ‘[a]n index... which allows the climate e?ects of
the emissions of greenhouse gases to be compared.
The GWP depends on the position and strength of
the absorption bands of the gas, its lifetime in the
atmosphere, its molecular weight and the time per-
iod over which the climate e?ects are of concern’
(Houghton, Jenkins, & Ephraums, 1990, p. 45).
Although very similar notions are to be found in
the scienti?c literature of the time (see, e.g., Lashof
& Ahuja, 1990), it was the IPCC itself that gave
‘global warming potential’ its canonical de?nition
GWP ¼
R
TP
O
a
x
½xðtÞ?dt
R
TP
O
a
r
½rðtÞ?dt
x designates the gas in question (e.g., HFC-23). a
x
is
an estimate of the e?ect on the radiation balance at
the tropopause (the boundary of the upper and lower
atmosphere) of an increase in the amount of gas in
the atmosphere, an e?ect measured in watts per
square meter per kilogram. x(t) is the mass of the
7
The mass of HCFC-22 produced (which is determined by
weighing shipping containers and storage tanks) determines the
‘eligible quantity’ of HFC-23: the quantity for the incineration of
which credits can be earned. For each tonne of HCFC-22
produced by the standard antimony pentachloride process, the
eligible quantity of HFC-23 is 0.0137 tonnes, corresponding to
the lowest recorded emission level from a process optimized to
minimize HFC-23 production (see McCulloch, 2005, p. 11). The
mass of gas fed into the incinerator is determined from the
readings of a ?ow meter, and the concentration of HFC-23 in it is
determined by gas chromatography of periodic samples. (A
correction for leakage is also applied.) The product of mass of gas
(in tonnes) and HFC-23 concentration, up to the maximum given
by the eligible quantity, is, as noted in the text, then multiplied by
11,700 to give the quantity of Certi?ed Emission Reductions
earned (SGS United Kingdom Ltd., 2007).
8
I am grateful to Thomas Grammig and to members of the
audience of a talk I gave at the University of Oxford for sparking
my interest in how gas equivalents are brought into being.
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 445
gas that will remain in the atmosphere at time t from
l kg released at time zero. TP is the overall time per-
iod in question: in the calculation in the HFC-23
commensuration, it is 100 years. The denominator
is the equivalent integral for the reference gas, CO
2
.
The expressions in this equation inscribe complex
processes. r(t), for example, is not (and obviously
could not be) determined by releasing a kilogram
of carbon dioxide and measuring what happens over
a century: it is a mathematical function generated
from a standard model (the Bern model: see, e.g.,
Siegenthaler & Joos, 1992) of the exchange of car-
bon between the atmosphere, the oceans and the ter-
restrial biosphere. a
x
and a
r
, likewise, are in part the
products of sophisticated spectroscopic studies,
recorded largely in a database managed by the Har-
vard–Smithsonian Center for Astrophysics. (The
database was originally a military project, designed
to enhance understanding of absorption of infrared
radiation with a view to improving the detection of
heat sources: see Taubes, 2004.) But a
x
and a
r
also
assume a scenario that is believed to be helpful in
predicting the climatic impact of a gas. In this sce-
nario, temperatures in the stratosphere, which are
understood as adjusting relatively quickly to such
perturbations, have done so, while temperatures in
the lower atmosphere and at the earth’s surface
(which adjust only slowly) have not.
9
Again, the sce-
nario cannot be observed empirically, so modelling
as well as spectroscopy is involved in the determina-
tion of a
x
and a
r
.
In 1990, the IPCCfelt able to o?er estimates of the
GWPs of only 19 gases, not including HFC-23, and it
labelled the ?gures ‘preliminary only’ (Houghton
et al., 1990, p. 59 & Table 2.8, p. 60). By 1995–1996,
the list had expanded to 26, and included HFC-23,
the GWP of which was estimated as 11,700 (Hough-
ton et al., 1996, table 2.9, p. 121). Both the notion of
‘global warming potential’ and the IPCC’s mid-1990s
estimates of GWPs were then inscribed into the
Kyoto Protocol, which laid down that they should
be used to translate emissions of other greenhouse
gases into their equivalents in CO
2
and that the
IPCC’s mid-1990s estimates should be used until
the end of the 2008–2012 commitment period.
10
The ‘exchange rate’ of 11,700 used to translate
HFC-23 into CO
2
is thus an example of ‘black-box-
ing’ in the sense of Callon and Latour (1981) and
MacKenzie (1990, p. 26). GWPs could be contested
in at least two senses. First, whether GWPs really
give the best estimates of the climatic e?ects of dif-
ferent gases could be and has been challenged (see
Shackley & Wynne, 1997, and also Shine, Fug-
lestvedt, Hailemariam, & Stuber, 2005, and the liter-
ature cited in the latter): for example, the choice of a
100-year time period is in a sense arbitrary, and very
di?erent GWPs can be generated if, for example, 25,
50 or 500 years is used.
11
Second, GWP estimates
were acknowledged to be subject to signi?cant
uncertainties, of the order of ±35% (Houghton
et al., 1996, p. 73, 119). By 2007, for example, the
consensus estimate of the global warming potential
of HFC-23 had increased from 11,700 to 14,800
(Intergovernmental Panel on Climate Change,
2007, p. 212). Neither of these two forms of chal-
lenge, however, has spilled over into the carbon
market. GWPs, with their apparent simplicity and
the black-box ‘possibility of use by policy-makers
with little further input from scientists’ (Shine
et al., 2005, p. 297) remain the way in which di?erent
gases are made commensurable, and the inscription
of the mid-1990s’ estimates of GWPs into the Kyoto
Protocol means that uncertainties and the changing
estimates of GWPs remain inside the black-box: a
matter for technical specialists, not carbon traders.
This black-boxing is crucial to allowing carbon
markets to encompass greenhouse gases other than
CO
2
: liquidity in such markets would be greatly
reduced if the relevant ‘exchange rate’ between
gases had to be negotiated ad hoc for each
9
‘The long-term forcing is. . .more accurately represented by
that acting after the stratosphere has returned to a state of global-
mean radiative equilibrium. Studies with simple models show that
the climate response, that is, the surface temperature change, is
proportional to the radiative forcing when the radiative forcing is
de?ned in this way. . . Importantly, the proportionality constant
is found to be the same for a wide range of forcing mechanisms’
(Pinnock, Hurley, Shine, Wallington, & Smyth, 1995, p. 23227).
10
See article 5, paragraph 3 of the Kyoto Protocol, the text of
which is available athttp://unfccc.int/resource/docs/convkp/
kpeng.html. Accessed 24.03.06.
11
Amongst other criticisms is ‘the fact that, despite its name, the
global warming potential does not purport to represent the
impact of gas emissions on temperature. The GWP uses the time-
integrated radiative forcing and this does not give a unique
indication of the e?ect of pulse emissions on temperature, because
of large di?erences in the time constants of the various
greenhouse gases. Although a strong greenhouse gas with a short
lifetime could have the same GWP as a weaker greenhouse gas
with a longer lifetime, identical (in mass terms) pulse emissions of
the two gases could cause a di?erent temperature change at a
given time. Economists have also criticised the GWP concept for
not being based on an analysis of damages caused by the
emissions’ (Shine et al., 2005, p. 282).
446 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
transaction. Note that the black-boxing rests upon
a ‘social’ factor: the authority of the Intergovern-
mental Panel on Climate Change. Although that
authority has been challenged by climate change
‘sceptics’ and ‘deniers’, public controversy has
focused on the reality, extent of and evidence for
anthropogenic climate change, not on matters of
‘detail’ such as GWPs, debate over which has
taken place only in much more limited circles.
The IPCC’s authority in such detailed matters is
thus an essential part of ‘making things the same’
in carbon markets, by keeping the ‘exchange rates’
between gases inside the black-box and separate
from political and economic disputes.
It is perfectly possible, however, that this black-
boxing may become harder in the future. At the
time of the Kyoto Protocol, it is unlikely that any-
one imagined that the ?gure of 11,700 for the global
warming potential of HFC-23 would determine a
?ow of funds of the order of $3.5 billion (the likely
total value of credits from HFC-23 decomposition
up to 2012: see Wara, 2007). As negotiations begin
over a successor to Kyoto, however, the ?nancial
consequences of such ?gures can now be seen. It is
possible that GWPs will remain in practice unchal-
lenged – it would be very hard, given the diversity
of economic interests involved, to get agreement
on a measure other than GWPs, or on anything
other than the IPCC’s estimates of them (which
are a ‘focal point’ in game-theoretic terms), so no
party to the negotiations may attempt to do so –
but it is not a foregone conclusion.
Making things the same: ‘Emission rights’
Gases are thus made the same by a combination
of measurement devices, complex natural science,
and the capacity (at least so far) of the Intergovern-
mental Panel on Climate Change to keep the estima-
tion of global warming potentials bracketed o?
from carbon-market politics. But practices of many
other kinds are also needed to make ‘carbon’ fungi-
ble, and amongst these accounting is of particular
importance.
12
The European allowances that Edin-
burgh University needs to emit carbon dioxide
and the CERs generated by Zhejiang Juhua Co.
are items that Europe’s (or indeed China’s) accoun-
tants have not previously encountered. What kind
of items are they? What accounting treatment
should they receive? These questions are signi?cant
for the operation of carbon markets, since account-
ing makes economic items visible, and whether and
how it does so is consequential.
13
Hatherly et al. (submitted for publication) argue
that a ‘?nitist’ perspective is useful for the analysis
of accounting, especially of accounting classi?ca-
tion, and it is particularly appropriate here. In this
perspective, how to classify an item (not just an
accounting item, but an item of any kind) is always
implicitly a choice. Past classi?cations – which are
always ?nite in number, hence ‘?nitism’ – in?uence
present classi?cations by analogy (‘this item is like
previous items we classi?ed as X, so this should be
classi?ed as an X’), but do not determine them.
Of course, classi?cation often does not feel like a
choice. Classi?ers – bookkeepers, accountants, orni-
thologists, botanists, and so on – often, probably
normally, come across items that seem familiar
and simply ‘see’ them as an X (‘this is an X’, not
‘I am classifying this as an X’). Items that seem to
classi?ers to be unfamiliar are thus of particular
analytical interest, because they make implicit
choice explicit. Instead of relying on habit and rou-
tine, those involved have consciously and explicitly
to decide what classi?cation is appropriate, and
the debate that is often sparked can reveal the con-
tingencies that a?ect classi?cation.
In the run-up to the launch of the European
Union Emissions Trading Scheme, the International
Financial Reporting Interpretations Committee
(IFRIC), a subsidiary body of the International
Accounting Standards Board, discussed how to
apply accounting standards to the new items, which
it called ‘emission rights’, which were about to come
into being. What kind of items were they? For
example, were they indeed ‘rights’? The IFRIC con-
cluded that they were not: ‘an allowance itself does
12
I am deeply grateful to Allan Cook, who served as Co-
ordinator for the International Financial Reporting Interpreta-
tions Committee at the end of the period in question for his help
in the research underpinning this section. Cook (forthcoming) is
his own account of these events. For broader legal debate over
the nature of carbon credits and allowances, see, Wemaere and
Streck (2005).
13
The issue of devising appropriate frameworks for making
carbon emissions ‘visible’, for example in corporate accounts, has
received considerable attention: see, for example, the work of
Fred Wellington and his colleagues at the World Resources
Institute (such as Lash & Wellington, 2007) and The Prince’s
Charities (2007). How to account for emissions allowances,
however, has received much less attention: see Cook (forthcom-
ing) and Casamento (2005).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 447
not confer a right to emit. Rather it is the instru-
ment that must be delivered in order to settle the
obligation that arises from emissions’ (IFRIC,
2004, p. 19).
14
An allowance was, however, in the IFRIC’s view
clearly an asset. But what was its nature? Was it an
‘intangible asset’ – ‘An identi?able non-monetary
asset without physical substance’ (IASB, 2005, p.
2227) – and thus within the scope of International
Accounting Standard (IAS) 38? Or was it a ‘?nan-
cial instrument’ – a ‘contract that gives rise to both
a ?nancial asset of one entity and a ?nancial liability
or equity instrument of another entity’ (IASB, 2005,
p. 2219) – and thus within the scope of the standard
governing such instruments, IAS 39? Some of those
who commented on the IFRIC’s initial draft argued
that an allowance was indeed a ?nancial instrument,
but the IFRIC disagreed: though allowances ‘have
some features that are more commonly found in
?nancial assets than in intangible assets’ – such as
being ‘traded in a ready market’ – they were not
?nancial instruments (IFRIC, 2004, p. 21).
An allowance was thus, in the IFRIC’s view, an
intangible asset, and therefore governed by IAS
38. If governments issued allowances at less than
their market value (most have issued them free-of-
charge) the di?erence was, IFRIC decided, a ‘gov-
ernment grant’, and its accounting treatment should
therefore follow the relevant standard, IAS 20.
Emissions themselves – as noted, previously outside
an economic or accounting frame – now had to
come within it. The emissions of those governed
by cap-and-trade schemes should, said the IFRIC,
be treated as giving rise to liabilities that were ‘pro-
visions’ whose treatment should follow IAS 37
(IFRIC, 2004, p. 7).
The IFRIC’s conclusions – crystallised in IFRIC
Interpretation 3: Emission Rights, issued in Decem-
ber 2004, on the eve of the start of the European
Union Emissions Trading Scheme – thus made
‘emission rights’ the same by laying down a homo-
geneous approach to accounting for them, in which,
for example, an allowance received free by an indus-
trial company or bought by an investment bank
were both treated in the same way as intangible
assets. However, IFRIC 3 encountered strong oppo-
sition, with critics arguing that the relationship of
IFRIC 3 to the three relevant standards – IAS 20,
37 and 38 – would create accounting mismatches,
especially in the light of anticipated changes to
IAS 20, which if made will mean that non-repayable
government grants have to be recognized when they
are received (see Cook, forthcoming). For example,
the fair value of the allowances that a company
received free would have to be recognized immedi-
ately as income, while the costs of the corresponding
emissions would be recognized only gradually as
they accumulated.
Re?ecting the criticism of IFRIC 3, the Euro-
pean Financial Reporting Advisory Group told
the European Commission in June 2005 that the
interpretation ‘will not always result in economic
reality being re?ected’, and recommended that the
Commission not endorse it.
15
The following month,
the International Accounting Standards Board,
while defending IFRIC 3 as ‘an appropriate inter-
pretation’ of existing accounting standards,
acknowledged that it ‘creates unsatisfactory mea-
surement and reporting mismatches’ and withdrew
it.
16
There was, of course, a ‘bottom line’ issue under-
pinning the controversy surrounding IFRIC 3. Cor-
porations generally fear earnings volatility: there is
a widespread conviction that investors prefer earn-
ings that rise smoothly to those that ?uctuate, even
around the same underlying trend. IFRIC 3 threa-
tened to produce volatility that, in its critics’ eyes,
would be arti?cial. For example, the advantage,
for corporations, of classifying an ‘emission right’
14
‘It therefore follows that a participant in a cap and trade
scheme does not consume the economic bene?ts of an allowance
as a result of its emissions. Rather a participant realises the
bene?ts of that allowance by surrendering it to settle the
obligation that arises from producing emissions (or by selling it
to another entity). Therefore, the IFRIC observed that amorti-
sation, which is the systematic allocation of the cost of an asset to
re?ect the consumption of the economic bene?ts of that asset
over its useful life, is incompatible with the way the bene?ts of the
allowances are realised. Although the IFRIC agreed that this
observation pointed to precluding amortisation, it agreed with
those respondents who highlighted that in some cases such a
requirement could be inconsistent with the requirements of IAS
[International Accounting Standard] 38. The IFRIC therefore
decided not to proceed with its proposal... that allowances should
not be amortised. Nonetheless, for most allowances traded in an
active market, no amortisation will be required, because the
residual value will be the same as cost and hence the depreciable
amount will be zero.’ (IFRIC, 2004, pp. 22–23).
15
Letter from Stig Enevoldsen to Alexander Schaub, 6 May
2005. Available fromhttp://www.iasplus.com/interps/
ifric003.htm. Accessed 11.07.07.
16
International Accounting Standards Board, ‘IASB withdraws
IFRIC Interpretation on Emission Rights’, available from http://
www.iasplus.com/interps/ifric003.htm. Accessed 11.07.07.
448 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
as a ?nancial instrument would have been that it
would make available the ‘hedge accounting’ treat-
ment permitted under IAS 39. If allowances could
‘be treated as the hedging instrument of a forecast
transaction (ie future emissions)’ (IFRIC, 2004, p.
20), then allowances and the corresponding emis-
sions would o?set each other. If a company received
N free allowances, forecast emissions of N tonnes of
carbon dioxide, and emitted N tonnes, then its earn-
ings would at no point be a?ected. ‘Carbon’ would
thus remain invisible.
The withdrawal of IFRIC 3 means that it
remains permissible to treat carbon in this way: as
inside an economic frame, but in a sense invisibly
so, since no accounting recognition is needed if the
above conditions are met. A survey by Deloitte
(2007) found that some market participants were
doing just that. Others were in e?ect following
IFRIC 3, while others again were doing so partially,
treating the provision for the liability created by
emissions in a di?erent way.
17
The attempt to make
‘emission rights’ the same has, in this sense, so far
failed.
The partial invisibility of carbon also means that
the incorporation of the carbon price into the mar-
ket’s ‘calculative mechanisms’ (Callon & Muniesa,
2003) is only partial. Although it is impossible to
be certain, there is tentative evidence from my inter-
view data of e?ects of both the accounting visibility
of carbon in some ?rms and its invisibility in others.
Consider, for example, the e?ect of the European
Union Emissions Trading Scheme on electricity
prices. If allowances are distributed free, one might
na? ¨vely think that they should have no e?ect on the
price of electricity. If a generator is given enough
allowances to cover its emissions (most generators
have actually had to buy some allowances, but let
me set that aside), what it charges customers surely
shouldn’t change? An economist will quickly tell
you what’s wrong with that argument. As already
noted, there’s an opportunity cost involved. In a
‘perfect market’, a pro?t-maximizing ?rm will pro-
duce electricity only if the price it receives is greater
than what it can earn by not generating electricity
and selling its stocks of the required inputs: its coal,
its gas, and now its carbon allowances (Point Car-
bon, 2007, pp. 24–25). If its allowances can com-
mand a non-zero price, the price of electricity
must rise correspondingly.
According to an interviewee in the electricity
market, however, it has required accountants to give
force to this economists’ reasoning. The ‘na? ¨ve’ view
prevailed in the industry until explicit valuations of
allowances started to be made. The price e?ect
‘should’ have been manifest in forward contracts
covering supply from January 2005 (the start of
the ETS) onwards, but apparently it initially was
not.
18
The e?ect began in the UK only once January
2005 was reached, and analysis by the consultancy
Point Carbon (2007) suggests it was even slower to
appear on the Continent. (Once the e?ect began,
the result in the UK was, for example, an increase
in domestic electricity prices in 2005 of around
7%
19
– for example, about £20 on a £300 annual bill
– and it is increases of this kind that are the source
of the much-criticized ‘windfall pro?ts’ that electric-
ity generators have made from the Emissions Trad-
ing Scheme.)
Carbon has thus been ‘visible’ for some time in
the electricity sector. When, in contrast, carbon is
kept invisible in accounting terms e?ects of three
17
Deloitte (2007) does not estimate the relative prevalence of the
three forms of accounting treatment. Those in the third category
‘recognise a provision on the following bases: To the extent that
the entity holds a su?cient number of allowances, the provision
should be recognized based on the carrying value of those
allowances (i.e., the cost to the entity of extinguishing their
obligation). To the extent that the entity does not hold a su?cient
number of allowances, the provision should be recognized based
on the market value of emission rights required to cover the
shortfall; and the penalty that the entity will incur if it is unable to
obtain allowances to meet their obligations under the scheme,
and it is anticipated that the penalty will be incurred (note that
the obligation to deliver allowances must still be ful?lled).’
(Deloitte, 2007, p. 3)
18
This is an interviewee’s assertion. Unfortunately, I do not
have access to the price data needed to test it quantitatively.
19
Calculation by Karsten Neuho?, quoted on BBC Radio 4,
‘File on Four’, 5 June 2007. Controversy is growing across
Europe about these ‘windfall pro?ts’. In the UK, for example, the
energy regulator Ofgem has called for the windfall pro?ts of the
UK’s electricity generators in the 2008–2012 phase of the
Emissions Trading Scheme – which Ofgem estimates at £9 billion
– to be used to help customers in fuel poverty (Crooks, 2007). In
Germany, the Bundeskartellamt (Federal cartel o?ce) charged
electricity generator RWE with behaving illegally by incorporat-
ing in the price it charged industrial consumers the market value
of permits it had received free. The case was settled out of court in
September 2007, with RWE continuing to defend its pricing but
agreeing that in 2009–2012 it would hold annual auctions of
quantities of power almost equivalent to its annual sales to
German industry (46,000 GWh in total over the 4 years) and
transfer to the purchasers, free of charge, the corresponding
carbon allowances if it had received these at no cost (RWE AG,
2007).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 449
kinds can be anticipated. The ?rst, which is hypo-
thetical (I have no direct evidence on the point),
would be to undermine a major desired e?ect of
a carbon market: incentivizing even those compa-
nies which have ‘enough’ allowances to cut their
emissions so as to generate income by selling
allowances. For this e?ect to be realized, allow-
ances need to be seen as assets with potential
monetary value, not simply as means of complying
with regulatory requirements. The second, related
e?ect (of which there is some tentative evidence)
is to delay the sale of allowances by those who,
even without abatement, have more allowances
than they need. The sale of allowances – and also
lending allowances for short sale – means that
they can no longer be kept invisible. They must
be recognized in accounting terms, and, for exam-
ple, a tax liability may be crystallized. This disin-
centive may reinforce other reasons for not
selling, such as the fact that emission levels will
in general be known in advance only approxi-
mately and the lack of a culture of proprietary,
risk-taking trading in many industrial companies
(in contrast to electricity suppliers, which are
active traders) that would permit the sale of allow-
ances that probably – but not certainly – will not
be needed.
My interview data do not permit me to judge
the relative importance of the various reasons for
postponing the sale of allowances that are likely
to be surplus to requirements, but those intervie-
wees with whom I explored the topic all believed
delayed sale to be a real phenomenon. It has been
consequential because the complex process of set-
ting national allocations for the ?rst phase (Janu-
ary 2005–December 2007) of the European scheme
led to over-allocation of allowances. The extent of
over-allocation was, however, not clear initially,
and the failure of those who were ‘long’ allow-
ances to bring them to market led to a constric-
tion of supply, which helped market prices to
rise to €31/tonne (see Fig. 2). Curiously, when
the extent of over-allocation became clear in the
spring of 2006, prices – though plunging dramati-
cally – did not initially fully re?ect the fact that
allowances no longer had any signi?cant economic
value. It took several months for the market price
of a phase-one European allowance to fall close to
zero (only in 2007 did prices become in e?ect zero,
with allowances towards the end of the year cost-
ing less than €0.10/tonne). Interviewees suggested
that delayed sale by those who were ‘long’ allow-
ances accounts for this paradoxical behaviour of
the carbon price. Even though it was clear that
allowances were intrinsically close to worthless
(because, in aggregate, there were more of them
than would be needed), they still commanded a
price of several euros, because not enough were
brought to market.
20
The third – again hypothetical – e?ect of the
accounting invisibility of carbon may be to
strengthen the hand of managers whose interests
lie in protecting market share by not passing on to
customers the opportunity cost of allowances that
have been allocated free, even when passing on the
cost is pro?t-maximizing for their ?rms. The extent
to which ?rms pass on the opportunity cost is cru-
cial to the environmental e?ects of a cap-and-trade
market – if they pass it on, there is likely to be car-
bon ‘leakage’ from the scheme, as imports from out-
side its boundaries become more attractive – and
there is ?erce controversy over likely behaviour in
this respect. Economists tend to predict pro?t-max-
imization, cost pass-through and thus leakage, while
?rms themselves tend to argue that market share
will be protected and costs will not be passed
through, at least in full. Unfortunately, empirical
analysis of the Emissions Trading Scheme so far is
too limited to be con?dent how ?rms outside the
electricity sector have behaved in this respect: see
Carbon Trust (2008).
Fig. 2. Price history of allowances, phase I of European Union
Emissions Trading Scheme. Courtesy Point Carbon.
20
One interviewee, at a hedge fund, reported making a
considerable amount of money by taking a short position in
allowances in this period.
450 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
The politics of carbon markets
One could go deeper into the issue of fungibility,
of making things the same. A trade, for example, is
a legal transaction requiring documentation, and
with three bodies (the International Swaps and
Derivatives Association, the European Federation
of Energy Traders, and the International Emissions
Trading Association) competing in this sphere,
interviewees reported that it has taken orchestrated
action to reduce the di?erences to a level at which a
trade documented in one format can be regarded as
similar enough to one documented in another, for
example for one to be used to hedge the other.
There has also, for instance, been sharp criticism
from competitors of the e?orts by Barclays Capital,
a leading player in the carbon market, to standard-
ise CERs via its SCERFA (Standard CER Forward
Agreement). The competitors regard a SCERFA as
speci?c to Barclays, not as a ‘standard’ entity.
Instead, however, let me consider the question of
the attitude to be taken to carbon markets. There is
a great deal of suspicion of them, ranging from right-
wing distaste for emissions caps to leftwing hostility to
an extension of market relations. The e?orts at market
construction so far have led to some environmental
bene?ts – for example, because of HFC-23’s potency,
curbing emissions of it is very valuable – but also sig-
ni?cant problems. There has, for example, been only
modest abatement by Europe’s electricity producers
(the sharp rise in gas prices in 2005–2006 swamped
any carbon-price incentive to switch from coal to
gas), while the mechanism discussed above led them,
as noted, to make substantial windfall pro?ts.
Similarly, the large sums that can be earned by
decomposing HFC-23 also create substantial pro?ts,
because the costs of decomposition are modest. A
specialized incinerator of the kind needed costs
around $4–5 million to install and $20,000 a month
to run (McCulloch, 2005, p. 12). Even with China’s
65% tax, a large HCFC-22 plant can recoup the
installation cost in a few months and go on to earn
revenues of well over a million dollars a month.
There is debate over just how much the subsidy
increases HCFC-22 production: McCulloch (2005)
argues that because the cost of HCFC-22 is only a
small proportion of the costs of the products in
which it is used,
21
a reduction in the price of
HCFC-22 will not expand the market for it very
much. However, the de facto subsidy may slow the
replacement of HCFC-22 by more environmentally
friendly refrigerants. (HCFC-22 is an ozone deplet-
er, the use of which as a refrigerant will eventually
be phased out under the Montreal Protocol govern-
ing such substances, and it is also a greenhouse
agent, though not as potent as HFC-23.) Because
of fears of this kind, ‘new’ HCFC-22 production
(i.e., over and above 2000–2004 levels) is currently
not eligible for CDM credits, but the consequence
is that there is no economic incentive not simply to
discharge HFC-23 from such new production into
the atmosphere rather than decomposing it.
In the light of issues such as these, it is tempting
to conclude that carbon markets are inherently
?awed means of achieving abatement. As Callon
(1998) points out, constructing a market requires
an enormous degree of ‘cooling’: of knowledge, of
metrologies, of actors, of identities, of interests. In
a perceptive article, Lohmann (2005, pp. 211 and
229) applies Callon’s analysis to the carbon market
and essentially concludes that market construction
will indeed fail: ‘conditions are not cool enough
for the spadework for commercial relations’, and
‘an unstoppable fount of complexity’ has been
uncorked.
Indeed, much of what I have described is consis-
tent with a bleak, essentialized view of capitalism, as
inherently irresponsible and environmentally dam-
aging, rather than Callon and Latour’s more opti-
mistic perspective. Yet the conclusion that carbon
markets are inherently ?awed carries a risk. Aban-
donment of such markets might well mean no seri-
ous international abatement e?orts, rather than
abatement by other means. If the Emissions Trading
Scheme were abandoned, could the European
Union ?nd a viable alternative, and how long would
it take? The political viability of a harmonized car-
bon tax, the obvious other route, remains question-
able, because of the unanimity required.
Similarly, political constraints mean that if inter-
national agreement on a replacement for the Kyoto
Protocol can be reached, it is likely to include some-
thing similar to the Clean Development Mechanism.
The CDM is, as noted, a result of the need to secure
developing-country participation in abatement
e?orts in a context in which the developing world
was and is unwilling to take on caps: even caps post-
poned to a later date, given the risk that by then
many of the cheaper opportunities for abatement
might be exhausted. The reluctance is understandable,
21
An air conditioning unit retailing at $500–1000 needs less than
a kilogram of HCFC-22, which costs around $1–$2 (McCulloch,
2005, p. 7).
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 451
given the desire not to allow a problem caused by
the industrialized countries to serve as a brake upon
development, and it is likely to persist – even in a
context in which China, in particular, no longer ?ts
the traditional template of a developing country.
Abatement e?orts in the developing world are thus
likely to continue to require funding from the devel-
oped world. Of course, such funding could be
achieved by direct government aid – Wara (2007)
points out that HFC-23 decomposition could have
been achieved far more cheaply via this route than
via the CDM – but that again raises the question
of whether governments would in practice make
the requisite large transfers of resources.
22
To conclude that carbon markets must fail may
also be unduly pessimistic, in that it would miss
the extent to which carbon markets hitherto have
been experimental, in the case of phase 1 of the
European Union Emissions Trading Scheme, quite
explicitly so: interviewees involved in establishing
it reported the many compromises that had to be
made to get it up and running, such as the fact that
it was possible to challenge only the most egre-
giously over-generous national allocations of allow-
ances. While existing carbon markets
unquestionably have major ?aws, those ?aws are
increasingly becoming manifest, and ways of reme-
dying them are available. Thus, windfall pro?ts
within the European scheme could be eliminated
by moving from free allocation to full auctioning
(Dales’s original proposal), and there is now a real
possibility that this will happen from 2013 on, at
least in the electricity sector.
If carbon markets are here to stay, can they be
improved? One example of a successful intervention
is of particular interest from the viewpoint of this
paper, because it involves making things not the
same. NGOs, especially the World Wildlife Fund,
have sought to create a separate category of ‘gold
standard’ CERs, covering only renewable energy
and energy conservation projects, and excluding
industrial gas projects such as HFC-23 decomposi-
tion.
23
The gold standard is a form of cooling in
Callon’s sense (as with the CDM as a whole, there
is a formal methodology, automated tools, a role
for auditors, and so on), and there are ‘bottom-line’
e?ects. Although from the viewpoint of the Kyoto
Protocol or of monetizability via the European
Emissions Trading Scheme, an ordinary and a gold
standard CER are identical, my interviewees
reported that the market price of the latter is now
around 10–20% higher. (They suggest that the cause
of the higher price is that those who are buying
CERs not just for compliance but to achieve ‘car-
bon neutrality’ or other forms of o?setting fear rep-
utational risk if it is discovered that ‘neutrality’ is
being achieved via industrial gas projects such as
HFC-23.) ‘Multiple monies’ have emerged in the
carbon market, as a result of intervention by
activists.
The intervention by the World Wildlife Fund and
other NGOs was informal: it did not alter the for-
mal procedures of the CDM. However, NGOs are
also seeking to practise a politics of market design
in a more formal sense, seeking to alter rules and
procedures. That, indeed, is precisely the course of
action that Callon and Latour’s perspective implies.
If markets are plural – Callon’s best-known work is
titled The Laws of the Markets (Callon, 1998) – and
‘capitalism’ has no unalterable essence, then this
may indeed be productive.
Such e?orts are too recent and too limited to
know whether they will be successful. However, it
is worth noting that changes in market design of a
kind that seem potentially achievable could be con-
sequential. Take the underlying issue of a carbon
market versus a carbon tax. Many environmental
activists prefer the latter, as do some economists
such as Nordhaus (2007). Nordhaus argues that
the classic analysis by Weitzman (1974) of the con-
ditions that in?uence the relative e?ciency of ‘quan-
tity-based’ instruments (such as a cap-and-trade
scheme) and ‘price-based’ instruments (such as a
carbon tax) suggests, given the speci?c cost-bene?t
features of combating global warming, the superior
e?ciency of a carbon tax.
Yet carbon markets seem politically feasible,
even in the US; carbon taxes may not be, even in
Europe. Intriguingly, however, a cap-and-trade
market, with full auctioning rather than free alloca-
tion, can be equivalent to an optimally set tax. In
both, polluters pay, either by having to buy permits
or by paying the carbon tax. Indeed, under admit-
tedly ‘idealized conditions’ (Hepburn, 2006, p.
229) they pay the same amounts, and the environ-
mental outcomes are the same. Thus, if the relation-
ship between emission levels and the carbon price is
known with certainty, either a cap-and-trade market
22
For an intriguing suggestion of a means of achieving north–
south transfers at a su?cient level to make a signi?cant impact on
developing countries’ needs to adapt to climate change, see
Mu¨ ller and Hepburn (2006).
23
Seehttp://www.cdmgoldstandard.org. Accessed 17.01.08.
452 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
or a correctly set tax can achieve a required level of
abatement, and the necessary tax rate will be the
same as the allowance price. Of course, the relation-
ship between emission levels and the carbon price is
not known with certainty, and for that and other
reasons the full equivalence between tradable per-
mits and a tax does not pertain in the real world.
However, economists’ analyses suggest ways of
designing a carbon market that might make it and
a tax more closely equivalent in practice. These
include rules facilitating the ‘banking’ of permits
for future use and the ‘borrowing’ of permits from
future years, regulated perhaps by an adjustable
requirement for ?rms to hold a certain amount of
permits in reserve, analogous to the adjustable
reserves that banks are required to hold (Newell,
Pizer, & Zhang, 2005).
Precisely because of the similarity of auctioning
to a carbon tax, emissions markets seem almost
always initially to involve free allocation, because
this reduces lobbying against them and political
opposition. However, once markets are well-estab-
lished, as the European Union Emissions Trading
Scheme now is, shifting to auctioning may become
easier (especially now the ‘economic experiment’
of Phase I of the ETS has made publicly visible
the problems that free allocation leads to). For
example, in October 2007 Sweden announced that
it was ending free allocation of allowances to its
electricity and heat sectors.
24
Indeed, as noted
above, it seems increasingly likely that in the third
phase of the ETS, from 2013 onwards, auctioning
may be much more heavily employed, at least for
sectors such as electricity that cannot in practice
easily move production outside of the European
Union.
The e?ort to shift the ETS to auctioning is ‘pol-
itics’ of a classic, recognizable kind, involving gov-
ernments, the policy-makers of a supranational
body, nation-state representatives, ?erce industry
lobbying against auctioning, and so on. Not all
the politics of carbon markets, however, ?ts that
recognizable template. Neither the IPCC nor the
International Accounting Standards Board see
themselves as political bodies, and indeed it is of
particular importance that the former not be seen
as political, despite the e?orts of its critics to paint
it as such. Yet they are arguably locales of ‘sub-
politics’ in Beck’s sense: politics ‘outside and
beyond the representative institutions of the polit-
ical system of nation-states’ (Beck, 1996, p. 18; see
Holzer & Sørensen, 2003). For example, the
IFRIC and now the International Accounting
Standards Board (which is turning its attention
to emission rights) have to contend with pressure
that has had the e?ect of blocking e?orts to ‘make
things the same’ in carbon markets. In the case of
the IPCC, the key ‘subpolitical’ matter is, para-
doxically, preserving the boundary between ‘sci-
ence’ and ‘politics’, since that boundary is
precisely what is needed to facilitate political
action, because it matters that action can be seen
as based upon ‘sound science’.
The subpolitics of carbon markets may seem
esoteric, and it is certainly not simple, but it is
important. Clearly, such markets are only one tool
for combating global warming, and other tools are
also important: direct regulation, carbon taxes
(where these are feasible), greatly increased public
expenditure on research and development and on
necessary infrastructure (for example, the electric-
ity grid changes needed to make increased renew-
ables production more attractive economically),
the removal of the many subsidies for fossil-fuel
extraction and use, and so on (see, for example,
Lohmann, 2006; Prins & Rayner, 2007). Neverthe-
less, making carbon markets more e?ective is cru-
cial, and the esoteric nature of their subpolitics
means that researchers have a particularly salient
role to play in bringing to light matters of appar-
ent detail that in fact play critical roles in this
respect.
It is this author’s hope that this paper will
encourage the work of this kind that is so badly
needed. The existing and planned experiments in
changing capitalism’s bottom line are heteroge-
neous, widely di?used worldwide, and involve many
aspects – scienti?c, technological, political, account-
ing, sociological, anthropological, geographical –
beyond economics as narrowly conceived. The
experiments need ‘witnesses’ (Shapin & Scha?er,
1985), and those witnesses must be multiple: lay as
well as professional, from many countries, and if
they are academics from many disciplines.
25
Carbon
markets need to become part of a process of ‘social
learning’ (qv Williams, Stewart, & Slack, 2005), in
which institutions to mitigate climate change are
24
Announcement of Environment Minister Anders Calgren,
reported by news service Point Carbon (www.pointcarbon.com).
11.10.2007.
25
I owe this way of formulating the matter to Andrew Barry.
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 453
created, evaluated and reshaped.
26
Such multiple
witnessing and social learning needs to concern
not just the overall features of carbon markets,
but the crucial ‘nuts and bolts’ of their construction,
questions such as how di?erent carbon sources and
sinks are commensurated, how allowances are trea-
ted in accounting terms, and many other such mat-
ters that I have been unable to discuss for space
reasons. If this modest paper recruits others to take
part in this multiple witnessing and social learning,
then it will have achieved its goal.
References
Agrawala, S. (1998a). Context and early origins of the intergov-
ernmental panel on climate change. Climatic Change, 39,
605–620.
Agrawala, S. (1998b). Structural and process history of the
intergovernmental panel on climate change. Climatic Change,
39, 621–642.
Barnes, B., Bloor, D., & Henry, J. (1996). Scienti?c knowledge: A
sociological analysis. Athlone and Chicago, London: Chicago
University Press.
Beck, U. (1996). World risk society as cosmopolitan society?
Ecological questions in a framework of manufactured uncer-
tainties. Theory, Culture and Society, 13(4), 1–32.
Bumpus, A. G., & Liverman, D. M. (submitted for publication).
Accumulation by decarbonisation and the governance of
carbon o?sets. Economic Geography.
Burtraw, D., Evans, D. A., Krupnick, A., Palmer, K., & Toth, R.
(2005). Economics of pollution trading for SO
2
and NO
x
.
Annual Review of Environmental Resources, 30, 253–289.
Callon, M. (Ed.). (1998). The laws of the markets. Oxford:
Blackwell.
Callon, M. (2007). What does it mean to say that economics is
performative? In D. MacKenzie, F. Muniesa, & L. Siu (Eds.),
Do economists make markets? On the performativity of
economics (pp. 311–357). Princeton, NJ: Princeton University
Press.
Callon, M., & Latour, B. (1981). Unscrewing the big Leviathan:
How actors macro-structure reality and how sociologists help
them to do so. In K. K. Cetina & A. V. Cicourel (Eds.),
Advances in social theory and methodology: Toward an
integration of micro- and macro-sociologies (pp. 277–303).
Boston: Routledge and Kegan Paul.
Callon, M., & Latour, B. (1997). ‘Tu ne calculeras pas!’ Ou
comment syme´triser le don et le capital. Revue du MAUSS, 9/
1, 45–70.
Callon, M., & Muniesa, F. (2003). Les marche´s e´conomiques
comme dispositifs collectifs de calcul. Re´ seaux, 21(122),
189–233.
Carbon Trust (2008). EU ETS impacts on pro?tability and trade:
A sector by sector analysis. London: Carbon Trust.
Casamento, R. (2005). Accounting for and taxation of emission
allowances and credits. In D. Freestone & C. Streck (Eds.),
Legal aspects of implementing the Kyoto Protocol mechanisms:
Making Kyoto work (pp. 55–70). Oxford: Oxford University
Press.
Cass, L. (2005). Norm entrapment and preference change:
The evolution of the European Union position on
international emissions trading. Global Environmental Pol-
itics, 5(2), 38–60.
CDM Executive Board. (2007). Project 0868: No. 2 HFC-23
Decomposition Project of Zhejiang Juhua Co., Ltd., PR
China. . Accessed 11.06.07.
Christiansen, A. C., & Wettestad, J. (2003). The EU as a
frontrunner on greenhouse gas emissions trading: How did it
happen and will the EU succeed?. Climate Policy 3, 3–18.
Coase, R. H. (1960). The problem of social cost. Journal of Law
and Economics, 3, 1–44.
Cook, A. (forthcoming). Emission rights: From costless activity
to market operations. Accounting, Organizations and Society.
Crooks, (Ed.). (2007). Watchdog wants £9bn windfall electricity
pro?ts clawed back. Financial Times 17 January, p. 1.
Dales, J. H. (1968a). Land, water, and ownership. Canadian
Journal of Economics, 1, 791–804.
Dales, J. H. (1968b). Pollution, property and prices: An essay in
policy-making and economics. Toronto: University of Toronto
Press.
Damro, C., & Me´ndez, P. L. (2003). Emissions trading at Kyoto:
From EU resistance to Union innovation. Environmental
Politics, 12(2), 71–94.
Deloitte. (2007). Accounting for emission rights. . Accessed 31.07.07.
Ellerman, A. D., Schmalensee, R., Bailey, E. M., Joskow, P. L.,
& Montero, J.-P. (2000). Markets for clean air: The US Acid
Rain Program. Cambridge: Cambridge University Press.
European Parliament, Council. (2004). Directive 2004/101/EC of
the European Parliament and of the Council. O?cial Journal
of the European Union, 338, 13 November, 18–23.
Greimas, A. J. (1987). On meaning: Selected writings in semiotic
theory. London: Pinter.
Grubb, M. (1999). The Kyoto Protocol: A guide and assessment.
London: Royal Institute of International A?airs.
Hahn, R. W. (1989). Economic prescription for environmental
problems: How the patient followed the doctor’s orders.
Journal of Economic Perspectives, 3(2), 95–114.
Hatherly, D., Leung, D., & MacKenzie, D. (submitted for
publication). The ?nitist accountant: Classi?cations, rules and
the construction of pro?ts. In T. Pinch & R. Swedberg (Eds.),
Living in a material world: On the mutual constitution of
technology, economy, and society. Cambridge, MA: MIT
Press.
Hepburn, C. (2006). Regulation by prices, quantities, or both: A
review of instrument choice. Oxford Review of Economic
Policy, 22, 226–247.
Holzer, B., & Sørensen, M. P. (2003). Rethinking subpolitics:
Beyond the ‘Iron Cage’ of modern politics. Theory, Culture
and Society, 20(2), 79–102.
Houghton, J. T., Jenkins, G. J., & Ephraums, J. J. (Eds.). (1990).
Climate change: The IPCC scienti?c assessment. Cambridge:
Cambridge University Press.
26
There is of course a trade-o? between the need to evaluate and
improve a market’s design and the need for rules that are stable
over reasonably long time periods. The European Union’s trade-
o? seems reasonable – an explicitly experimental three-year initial
phase, then a ?ve-year second phase, followed by a third phase
that is likely to last eight years (2013-20).
454 D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455
Houghton, J. T., Meira Filho, L. G., Callander, B. A., Harris, N.,
Kattenberg, A., & Maskell, K. (Eds.). (1996). Climate change,
1995: The science of climate change. Cambridge: Cambridge
University Press.
IASB (International Accounting Standards Board) (2005). Inter-
national Financial Reporting Standards (IFRSs) 2005. Lon-
don: IASB.
IFRIC (International Financial Reporting Interpretations Com-
mittee) (2004). IFRIC interpretation 3: Emission rights.
London: International Accounting Standards Board.
Intergovernmental Panel on Climate Change. (2007). Climate
change 2007: The physical science basis. . Accessed 31.08.07.
Lash, J., & Wellington, F. (2007). Competitive advantage on a
warming planet. Harvard Business Review, 85(3), 95–102.
Lashof, D. A., & Ahuja, D. R. (1990). Relative contributions of
greenhouse gas emissions to global warming. Nature, 344(5),
529–531.
Latour, B. (2005). Reassembling the social: An introduction to
actor-network theory. Oxford: Oxford University Press.
Latour, B. (submitted for publication). Re´sume´ d’une enqueˆte sur
les modes d’existence, ou Bref e´loge de la civilisation qui vient.
Unpublished typescript.
Lecocq, F., & Ambrosi, P. (2007). The clean development
mechanism: History, status, and prospects. Review of Envi-
ronmental Economics and Policy, 1, 134–151.
Levin, P., & Espeland, W. N. (2002). Pollution futures: Commen-
suration, commodi?cation, and the market for air. In A. J.
Ho?man & M. J. Ventresca (Eds.), Organizations, policy, and
the natural environment: Institutional and strategic perspectives
(pp. 119–147). Stanford, CA: Stanford University Press.
Lohmann, L. (2005). Marketing and making carbon dumps:
Commodi?cation, calculation and counterfactuals in climate
change mitigation. Science as Culture, 14, 203–235.
Lohmann, L. (2006). Carbon trading: A critical conversation on
climate change, privatisation and power. Development Dia-
logue, 48(September), 4–359.
MacKenzie, D. (1990). Inventing accuracy: A historical sociology
of nuclear missile guidance. Cambridge, MA: MIT Press.
MacKenzie, D. (2007). The political economy of carbon trading.
London Review of Books (pp. 29–31). UK: Nicholas Spice.
MacKenzie, D. (submitted for publication). Material markets:
Facts, technologies, politics. Oxford: Oxford University Press.
McCulloch, A. (2005). Incineration of HFC-23 waste streams for
abatement of emisisons from HCFC-22 production: A review
of scienti?c, technical and economic aspects. Comberbach,
England: Marbury Technical Consulting. . Accessed
11.10.07.
Michaelowa, A. (2005). Determination of baselines and addi-
tionality for the CDM: A crucial element of credibility of the
climate regime. In F. Yamin (Ed.), Climate change and carbon
markets: A handbook of emission reduction mechanisms
(pp. 289–304). London: Earthscan.
Mu¨ ller, B., & Hepburn, C. (2006). IATAL: An outline proposal for
an international air travel adaptation levy. Oxford: Oxford
Institute for Energy Studies.
Muniesa, F., & Callon, M. (2007). Economic experiments and the
construction of markets. In D. MacKenzie, F. Muniesa, & L.
Siu (Eds.), Do economists make markets? On the performativity
of economics (pp. 163–189). Princeton, NJ: Princeton Univer-
sity Press.
Newell, R., Pizer, W., & Zhang, J. (2005). Managing permit
markets to stabilize prices. Environmental and Resource
Economics, 31, 133–157.
Nordhaus, W. D. (2007). To tax or not to tax: Alternative
approaches to slowing global warming. Review of Environ-
mental Economics and Policy, 1, 26–44.
Pinnock, S., Hurley, M. D., Shine, K. P., Wallington, T. J., &
Smyth, T. J. (1995). Radiative forcing of climate by hydro-
chloro?uorocarbons and hydro?uorocarbons. Journal of
Geophysical Research, 100, 23227–23238.
Point Carbon (2007). Carbon 2007: A new climate for carbon
trading. Oslo: Point Carbon.
Prins, G., & Rayner, S. (2007). Time to ditch Kyoto. Nature,
449(25), 973–975.
RWE AG. (2007). RWE erzielt Einigung mit Bundeskartellamt in
CO
2
-Verfahren. Essen: RWE AG, 27 September. . Accessed 28.09.07.
Schellnhuber, H. J. (Ed.). (2006). Avoiding dangerous climate
change. Cambridge: Cambridge University Press.
SGS United Kingdom Ltd. (2007). Veri?cation and certi?cation
report for HFC23 decomposition project of Zhejiang Juhua
Co., Ltd., PR China. . Accessed 11.05.07.
Shackley, S., & Wynne, B. (1997). Global warming potentials:
Ambiguity or precision as an aid to policy? Climate Research,
8(2), 89–106.
Shapin, S., & Scha?er, S. (1985). Leviathan and the air-pump:
Hobbes, Boyle, and the experimental life. Princeton, NJ:
Princeton University Press.
Shine, K. P., Fuglestvedt, J. S., Hailemariam, K., & Stuber, N.
(2005). Alternatives to the global warming potential for
comparing climate impacts of emissions of greenhouse gases.
Climatic Change, 68, 281–302.
Siegenthaler, U., & Joos, F. (1992). Use of a simple model for
studying oceanic tracer distributions and the global carbon
cycle. Tellus, 44B, 186–207.
Taubes, G. (2004). An interview with Dr. Laurence Rothman,
.
Accessed 31.05.07.
The Prince’s Charities (2007). Accounting for sustainability.
London: The Prince’s Charities.
Wara, M. (2007). Is the global carbon market working?. Nature
445(8), 595–596.
Weitzman, M. L. (1974). Prices vs. quantities. Review of
Economic Studies, 41, 477–491.
Wemaere, M., & Streck, C. (2005). Legal ownership and nature
of Kyoto units and EU allowances. In D. Freestone & C.
Streck (Eds.), Legal aspects of implementing the Kyoto
Protocol mechanisms: Making Kyoto work (pp. 35–53).
Oxford: Oxford University Press.
Wettestad, J. (2005). The making of the 2003 EU emissions
directive: An ultra-quick process due to entrepreneurial
pro?ciency?. Global Environmental Politics 5, 1–23.
Williams, R., Stewart, J., & Slack, R. (2005). Social learning in
technological innovation: Experimenting with information and
communication technologies. Elgar: Cheltenham.
World Bank (2007). State and trends of the carbon market, 2007.
Washington, DC: World Bank.
Zapfel, P., & Vainio, M. (2002). Pathways to European green-
house gas emissions trading history and misconceptions.
. Accessed 19.09.06.
D. MacKenzie / Accounting, Organizations and Society 34 (2009) 440–455 455
doc_341600365.pdf