Toward a different debate in environmental accounting: The cases of carbon and cost–benefi

Description
Many champions of environmental accounting suggest that calculating and internalizing ‘externalities’ is the solution to
environmental problems. Many critics of neoliberalism counter that the spread of market-like calculations into ‘non-market’
spheres, is, on the contrary, itself at the root of such problems. This article proposes setting aside this debate and
instead closely examining the concrete conflicts, contradictions and resistances engendered by environmental accounting
techniques and the perpetually incomplete efforts of accountants and their allies to overcome them. In particular, it
explores how cost–benefit analysis and the carbon accounting techniques required by the Kyoto Protocol, the European
Union Emissions Trading Scheme and other carbon trading mechanisms ‘frame’ new agents, spaces, relations and objects,
and what the consequences have been and are likely to be.

Toward a di?erent debate in environmental accounting:
The cases of carbon and cost–bene?t
Larry Lohmann
The Corner House, Station Road, Sturminster Newton, Dorset DT10 1YJ, UK
Abstract
Many champions of environmental accounting suggest that calculating and internalizing ‘externalities’ is the solution to
environmental problems. Many critics of neoliberalism counter that the spread of market-like calculations into ‘non-mar-
ket’ spheres, is, on the contrary, itself at the root of such problems. This article proposes setting aside this debate and
instead closely examining the concrete con?icts, contradictions and resistances engendered by environmental accounting
techniques and the perpetually incomplete e?orts of accountants and their allies to overcome them. In particular, it
explores how cost–bene?t analysis and the carbon accounting techniques required by the Kyoto Protocol, the European
Union Emissions Trading Scheme and other carbon trading mechanisms ‘frame’ new agents, spaces, relations and objects,
and what the consequences have been and are likely to be.
Ó 2008 Elsevier Ltd. All rights reserved.
‘A re-examination of accounting systems and
measurement protocols to include the environ-
ment in the routine, everyday calculations by
which our economy is governed comes about as
close as you can get to the heart of why we have
this crisis. . . Accounting systems are required to
hold routinely in mind factors that are deemed
to be important and signi?cant in weighing the
pros and cons of any decision. There has been
progress to reform and redesign the accounting
system. But not nearly enough.’ (Al Gore, Inde-
pendent (London), 7 July 2007).
‘. . .the source of environmental problems lies in
part in the spread of markets both in real geo-
graphical terms across the globe and through
the introduction of market mechanisms and
norms into spheres of life that previously have
been protected from markets. . . The neoclassical
project of attempting to cost all environmental
goods in monetary terms becomes an instance of
a larger expansion of market boundaries. The
proper response is to resist that expansion, be this
in the spirit of resistance to market society or
more modestly to maintain the proper boundaries
between spheres.’ (O’Neill, 2007, pp. 21–22):
In the statement above, Al Gore expresses what
has arguably become conventional policymaking
wisdom about the role of accounting in environmen-
tal sustainability. On this view, environmental crises
are ine?ciencies deriving from a failure to calculate
all social costs, internalize externalities, or design
enough properly-functioning markets. Even global
0361-3682/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aos.2008.03.002
E-mail address: [email protected]
Available online at www.sciencedirect.com
Accounting, Organizations and Society 34 (2009) 499–534
www.elsevier.com/locate/aos
warming, according to this perspective, is a mere
‘market failure’ (Stern, 2007, p. viii), correctable
through improved pricing and information ?ow.
Introduction
Environmental accounting is supposed to address
such crises in two ways. First, it makes environmen-
tal crises more visible to decision-makers, by classi-
fying them in a way that makes explicit pre-existing
equivalences or quanti?able relationships with com-
modities and other economic objects. Coding the
statement ‘we must pay more attention to the envi-
ronment’ as ‘we must calculate the value of the envi-
ronment’, it provides a ‘guide to analysis and a
language of debate’ (Porter, 1995, p. 86) that
enables decision makers to trade one thing o? for
another more con?dently by providing ‘a clearer
sense of the stakes’ (Sunstein, 2005, p. 129). Second,
environmental accounting helps transform environ-
mental objects into commercial ‘goods and services’
whose value can be ‘discovered’ in markets them-
selves. Trade itself becomes comparative valuation
and environmentalist action.
Like most received opinion, the view expressed by
Al Gore has attracted its share of standard critiques.
One of the most important is articulated in the epi-
graph from O’Neill. This is that the problem has
been mistaken for the solution. Environmental crises
are rooted not in inadequate costing, insu?cient
commodi?cation or incomplete accounting, ‘but in
the very spread of market mechanisms and norms’
into putative non-market spheres of society or nat-
ure (O’Neill, 2007, p. 21). Environmental account-
ing, on this view, does not reveal what has
previously been only implicit, but rather misrepre-
sents, and thereby endangers, a ‘‘‘free” unpriced
world of knowledge, the body and so on’ (O’Neill,
1997, p. 550). ‘Protection of our environment is best
served, not by bringing the environment into a surro-
gate version of the commercial world, but by its pro-
tection as a sphere outside the world of commodity
exchange and its norms’ (O’Neill, 1997, p. 550).
As often happens, received wisdom and standard
critique revolve around a shared metaphor. The met-
aphor in this case pictures an ‘economy’ as a territory
whose boundaries can be contracted or expanded by,
among other actions, delimiting or expanding
accounting practices (Dove, 1999, pp. 2–3). Alterna-
tively, environmental objects such as land or climatic
stability can be transported across a boundary into
the territory of an economy with the help of new
accounting and other technical and legal practices.
Received wisdom and standard critique di?er only
about what happens to these invariant objects when
they are shipped over the border. According to
received wisdom, they bene?t when calculation
reveals their intrinsic value or at least a useful ‘proxy’
thereof (Barnes, 2001, p. 88), leaving behind at most
only a ‘philosophical’ (Independent, 2007) residuum.
According to the standard critique, however, the
value of such objects is intrinsically or constitutively
incalculable, meaning that their survival itself can be
threatened when they are treated otherwise.
The metaphor of a territorial ‘economy’ in?u-
ences a great deal of both scholarly and popular dis-
course, inspiring important work on all sides. Its
in?uence can be detected in commonplace expres-
sions ranging from ‘in a world ruled by markets, a
market solution must be made to work for the envi-
ronment’ (Evangelista, 2007) to ‘our preachings and
sermons will be for naught if we don’t inscribe them
on tablets that markets can understand’ (Barnes,
2001, p. 88) to ‘the market economy is not a neutral
medium for conservation but rather a corrosive acid
bath which dissolves many conservation practices it
comes into contact with’ (Lohmann, 1991,p. 100).
However, like all metaphors, it opens one path of
inquiry only by obscuring others.
Critical sociologists and anthropologists of mar-
kets such as Callon (1998a, 1998b, 1999, 2005) and
Mitchell (2002) have recently proposed a fresh meta-
phor, that of ‘framing’, which, they suggest, helps
open newpaths of inquiry. Market exchange becomes
possible, they argue, only through a laborious and
ongoing process of construction of spaces for calcula-
tion and transaction, of accounting systems that
determine bothwhois accountable and howand what
to count and not to count, and of simpli?ed, uncon-
troversial owners, products and modes of ownership.
‘Agents and goods involved in calculations’, Callon
says, ‘must be disentangled and framed if calculations
are to be performed and completed’ (Callon, 1999, p.
186). Nineteenth-century Chicago grain futures, for
example, could emerge only when it became physi-
cally and socially possible to standardize grain and
commensurate its present and future incarnations,
transporting it in railroad cars and storing it in
steam-powered grain elevators instead of the sacks
that had previously entangled each bit of grain with
its grower until it reached its ?nal buyer (Cronon,
1991, pp. 97–147, cited in MacKenzie, 2009, 440–
455; Espeland & Stevens, 1998, p. 318). Similarly,
the automobile market is possible onlybecause buyers
500 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
and sellers alike take it for granted that it is the car
company that owns the product. Any possible claims
of partial ownership by workers or communities near
sources of raw materials are elided, along with other
potential nuisances such as medieval notions of ‘fair
price’. In addition, many costs associated with the
automobile sector – certain kinds of pollution, prob-
lems associated with forms of social organisation
dependent on personal mobility, and so on – are put
o? on the community as a whole.
The metaphor of framing is clearly indebted to the
insight of Polanyi (2002 [1944], p. 146) that ‘[t]he road
to the free market was opened and kept open by an
enormous increase in continuous, centrally-organized
and controlled interventionism’. But instead of pictur-
ing a ‘self-regulating’ market being ‘disembedded’ or
liberated from a larger social ground that previously
contained and constrained it (Polanyi, 2002, [1944],
p. 144), the metaphor pictures economies as ‘embed-
ded in economics’, their every aspect – property, mer-
chandise, actors, contract, product quality – not only
described, de?ned and measured but constituted, nur-
tured, ‘performed’ and transformed by a multitude of
practices of calculation and governmentality originat-
ing both in academia and ‘in the wild’ among eco-
nomic agents at large (Callon, 2005, p. 9). ‘Expert
knowledge’, in the words of Mitchell, ‘works to for-
mat social relations, never simply to report or picture
them’ (Mitchell, 2002, p. 118). One instance of such
expert knowledge is accounting procedures, which,
as Peter Miller argues, are ‘intrinsic to and constitu-
tive of social relations, rather than secondary and
derivative’ (Miller, 2001, p. 392).
Double-entry bookkeeping, for example, ‘was
devised to account for business transactions, but
once established, it altered these transactions by
changing the way businessmen interpreted and
understood them’ (Carruthers & Espeland, 1991, p.
36). It in?uenced the premises of decision making
rather than just being a tool for implementing them.
Similarly, if application of economic theory often
‘makes economic processes more like their depiction
by economics’ (MacKenzie, 2009, pp. 440–455), so
Homo economicus traits can often be fostered in
human beings merely through the commensuration
involved in simple accounting innovations. Fining
parents for showing up late to pick up their children
from school, for instance, can paradoxically incen-
tivize parental delinquency by replacing a moral
stigma with a ?nancial penalty (Gneezy & Rustichi-
ni, 2000). By the same token, making good behav-
iour a matter for ?nancial reward (as when people
are paid for giving blood) can discourage it (Titmuss,
1996 [1972]). Net present value calculations and
cost–bene?t analysis have a similar potential to
mould the behaviours that, ironically, they posit as
invariant, in ways that increase (or decrease) their
‘?t’ with theory (Anderson, 1993; Radin, 1996; cf.
Nussbaum, 2001, p. 195). As Plato understood, com-
mensuration is often a social change and an achieve-
ment, rather than a report on the status quo
(Nussbaum, 1986). ‘We see the world remade, not
the world we live in’, Nussbaum remarks (Nuss-
baum, 1997, p. 1200).
By the same token, just as census work helps cre-
ate categories such as ‘the Hispanic vote’ (Anderson,
1999, p. 43; Petersen, 1987, pp. 223–229) or ‘the
unemployed’ that become e?ective collective politi-
cal agencies subject to their own discipline (Espeland
& Stevens, 1998, p. 331), so too, accounting helps
produce agents and other entities. Thus the Kyoto
Protocol’s carbon monitoring system, which catego-
rizes emissions sources according to physical loca-
tion on national territories, helps ensure that
nation–states are treated as the agents of global
warming despite the fact that transnational entities
such as multinational corporations, international
?nancial institutions or social classes are, on some
views, equally plausible candidates. Similarly, while
the category of ‘water quality’ as used in accounting
for the costs and bene?ts of hydroelectric dams
comes into being through the rather ad hoc aggrega-
tion of attributes such as temperature, amount of dis-
solved solids, turbidity and pH (Espeland & Stevens,
1998, p. 317), it ultimately becomes an entity as ‘real’
in policy deliberations as any other. As will be argued
below, such globally-recognized objects as ‘certi?ed
emissions reductions’ likewise exist only by virtue
of a chain of disentangling, commensurating, simpli-
fying and boundary-drawing calculating practices.
Framing, unlike boundary-crossing, is a never-
ending process. Each act of framing, because it
‘mobilizes or concerns objects or beings endowed
with an irreducible autonomy’ (Callon, 1998a, p.
39), is also a source of what Callon (1998a, p. 39)
terms ‘over?owing’. There are ‘always relations
which defy framing’. The ‘constraints, understand-
ings and powers that frame the economic act. . .and
thus make the economy possible, at the same time
render it incomplete’ (Mitchell, 2002, p. 191). ‘t
is one and the same movement which causes calcula-
tive agencies to proliferate, while reinscribing them
into spaces of noncalculability’ (Callon, 1998a, p.
39). Only by the creation of over?ows and new
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 501
entanglements is framing even possible. Every
attempt to bring something ‘inside’ creates new ‘out-
sides’. Market agents and goods are always ‘bound-
ary objects’ (Star & Griesemer, 1989) which, while
partly resynthesized for a market, at the same time
maintain and continue to develop characteristics
relating to other contexts, as an actor plays a role
without ever becoming it. Individuals framed as the
mute, maximizing bundles of preferences of eco-
nomic theory, for instance, are constantly – fortu-
nately for the market – reasserting themselves as
persuasive negotiators with voices and relationships
(McCloskey, 1998, pp. 95–97). Similarly, money,
framed as a unitary solvent of social ties, is, in the
hands of its users, constantly fragmented into dis-
crete, incommensurate categories – a process that
turns out to be essential for accounting itself (Callon,
1998a; Zelizer, 1997). Indeed, framing institutions
themselves cannot be separated from what they
frame with any guarantee of stability. Frames for
market negotiation are themselves negotiable. On
close examination, the purported border of a market
is ‘not a line on a map, but a horizon that at every
point opens up into other territories’ (Mitchell,
2002, p. 292). Spaces of calculation and noncalcula-
tion cannot be walled o? in rigid, mutually-exclusive
spheres (cf. Walzer, 1983).
It follows that every attempt to identify and
frame over?ows themselves, or to internalize exter-
nalities, creates further over?ows or externalities.
What economic theory refers to as the category of
externalities is not incidental and residual, but cen-
tral and enduring. Full cost accounting is an ever-
receding mirage. Just as the successful corporation,
in the words of investment banker Robert Monks,
must always play the role of an ‘externalising
machine’ (quoted in Bakan, 2004, p. 70), so markets
themselves ‘would be impossible if people were
made to account for every cost’ (Mitchell, 2002, p.
290). Every market transaction must exclude ‘fea-
tures of the world that actors do not have to take
into account’, indirectly revealing ‘all the work that
has to be done, all the investments that have to be
made in order to make relations calculable in a net-
work’ (Callon, 1999, p. 188).
1
In a sense, projects
such as Al Gore’s or Nicholas Stern’s can never be
completed.
2
In extreme cases, or what Callon calls ‘hot’ situ-
ations, even negotiations aimed merely at identify-
ing over?ows are incomplete or unachievable,
interests are unstable, and the identity of actors is
unclear, making continued framing exercises impos-
sible or premature. Although ‘externalities are at the
centre of public debates’, conditions are not ‘cool’
to carry out the spadework needed to establish com-
mercial relations. In ‘cooler’ situations, processes
through which products, their owners, and the rules
of calculation are framed are often not only carried
out, but can be ‘black-boxed’ as well. In certain
commodity markets, for instance, it becomes possi-
ble to refer to the ‘e?ciency’ of economies of scale
while eliding the ‘ine?cient’ violence or legal action
that created the possibility of scale (e.g., large
reserves of land or labour) and the requisite large
demand in the ?rst place (Lohmann, 1995), even
Whiggishly assuming that such markets came about
because they were ‘e?cient’ ab ovo. Practical suc-
cesses in framing living, breathing approximations
of Homo economicus help make it plausible to ‘nat-
uralize’ the species’ traits, reading them back into
human nature. Similarly, practical successes in
framing land, forests or wild ?sh as ‘manageable
natural resources’ help make it possible to regard
trade in ‘ecosystem goods and services’ as uncontro-
versial, even inevitable (Holm, 2007). Nevertheless,
even when the stage-setting for acts of calculation
and exchange is fairly successful, the ‘black-box’ in
which the inevitable entanglements have been
1
While the Marxist tradition has stressed the priority of
exchange to commensuration, it has perhaps neglected exploring
the way categories created by new commensurations help make
possible new forms of exploitation and wealth, preferring instead
to follow out a more classical attempt to locate a supposedly
calculable ‘surplus value’.
2
What is ‘external’ to one framing is, in addition, often itself
the product of a previous framing. For example, the ‘intrinsic
value’ of a wetland or wood that environmental economists and
deep ecologists commonly speak of as resistant to accounting
procedures is arguably itself a residuum of the previous framing
of the wetland or wood as a fungible, calculable global
commodity and the reduction of its role in speci?c local
livelihoods; Raymond Williams famously remarked of the related
term ‘landscape’ that a ‘working country is hardly ever a
landscape. The very idea of landscape implies separation and
observation’ (Williams, 1973, p. 120). As Espeland and Stevens
(1998, p. 327) note, ‘‘The importance of incommensurable
categories. . .depends. . .on the relative status of their oppositional
form, commensuration. The extension of commensuration into
more spheres of life may make incommensurable categories more
meaningful, their defense more necessary. This extension may
produce paradoxical e?ects, as when pricing children in law,
labour and insurance shifted the terms of their value from
primarily economic to moral and emotional. Children became
priceless.”
502 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
stu?ed and concealed has a permanent potential to
become a ‘jack in the box’, complicating or disrupt-
ing hitherto smooth exchanges. Rural villagers and
other non-professionals often contest framing pro-
cesses in which they have not been enlisted, if they
damage their perceived interests, dismissing envi-
ronmental accountancy out of hand or even driving
technical experts from their localities by force. Aca-
demics with an outsider’s perspective and a willing-
ness to ‘regard our own world as a problem, a
proper site for ethnographic inquiry’ (Comaro? &
Comaro?, 1991, p. 6) meanwhile often busy them-
selves with excavating and opening the black-boxes.
Thus Espeland and Stevens (1998, p. 327) empha-
size the ‘largely invisible’ yet ‘vast resources, disci-
pline and organization’ that go into
commensuration, while Mitchell (2002, p. 299) tax-
onomizes the diverse and unanticipated results of
the ‘violence and theory’ that go into framing
‘Egypt’s economy’ – including military intervention
and the theory of economics itself. Even scholars
such as Desrosieres (1996, pp. 336–367), who are
more prone to emphasize the social value of keeping
the ‘black-boxes’ closed and the costs of opening
them, acknowledge, in e?ect that the dialectic of
framing and over?owing is continuous.
As Espeland and Stevens (1998, p. 317) empha-
size, commensuration can be understood as a sys-
tem for ‘absorbing uncertainties’. But where
accounting practices required for a new market
encounter complexities, uncertainties, nonlinearities
and indeterminacies that they cannot immediately
accommodate, they also often actively rework their
objects, whether they involve human or non-human
objects, to try to make them more ‘passive’ and trac-
table to the agencies of calculation. As Bowker and
Star (2005, p. 254) remark of classi?cation proce-
dures generally, ‘it is not a question of mapping a
pre-existing territory but of making the map and
the territory converge’. Commensuration in particu-
lar, as Espeland and Stevens note, ‘has the power to
transform what it measures’ (Espeland & Stevens,
1998, p. 334). In recent years, Scott (1999) has been
particularly energetic in documenting the mecha-
nisms of, and blowbacks from, ‘state simpli?cations’
of human and non-human structures alike, from
managed woodlands to village layouts.
In short, the metaphor of framing challenges the
picture the objects of accounting as stable, pre-exist-
ing and transportable across borders. Instead of
focusing on imagined pre-existing or intrinsic prop-
erties of environmental objects and agents, it
focuses on what produces and sustains the objects
and agents. Rather than picturing essentialized
objects moving across sharply-delineated bound-
aries between what is internal and external to an
economy, the framing metaphor sees objects con-
stantly being made and remade, and boundaries as
?uid or poorly-de?ned. Correspondingly, it sees
‘failure’ of quanti?cation as a matter of the social
problems connected with achieving commensura-
tion rather than ?owing from the intrinsic proper-
ties of objects (Radin, 1996).
This casts doubt on the common notion that
there is a monolithic entity called ‘the market econ-
omy’ or ‘capitalism’ that might someday expand to
annex everything that is outside it, or whose hege-
mony is so complete that any environmental solu-
tion must be cast in its mold. As Callon stresses,
‘speeches – optimistic as well as pessimistic – on
‘‘the inexorable growth of the marketplace” have
no foundation in fact. . .the market must be con-
stantly reformed and built up from scratch: it never
ceases to emerge and re-emerge in the course of long
and stormy negotiations in which the social sciences
have no choice but to participate’ (Callon, 1999, p.
266). But it also casts doubt on the idea that there
are things that are by their nature resistant to such
imaginary monoliths, or that criticism of one or
another type of incipient market presupposes essen-
tialism about markets. According to the new meta-
phor, it may not be fruitful to analyze
environmental protection as a matter of either inte-
gration into, or isolation from, market economies.
Rather, it suggests that closer attention be paid to
speci?c contexts and nets of practices.
One test of a metaphor in inquiry is how it fur-
thers concrete debate. Can restating current contro-
versies over environmental accounting in terms of
the metaphor of framing and over?owing rather
than the metaphor of territories, boundaries and
essentialized objects lead to new insights and resolu-
tions? This paper considers what implications the
analytical use of the metaphor might have for the
strategies of both defenders and critics of two of
the most ambitious attempts of the past half-century
to expand accounting’s scope in the service of envi-
ronmental sustainability. These are the e?ort to
evolve the accounting procedures required for a glo-
bal carbon market and the e?ort to institute work-
able and uncontroversial forms of cost–bene?t
analysis. The paper proposes that the metaphor of
framing and over?owing can help defenders and
critics of both these relentlessly contested projects
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 503
better engage with, rather than bypass, each other’s
concerns. The metaphor o?ers multiple challenges
to the old ?gure of territories and borders, putting
productive pressure on both defenders and critics
of the received wisdom expressed by Al Gore to
reformulate their positions.
Constructing marketable ‘emissions reductions’,
1976–2007
The past decade’s attempt to create a global car-
bon market as the centrepiece of o?cial e?orts to
address global warming has entailed and presup-
posed some of the most potentially far-reaching
innovations in accounting in modern times. These
innovations were pre?gured, like some others, by
the work of economist Coase (1960, 1988), one of
the ?rst to urge that pollution be ‘optimized’ by
being integrated into a market calculus. In a market
without transaction costs and with perfect informa-
tion, and inhabited by properly calculating, maxi-
mizing economic agents, Coase concluded,
pollution dumps, as one ‘factor of production’
among many, would automatically be bargained
into the hands of those who could produce the most
wealth from them (or best ‘improve’ them, to use
17th century terminology), and thus the greatest
good for society. Coase’s successors, such as the
economist Dales (1968), while continuing to empha-
sise the importance of allowing polluters formal
rights to pollute, suggested that states would be bet-
ter placed than an imaginary ‘perfect market’ to set
overall pollution levels. In this way, pollution trad-
ing became mainly a way of ?nding the most cost-
e?ective way for businesses to reach an emissions
goal that had been set beforehand. The principle
was simple. Facilities with high abatement costs
would buy pollution rights from facilities with lower
abatement costs, saving themselves money. Facili-
ties for whom reductions come cheaper could mean-
while make money by cutting pollution and selling
the unused pollution rights they were thus enabled
to stockpile. The system would reward both sellers
and buyers and result in reductions being made
where they were least expensive.
Early in the history of pollution trading, govern-
ments and private ?rms sought ways of injecting still
more inexpensive pollution permits, but of a di?erent
kind, into the market, to make meeting targets even
easier (Liro?, 1986). In 1976, the US Environmental
Protection Agency promulgated a policy allowing
major new pollution sources to be sited in locations
where standards were not being attained as long as
they obtained ‘o?set’ pollution credits generated
from other projects that saved or reduced emissions.
Similarly, some 20 years later, the US successfully
demanded that the Kyoto Protocol include mecha-
nisms (the Clean Development Mechanism and Joint
Implementation) that would o?er inexpensive extra
emissions permits for sale to industrialised countries
and corporations from special carbon-saving or car-
bon-sequestering projects – schemes that capture
and destroy greenhouse gases, put them out of
harm’s way in trees or underground reservoirs, use
fossil fuels more e?ciently, displace fossil fuel-?red
power generation and so on. Such credits would have
the e?ect of expanding the pollution ‘cap’ of the asso-
ciated ‘cap and trade’ scheme. One objective, again,
was to make target attainment cheaper for the pri-
vate sector by allowing large emitters with high sunk
costs in fossil fuel plant to delay costly structural
reinvestment, while also supporting export of cli-
mate-friendly technology from industrialised coun-
tries to the private sector in Southern countries. As
Karin Backstrand and Eva Lovbrand remark, ‘from
a critical international relations perspective the car-
bon o?set market epitomizes continued neoliberal
governance building on a capitalist compact between
business and government elites in industrialised and
developing countries’ (Backstrand & Lovbrand,
2006, p. 70). This is the sector of the carbon market
whose accounting procedures will be the particular
focus of this section of the paper.
In order to become tradable for emissions allow-
ances, ‘o?set’ credits had to be made equivalent to
emissions reductions. Just as present and future
wheat of varying provenances and locations had
to be made equivalent in the US Midwest in the
years following the American Civil War, so too, in
the 1970s and 1980s, various US authorities and
regulated corporations eager to build a pollution
o?set market tried to commensurate reducing pollu-
tion from industrial installations with, say, buying
up and scrapping old cars or halting production
or making material process substitutions elsewhere
(Driesen, 1998, 2003a, 2003b; Drury, Belliveau,
Kuhn, & Bansal, 1999; Liro?, 1986). Environmen-
tally, the experiment failed. For example, entrepre-
neurs sold credits for destroying cars that in fact
already had been abandoned, while states lured
industry by providing it with ‘o?sets’ created
through substitution processes that were already
occurring for non-environmental reasons (Drury
et al., 1999; Liro?, 1986, pp. 16, 117). Under one
504 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
California smog trading program, the Sacramento
Metropolitan Air Quality Management District
issued 5 tonnes/year of volatile organic compound
pollution credits created by the decommissioning
of B-52 bombers that had been based in the region.
The credits were bought by companies such as Intel,
Campbell’s Soup and Aerojet, who were able to
avoid installing pollution control equipment as a
result. However, the credits arguably functioned to
increase pollution above what it would have been
otherwise, because the bombers had been slated
for destruction anyway under the terms of the
START treaty. In that sense, the B-52s continued
to ‘pollute from the grave’ (Drury et al., 1999, p.
264). Such credits quickly earned the sobriquet ‘any-
way tonnes’, meaning that they represented actions
that would have happened anyway.
In the 1990s, such pollution o?set accounting pro-
cedures went global with the advent of carbon trad-
ing. Traders, economists, consultants, non-
government organizations and UN technocrats
began to collaborate to set up institutions creating
a hybrid emissions-permit commodity that mixed
carbon emissions allowances with credits generated
by greenhouse gas-saving projects. Economist and
trader Richard Sandor, who in the 1970s had helped
design and standardize ?nancial derivatives such as
interest rate futures at the Chicago Board of Trade,
was one prominent ?gure who, in the 1990s, optimis-
tically outlined a sequence of carbon market devel-
opment which would provide capital for ‘structural
change’ to tackle climate change. Sandor’s main
interest was in the development of informal spot
and forward markets, followed by exchanges, orga-
nized futures and options markets, and over-the-
counter trading. Together with most early players
in the carbon market, he apparently regarded the
foundational step of creating a uniform commodity
(together with legal instruments providing evidence
of ownership) as largely unproblematic except inso-
far as the standardization work involved might entail
considerable ‘transaction costs’. While such insouci-
ance might be expected, as Espeland and Stevens
argue, where the ‘complex technical feats’ involved
in such ‘highly elaborated modes of commensura-
tion’ are already built into ‘practical organization
of labour and resources’ (1998, pp. 318, 329), what
was remarkable about the carbon trading instance
was that the commensurating institutions on which
the existence of the commodity depended were just
beginning to be established. A more likely source
of the anticipatory ‘black-boxing’ practiced by San-
dor and others was the success of commensuration
in previous market-building e?orts, which may have
encouraged an image of intrinsic measurability that
has also been perennially evident among advocates
of cost–bene?t analysis (see below). However, most
of those who came to be responsible for framing
the market simply had their hand forced, however,
di?cult they may have regarded the task of commen-
suration. When, at the last minute of the Kyoto Pro-
tocol negotiations, the Clean Development Fund –
an essentially juridical system with ?nes for exceeded
emissions targets earmarked for green technology
for the South – was transformed under US pressure
into a Clean Development Mechanism, a trading
scheme, o?sets were made exchangeable with allow-
ances by ?at. It became necessary to accept, without
any discussion, that pollution permits granted to
industry in industrialised countries could be made
commensurable with Clean Development Mecha-
nism carbon credits generated by a gigantic variety
of projects in the global South. Indeed, by the time
of the 1997 Kyoto Protocol, well before most of
the necessary social and technological processes
had got under way, o?set credits generated by such
projects were already being referred to as ‘emissions
reductions’, as if they were not only already inter-
changeable with each other and with emissions
allowances, but were identical to them. On the basis
of this generally untested accounting assumption, a
page and a half of abstract text in the Protocol gave
rise, within ten years, to a bewilderingly diverse pipe-
line of carbon credit-generating projects numbering
over 5390 which would be capable, if implemented,
of licensing on the order of 4 billion tonnes of carbon
dioxide releases, or 50,000 megawatts of coal-?red
electricity generation over a period of 10 years, in
industrialised countries such as the UK or Japan.
These projects included, for example:
factories in Korea or India re?tted to capture or
destroy hydro?ourocarbons such as HFC-23 or
other powerful greenhouse gases such as nitrous
oxide;
projects taking methane from waste dumps in
South Africa, coal seams in China, pig farms in
Mexico or ?aring towers in Nigerian oil ?elds
and using it as a fuel for generating electricity;
hydroelectric dams in Guatemala or Brazil
‘replacing’ electricity generated by fossil fuels;
wind farms generating green electricity;
e?ciency projects distributing energy-frugal light
bulbs or rearranging tra?c signals;
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 505
biofuel plantations providing feedstocks ‘substi-
tuting’ for fossil fuels;
fuel switches; or
tree plantations.
The construction of institutions and careers
around a carbon market that by 2006 measured
over US$30 billion made it ever more di?cult to
question the coherence of the project of ‘making
such projects the same’ climatically, to adapt
Donald MacKenzie’s useful phrase (MacKenzie,
2009).
Commensuration, write Espeland and Stevens
(1998, p. 325), ‘makes possible precise comparisons
across vast cultural and geographical distances that
allow transactions fundamental to global markets.’
Yet while it ‘overcomes distance (by creating ties
between things where none before had existed)’,
commensuration also ‘imposes distance (by express-
ing value in. . .abstract, remote ways)’ (p. 324) and
implicitly denigrating ‘particularistic forms of
knowing’ in favour of the ‘rigourous methods of dis-
tant, if less informed, o?cials’ (p. 331). From the
start, carbon o?set markets were characterized by
huge ?gurative distances between traders’ concep-
tual, largely electronic universe of ‘abstract’, simpli-
?ed, fungible carbon credit numbers and the
universe of the ‘concrete’, diverse, particular, highly
complex, often obscure local projects that produced
them, together with the tangled chains of physical
and social relationships that connected them to
the ongoing history of the atmosphere. Those who
were acquainted with the one seldom had much
experience with the other. These ?gurative distances
re?ected, and were rooted in, the literal distances
between computers on desks in the urban o?ces
of carbon consultants, UN o?cials, bankers, hedge
fund managers and ministries on the one hand, and,
on the other, hydroelectric dam or wind farm sites
in less industrialised countries, together with the
social and technological micro-arenas in which
?ows of carbon dioxide and other GHG molecules
were imagined and negotiated by scientists and tech-
nicians. The Executive Board of the Kyoto Proto-
col’s Clean Development Mechanism (CDM),
negotiating with carbon entrepreneurs and private
consultants, strove to establish standardized, consis-
tent methodologies for calculating how much car-
bon had been ‘saved’ by various classes of project.
Carbon consultants and their employers lost no
time in taking advantage of CDM Project Design
Document (PDD) templates that allowed them to
economize on credit production through uniform
procedures that, while they entailed the production
of PDDs of invariably enormous length and techni-
cality, rigorously excluded discussion of issues such
as local regulatory politics, corporate reliability,
nonlinearity, economic uncertainty and climatolog-
ical unknowns. In India, for example, consultants
appear to have cut and pasted text about local con-
sultations from the PDD of one ?ourochemical pro-
ject to that of another hundreds of kilometres away,
with identical quotations appearing from villagers
and labour union leaders in each locale. The identity
of the wording in both documents was explained
away by suggesting that because the projects were
‘similar’, it stood to reason that the responses and
even names of local people would be as well (Indian
Express, 2005). This may seem merely a ?agrant
example of manipulation, but it is continuous with
the standard practice of expert ‘formatting’ of pub-
lic comments on proposed CDM projects. One
Indian social activist remarked on being confronted
with an o?cial UN form for submitting comments
on a CDM project that ‘the form for public input
is so full of technicalities there seems to be no space
for general comments’ (Lohmann, 2006, p. 194).
Such simpli?cations are of necessity the norm for
PDDs, and range from summary assumptions about
future currency ?uctuations (often needed in order
to justify claims that projects would not be pro?t-
able without carbon ?nance) to standard emissions
factors that generate scienti?cally questionable
‘equivalences’ between the climate forcing potential
of di?erent greenhouse gases such as HFC-23 and
carbon dioxide (see below). The distance between
numbers on computer screens, carbon projects in
remote rural areas and ongoing scienti?c debates
in research establishments became a resource for
black-boxing commensuration di?culties that in
many ways was key to market establishment.
3
3
The long-distance calculation that ‘carbon footprint’ account-
ing involves can also conceal a politics of historical power
imbalances. For example, in July 2007 Britain’s Guardian
newspaper reported that ‘Kenya’s Kikuyu farmers are preparing
for war with Britain. . . Behind the furore is the proposal by the
UK’s Soil Association to ban imports of organic produce from
poor countries like Kenya because of their ‘‘food miles” – the
carbon emitted by air transport. . . A ban would mean labelling
air-freighted products so that they e?ectively lost their organic
status. . . Such a move would destroy the livelihoods of tens of
thousands of smallholders across Africa’ (The Guardian, 15 July
2007).
506 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
Of course, just as the 19th-century Chicago grain
futures exchange relied on a corps of grain inspec-
tors to judge aggregated wheat shipments’ compli-
ance with new standards of assessment and
exchangeability, and thus connect a physical prod-
uct with a new system of categories, so too the
Kyoto Protocol carbon trading system – although
not the voluntary carbon market – early on set up
an inspectorate to assess the (less measurable and
‘physical’) carbon-credit ‘product’ and create and
calibrate instruments that might quantify it. Unlike
their counterparts in the grain futures market, how-
ever, carbon market-makers in the UN and else-
where felt that an independent inspectorate would
be impossible and pointless to insist on (Lohmann,
2006, p. 62). From the beginning, not only advisory
bodies such as the Intergovernmental Panel on Cli-
mate Change (Lohmann, 2001) but also regulatory
institutions such as the CDM Methodology Panel
and Executive Board were peopled by ?gures with
vested interests in lenient rule-setting, such as car-
bon consultants who stood to pro?t from high pro-
ject volume and o?cials from credit-buying
countries. ‘I don’t see us as police’, the chair of
the CDM Executive Board recently remarked
(Nicholls, 2007, p. S42). In the unregulated ‘volun-
tary’ markets for carbon credits, where buyers seek
credits for reasons other than legal compliance, atti-
tudes were even more relaxed. Laurent Segalen of
Lehmann Brothers expressed a wide consensus
when he a?rmed that ‘traders should be the ones
designing and determining the standards’ (Reklev,
2007, p. 27); the UK government was persuaded
to consider promulgating government standards
for the sector, which in 2006 had a worldwide
annual value of US$91 million, only in 2007 (Ham-
ilton, Bayon, Turner, & Higgins, 2007, p. 6). At the
same time, banks, brokers and buyers kept the pres-
sure on for still greater simpli?cation and streamlin-
ing of regulatory procedures for the compliance
sector. At the 2005 climate negotiations in Mon-
treal, Natsource head Jack Cogen reminded UN
o?cials – whom he regarded as still overly sensitive
to environmental and social questions – of the haz-
ards of trying to overload the bandwidth carrying
the price signal: ‘The carbon market doesn’t care
about sustainable development. All it cares about
is the carbon price’ (Lohmann, 2006, p. 296; see also
Olsen, 2007; Sutter & Parren˜ o, 2007). What few
whistleblowers emerged from this setting tended to
be industry ?gures concerned primarily about
whether persistent accounting ‘fraud’ and insu?-
cient self-regulation among developers, buyers, sell-
ers and the UN might ultimately cause the market
to self-destruct in a bubble similar to the Dutch
tulip bulb ?asco of the 17th century or the dotcom
craze of more recent times (Michaelowa, 2007, p.
34).
Some such tensions can be expected to arise
whenever a novel commodity is being created that
depends fundamentally on the development of new
accounting procedures. However, the framing of
an amalgam carbon commodity also faced many
entanglements and over?ows that were unfamiliar
to the bulk of 19th- and 20th-century trading sys-
tems, and that arguably were not susceptible to
treatment in a straightforward way by any amount
or degree of regulation. In large part, these entan-
glements and over?ows stemmed from encounters
involving innovation, path dependence, counterfac-
tual history, nonhalting calculations and radical
uncertainties and indeterminacies.
Accounting for innovation
Many parties to the emerging controversies over
carbon markets agreed that social or technological
innovation, particularly in industrialised societies,
was crucial to reducing the ?ow of fossil-origin car-
bon into the atmosphere. They di?ered, however, in
their views of the role of carbon markets and carbon
accounting in fostering innovation. Carbon market
proponents claimed that giving carbon a price
would give entrepreneurs hoping to sell emissions
permits (whether surplus allowances not needed to
cover their own emissions, or ‘o?set’ credits pro-
duced by carbon-saving special projects) incentives
to develop greener technologies. Carbon market
critics pointed out, however, that trading also pro-
vided incentives not to innovate among the biggest
carbon permit buyers, who tended to be concen-
trated in precisely the sectors in which climate-
friendly innovation was probably most important
(power generation, oil and gas, steel, cement, chem-
icals, and so forth) (Driesen, 2003a). Trading also
delayed the early start to structural reinvestment
and research and development that is usually seen
as a key precursor to change in such sectors, much
of whose capital plant may have a 30- to 50-year
lifetime (Stern, 2007). Such drags on innovation,
critics argued, outweighed innovation incentives
that might also be associated with trading (Driesen,
2003a, 2003c). They cited evidence from the US sul-
phur dioxide programme showing that innovation
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 507
was associated more with conventional regulation
than with trading (Taylor et al., 2005, p. 372). Other
critics pointed out, in addition, that innovation of
the type required for addressing global warming is
not historically associated with price, but with, for
example, attempts to capture market share or with
larger historical forces (Buck, 2006; Lovell, 2007).
Even if a stable global price emerged within ?ve to
ten years, it would be too modest to stimulate much
beside e?ciency gains, particularly in the absence of
a ‘signi?cant increase in publicly funded research
and development for clean energy technology and
changes to innovation policies’ (Prins & Rayner,
2007, pp. 973–974). Price instruments, it was noted,
may be ‘quite e?ective for introducing changes on
the margin, but there is little evidence of price incen-
tives inducing a fundamental transformation in the
economy or society’ (Banuri & Opschoor, 2007, p.
22). Carbon trading simply could not ‘deliver the
escape velocity required to get investment in techno-
logical innovation into orbit, in time’ (Prins & Ray-
ner, 2007, p. 974).
This debate had profound implications for car-
bon accounting, if only because of the ambitious
claims that had been made for it as a result of the
prior choice of hybrid carbon markets as the ‘magic
bullet’ of international climate policy. How much
carbon projects saved depended partly on their
e?ects on innovation. Yet no metrics were available
either to assess or to predict these e?ects. Insofar as
a carbon project spurred innovation in signi?cant
ways, its e?ects on carbon cycling were unquanti?-
able in the long term. Insofar as carbon crediting
militated against fundamental social and technolog-
ical changes in industrialised societies, each carbon
credit carried long-term carbon costs, accounting
for which was also beyond the current scope of
the discipline, and was recognized to be so, due both
to creativity’s unpredictability (Malloy, 2000) and
to its unquanti?able precursors and e?ects. No
CDM project validators or veri?ers were making
such calculations, nor thought it necessary to try,
in spite of the possible large carbon e?ects. To bor-
row a formulation of Mitchell’s (2002, p. 209),
excluding the e?ects on and of innovation from cal-
culations about how much carbon could be saved or
lost by particular projects was both ‘necessary and
impossible’. The need to disentangle carbon num-
bers from social history and con?ne them to, say,
molecule counts or numbers of patents, under the
assumption of a static technological structure,
meant that accounting could not achieve its objec-
tive. This tension faintly echoed the see-saw, 60-year
controversy over the use of discounting techniques
in management accounting and whether they were
realistic or led to suboptimal investment (Miller,
1999, p. 185). But the size of the potential over?ows
that resulted from the need to exclude innovation
from carbon accounting rendered the issues far
starker.
Accounting for path dependence and lock-in
Large-scale, di?cult-to-manage over?ows from
carbon accounting also stemmed from the centrality
of path dependence to the economics of climate
change mitigation. Ironically, just as carbon trading
was beginning to establish itself as a theoretical
approach, a number of economists were assembling
a body of theory that contested the assumption that
historical accidents and starting points were unim-
portant to economic outcomes, merely delivering
the economy, through a series of negative feed-
backs, to its inevitable equilibrium (Arthur, 1999,
p. 11). Starting points, positive feedbacks and multi-
ple equilibria were not marginal or negligible eco-
nomic phenomena, but often central, and nowhere
more so than in responses to climate change, where
‘locking in’ new social and technological patterns
was widely agreed to be crucial to overcoming a pre-
viously ‘locked-in’ fossil fuel dependence (Unruh,
2000, pp. 817–830). Work by economists such as
W. Brian Arthur suggested that in contexts in which
increasing returns were signi?cant, leaving research
and development of carbon reduction methods to
private ?rms incentivized by price – the supposed
wisdom of which was one of the premises of the car-
bon market – would not guarantee that the ‘?ttest
technology in the long run sense will be the one that
survives’ (Arthur, 1999, p. 27). Unlocking path-
dependent systems, as Prins and Rayner observed,
‘is usually initiated by quite unexpected factors
resistant to being accounted for in advance’ (2007,
p. 934).
This challenge to neoclassical assumptions from
within the economics profession proved to be
another implicit challenge to the stability of carbon
accounting. In Callonian terms, a carbon commod-
ity could be framed only by creating an important
over?ow the handling of which there existed no
accounting procedures for. Experts could not quan-
tify the role CDM projects had in foreclosing or
promoting structurally diverse long-term carbon
futures whose evolution was dependent less on price
508 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
than on the step changes associated with historical
accident or nonlinear positive feedbacks. The best
they could do was to quantify the role such projects
might have in making various arbitrarily chosen
‘business-as-usual’ pathways marginally more car-
bon-e?cient. The inappropriateness of a project of
commodity construction as a response to the cli-
mate problem had led to an internal incoherence
in the calculating procedures that the project
required.
Accounting for future politics
All carbon credit accounting, and thus carbon
accounting tout court in any market that attempts
to make allowances and credits fungible, relies on
expert assessment of counterfactual scenarios. The
credits generated by a greenhouse gas-saving project
built as a result of carbon ?nance are calculated by
subtracting the emissions of a universe with the pro-
ject from the emissions of a hypothetical ‘baseline’
or business-as-usual universe. Industrialised coun-
tries or corporations can then buy credits represent-
ing the emissions that are claimed to have been
saved over the ‘baseline’ in lieu of reducing their
own fossil fuel use.
To determine how much carbon a project is sav-
ing, and thus how many carbon credits it is allowed
to generate, carbon accountants thus must (in Cal-
lonian terms) disentangle the project from the ‘base-
line’ – that is, prove that it is feasible only by virtue
of carbon credit income. This is called proving that
the project is ‘additional’.
4
Carbon market actors
have by now made such attempts at disentangle-
ment thousands of times. Yet the controversy that
surrounds the resulting calculations mounts year
on year. The CDM Methodological Panel’s ‘Tool
for the Assessment and Demonstration of Addition-
ality’ provides one example. According to the tool,
project proponents must be able to prove that a pro-
ject would not be the ‘most economically or ?nan-
cially attractive’ investment among various
alternatives, or at least that ‘barriers’ exist that
would make it di?cult for the project to go forward
without carbon ?nance and that do not all apply to
all the alternatives. In addition, proponents should
be able to show that the project is not ‘common
practice’ in the region where it is being implemented
(UNFCCC, n.d., pp. 1–2). None of these criteria for
disentangling a project from its baseline, despite
having evolved through years of negotiation and
compromise, has shown much promise in stabilizing
the distinction against persistent and growing
criticism.
The ?nancial test, for instance, often compares
the internal rate of return (IRR) of a project with
and without carbon credit ?nance. But IRR ?gures
depend on the assumptions and calculation method
used. In addition, what counts as a viable IRR for
one actor may not appear so to another, even if
the number is strongly positive. A further problem
is that lenders agree openly that due to CDM credit
risks, they ‘do not lend to projects that are not good
investments on their own, without the CDM’
(Haya, 2007). Partly as a result, many carbon pro-
ject proponents, as one carbon banker, James Cam-
eron of Climate Change Capital, notes, ‘tell their
?nancial backers that the projects are going to make
lots of money’ at the same time they claim to CDM
o?cials ‘that they wouldn’t be ?nancially viable’
without carbon funds (Financial Times, 16 Febru-
ary 2005). Such deceptions are not particularly dif-
?cult (Michaelowa, 2007) and, as trading experts
had warned from the time the Kyoto Protocol was
promulgated, ‘every government and every com-
pany’ had an incentive to practice them as a way
of attracting top-up ?nance for projects that it was
already implementing (Grubb et al., 1999, p. 229).
Project proponents can claim that the data used
are commercial secrets, for example, or even forge
board minutes to claim that the CDM was consid-
ered in early stages of project development. The bar-
rier test has proved equally unsuccessful in
disentangling project from baseline. Any project
can be said to be hampered by barriers – it may
be in a remote region, it may face objections from
local o?cials, it may face unexpected supply prob-
lems requiring extra loans, and so forth. Whether
any of these would be decisive in the absence of car-
bon ?nance is resistant to veri?cation. The criterion
4
The Marrakech Accords of 2001 marked out three candidates
for project baselines without specifying which had to be chosen:
existing actual or historical emissions; emissions of an ‘econom-
ically attractive course of action, taking into account barriers to
investment; and ‘‘average emissions of similar projects under-
taken in the previous ?ve years, in similar. . .circumstances, and
whose performance is among the top 20% of their category’
(Michaelowa, 2005). By contrast, the o?set rule book of the
Chicago Climate Exchange (privileged to members, who are
required to pay US$5000 in membership fees) is reported to judge
‘reductions’ against calendar baselines, so that ‘if a member
company emits less carbon dioxide than it did a few years ago, it
gets to sell those reductions as o?sets – regardless of whether they
are the result of. . .declining sales, plant closures, routine main-
tenance, or, say, rising sea levels’ (Bright, 2008, p. 90).
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 509
that a project not be ‘common practice’ is similarly
open to a wide range of interpretation (Haya, 2007).
Unsurprisingly, carbon market experts have long
acknowledged that estimates of hypothetical ‘emis-
sions reductions’ for many projects can be expected
to di?er by hundreds of percent given small changes
in initial assumptions (Lazarus, 2003).
Even the most staunch supporters of the account-
ing procedures involved were commonly admitting
by 2007 that a quarter (Sutter & Parren˜ o, 2007), a
half (Schlup, 2005) or more of all CDM projects
were business as usual. One investigation of CDM
projects in India concluded that a third of the sam-
ple was not ‘additional’ (Channel 4, 2007). How-
ever, such assessments are likely to have
understated the problem. For example, the 402
hydroelectric projects in the CDM pipeline in China
were expected to contribute 5.1 gigawatts (GW) of
new generating capacity in 2007 alone, more than
half of the estimated 9 GW expected to come on line
that year. If all those projects were additional, as
required, it would follow that business as usual in
the Chinese hydropower sector had dropped by
65% from the 11.2 GW capacity that had come
on line in 2006. Yet a review of the PDDs for 70
of these projects by Barbara Haya of the University
of California found no evidence that China’s ability
to ?nance new dams without carbon credit income
had dwindled over the year. In addition, some
77% of the Chinese hydropower projects submitted
for CDM validation or currently registered are
expected to start generating credits within a year
of their validation comment period and 96% within
two years. Large hydroelectric projects typically
take 4–8 years to build (on top of several years of
project preparation). That suggests that the great
bulk of Chinese hydropower projects in the CDM
pipeline started construction prior to beginning the
CDM validation process. In Brazil, similarly, the
880 MW Campos Novos dam in Santa Catarina
state started construction in 2001 and began gener-
ating electricity in May 2007. Yet it applied for
CDM validation only in November 2007. While
the CDM Project Design Document describes in
length an unfriendly regulatory and investment
environment in which the project was commissioned
and built, it does not even attempt to argue that
Campos Novos required the CDM to go forward.
In Kenya, meanwhile, the Sondu Miriu Hydro-
power Project in Kenya had secured a loan in
1997 that enabled construction to begin in 1999,
and completion of the project appears su?ciently
assured that the Japanese Bank for International
Cooperation has signed an agreement to support
an additional plant that would extend Sondu Miriu
by using its outfall. Yet the project only entered the
CDM pipeline application process in 2007. The
additionality arguments in its PDD describe barriers
that hydropower dams face in general in Kenya, but
does not explain why Sondu Miriu requires CDM
registration to go forward (Haya, 2007). An investi-
gation by the UK Channel 4 Dispatches programme
in 2007 elicited admissions from managers of car-
bon ?nance-supported projects in Bulgaria and Brit-
ain that their schemes, too, would have been
instituted with or without carbon money (Channel
4, 2007; see also Central & Eastern Europe Bank-
watch, 2005; Van Vliet, Faaij, & Dieperink, 2003,
p. 154).
Because the net carbon e?ect of a successfully-
calculated o?set project whose credits license emis-
sions elsewhere is designed to be zero or thereabouts
(small margins of error are sometimes included in
the calculations), such ?ndings strongly suggest that
CDM projects are having a net negative e?ect on cli-
mate change mitigation. In many cases, in fact, such
projects are enabling increases in fossil emissions in
both industrialised nations and the global South
(Lohmann, 2006, p. 148). The threat of market col-
lapse implied by this failure to separate out project
and baseline has concerned market players as much
as market critics. So far, however, most market
experts, mindful of the 1997 stipulation in the
Kyoto Protocol that o?sets are identical to emis-
sions reductions, have maintained, at least publicly,
that the problems can be con?ned to inaccurate
technique, incompetence and fraud, and someday
reduced to tolerable levels through better methodol-
ogies and improved diligence and enforcement.
Thus Einar Telnes, a Det Norske Veritas executive
representing a forum of private ?rms that ‘validates’
and ‘veri?es’ carbon projects, expressed concern in
2006 that carbon accountants’ use of signi?cantly
di?erent ways of tallying up credits ‘could lead to
a lack of con?dence in the market as such. . . We
don’t want an Enron scandal where excess (CDM
carbon credits) are issued without the actual reduc-
tions taking place. . . It is crucial that those verifying
have the necessary knowledge. Many of them don’t’
(Point Carbon, 2006).
Many trading proponents have acknowledged
intermittently that the problem may go deeper and
may require a questioning of whether regulation
of a carbon credit commodity, or even the commod-
510 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
ity itself, is possible. Early on, trading expert
Michael Grubb stated unequivocally that ‘measur-
ing or even de?ning savings that are additional to
those that would have occurred in the absence of
emissions credits’, was an ‘impossibility’ (Grubb
et al., 1999, p. 138), a judgement shared by many
others (for example, Fischer, 2005, p. 1807). In
2006, Mark Trexler, a businessman with 15 years’
experience in the carbon trade, co-authored an arti-
cle that admitted that a resolution of the debate
over how to decide whether a project would have
happened anyway ‘seems as elusive as ever. . . There
is no technically ‘‘correct” answer. . . Never has so
much been said about a topic by so many, without
ever agreeing on a common vocabulary, and the
goals of the conversation’ (Trexler, Broekho?, &
Koslo?, 2006). However, analysts such as Grubb
typically retract such judgements within pages of
making them (Lohmann, 2001). Similarly, panellists
at conferences on carbon trading who admit in the
corridors that it is possible to show the additionality
of nearly any project often chastise colleagues for
saying so in public (Haya, 2007). The incipient mar-
ket is being provisionally stabilized partly by an
image of carbon accountancy whose veracity is pri-
vately doubted by many of its prominent
practitioners.
5
A second and more fundamental type of disen-
tanglement, also required to construct the concept
of veri?able ‘additionality’ or ‘nonadditionality’ –
and thus also required for carbon credit accounting
– is much less discussed among experts and policy-
makers. For accounting to be possible and carbon
credits to be saleable, each project must be framed
as generating a determinate number of credits. That
becomes possible only if the counterfactual scenario
of the ‘baseline’ world is framed as singular, that is,
separated out from a large number of other theoret-
ically possible without-project scenarios. Only then
can accountants quantify the emissions associated
with it. Without this framing, it is pointless to
attempt disentangling the project from the baseline,
and thus to try to distinguish additionality from
nonadditionality, or to speak of additionality at all.
Here again, there are strong incentives for data
manipulation. Buyers, sellers and intermediaries
alike have both motive and opportunity both to fab-
ricate conservative baselines and to exaggerate the
novelty of o?set projects: ‘the more conventional
the baseline, the more additional funds or credits. . .
can be recovered’ from a carbon project (Ott &
Sachs, 2000). But the deeper di?culty is that to dis-
entangle a single baseline necessitates framing the
political question of what would have happened
without projects as matter of technical prediction
in a deterministic system about which near-perfect
knowledge is in principle possible. ‘Social condition-
alities. . . that do not easily lend themselves to pre-
diction. . .(inter alia, socio-economic development,
demographic trends, future land use practices, inter-
national policy making)’ are reduced to ‘technical
and methodological uncertainties’ (Lovbrand,
2004, p. 451).
6
Project proponents, by contrast,
must be framed non-deterministically, as free deci-
sion-makers, if their carbon project initiatives are
to be seen as ‘making a di?erence’.
This attempt at disentanglement, while saddling
project validators with the unsustainable attribute
of being able to determine the future through tech-
nical means, also creates a political over?ow, pro-
voking understandable opposition among activists
with their own ideas of counterfactual possibilities
and their own desire to be counted as free deci-
sion-makers. One example of this over?ow can be
found in the actions of residents of an area of Minas
Gerais, Brazil, much of whose land a local com-
pany, Plantar, had been occupying to plant environ-
mentally destructive eucalyptus plantations to
produce charcoal to fuel its pig iron operations.
Backed by World Bank, Plantar had applied for
carbon credits on the ground that without its plan-
tations, the company would be forced to switch to
(less climate-friendly) coal as a fuel source. The res-
idents vociferously opposed the accounting proce-
dures involved:
5
As one analyst mordantly puts it, carbon credits are an
‘imaginary commodity created by deducting what you hope
happens from what you guess would have happened’ (Welch,
2007). One well-attended session of the International Emissions
Trading Association during the December 2007 UN climate
negotiations in Bali bore the title ‘Additionality: Never-Ending
Story or Workable Solution?’ (see also Lohmann, 2006, pp. 143–
152).
6
Such assumptions about predictability are also made, of
course, in cost–bene?t analysis (see below). For example, there
appeared in the 1990s a cost–bene?t manual for Britain’s
Department of Transportation evaluating road bene?ts accord-
ing to time savings for motorists, which forecasts the total
number of kilometers driven by di?erent kinds of vehicles, laid
out in four signi?cant ?gures for every year until 2025 (Lohmann,
1997).
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 511
‘The argument that producing pig iron from
charcoal is less bad than producing it from coal
is a sinister strategy. . . What about the emissions
that still happen in the pig iron industry, burning
charcoal? What we really need are investments in
clean energies that at the same time contribute to
the cultural, social and economic well-being of
local populations. . . We can never accept the
argument that one activity is less worse [sic] than
another one to justify the serious negative
impacts that Plantar and its activities have
caused. . . [W]e want to prevent these impacts
and construct a society with an economic policy
that includes every man and woman, preserving
and recovering our environment’ (FASE, 2003).
In a June 2004 letter to the CDM Executive
Board, some 143 local groups and individuals, after
insisting that ‘the claim that without carbon credits
Plantar. . .would have switched to coal as an energy
source is absurd’, went on to characterize the
accounting procedure as a ‘threat’: ‘It is comparable
to loggers demanding money, otherwise they will
cut down trees . . . [the CDM] should not be allowed
to be used by the tree plantation industry to help
?nance its unsustainable practices’ (Suptitz et al.,
2004).
7
In denying the plausibility of Plantar’s counter-
factual baseline (a switch to coal), project oppo-
nents need not be read as asserting that there was
a single ‘correct’ counterfactual alternative (that is,
that CDM accounting could have been plausibly
carried out but just happened to have been per-
formed incorrectly in this case). In context, they
are more reasonably construed as rea?rming the
political basis of such counterfactual claims. For
them, the translation of decision into prediction
was inextricable from environmental threats and
physical repression of alternative land uses. Carbon
accounting’s repression of knowledge of the plural-
ity of alternative futures was equated with an
attempt to repress popular participation in the tak-
ing of alternative decisions. Like other ‘anti-politics
machines’ (Ferguson, 1994) such as cost–bene?t
analysis, institutional e?orts at framing a carbon
o?set commodity may at ?rst seem to consolidate
expert power by allowing decision-makers to ‘neu-
tralise and hence legitimise politically charged deci-
sions’ undertaken in the absence of public debate
(Lovbrand, 2004, p. 451). But experiences such as
that of Plantar suggests that such a conclusion, by
overlooking the fact that this disentanglement is
radically incomplete, would be facile. As Sarewitz
(1996) observes, looking to science to resolve such
disputes about public choices is likely to be futile.
If there is no such thing as additionality or nonad-
ditionality, then further doubt must be cast on
claims that the carbon market will be a ‘fast track’
to climate change mitigation.
Nonhalting calculations
One perennial threat to the utility of certain
kinds of accounting is what might be called nonhalt-
ing calculations – ones that, by a process of self-iter-
ation, generate an unending series of signi?cantly
di?erent numbers instead of stopping at a single,
?nal ?gure. If the relevant calculations can literally
never be completed, the failure to frame a commod-
ity is likely to be complete: the over?ows created by
accounting have defeated its purpose.
8
In the form of moral hazard, the problem has
long been of at least theoretical relevance to the sta-
bility of, say, statistical tables used by ?re insurers.
Unless proper precautions are taken, a low predic-
tion of ?re incidence may itself a?ect the incidence
of ?res insofar as it provides incentives for policy-
holders to set their own ?res, which outcome of
course would necessitate a change in future predic-
tions, which could occasion further changes in the
frequency of ?res, and so on. In the insurance case
this can be handled by, for example, policing cus-
tomers taking out insurance to ensure that they con-
form as much as possible to their portrayal as
idealized ‘non-gaming’ agents in tables of risk statis-
tics, or by avoiding the calculation if such policing
cannot be carried out.
In carbon accounting, however (as in the contin-
gent valuation procedures used by cost–bene?t ana-
7
Recent moves by the World Bank and other UN agencies to
open up native forests to carbon accounting are similarly viewed
as providing an opening for governments to threaten to destroy
their forests if they are not granted carbon credits (Gri?ths,
2007).
8
This is analogous to what MacKenzie (2007, pp. 75–80) calls
‘counterperformativity’ in economics: the use of accounting
techniques makes accounting’s object less tractable to accounting
itself. In the case of accounting, counterperformativity is often
related to what Ian Hacking calls the ‘looping e?ect of human
kinds’, which occurs when ‘a causal understanding, if known by
those who are understood, can change their character, can change
the kind of person that they are. This can lead to a change in the
causal understanding itself’ (Hacking et al., 1995, p. 351).
512 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
lysts; see below), the problem is less tractable. First,
baseline accounting procedures set up perverse
incentives for credit seekers (including host govern-
ments, credit buyers and consultant validators seek-
ing future contracts) not only to postulate but also
to bring about ‘business as usual’ scenarios which
are the highest-emitting possible, in order to make
the proposed projects appear to be saving as much
carbon as possible (Wara, 2007). For example, in
many CDM host countries, the CDM is creating
incentives for emissions-related environmental laws
not to be enforced, since the greater the ‘baseline’
emissions, the greater the payo?s that can be
derived from CDM projects.
9
Logically, this is likely
to necessitate continual recalculation of the baseline
and continual alteration in the number of credits
calculated. CDM accounting, in other words, is
undermining its own stability.
Moreover, the accounting requirement that pro-
jects show that they are economically unviable with-
out carbon ?nance gives green technology
developers such as wind turbine manufacturers a
perverse incentive to keep their products just
slightly too expensive to be competitive with gas
or coal. It also encourages project developers to
pay high prices for the turbines (provided carbon
credit sales outweigh the bene?ts from putting a
competitive product on the market or paying lower
prices). This also entails continual recalculation of
the baseline, since the point at which wind is
expected to cease to be ‘additional’ will continually
recede (Austin, 2007, p. S46).
Finally, the material nature of the carbon
accounting discipline as well as the institutions that
practice it has so far ensured that carbon credits
?ow to well-?nanced, high-polluting operations
capable of hiring professional validators of counter-
factual scenarios, but not to non-professional actors
eager to preserve or extend low-emitting livelihood
practices or social movements actively working to
reduce use of fossil fuels. As Backstrand and Lov-
brand (2006, pp. 70–71) remark:
‘Rather than providing a multitude of bene?ts
for all involved, the CDM market appears to be
deeply embedded in global power structures that
e?ectively marginalize local actors from global
institutions and reproduce patterns of inequity.
In a similar vein the technocratic and administra-
tive control and veri?cation apparatus that has
been developed for CDM projects contradicts
the image of the international climate regime as
an expression of a less state-centric and just mar-
ket order. Instead of promoting active participa-
tion of nonstate actors on equal grounds, the
government-regulated and supervised seven-step
project cycle of the CDM privileges the manage-
rial perspective of Big Science and policy elites.’
What has been less noticed is that this privileging
has climatic as well as other social e?ects. In cases
such as that of Plantar or of CDM-supported
hydroelectric projects in India which are supplant-
ing low-emitting irrigation systems (Ghosh & Kill,
in press), carbon crediting is undermining existing
local, climate-friendly livelihoods. Hence this sector
of the carbon market is likely to generate large cli-
matic ‘opportunity costs’, again requiring, in princi-
ple, continual and dauntingly di?cult revisions in
accounting methodology to take into account the
e?ects on carbon ‘savings’.
Accounting for uncertainties, ignorance and
indeterminacies
As Callon, Mitchell and other writers have con-
sistently stressed, the interactions that must be con-
tained within what is framed as ‘the economy’ are
not merely those among human agents. Non-human
agencies and forces are also important. These non-
human elements are not as passive and tractable
as often assumed, instead often being akin to the
‘tricksters’ of folklore (Haraway, 1995). Among cli-
mate scientists this truth is recognized in the infor-
mal use of the term ‘monsters’ to designate
9
International carbon trading agencies are currently acting in a
way that blurs the distinction between price incentives and
prohibitions enshrined in legal codes, and thus between legality
and ‘e?ciency’, by normalizing the expectation that certain laws
will be obeyed only if it becomes possible to earn carbon credits
by doing so. In August 2007, the CDM Executive Board
published forms for the submission of applications for a new
type of carbon project called programmatic CDM or ‘pro-
grammes of activities’ (PoA). A PoA, it stated, could be
additional and thus acceptable as CDM even if a law already
existed that mandated the measures that the PoA would bring
about, if that law was not being ‘enforced as envisaged but rather
depend[ed] on the CDM to enforce it’, or if the PoA would ‘lead
to a greater level of enforcement of the existing mandatory
policy/regulation than would otherwise be the case’ (Figueres,
2007, p. S50). Oil companies have also applied for carbon credits
for not ?aring natural gas in Nigeria, a prohibition already
mandated by the environmental laws of that country. Just as
norms of commons regimes have historically been partly
supplanted in many places by prices, so too now are legal
safeguards (Lohmann, 2006, p. 148).
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 513
nonlinearities, uncertainties, indeterminacies, and in
general various unknowns (and unknowables) that
have large potential atmospheric consequences
(Pearce, 2006). In an unstable climate system, for
instance, runaway feedback e?ects triggered by
obscure factors such as the reduced capacity of
warming oceans to absorb carbon dioxide are capa-
ble of radically altering even such symbols of
unchangeability as the Indian monsoons. In the
past, climate change has often been characterized
by deterministic but unpredictable (or ‘chaotic’)
events and processes of extreme impact. These ren-
der problematic reliance on probabilistic bell curves
and conventional ‘risk management’, which assume
that individual variation averages out and no single
event is capable of changing overall trends.
10
One imperative of carbon credit accounting
(derived from both policy and economic theory) is
to reduce these ‘monsters’ to (or frame them as)
probabilities. Demand is strong for scientists to pro-
duce tidy clusters of ‘likeliest scenarios’ to feed into
economic or political models, complete with ‘prob-
abilities’ of, say, a 2 ° or 5 ° temperature rise by
2100. This is necessary for credit accounting involv-
ing future biotic sequestration (Lohmann, 2005,
2001), but also for accounting for carbon outcomes
more generally, and for cost–bene?t analysis of
action on climate change. For example, the Danish
statistician Bjørn Lomborg has calculated that the
cost of doing nothing about climate change would
be US$4.8 trillion (Monbiot, 2006), and the British
government’s Stern Report on Climate Change,
suggests that, depending on what discount rate is
chosen, each tonne of CO
2
causes social damage
worth ‘at least $85’ (Stern, 2007).
The Weberian drive to use such numbers to
tame chance, or to make an uncertain, complex,
nonlinear, largely unpredictable world amenable
to management and governance, goes ‘all the way
down’ into the technical work of climatologists
and UN-designated scienti?c panels. For example,
since the 1990s, scientists have been prevailed upon
to create a new climatic entity analogous to ‘water
quality’ as used in environmental accounting for
hydroelectric dams – namely, the ‘global warming
potential’ (GWP) of various greenhouse gases.
GWP is measured in carbon dioxide ‘equivalents’,
so that the climate-forcing power of greenhouse
gases such as nitrous oxide or methane can be
commensurated with that of carbon dioxide. In this
way, the decomposition of, say, the industrial gas
HFC-23 in CDM projects in refrigerant plants in
China can be commensurated and traded with car-
bon dioxide releases in Europe. In another example
of how climate economics is framed, immense
e?ort has been expended in trying to determine a
maximum ‘safe’ level of temperature rise (the by
now famous 2 °C ?gure) as well as the probabilities
that one or another course of action will keep tem-
peratures below that level. This framing arguably
follows the strictures of rational choice theory
more closely than those of atmospheric science. It
attempts to integrate di?erent types of value and
uncertainty as a prelude to evaluating alternative
outcomes based on probabilistic predictions about
their consequences.
Insofar as such actions are associated with the
imperatives of economic accounting for climate
change, they deserve Callon’s label of ‘disentangle-
ment’, and like other forms of disentanglement
they entail over?ows. For example, ?gures for
‘CO
2
equivalences’ emanating out of the Intergov-
ernmental Panel on Climate Change (IPCC), the
UN’s scienti?c climate advisory panel, are admitted
to be gross oversimpli?cations: the e?ects and life-
times of di?erent greenhouse gases in di?erent
parts of the atmosphere are so complex and multi-
ple that any straightforward equation is impossible.
The original GWP ?gure for HFC-23 of 11,700
molecules originally put forward by the IPCC in
1995–1996 was revised in 2007 to 14,800, and the
error band of this estimate is still an enormous plus
or minus 5000 (MacKenzie, 2009, pp. 440–455).
The practical e?ects of this new ?gure’s jumping
out of the ‘black box’ are considerable: HFC-23
destruction is the largest single credit earner in
the CDM, accounting for 67% of the credits gener-
ated in 2005 and 34% of those generated in 2006
(World Bank, 2007, p. 27). The attempt to base
climate change politics on ?ndings of likely ‘safe’
degrees of warming, similarly, drastically foreshort-
ens important scienti?c distinctions. In 2001, for
instance, a controversy erupted between scientists
themselves about the wisdom of presenting assess-
ments of climatic futures in terms of (subjective)
probabilities. Stephen Schneider of Stanford
University had argued that
10
In the 1920s, Frank Knight (1921) introduced a rough
distinction between risk and uncertainty that continues to be
neglected. Risk can be described probabilistically, but uncertainty
cannot be quanti?ed in the same way. Writers such as Harremoe¨s
et al. (2002) have drawn further important distinctions involving
ignorance and indeterminacy.
514 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
‘policy analysts needed probability estimates to
assess the seriousness of the implied impacts;
otherwise they would be left to work out the
implicit probability assignments for them-
selves. . .. a policymaker concerned with ‘avoid-
ing dangerous anthropogenic interference in the
climate system’ would propose stronger policies
and measures if there was a 39% chance of
exceeding the 3.5–7 °C warming ‘threshold’ than
if the ?gure was 23%.’
But, Schneider himself asked, ‘what do these ?g-
ures actually represent? Unless probabilities are
assigned to individual scenarios and Global Circula-
tion Model climate sensitivities, their joint distribu-
tion. . .will depend on the particular selection of
scenarios and models’ (Schneider, 2001, p. 18; see
also Hall, Fu, & Lawry, 2007; Pittock, Jones, &
Mitchell, 2001; Schneider, 2002). Other scientists
raised the question more strongly: ‘This condition
of deep uncertainty di?ers from many risk-manage-
ment problems, in that little solid information exists
to inform subjective probabilities for the long-term
economic, social and technological trends underly-
ing di?erent greenhouse-gas emission scenarios. It
is unlikely that scienti?c evidence will soon resolve
the assumptions about the socio-economic future
made by di?erent groups’ (Lempert & Schlesinger,
2001, p. 375; see also Hansen, 2007; Pielke & Sare-
witz, 2000; Sarewitz, 1996; Sarewitz & Pielke, 2007).
‘The concept of probabilities as used in natural sci-
ences should not be imposed on the social sciences’,
warned other scientists insisting on a frequentist
view. ‘There is a danger that Schneider’s position
might lead to a dismissal of uncertainty in favour
of spuriously constructed ‘‘expert” opinion’ (Gru-
bler & Nakicenovic, 2001, p. 15; see also Shackley,
Young, Parkinson, & Wynne, 1998, p. 176). By
the same token, the IPCC has generally voted to
suppress in its reports what is called ‘Type II’ cli-
mate change – the abrupt, messy, chaotic, surprising
kind that results from the crossing of hidden ‘tip-
ping points’. Instead, it tends to stress ‘Type I’ cli-
mate change, which follows smooth, well-behaved,
accounting-friendly, global temperature curves.
But this stance too is increasingly sparking scienti?c
criticism, as has the position adopted by the Stern
Report (Cole, 2007).
The tension between the imperatives of carbon
accounting and the need to accommodate conceptu-
ally the unknowables of future climate change is
paralleled in the contrast between di?erent senses
and contexts of the concept of ‘conservatism’ –
one used by carbon accountants and another by
many small farmers and indigenous peoples. When
faced with uncertainties and ignorance, carbon
accountants tend to hedge their calculations toward
the ‘conservative’ side by adding a margin of, say,
25%. Where probabilities cannot be quanti?ed,
however, the concept of accuracy/inaccuracy comes
into question and the appropriateness of such mar-
gins cannot be veri?ed, leading to inevitable over-
?ows. This becomes important especially with
respect to events and processes of extreme impact,
which, however unexpected, could overwhelm
almost any margin likely to be added. By contrast,
many small farmers and indigenous peoples, espe-
cially in the global South, tend in contexts in which
‘conservatism’ is important to value resilience and
‘safety ?rst’ practices over probabilistic calculations
of gain and loss or arbitrary, numerical ‘safety mar-
gins’ as ways of handling unknowns, as has been
noted by many scholars of rural society (Berger,
1979; Scott, 1976; Thompson, 1990). To a certain
extent this re?ects the rough contrast between a
resource or accumulation conception of livelihood
and one oriented around commons regimes and
community survival. The processes by which such
small farmers and indigenous peoples are likely to
frame their own responses are certain, of course,
to have their own over?ows, but the di?erences of
approach merit further study.
Attempts to measure and account for biotic car-
bon sequestration engendered over?ows of a di?er-
ent kind. At ?rst, forestry specialists imagined that
they could measure precisely the amount of carbon
a plantation project, say, was ?xing, and therefore,
the volume of pollution rights it could generate,
simply by doing periodic measurements of tree
growth, gas transfer in the canopy, and so forth.
But it quickly became clear that quantifying the cli-
matic impact of such projects would also necessitate
investigating their e?ect on soils’ carbon production
both inside plantation boundaries and downstream,
requiring the hiring of additional experts. At the
same time, sobering evidence emerged that error
bars in such relatively simple matters as forest
inventories and physical ?uxes of carbon into and
out of forests were so wide that they swamped the
signal required for the establishment of a biotic car-
bon market. Unknowns concerning the response of
soil biology and chemistry to global warming itself
also became a signi?cant consideration. Moreover,
in order to complete their calculations, accountants
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 515
realized, they would have to monitor the e?ects of
plantations on the human groups displaced or
otherwise a?ected. For example, communities
evicted by carbon plantations might clear forests
elsewhere, migrate to cities where they might adopt
lifestyles with a di?erent carbon budget, and so on.
Due to the persistence of greenhouse gases in the
atmosphere, the activities of such groups would
have to be monitored over a signi?cant time period
(between 42 and 150 years) whose length itself was a
contested issue (Marland et al., 2001, p. 259;
Dutschke, 2002, p. 381). In formulating a counter-
factual baseline for carbon production without the
project, accountants would also have to venture into
economic predictions about trade patterns involving
commodities produced on forest lands, such as soy,
as well as predictions about future currency
exchange rates. Not surprisingly, as such di?culties
mounted, the very concept of ‘project boundary’
became increasingly disputed within the community
of sequestration experts (Lohmann, 2001, pp. 36–
45) – and with it the concepts of ‘carbon o?set pro-
ject’ itself and the status of the experts who
attempted to frame it.
Constructing preferences, subjects, agencies and
community through cost–bene?t analysis, ca. 1934–
2007
While it also originates from the US, cost–bene?t
analysis (CBA) is of much longer standing than the
accounting techniques associated with carbon trad-
ing. The technique came into political prominence
before the Second World War largely as an attempt
by US engineering bureaucracies, particularly the
Army Corps of Engineers, to cope ‘objectively’ with
questions from a wide range of other interest groups
about the big water developments then being pio-
neered in the wake of the Depression and the great
Mississippi ?ood of 1927. In the US, the personal
judgements of the technocratic elite, based largely
on di?cult-to-communicate craft knowledge and
institutional interest, were not as insulated from
challenge by elected elites and other agencies as they
were in France or the UK. Cost–bene?t numbers
both promised a way of giving ‘evidence of fairness
in the selection of water projects’ (Porter, 1995, p.
149) and helped justify the rejection of projects the
Corps did not want to build. The di?culty was that
in the early 1930s, these numbers were sometimes
too narrow in scope – con?ned to tangible, rela-
tively local costs and bene?ts such as capital outlay
and ?ood damage or maintenance expenditures
avoided – to justify investments on a New Deal
scale. Thus in 1934, a National Resources Board
committee headed by Secretary of the Interior Har-
old Ickes, friendly to the idea of adding new projects
to the list of candidates for federal funding, recom-
mended what it called a ‘striking revision of costing
technique’. The idea, the Board explained, was to
‘include not only private but social accounting’.
‘Intangible factors’, it proposed, could be quanti?ed
and added to projected water project bene?ts
according to ‘a generalized formula’. The damage
a project did to homes and the resulting loss in
tax revenue, according to this idea, could be com-
mensurated with and balanced against not just irri-
gation or power production, but also increased
recreation opportunities, aesthetic improvements,
and other unmarketed factors. Two years later the
Flood Control Act of 1936 repeated these notions,
but also emphasized the potential of expanded
accounting techniques to check pork-barrel schemes
and control federal spending, warning that the gov-
ernment could participate in schemes only ‘if the
bene?ts to whomsoever they may accrue are in
excess of the estimated costs’ (Hammond, 1960, p.
5).
In order to be able to argue that an ‘impartially’-
determined public good would be served by a pro-
ject, proponents thus had to learn to assign numbers
to numerous indirect, ‘intangible’ and far-?ung ben-
e?ts and then, after subtracting or dividing by the
costs, present the resulting number to oversight
bodies. The Bureau of Reclamation once credited
a dam it wanted to approve not only with the value
of the wheat grown on the land to be irrigated, but
also the gross value of the bread that might be
baked from the wheat, as well as increased atten-
dance at local cinemas. To justify a development
on the West Coast, an agency might need to quan-
tify its possible e?ects on the state of Maine, or esti-
mate average prices for goods expected over a
project’s entire 50- to 100-year life (Krutilla & Eck-
stein, 1958, pp. 199–264; Hammond, 1960, pp. 22–
23). As with carbon o?set calculations, CBAs per-
formed by di?erent teams for the same project could
yield staggeringly di?erent results – a problem that
was destined to plague the technique throughout
the next 70 years.
Unsurprisingly, in the 1940s, amid e?orts to rein
in spending, the Corps of Engineers’ cost–bene?t
numbers came increasingly under ?re from compet-
ing railroad companies, utilities and other govern-
516 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
ment departments, and di?erences among CBA
techniques became an embarrassment to the govern-
ment. Yet an attempt by the Federal Inter-Agency
River Basin Committee and the Bureau of the Bud-
get to harmonize them only made things worse. The
more explicitly each new ‘universal’ basis for CBA
was set out, the less credible it became. The River
Basin Committee’s report of 1950, for example, rec-
ommended that the opportunity cost of a good
always be set equal to that good’s market price. This
was criticized not only for equating the market price
of a good to the value people currently placed on it,
but also for attributing to the market the ability to
prophesy future uses and valuation of the good.
Yet the alternative was to fall back on what econo-
mist Hammond termed ‘imaginary’ data. In 1955,
the Second Hoover Commission advised backing
o? from the quanti?cation of intangibles altogether.
‘Verbal discussion of the intangible bene?ts and
costs will communicate the facts to Congress more
clearly than invalid bene?t estimates’, concluded
another critic. ‘Relevant ?gures may be submitted
without forcing them into the bene?t–cost frame-
work’ (Krutilla & Eckstein, 1958, p. 41).
But CBA had become too well-entrenched to
retreat, even if other supposedly ‘impersonal’ prac-
tices had been ready to take its place. Growing
demand for techniques to commensurate unmarket-
ed goods was creating the conditions for a more
professionalized, identi?able CBA community. The
Corps of Engineers started hiring more economists,
who began to perform CBAs for other government
agencies as well. At the same time, the new welfare
economics was increasingly making its in?uence felt.
The idea that a worthy project could be unfavour-
ably assessed by a competent CBA became more
di?cult for o?cials and other professionals to coun-
tenance publicly than it had been in the 1940s. By
the 1960s, as increasingly ambitious forms of
accounting became established in the US Depart-
ment of Defense under the championship of Robert
McNamara, it was possible for an economist to
write that the ‘economic valuation of bene?ts and
costs of an institution, plan or activity must attempt
to take account of values of any sort’ (Dorfman,
1965). As PDDs would later do, CBAs grew to
immense size. Discussion grew intense about such
issues as techniques of valuing lives. Should one
(for example) tot up discounted future earnings,
production or consumption? Calculate the dis-
counted present value of expected losses others suf-
fer as a result of one’s death? Look at jury awards in
compensation for death? As CBA became a serious
academic economic specialty, some economists
began to envisage CBA as a routine of such poten-
tial legitimacy that, ‘once set in motion by appropri-
ate value judgments on the part of those politically
responsible and accountable’, it ‘would – like the
universe of the deists – run its course without fur-
ther interference from the top’ (Sen, Das Gupta, &
Marglin, 1972, quoted in Porter, 1995, p. 150).
Presidents Nixon and Carter applied CBA to a
growing body of regulatory actions, but it was not
until President Ronald Reagan’s Executive Order
12291 of 1981 requiring ‘regulatory impact analysis’
of all ‘major’ rules, and advising against adoption of
any that did not pass a cost–bene?t test, that it
became entrenched throughout the regulatory appa-
ratus of the US government. Under the Bill Clinton
regime, CBA was sometimes used to advance rather
than inhibit regulation, but George Bush reverted to
the Reagan strategy of using it as an anti-regulatory
device. In 2003, the O?ce of Management and Bud-
get – charged with overseeing most government reg-
ulation – went so far as to consider assigning a
monetary value to the liberty and privacy lost to
new anti-terrorism legislation, while Harvard cost–
bene?t analysts asked subjects if they were willing
to accept racial pro?ling at airport security check-
points in exchange for saving time standing in line
(Viscusi & Zeckhauser, 2003, pp. 104–105). Today
CBA remains part of the ‘regulatory reform’ agenda
at the individual US state level as well (Hahn, 2001,
pp. 57–75). It is also widely used to evaluate policies
and projects in other countries and at international
?nancial institutions. Across the world, a range of
quanti?able values have become regularly attached
to health, biodiversity, noise, scenery, time and
human life in a way that enables them to be
inspected and weighed against each other from the
convenience of spreadsheets in o?ces. Yet the more
it is informed about such procedures, ‘the more the
public seems likely to distrust it, again bringing the
method’s legitimacy into question’ (Verchick, 2005).
In the view of one of its prominent contemporary
apologists, CBA was designed ‘to induce govern-
ment to simulate market outcomes’ (Posner, 2001,
p. 323). In Callonian terms, what it helped ‘frame’
was not a literal market but rather concrete, physi-
cal market-like arenas of coordination, negotiation,
persuasion, practical reasoning, decision-making
and resource allocation, primarily but not
exclusively situated in, and for use by, state agen-
cies. Recasting public choice as virtual exchange
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 517
undertaken by bureaucracies or decision-makers,
CBA, like carbon accounting, was all about creating
a new commodity – not for trade, but rather for
‘trade-o?s’. Lewis Kornhauser characterizes this
commodity (or, rather, quasi-commodity – it is only
quasi-bought and quasi-sold) as policies: ‘CBA
prices policies, not the consequences of those poli-
cies’ (Kornhauser, 2001, p. 221). However, the
quasi-commodity should perhaps be characterized
more broadly, since CBA is most often used for
comparative pricing or commodi?cation of ‘worlds’
– worlds with and without a given policy, project or
event. (Whereas in carbon o?set accounting, the
hypothetical world being quanti?ed is a world with-
out a given project, in CBA it is most often a world
with a proposed project.) This quasi-commodity is
not literally priced – the whole point of CBA is to
go beyond existing prices, not just to say how much
a policy or a project would cost in actual markets –
but rather quasi-priced in a distinct currency con-
sisting of overall welfare phrased in monetary equiv-
alents. In a regulatory quasi-marketplace, o?cials
try to aggregate the bene?ts of a public policy and
compare them to the costs. More speci?cally, ‘[t]he
very idea behind CBA is to commensurate winners’
gains and losers’ losses’ (Adler & Posner, 2001, p.
273) as a step toward determining whether there
could be a ‘hypothetical costless lump-sum redistri-
bution in the project world, from winners to losers,
such that this amended project world is Pareto e?-
cient relative to the status quo’ (Adler & Posner,
2001, p. 272). As in any market, consumers (in this
case policymakers) were free to be foolish shoppers
if they so chose, and pay a high quasi-price (or cause
the public to pay it) for an inferior good (or policy,
or project) – or a good that delivered an overall loss.
But smart shoppers would know better, and make
the right choice for their societies.
To make this quasi-market work, a numeraire
had to be framed which was, in one form or
another, welfare-commensurated-with-money
(which on an extreme economistic view is identical
with welfare simpliciter). Traditionally, this was
construed as the sum of compensating variations
for a project or policy, or the dollar amounts hypo-
thetically paid to or from individuals in the ‘project
world’ such that given their ‘preferences’ they would
be indi?erent between that world and the status quo
(Adler & Posner, 2001, p. 270); like carbon o?set
accounting, CBA depends heavily on reasoning
involving counterfactual conditionals. In turn, these
‘preferences’ themselves had to be framed or con-
structed using techniques that would result in their
being calculable and aggregatable – that is, well-
behaved in the centres in which they were supposed
to perform. ‘In practice, CBA proceeds by assuming
that consumer purchasing decisions are a proxy for
preferences and preferences are a proxy for utility’
(Heinzerling, 2002, p. 2314).
Two of the leading techniques developed after
the Second World War for producing or eliciting
such preferences or prices were hedonic pricing
and ‘contingent valuation’ (CV). Using statistical
techniques, hedonic pricing infers preferences from
observable market behaviour. For example, prefer-
ences for workplace safety can be inferred or con-
structed by comparing wage levels of various jobs
with their work-related injury rates. Highly inten-
sive in its use of ingenuity, expertise and time, this
procedure generally sidesteps the need to involve
laypeople directly in the compilation of their prefer-
ences. Not so with contingent valuation, which
involves a larger number of preference production
zones and proceeds in several stages.
In CV, an early round of disciplining takes place
in interview rooms or on questionnaire sheets. A
sample of people are quizzed individually to deter-
mine the maximum amount of money they would
be willing to pay, as individuals, for, say, clean
air, the conservation of local lakes, the survival of
blue whales – or, alternatively, the minimum
amount they would accept for the loss of some such
good. A survey seeking to monetize the sub-clinical
health e?ects of increased levels of ozone, for exam-
ple, may ask people how much money they would
be willing to pay to have avoided having been tired
easily one time during the last month. In general,
questions are designed to format responses, includ-
ing expressions of citizen responsibility (Sago?,
1988), into a calculable pattern of quasi-consumer
preferences. Questions about whether a price should
be assigned at all, for instance, are not asked.
Unwieldy compromises can result. For example,
survey subjects who at ?rst balk at answering ques-
tions about how much money they would be willing
to accept for the loss of their homes may ultimately
be forced to enter a bid for in?nite compensation.
(Ironically, CV theory ‘performs’ a quasi-market-
place that strives to be even less entangled with
other arenas of social relations than the real market-
places it supposedly takes as a model. CV subjects
are not permitted to exit from the designated choice
without their views being discarded, whereas buyers
in actual markets are generally allowed to leave
518 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
rather than sell or buy. In addition, subjects in CV
surveys are supposed to place their decisions in the
context of other spending decisions, whereas many
actual markets depend on a context of stimulation
of ‘impulse buying’ (Heinzerling, 2002, pp. 2334–
2335).)
A second stage of framing takes place at the sur-
veyor’s or economist’s o?ce, where the mass of
emergent ?gures, agencies and objects emerging
from the original survey are ?rmed up and resculp-
ted. Responses that may have re?ected a heteroge-
neous mix of considered views, hesitations,
strategic gambits, free-association or expressions
of frustration or boredom are reinterpreted and fur-
ther simpli?ed and edited, and controversies, uncer-
tainties and con?icts black-boxed to the extent
possible. As elsewhere, to borrow the words of
Espeland and Stevens, ‘commensuration can be
understood as a system for discarding information
and organizing what remains into new forms’
(1998, p. 317). To take one obvious case, demands
for in?nite compensation emerging from the origi-
nal survey must be thrown out or reinterpreted; they
cannot ‘realistically’ be entered into an agency’s cal-
culus since they would trump all others automati-
cally, leaving no room for weighing alternatives
against each other and thus no policy pointers
(Helm & Pearce, 1991; Turner, 1991; Viscusi,
Magat, & Huber, 1987).
11
For example, in the late
1960s, the Roskill Commission considering sites
for a third London airport reinterpreted bids for
in?nite compensation for homes that would be
destroyed as demands for compensation of 200%
above market value (Adams, 1989). Alternatively,
a new interview team can be dispatched to brief
recalcitrant subjects more thoroughly so that when
they are questioned again they can help produce
preferences which can be fed more easily into the
computational process (Hanemann, 1994, p. 24).
Just as various ‘monsters’ of climate change can
be framed or tamed by running together uncer-
tainty, ignorance and indeterminacy with probabil-
ity, other so-called ‘outliers’ that result from
rebellious subjects’ e?orts to wrench apart the origi-
nal CBA ‘frame’ can also be eliminated from the
data set using justi?cations from statistical theory.
A third arena of disentanglement is one also used
in hedonic pricing: meeting rooms or printed docu-
ments where experts or their patrons summarize
CBAs to the public, at which point the now-hard-
ened numbers take on their own life, often setting
in train a ‘bandwagon e?ect’ capable of changing
those opinions. These arenas, too, are structured
in a way that helps reduce the authority of the ori-
ginal subjects to complicate the data by engaging
in bargaining over how their actions are to be inter-
preted. In 1995, for example, economists in Work-
ing Group III of the Intergovernmental Panel on
Climate Change, when criticized for having calcu-
lated the value of a statistical life of a US citizen
at US$1.5 million and that of a statistical life of a
‘developing country’ citizen at $100,000, responded
that they had merely been reading o? ‘people’s
appreciation for a risk-free environment’ using pub-
lished economic data in accordance with established
procedures (Fankhauser, 1995, p. 167).
12
It is part
of CBA’s framing function, like that of the audit
techniques described by Michael Power, that it
brings about a ‘loss of social thinking’. Represented
as one embodiment of democracy, it tends to allo-
cate problems ‘to a particular class of experts who
may pro?t from its own abuses’ (Power, 1988, p.
312).
Like carbon accounting, in short, CBA has had
to build up a large disciplinary infrastructure with
a variety of social implications in order to ‘make
things the same’. As such, it can be seen as another
example of the ‘state simpli?cation’ or ‘high mod-
ernism’ surveyed by James C. Scott. Like the Prus-
sian forestry practices Scott reviews (Scott, 1999,
pp. 11–22), it involves processes that do not just
redescribe, but also, as centralized calculation
becomes more important, remold their objects in
ways that render them more ‘legible’ to state agen-
cies pursuing speci?c sets of goals. Nikolas Rose
11
In a parallel case in carbon accounting, carbon credit retailers
such as Mike Mason of the UK ?rm Climate Care argue that
aviation o?set prices must be kept within the reach of ordinary
consumers regardless of scienti?c ?ndings that suggest jet ?ights
have especially high climate forcing e?ects due to the altitude at
which they release greenhouse gases. ‘You might want to be
conservative. . .but if what you do is push it out of reach of many
people, so that many fewer people buy it, you haven’t actually
done the planet a favour’ (Channel 4 Dispatches, 2007).
12
CBA was attractive to US and other industrial-country elites
in this forum partly because its weaknesses in dealing with
distributional and narrative issues were politically useful in a
situation in which accusations of responsibility for global
warming were tied to demands for proportional contributions
toward alleviation. The Working Group III ?gures were used in
further calculations which suggested that climate change would
cost twice as much ‘socio-economic’ damage to the industrialised
countries as to the rest of the world. For further discussion of the
politics of ‘statistical people’ see Heinzerling (2000).
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 519
describes such processes in more abstract terms
when he writes that ‘numbers do not merely inscribe
a pre-existing reality. They constitute it’:
‘. . .the collection and aggregation of numbers
participate in the fabrication of a ‘‘clearing”
within which thought and action can occur.
Numbers here help to delineate ‘‘irreal spaces”
for the operation of government, and to make
them out by a grid of norms allowing evaluation
and judgement. . . These ‘‘calculable spaces” have
made up the ?elds of government at both macro-
and micro-levels. . .as these accountancy-shaped
spaces are thrown over a whole range of other
institutions. . .a whole variety of new calculable
spaces are brought into existence.’ (Rose, 1999,
pp. 212–213).
The ‘historical and social processes of construct-
ing and solidifying equivalences’ involved in CBA,
to adapt a phrase of Desrosieres, were no doubt
expected by some experts to be able eventually to
create ‘things that hold up well, independent of par-
ticular interests, in order to be able to act on them’
(Desrosieres, 1996, p. 336). Historians have habitu-
ally seen numbers as a response to turbulence (Cline
Cohen, 1982) and the standardization provided by
CBA a measure of accountability-de?ecting ‘objec-
tivity’ (Porter, 1995). Yet for those closer to the coal
face, the ‘framings’ associated with CBA, like those
associated with carbon accounting, have persis-
tently and inevitably involved destabilizing over-
?ows of many kinds. As economists Hanke and
Walker wrote a quarter of a century ago:
‘In spite of years of re?nement in the theory of
cost–bene?t analysis no one has succeeded in
making it impartial or indisputable... no amount
of technical wizardry will succeed in absolving us
of the need to resolve... con?ict through political
processes.’ (Hanke & Walker, 1974, p. 908).
‘Far from resolving controversy’, geographer John
Adams observed 20 years later, ‘cost–bene?t analy-
sis generates it.’ (Adams, 1995; see Herbst, 1993 for
a parallel discussion of opinion polling).
At the same time, CBA’s failure to stabilise a
market-like arena for policy choice has seldom
given pause to, and certainly not overwhelmed,
the institutions that produce CBA. The result, to
revert to Callon’s words, has been a space that
‘must be constantly reformed and built up from
scratch’ and that ‘never ceases to emerge and re-
emerge’ from the entailments and entanglements it
itself helps bring about. For more than half a cen-
tury, CBA and the decision-making arenas it ‘per-
forms’ have engendered con?icts and dilemmas in
ever-evolving but recurring patterns. Whatever
strategies they pursue, specialist economists com-
mitted to re?ning and deploying CBA have found
themselves at odds with public, governmental and
academic opponents, and vice versa. Many of the
arguments and rhetorical stances evident in the
debates of 1955 have been echoed in 1965, 1975,
1985, 1995 and 2005; Amartya Sen alludes to a
continuing ‘conversation between great soliloquists
– very skilled in making their points, and somewhat
less troubled than Hamlet (to be, say some, and not
to be, announce the others)’ (Sen, 2001, p. 95). This
pattern’s persistence calls into question both the
idea that CBA technique might someday be eluci-
dated or ‘puri?ed’ to the point at which it is no
longer controversial and the mirroring idea that it
exempli?es a creeping and inexorable commodi?ca-
tion of society and nature. Perhaps unlike carbon
accounting, which at the present moment seems vul-
nerable to collapse, CBA has survived many
moments of crisis. Yet the succession of entangle-
ments and over?ows visible in its history is structur-
ally similar to those that are appearing around
carbon accounting. This article will brie?y consider
some over?ows that result from framing prefer-
ences, subjects, surveyors and the general public.
Framing welfare, framing preferences
Framing a quasi-market for public choice means
constructing quanti?able, aggregatable preferences
through techniques such as contingent valuation.
Such techniques temporarily constrain their sub-
jects’ exercise of rationality in a number of ways, ?t-
ting it, for as long as they are applied, to the
Procrustean bed of what Amartya Sen calls a ‘mar-
ket analogy’ (Sen, 2001, p. 111) (although, as we
have seen, only certain parts of the analogy are
stressed). For instance, for reasons of convenience
in calculation, acts of valuation under CV are com-
monly not iterative. That is, the weights subjects
give to the options on o?er are not easily recorded
within CV as tentative or open to revision, or as
evolving in the very process in which they are given
expression (Sen, 2001, p. 107), or as requiring a pro-
cess of collective deliberation (Richardson, 2001,
pp. 141 ?.), as many evaluations are in other social
contexts. As Richardson points out, a number of
types of ‘intelligent refashioning of ends. . .are avail-
520 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
able to ordinary deliberation but foreclosed by any
model of decision that takes as its contentful norma-
tive basis a given set of willingness-to-pay informa-
tion’ (2001, p. 153; see also Richardson, 1994). In
the words of Adler and Posner, preferences shaped
and collected for CBA purposes are ‘non-adaptive’;
that is, they are not treated as subject to change
pending the outcome of the project being evaluated,
and thus not as if they might be partly dependent on
the CBA itself (Adler & Posner, 2001, pp. 284–285).
Yet even if ‘sensible responses to collective con?icts’
could be ‘captured in terms of revised preferences’,
eliciting or constructing individual preferences still
leaves out of the frame the kind of practical intelli-
gence needed to generate those responses (Richard-
son, 2001, pp. 157–158). In addition, the
hypothetical ‘willingness to pay’ of the CV subject
must usually be interpreted as independent of what
other subjects are willing to pay. This makes fewer
considerations available to CV subjects during their
deliberations than is usually the case during discus-
sions about social policy. What one person is willing
to pay to clean up an oil spill, for example, will usu-
ally vary depending on what others are willing to
pay or on the conclusions of public consultations
over what government action would be appropriate.
Normally, people’s views on the environment can be
expected to be elicited in a context of being ‘told
what the real alternatives are, involving speci?ca-
tion of what will be done by the others’. Yet this
is ‘simply not provided by the market-based ques-
tioning (either in the form ‘‘How much would you
pay if you could singlehandedly bring about the
change?” or the form ‘‘How much would you con-
tribute, assuming whatever you want to assume
about what others are doing?”’ (Sen, 2001, p.
114).)
13
Similarly, to use non-iterative individual prefer-
ences to calculate welfare is leave open the possibil-
ity of terminating the calculation before all
information has been collected that subjects them-
selves, or others, might consider relevant to the val-
uation (Adler & Posner, 2001, p. 278), or before any
collective decision has been made about what sort of
information will count as relevant. It is also to
exclude processes of censorship or weighting that
are applied in most other circumstances to individu-
als’ views that are rooted in psychological problems
or considered sadist, racist or otherwise socially
unacceptable (Adler & Posner, 2001, pp. 295–297).
Finally, insofar as it does not discount the willing-
ness-to-pay ?gures of wealthy individuals, it
excludes the egalitarian considerations commonly
appealed to in democratic societies.
Applying CBA to environmental decision-mak-
ing necessitated carving out spaces for calculation
in still further frontier zones. For instance, accord-
ing to many CBA practitioners, moral attitudes
about ‘nature’ can be framed as compensating vari-
ations for the existence of environmental goods not
enjoyed directly and then commensurated with
prices for other goods. According to others, disin-
terested moral reasons must somehow be placed
altogether outside the ‘frame’ used to elicit or de?ne
preferences, because they are irrelevant to welfare
(Adler & Posner, 2001, pp. 276–277, 290–291). In
general, CBA tends to exclude everyday reasoning
about clusters of interlocked, mutually irreducible
ends and how to develop them in light of the means
available; reasoning about if and how a rule applies;
and reasoning that involves acquiring, like a critic,
student, artist or revolutionary scientist, a new lan-
guage, taste, perception or goal that recontextualiz-
es or comments on older ones (Isenberg, 1949;
Rorty, 1979). To treat such patterns of rationality
on the model of, say, comparing body weights to
check babies’ health or comparing prices when
deciding on investments in land, corn, iron or
microchips is a radical social innovation. It strives
to displace those types of rational decision-making
in which each person or group brings into play
mutually incommensurable considerations (which
normally encourages treating experienced and per-
ceptive persons, not just sets of criteria or numbers,
as touchstones of rational choice) or those even
more complex types of practical reasoning involving
speakers of di?erent languages who do not share
common adjudication procedures when looking at
the same set of alternatives (which normally encour-
ages treating intercultural conversation as a further
important touchstone of rational choice).
At the same time that certain types of reasoning
are excluded by the ‘framing’ performed by CV and
13
In the 1980s, a Karen villager in Northern Thailand simply
rejected a Finnish forestry company’s request that he indicate his
land use preferences, indicating that such a question could
intelligibly be posed only to the community as a whole (Ann
Danaiya Usher, personal communication, 1989; see also Hein-
zerling, 2002, p. 2324). The parallel with opinion polling is
relevant: as Andrew Ross (1998, p. 152) observes, ‘how people
respond to a slate of surgically prepped questions tells virtually
nothing about the opinions they might ?nd they held in common
if the conditions of a properly radical democracy permitted them
to do so’.
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 521
other CBA techniques, others are made mandatory.
Subjects are required to commensurate and price
many events, states or goods that are not commens-
urated or priced in everyday life – or whose com-
mensuration is at present socially circumscribed or
disallowed in other contexts. These include lifespan,
biodiversity, disaster avoidance, terrorism preven-
tion measures, noise, visibility, the existence of wild-
life, pollution, scenery, ozone damage, pension
provision, rights, futures with or without various
development projects, and so on. In one experiment
designed to test hypotheses about how to measure
the ‘intrinsic’ value assigned to life, interviewees
were even asked how much they would pay not to
have a researcher kill a potted Norfolk Island pine
tree shown to them. Any context of reasons that
could help subjects make sense of the unfamiliar
demand to commensurate such entities with each
other and with money is sometimes explicitly
pushed outside the ‘frame’. An analogy with opin-
ion polling is useful: a Mori handbook instructs
opinion pollsters who are asked what a question
means to reply, ‘whatever you want it to mean’
(Barnard, 1992). Another pollster explains that
any answers outside the choices on o?er ‘get
dumped or written o? as ‘‘other”’ (Hitchens, 1994,
p. 48).
Such processes of framing preferences in ways
that are intended to ensure that they are well-
behaved in centres of cost–bene?t calculation have
well-documented over?ows. For example, prefer-
ences constructed through contingent valuation
often turn out to be so badly-behaved that they
are useless in welfare calculations. So much of the
familiar contexts of practical reasoning have been
removed by the framing process that some survey
subjects, ?nding that the questions do not make
sense, register zero or arbitrary valuations. Others,
objecting to the commensuration of goods that they
do not want to commensurate, enter, as noted
above, in?nite ‘protest bids’ for the good being
quanti?ed, or produce numbers re?ecting ‘a defen-
sive reaction to a perceived threat’ (Clark, Burgess,
& Harrison, 2000). Still other subjects
‘provide valuations that are invariant across
large and small parcels of wilderness or quantities
of wildlife or that are inconsistent or intransitive.
Their answers depend on the order in which
questions are asked and are sensitive to the word-
ing of the questions.’ (Adler & Posner, 2001, p.
290; see also Desvousge, 1993; Harvard Law
Review, 1992; Plott, 1993; Rosenthal & Nelson,
1992; Sen, 2001).
One economist seeking to measure the economic
value of recreational inner-tubing on Arizona’s Salt
River as part of an environmental assessment of the
Orme dam tried and failed to build a regression
model synthesizing a demand curve tracking both
real payments inner-tubing enthusiasts made for
their experience and willingness-to-pay data: ‘The
curve didn’t work. It didn’t turn out to be measur-
able.’ The value of tubing was duly excluded from
the ?nal analysis (Espeland, 1999).
The presence of a large preponderance of data
that must be discarded suggests that even the bet-
ter-behaved data that emerge from contingent value
surveys should be handled with caution (Adler &
Posner, 2001, pp. 290–292). The process of framing
preferences often removes so much social context
that it becomes di?cult to resolve the classic ‘radical
interpretive’ dilemma described by Donald David-
son (Davidson, 1984) – that is, to make the choice
between deciding that one is misinterpreting a
speaker and deciding that the speaker has false
beliefs.
14
The need to deliver ?gures pushes analysts
toward the latter, undermining the credibility of CV
?gures and setting the stage for further instability –
as will be explored below – by constructing CV sub-
jects as stupid and CV surveyors as arrogant.
Information de?cits among survey subjects may
lead to further calculational di?culties. Moreover,
the problem of accounting for innovation that was
evident in carbon accounting resurfaces in CBA.
‘Regulatory analysis is notorious for failing to take
into account the technological innovations that ulti-
mately make many regulations cheaper to imple-
ment than regulators anticipate’ (Heinzerling,
2002, p. 2314). The di?culty of nonhalting calcula-
tions also reappears. Projects implemented partly as
a result of CV can themselves change preferences
about them (Sen, 2001, p. 109; see also Sunstein,
14
Depending on context, the question ‘How much are you
willing to pay for X?’ may be interpreted in a great variety of
ways: for example, as an extortion demand; ?eamarket vendor’s
gambit; challenge to honor, prestige or some other form of
‘symbolic capital’; corrupt judge’s request for a bribe (or honest
judge’s test of a defendant’s values); prostitute’s solicitation;
invitation to discuss the ‘imaginary market’ a surveyor is
attempting to construct; close friend’s idle query on a holiday
walk; or obnoxious joke. Correspondingly, ‘How much are you
willing to accept for place X?’ may be received as a di?erent
question depending on, say the profane or sacred standing of the
X in question.
522 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
1993). As Sen sums it up, ‘When all requirements of
market-centred evaluation have been incorporated
into procedures of CBA, it is not so much a disci-
pline as a daydream’ (Sen, 2001, p. 116). ‘Because
it shuts o? practical intelligence’, Richardson adds,
‘CBA’s interpretation of what people want is
accordingly implausible’ (Richardson, 2001, p.
155). Adler and Posner, meanwhile, although they
attempt to defend certain types of CBA, admit that
the satisfaction of what they call ‘actual prefer-
ences’, or, in the terms of this paper, preferences
constructed through textbook CBA, is simply ‘not
equivalent’ to the ‘maximization of well-being. . .
Actual preferences are not necessarily constitutive
of welfare’ (Adler & Posner, 2001, pp. 270–271).
The instability and limited utility of the prefer-
ences framed by CV and other techniques means
that preferences must be continually ‘reformed and
built up from scratch’ through new framing exer-
cises. Adler and Posner (2006) propose a process
that they frankly call ‘laundering’, by which suppos-
edly ‘distorted’ preferences
15
are made more
rational (the sum of compensating variations being
transformed into ‘welfare equivalents’), yet retain
their technical suitability for the quasi-market for
policy options that CBA formats. Information that
subjects would bene?t from knowing but do not
have at their disposal in the forums in which CVs
are calculated is imported and preferences corrected
accordingly. Preferences that are ‘objectively wrong’
representations of an individual’s valuations are
also corrected. Racist or sadist preferences are cen-
sored to ensure that so-called ‘objective values’ are
tracked, or are registered as 0. Adaptive values are
frozen at a certain level to ensure that calculations
halt at a unique ?gure. In other words, attempts
are made to (re-)frame entanglements or over?ows
that were pushed outside the original frame, yet
on which the framed ‘preferences’ depend for their
intelligibility and utility. Other theorists have pro-
posed expanding the concept of preferences to
include those based on ‘commitment’ (Sen, 1977)
or ‘ethics’ (Harsanyi, 1955). What Richardson calls
‘intelligence’ is reintroduced by the back door.
16
In fact, preferences are already routinely being
‘laundered’ in ad hoc ways in the everyday contem-
porary practice of CBA. Agencies deploying the
procedure preserve CBA’s appearance of ‘tracking
overall well-being’ or sustain the ?ction that prefer-
ences revealed in hedonic pricing or contingent val-
uation re?ect ‘complete thinking’ (Richardson,
2001, p. 167) only by ?outing textbook norms.
For example, in evaluating a proposed regulation
involving labeling of meat and poultry products,
the USDA relied not on subjects’ contingent valua-
tions of nutrition disclosure, but rather on contin-
gent valuation of the health bene?ts they would
enjoy if they altered their behaviour in response to
the labels. Similarly, the premise of modern work-
place regulation is that workers are ‘uninformed
about risks’ (Adler & Posner, 2001, p. 283), making
CV-based calculations inappropriate. Many agency
programmes discount or discard the preferences of
various social groups – for example, the preferences
of drug addicts not to implement antidrug pro-
grammes for airline employees. They also routinely
ignore sadistic preferences: the FDA’s CBA of a
regulation designed to curb distribution of ciga-
rettes to children did not include as a cost the lost
pro?ts to industry, ‘because this pro?t stems from
illegal sales to youths’, nor did it count the chil-
dren’s lost consumer surplus. The FAA’s CBA for
an airline security program, by the same token,
did not consider the possibility that a system of
racial pro?ling might be less costly than the system
15
The assumption that ‘undistorted preferences’ ready-made for
accounting use exist in individuals is evident in the statement of a
prominent environmental economist that ‘the absence of markets
in environmental services creates a practical problem of mea-
surement – i.e. one of ?nding out what people’s preferences
actually are in a context where there are no apparent markets –
but it does not create a conceptual problem of measurement’
(Pearce, 1991, p. 3).
16
Alternatively, the scope of CBA itself can be restricted so that
it is simply not used to resolve issues of fairness, justice and
deontological rights or commensurate moral commitments with
other preferences. In practice, the application of CBA has always
been limited by the political process and often by accountants
themselves (Dove, 1999). In the 1940s, the US Army Corps of
Engineers had no di?culty in maintaining that water projects in
Michigan, Alaska and Pennsylvania were justi?ed given the
‘welfare of the communities’ a?ected or the ‘importance of future
development in the region’ despite low bene?t/cost ratios ranging
from 0.53 to 0.82 (Porter, 1995, p. 160). More than two decades
later, the Roskill Commissioners found that the government
simply disregarded their recommendation on the site for a third
London airport following public ridicule of a CBA that
attempted to aggregate costs of rail construction, noise, property
and recreation losses, air travel, tra?c growth, loss of a Norman
church and human fatalities (Self, 1975). This was despite the fact
that the CBA had been designed precisely to reduce the political
heat that government o?cials would have to absorb for making a
decision on the airport.
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 523
it ultimately endorsed. O?cial agencies correct for
wealth distortions, meanwhile, by using a constant
?gure for the monetized value of life rather than
the willingness to pay of individuals, which is likely
to vary with wealth. Many such routine undermin-
ings of textbook CBA are likely to stem from agen-
cies’ fear of the ‘over?ow’ consisting of public
outrage that would result from uncompromising
deployment of hedonic pricing, contingent valua-
tion and commensuration across the board (Adler
& Posner, 2001, pp. 285–287, 288–289) – the sort
of outrage that surfaced in the case of the Intergov-
ernmental Panel on Climate Change discussed
below. As Vatn and Bromley put it, ‘the most fun-
damental environmental choices will continue to
be made without prices – and without apologies’
(Vatn & Bromley, 1994, p. 145).
In short, with each reframing of the public policy
quasi-market by CBA come further over?ows
(Heinzerling, 2002, p. 2329). CBA’s project of e?-
ciently satisfying the preferences of individuals can
never be completed, partly because CBA cannot
specify those preferences (Richardson, 2001, p.
155). The more rigorous that attempts to capture
such over?ows become, the more unwieldy the con-
struction process gets. As Sen (2001, p. 95) notes,
there is a ‘tradeo? here between easier usability
(through locked-up formulae) and more general
acceptability (through allowing parametric varia-
tions)’. For example, replacing the question ‘How
much would you pay for X?’ with the question
‘What should be done for X using tax money?’ in
contingent valuation surveys would avoid some of
the di?culties connected with the attempt to force
respondents to consider X merely as a commodity
of private interest, yet would threaten to defeat
the purpose of the contingent valuation exercise
(Jacobs, 1997, p. 219).
In a sense, therefore, the ‘di?erent debate’ pro-
posed in the title of this article has already been qui-
etly proceeding, only under the radar of more
theoretical disputes. While many CBA theorists still
speak of ‘distorted’ preferences, as if there existed or
should exist in each individual an underlying stra-
tum of ‘undistorted’, pristine, invariant preferences
waiting to be made explicit through expert tech-
nique (Adler & Posner, 2001; Harvard Law Review,
1992; Pearce, 1991), in practice government agencies
deploying CBA do what they have to do to balance
a wide range of political pressures without much
regard for whether they are collaborating in the cre-
ation of preferences or not. Cost–bene?t analysis is
itself reformatted and retranslated by less con-
strained forms of discussion.
By the same token, while writers such as O’Neill
(2007, p. 130) and Raz (1986, p. 345) contend that
it is part of the ‘constitution’ of certain goods that
they are incommensurable with others, and writers
such as Sen (2001, p. 113) and Nussbaum (1990,
pp. 60–61) complain that standard CV cannot cap-
ture the ‘nature’ of certain types of valuation, again,
it may be more illuminating, rather than pretending
to be able to lay down the law about the violated
‘nature’ of various goods and kinds of valuation,
to take seriously the reality that CBA is proceeding
apace and helping to contrive real subjects and
objects even if it entails complex over?ows and will
be forever incomplete. Surveyors can and do some-
times begin to get commensuration practices going
during surveys, and subjects do ‘learn’ to price new
things. As Viviana Zelizer documents, the borders
between what can and cannot be priced or otherwise
commensurated are in constant historical ?ux (Zeliz-
er, 1985, 1997), and survey sessions and their associ-
ated practices are part of that ?ux. In describing this
?ux, many CBA critics run the risk of stalling the
debate when they attribute conceptual ‘errors’ to
CBA defenders. Defenders do the same when, as is
typical, they fail to grasp that, in MacKenzie’s
phrase, CBA is all about ‘making things the same’
– and that hardening a CV subject’s responses into
the useful objects called ‘preferences’ is invariably a
more complex, di?cult, time-consuming and alto-
gether di?erent matter than they acknowledge. An
interesting parallel is the Plantar case, where the
political issue revolves less around the formally
undecidable question of the ‘accuracy’ or otherwise
of the unique without-project future scenario for
Minas Gerais delineated by experts, but rather about
what local residents will stand for in practice in their
battle with carbon project accountants and propo-
nents. To talk about ‘natures’ and ‘constitutions’ in
the style in which, in a past era, analytic philosophers
used to talk about ‘category mistakes’, not only is
limiting analytically, but also obscures or foreshort-
ens complex historical processes as well as rivalries
among accounting technique users that political
and environmental activists need to consider in
greater detail (Dove, 1999).
Framing subjects
At the same time that it constructs preferences,
CV, like opinion polling, works to shape and disci-
524 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
pline expert-friendly subjects. While there is no need
for CV to demand that its subjects transform them-
selves into Homo economicus to the extent of treat-
ing all values as homogeneous – assigning a
monetary value to an option does not imply that
is the only value it has, nor necessarily set in train
a shift in the way it is valued overall (Nussbaum,
2001, p. 195; cf. Anderson, 1993; Verchick, 2005)
– it does set up, at least temporarily, an often unfa-
miliar pricing practice that ‘formats’ survey subjects
even when they ?nd it di?cult to interpret or
respond to (Farber & Hemmersbaugh, 1993, p.
301). The pared-down, uniform structure of CV sur-
veys, like that of CDM consultation forms, neces-
sarily leaves little room for respondents to say
‘that’s an odd question’ or express discomfort with
questions that seem contextless, unanswerable, triv-
ial, or misleading. As in opinion polling, surveyors
are trained not to conduct an everyday conversation
with their subjects but to follow a simpli?ed form of
interaction that constrains both parties in novel
ways. Still, the narrow range of questions that are
asked are capable of guiding conversation and sub-
ject behaviour after the survey – a fact well-known
to pollsters who are employed to shape as well as
to report ‘public opinion’. According to one socio-
logical study of CV, ‘the great majority of respon-
dents will subordinate themselves and their ways
of making sense to those of the survey designers.
If [the survey] does not make sense. . .respondents
will see these ‘‘failings” as their own’, many of them
disgorging arbitrary numbers largely out of a desire
to please or impress (Clark et al., 2000; Fischo?,
1991). Such surveys constitute one of a battery of
practices (together with economics courses, political
speeches, and so forth) that, in some societies, natu-
ralize cost–bene?t calculations to such an extent
that they become identi?ed, at least in abstract the-
ory, with practical reasoning itself. CBA-like ritual
is read back into all individual decision-making as
an unarticulated origin myth, justifying further
extensions of the technique.
17
However, resistance to this ‘formatting’ of sub-
jects is also widespread, as is acknowledged by
many CBA practitioners and defenders (Adler &
Posner, 2001, p. 290). When one survey asked what
Wyoming residents would accept in monetary com-
pensation for loss of visibility due to pollution from
a power plant, most interviewees rejected what they
saw as the surveyors’ assumption that they could be
‘bought o? to permit pollution’ and either refused to
cooperate or, again, entered ‘protest bids’ requiring
in?nite compensation (Rowe, D’Arge, & Brook-
shire, 1980). Zero bids constitute another destabiliz-
ing response from subjects disgusted with the
pricing game (Levy, Hammitt, Duan, Downes-
LeGuin, & Friedman, 1984). One subject of a con-
tingent valuation survey, after being asked ques-
tions aimed at ?nding out how much households
would be willing to pay for a wildlife enhancement
scheme in the UK’s Pevensey Levels, protested: ‘I
think you can put a value on nature but not a value
in money terms. A value is what we teach our chil-
dren’ (Clark et al., 2000). In six Scottish contingent
valuation studies, the surveyors themselves con-
ceded that between one-quarter and one-third of
respondents could not be made to take the questions
seriously (Hanley, 1991). Most respondents to a
contested New Zealand survey testing people’s will-
ingness to pay to prevent development of an island
o?ered ?gures that they clearly regarded as ‘gestures
in a political process’ (Vadnjal & O’Connor, 1994,
p. 375). In such cases, CV subjects appear not only
unaccustomed but also unwilling to be framed as
quasi-consumers, or to participate in unfamiliar or
circumscribed commensuration practices, even in
the restricted language-game of the survey, instead
expecting or demanding to be treated as citizens
(Heinzerling, 2002, pp. 2330–2331; Sago?, 1988;
see also Radin, 1996). E?orts to exclude power rela-
tions and property rights from the CBA equation in
favour of price adjustments and revisions of the ‘cal-
culations by which our economy is governed’, or to
reduce objections to the status of mere ‘philosophi-
cal’ musings about ‘intrinsic’ value, have often pro-
duced considerable opposition.
Proponents of CBA frequently frame this resis-
tance in ahistorical terms, attributing it to subjects’
ignorance of the pre-existing ‘nature’ of their prefer-
ences or of rationality itself. Subjects’ reluctance to
commensurate certain goods, they claim, is rooted
in their lack of understanding of the fact that ‘every
decision implies a monetary evaluation’ (Barde &
Pearce, 1991, p. 1), that tradeo?s are always possible
17
Arguably, the spread of CBA has helped nudge the academic
image of ‘rationality’ away from what Richard Rorty calls
‘reasonableness’ – tolerance, respect for the opinions of those
nearest one, willingness to learn, nondefensiveness, and reliance
on persuasion rather than force, which are traits over which no
one has a monopoly – toward that of a ritual of measurement,
calculation and aggregation which is the special province of an
economic and bureaucratic priesthood (Rorty, 1991, p. 37).
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 525
and usually necessary (‘measurement is essential,
since trade-o?s are inescapable’ (World Bank,
1992; see also Frank, 2001)), or in their underesti-
mation of CBA’s ability to clarify choices by high-
lighting ‘stakes’, values or states of a?airs which
are, again, assumed to be pre-existing (Sunstein,
2005, pp. 129–130). Rational choices, the Whiggish
assumption goes, have always involved at least
implicit quanti?cation of the alternatives. As one
environmental journalist puts it:
‘Knowingly or unknowingly, people who decide
that they would rather pay more for electricity
than destroy a forest to build a dam are implying
a valuation of the forest – crudely put, some-
where between the increased cost of electricity
and ‘‘priceless”’ (Ungphakorn, 1988).
To insist that placing a dollar value on human
life is morally illegitimate, claims Frank (2001, p.
71), carries the ‘implication’ that any measure that
would prevent deaths should be instituted ‘no mat-
ter how much it costs or no matter how little it
a?ects the risk of death and injury’ – an obvious
absurdity. On the extreme view expressed by Frank
and many other economists and scholars (e.g.,
Common, 1988), CBA only makes explicit what
everyone already does implicitly, and with enough
education or self-awareness would recognize that
they do.
Views such as Pearce’s or Frank’s, which con-
struct resisters to cost–bene?t accounting as igno-
rant or irrational, have provoked ?erce retorts
from a long line of Aristotelian commentators from
Otto Neurath (O’Neill, 1995) to Nussbaum (1986).
As Nussbaum argues, many if not most choices
are made on a basis which is ‘qualitative and not
quantitative, and rational just because it is qualita-
tive’ (Nussbaum, 1990, pp. 60–61; see also Ander-
son, 1993; O’Neill, 2007; Radin, 1996; Wiggins,
1987). Questioning the practice of placing a dollar
value on, say, human life, is compatible with declin-
ing to expend inde?nite amounts on extravagant
injury-preventative measures (Heinzerling, 2002).
If di?cult, even tragic, choices have to be made, it
does not follow that they have been made by com-
paring quantities:
‘we often make comparisons among diverse val-
ues, and choices among options involving di?er-
ent values, without commensurating in this
reductive way. . .the bare fact that we are able
to make these comparisons of overall good and
bad does not, it seems to me, imply that we have
all along been reducing them to a single metric of
value’ (Nussbaum, 2001, pp. 194–195).
Willingness to pay ?gures, in Henry Richardson’s
words, ‘provide a poor interpretation of what peo-
ple want’ (Richardson, 2001, p. 155) in that they
exclude the process by which human beings ‘regu-
late and revise their aims’ in light of information
that surfaces in the course of social interaction
(pp. 153, 155). If there is ‘stupidity’ here, Richard-
son suggests, it lies not with CV resisters but with
the view that CBA could e?ciently satisfy the pref-
erences of individuals or provide a standard of pub-
lic choice (Richardson, 2001). Here again, a
venerable tradition of scholarly argument can easily
be marshaled: to take just a few examples of many,
Michael Oakeshott’s or John Dewey’s assault on the
notion that ‘‘‘rational” activity is behaviour in
which an independently premeditated end is pur-
sued and which is determined solely by that end’
(Oakeshott, 1962, pp. 83, 89), Lindblom’s (1959)
skepticism about whether working bureaucratic
practice could be made more ‘rational’ through ?rst
isolating ends, then seeking and comparing all avail-
able means, or David Wiggins’s insistence that the
‘deliberative speci?cation of ends’ constitutes ‘most
of what is interesting and di?cult in practical rea-
son’ (Wiggins, 1987). It should not be surprising
that lay resisters to CV or hedonic pricing tech-
niques, too, are capable of replying in kind when
CBA practitioners frame them as stupid or irratio-
nal. The Pevensey survey’s respondents, for exam-
ple, reckoned that asking individuals for their ‘ill-
considered preference for one site in isolation’
was ‘to insult their intelligence’ (Clark et al.,
2000). Other places and the communities who lived
in them also had valid claims on available resources
– claims that the Pevensey subjects would need to
learn about and discuss with the people concerned
before working out and expressing their values to
a central authority, rather than submitting pre-for-
matted preferences to experts for aggregation. In
the same vein, the above-described attempt of econ-
omists in IPCC Working Group III to attribute
opposition to their hedonic pricing of statistical
lives (and thus to their interpretation of ‘people’s
appreciation for a risk-free environment’) to igno-
rance of economics methodologies contributed to
a furore among Southern UNFCCC delegates that
ultimately led to a permanent loss of prestige and
power for the late David Pearce, one of the econo-
526 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
mists leading the attempt. Again, the attempt to
‘contain’ the over?ow consisting of resistance to
CBA within a newly-drawn frame for supposedly
‘objective’ calculation and the construction of
strictly ‘statistical’ lives immediately occasioned
fresh over?ows.
Subjects often push back, too, when the ‘frame’
excludes them, as it did the Salt River inner-tubers
when their values could not be translated into mar-
ket values in methodologically acceptable ways
(Espeland, 1999), or as it does many groups when
it does not take into account their views. Indigenous
groups often react when the frame disentangles land
from their lives and identities by valuing it princi-
pally in terms of title or resource values (Espeland,
1999; Shanks, 1974). ‘Without the land, the Indian
cannot survive and with the Indian the land cannot
be land, because the land needs to be taken care of
in order to survive life’, as one Yavapai from Ari-
zona put it (Espeland, 1998, p. 201).
A similar dynamic results when economists trans-
late reluctance to accept results of a CBA into a
desire to ‘opt out of the debate’ (Common, 1992).
This is an interpretation that follows naturally from
the assumption that preferences revealed in hedonic
pricing or CV already re?ect ‘complete thinking’,
making further conversation super?uous. Such a
dynamic can also occur when economists interpret
public skepticism about the validity of the ‘prefer-
ences’ supposedly revealed by CBA as ‘opposition
from those who do not want preferences to count,
because the majority sometimes does not want what
they want’ (Pearce & Moran, 1994). One Pevensey
survey subject sums up a common reaction to such
expert translations:
‘The way they can manipulate this is basically by
saying, ‘‘Well, right. We’re not going to fund [this
conservation scheme] any more”. Then there’ll be
a public outcry saying ‘‘Why not? You’re sup-
posed to be looking after the environment!”
And they could directly turn round and say,
‘‘well, we took public opinion”’ (Burgess, Clark,
& Harrison, 1998).
The commensurating procedures of CBA, in
enabling centrally-located o?cials to ventriloquize
the ‘common will’ not only within but across socie-
ties, sometimes provoke even more powerful reac-
tions. In countries such as Thailand,
environmental accountants have on more than one
occasion been physically driven out of local areas
where they were doing nothing more than gathering
data for CBAs for proposed power projects whose
realization still lay far in a hypothetical future.
Where such blunt tactics are likely to be politically
ine?ective, laypeople may seek other means of cir-
cumventing the techniques that would translate
their views into calculable ‘preferences’. In Arizona,
the Yavapai saw no choice but to participate in the
formal environmental assessment of the Orme dam
that would have disrupted their lives, yet simulta-
neously also changed the subject and opened other
forums, re-enacting for journalists the Yavapai’s
earlier brutal resettlement and explaining to those
who would listen why money was an inappropriate
way to express the value of their land and culture.
‘White men like to count things that aren’t there.
We have a way of life that will be destroyed if that
dam comes through. Why don’t they just say that?’
The Yavapai also sought to communicate outside
the CBA framework using non-market analogies
meaningful to their white interlocutors. ‘The land
is our mother’, one Yavapai teenager said. ‘You
don’t sell your mother’ (Espeland, 1999).
Such types of opposition are akin to those fol-
lowed by opponents of the Plantar project discussed
above in that they challenge the accountants’ fram-
ing processes themselves, bursting them open by vis-
ibly ‘re-entangling’ the subjects and objects
formatted by CBA with the world outside, the revis-
ability of ends, other people whose desires are not
yet known, or with the history and futures they have
been accustomed to debating within other frame-
works. In this respect, these modes of opposition
are also akin to claims that carbon o?set accounting
is incoherent because the way it selects for ‘e?-
ciency’ at the margin cannot overcome challenges
posed by historical ‘lock-in’ of fossil fuel-intensive
structures.
CBA’s success in ‘performing’ a new public of
individuals legible to, responsive to, and instructed
by centrally-located experts and o?cials, then, will
probably always be limited. Entanglements with
other sites and forms of social reasoning cannot
be sliced away or prevented so easily. Al Gore’s
assumption that the ‘factors deemed to be impor-
tant’ to the environment are accounting-ready,
patiently waiting to be calculated, turns out itself
to provoke resistance. Hence one limitation of the
‘trust in numbers’ thesis that views supposedly
‘objective’ quantitative techniques such as CBA as
providing solutions to problems of distrust of dis-
cretionary decision-making by government agen-
cies. The historical record demonstrates that
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 527
numbers generated by CBA are unlikely to be
trusted, either. It is not that they are merely a cover
for business as usual – the institutions involved gen-
erate their own political centre of gravity, and the
numbers do intermittently help ‘change the subject’
of political discourse – but rather that they are per-
petually incomplete and partial, often unbelievable,
and of limited potency if not backed up by other
bureaucratic or political practices. The more CBA
is used, the more evident it becomes that it ‘is no less
dependent on o?cials’ virtues than are its more
intelligent alternatives’ (Richardson, 2001, p. 154)
and that, however, it is construed, the technique
itself engenders movements of opposition and self-
rede?nition.
Framing surveyors
In formatting subjects, cost–bene?t analysts, like
opinion pollsters, inevitably also format themselves.
Simultaneously constituted, empowered and con-
strained by the practices they participate in, they
produce numbers that make their actions, and those
of their political patrons, vulnerable to scrutiny by
experts while protected from challenge by many
others. Yet the ‘gated community’ protection that
CBA o?ers to its practitioners and patrons – and
the privileges, centralized location and exclusions
that go with it – itself ultimately opens up new vul-
nerabilities and negotiations for identity. What
‘works’ in this attempt to frame new economic/pol-
icy transactions also renders it incomplete. The
frame itself turns out to be an unstable border
object vulnerable to exchange.
For example, as Espeland (1999) notes, the legit-
imacy of commensuration often hinges on the dis-
avowal of its constitutive e?ects. Its authority
depends on being seen as a representation of ‘some-
thing that is already out there’. Accordingly, cost–
bene?t analysts, insofar as they follow a textbook
model, must construct themselves, as much as possi-
ble, as beings that are ‘not there’ except as passive
conduits or inscribers of knowledge originating else-
where. In the CV interview room, this plays out, as
suggested above, in a stance of conversational semi-
passivity that attempts to avoid any ‘perception of
interviewer pressure’ (Hanemann, 1994, p. 24).
Interviewers pose questions but strive to exclude
context. However, in a survey, as in any other con-
versation, complete passivity is unattainable. As
geographer Jacquelin Burgess and her colleagues
point out, questioner and respondent will always
be ‘locked in dialogue’, each acting on the other
(Clark et al., 2000). In a dialogue, no one can
black-box herself into invisibility or avoid in?uenc-
ing the person she is listening to (Lewontin, 1995,
pp. 43–44). Even silence conveys a meaning, and
some context, or set of contexts, is always implied.
The question is not whether researchers act on their
subjects, but how. Every interview is a step in the
evolution of society’s views. Holding detachment
up as an ideal is likely to increase not only di?cul-
ties of interpretation for both parties, but also awk-
wardness and possibilities for hostility and
perceptions of disrespect.
In addition, cost–bene?t analysts construct them-
selves as arrogant and elitist whenever, during fur-
ther stages of preference processing, they discard
‘protest bids’ or statistical inconsistencies as mani-
festations of subjects’ irrationality, ignorance, stu-
pidity or ?ckleness; edit them so that they re?ect
what subjects ‘must have really meant’; or disregard
certain kinds of identities, all the while attributing
to themselves privileged access to subjects’ values
or the inner structure of practical reasoning.
18
The
result can be further resistance and di?culties in
carrying out CBAs. ‘Trust in numbers’ turns out,
again, to be more di?cult to construct among lay-
people in ‘the ?eld’ than in the university lecture hall
or textbook page. Because conversation is by its
nature a social activity, the contingency of the ‘code
books’ through which subject responses are trans-
lated into ?gures cannot be hidden for long (Espe-
land, 1999).
The di?culty facing the new policy-making space
that CBA promised to frame, in short, was not that
not enough ‘externalities’ were being, or could be,
‘internalized’, or brought across the economic bor-
der. It lay, rather, in the notion that an ‘internal’
and an ‘external’ could be ?xed at all. Like the
‘frame or border of the economy’, the boundary
of the CBA’s quasi-market for policy choice proved
to be ‘not a line on a map, but a horizon that at
every point opens up into other territories’ (Mitch-
ell, 2002). The framing operation that brought into
existence both a newly centralized expertise on the
one hand, and a new market-inspired decision space
on the other, turned out to be inde?nitely problem-
atic. As Callon urges, ‘we need to ask what the cost
is, in terms of practical operations, for this neoclas-
18
Opinion pollsters also tend to attribute irrationality or
?ckleness to the public when their predictions fail, as they did
in the New Hampshire democratic primary election of 2008.
528 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
sical economics to gather the instruments, prosthe-
ses and devices needed to constitute the agencements
required in the emergence and survival of the very
particular (and highly improbable) agencies that it
promotes’ (Callon, 1998b).
Conclusion
‘Every category’, Bowker and Star (2005) write,
‘valorizes some point of view and silences another.’
The categories created through commensuration for
purposes of environmental accounting are no excep-
tion. For many years social scientists have enriched
their analyses of power by studying how new equi-
valences and omnibus categories are created that
compress space and time (Harvey, 1989), ‘simplify’
nature or culture in the service of the ‘high modern
state’ (Scott, 1999), create ‘bound and unbound seri-
alities’ (Anderson, 1999) and so on. Mingled with
such analyses have come a growing number of
investigations of ‘borderlands’ (Tsing, 2007) and
of limits to the ‘naturalization’ of categories (Bow-
ker & Star, 2005; Collins, 1992; Geertz, 1973;
Latour, 1994; Wittgenstein, 1953) and of commen-
suration (O’Neill, 2007; Rose, 1999; Thompson,
1990). Environmental accounting, with its relentless
generation of new equivalences and categories – as
well as resistances to them – deserves a prominent
place in such studies. Even on the narrowest concep-
tions of politics, environmental accounting both
comes out of politics (for example, political pres-
sures for a neoliberal ‘market solution’ to climate
change, competition between government agencies
and a need to shift accountability for public choice
to centralized, supposedly impersonal mechanisms)
and returns to it (for example, carbon accounting
institutions giving rise to local opposition, CBAs
being used for competing purposes). On a broader
conception, environmental accounting never left
politics; in its beginning is its end.
In exploring this theme, this article has distanced
itself from the questions of whether ‘internalizing’
the ‘externalities’ of climate or social welfare into
‘the economy’ is a solution or a contribution to eco-
logical crisis, and of whether it is better in the
abstract to ‘mend or end’ e?orts to ‘mainstream’ cli-
mate, welfare or environmental considerations by
engaging in, say, carbon o?set accounting and
cost–bene?t analysis. Instead, it has problematized
the ‘internal’/‘external’ metaphor itself in the course
of suggesting that there may be a more fruitful
debate to be had. For example: how disruptive
and damaging are the practical consequences of car-
bon accounting’s attempts to frame a new omnibus
category of ‘emissions reductions’? Of cost–bene?t’s
attempts to frame a new policymaking space and the
individuals that are to inhabit it? Are there better
ways for critics of environmental accounting to
argue that it is a central source of environmental cri-
sis than by warnings about the dangers of environ-
mental goods being assimilated into a sharply-
bordered, ever-expanding ‘market sphere’? Would
it not be more fruitful and historically informed
for proponents of increased environmental account-
ing, too, to abandon the metaphor in favour of a
programme for anticipating, evaluating and dealing
with the perpetually proliferating over?ows and
entanglements that, for example, carbon accounting
and cost–bene?t analysis entail?
Carbon accounting and cost–bene?t analysis,
this article has shown, are parallel in many ways.
Both are pioneer techniques, performing multi-
tudes of new spaces, subjects and objects in their
work to ‘make things the same’. Both are inde?-
nitely ambitious, calculating worlds with and with-
out certain policies or projects. Both will remain
perpetually incomplete while always holding out,
to some of their proponents, the promise of com-
pletion, thus encouraging inde?nite further invest-
ment in centres of calculation. Accordingly, both
require the contributions of ever-expanding bodies
of expertise, and the documentation connected
with each technique is invariably enormous. Both,
too, can be ‘adopted by many warring factions’
(Sen, 2001, p. 115), and crafted to yield vastly dif-
ferent results (Stirling, 1992; Sunstein, 2005, pp.
132–148). Yet while both techniques might seem
destined to reinforce the political power of
accounting expertise, each in fact helps engender
intractable reactions and resistances to itself.
Although each technique to a certain extent uses
the public’s distance from its centres of calculation
to ‘black-box’ areas of measurement controversy,
and thus maintain some public faith in the abstract
idea of computability, in both cases, the more inti-
mately acquainted people become with the relevant
accounting practices, the less plausible and more
contested they become.
Both carbon accounting and cost–bene?t analy-
sis work to frame major new market or purportedly
market-like spaces in which conventional distinc-
tions between physical science and economics and
among legal, political and price incentives become
blurred. Carbon trading’s requirement to commen-
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 529
surate disparate properties, actions and potentials in
the service of making carbon pricing possible paral-
lels CBA’s need to isolate well-behaved, commensu-
rable preferences in the service of calculating
welfare. Both imperatives generate zones of igno-
rance and ‘stupidity’ (Richardson, 2001) that are,
in the long term, di?cult to maintain given the goals
each technique was ostensibly designed to help
achieve. For example, carbon accounting’s indi?er-
ence to where or how emissions cuts are made
discourages attention to path dependence, positive
feedbacks and innovation; its con?ation of
reductions and o?sets leads to a running together
of probability with uncertainty, ignorance and inde-
terminacy; and its focus on means of achieving
short-term e?ciency obstructs social thinking about
long-term directions and the drawbacks of having to
monitor geographically distant e?ects. By the same
token, techniques for constructing preferences for
use in CBA typically exclude from their ‘epistemic
probe’ (Sen, 2001, p. 114) alternatives that require
public discussion to identify, while interpreting sub-
ject resistance as irrationality.
All translations have biases. So does the refra-
ming of disputes over environmental accounting
that has been suggested in this article. While this
reframing aims to stimulate new approaches in what
often seems a stalled debate, it is also intended to
provide suggestions for strategic self-interpretation
on the part of social activists, critical scientists and
other intellectuals disturbed by the erasures, con-
?icts and exaggerated claims generated by environ-
mental accounting techniques, or curious to see to
what extent the extension of such techniques can
be defended. Whether the article succeeds in this
job is a matter for them to judge.
Acknowledgement
I am grateful to the following institutions: the
Foundation for Ecological Recovery, Bangkok;
Yale University’s Program in Agrarian Studies,
which supported much of the research on cost-ben-
e?t analysis during 1996–1997; and Durham Uni-
versity’s Institute of Advanced Study, which
supported the workshop ‘Carbon Markets in Social
Science Perspective’, 7 November 2007, at which a
related presentation was made. I would also like
to express thanks to Joan Martinez-Alier, Wendy
Espeland, Robin Grove-White, Stephen Gudeman,
Steven Marglin, Martin O’Connor, John O’Neill,
Carol Rose, James C. Scott, Thomas Summerhill,
members of the Durban Group for Climate Justice
and the World Rainforest Movement, and my Cor-
ner House colleagues Sarah Sexton and Nicholas
Hildyard.
References
Adams, J. (1989). London’s green spaces: What are they worth?.
London: London Wildlife Trust and Friends of the Earth.
Adams, J. (1995). Cost–bene?t analysis: Part of the problem not
the solution. Oxford: Green College.
Adler, M. D., & Posner, E. A. (2001). Implementing cost–bene?t
analysis when preferences are distorted. In M. D. Adler & E.
A. Posner (Eds.), Cost–bene?t analysis: Legal economic and
philosophical perspectives (pp. 269–311). Chicago: University
of Chicago Press.
Adler, M. D., & Posner, E. A. (2006). New foundations of cost–
bene?t analysis. Cambridge, MA: Harvard University Press.
Anderson, B. (1999). The spectre of comparisons: Nationalism
Southeast Asia and the world. London: Verso.
Anderson, E. (1993). Value in ethics and economics. Cambridge,
MA: Harvard University Press.
Arthur, W. B. (1999). Increasing returns and path dependence in
the economy. Cambridge: Cambridge University Press.
Austin, M. (2007). Meeting the post-2012 challenge. Environ-
mental ?nance, November.
Backstrand, K., & Lovbrand, E. (2006). Planting trees to mitigate
climate change: Contested discourses of ecological modern-
ization, green governmentality and civic environmentalism.
Global Environmental Politics, 6(1), 50–75.
Bakan, J. (2004). The corporation. New York: Free Press.
Banuri, T., & Opschoor, H. (2007). Climate change and sustain-
able development. United Nations Department of Economic
and Social A?airs Working Paper No. 56, ST/ESA/2007/
DWP/56, October, New York: United Nations.
Barde, J. P., & Pearce, D. W. (Eds.). (1991). Valuing the
environment. London: Earthscan.
Barnard, P. (1992). So that’s your opinion, is it? London times, 20
March.
Barnes, P. (2001). Who owns the sky? Our common assets and the
future of capitalism. Washington: Island Press.
Berger, J. (1979). Pig earth. New York: Pantheon.
Bowker, G. C., & Star, S. L. (2005). Sorting things out:
Classi?cation and its consequences. Cambridge, MA: MIT
Press.
Bright, A. (2008). Buy now, pay later: Is it too late to buy o? our
carbon debt?. Good 8, 86–91.
Buck, D. (2006). The ecological question: Can capitalism survive?
In L. Panitch, & C. Leys (Eds.), Coming to terms with nature:
Socialist register 2007 (pp. 60–71) .
Burgess, J., Clark, J., & Harrison, C. M. (1998). Culture,
communication and the ‘Information Problem’ in contingent
valuation. Environment and Planning C: Government and
Policy, 18(5), 505–524.
Callon, M. (1998a). The embeddedness of economic markets in
economics. In M. Callon (Ed.), The laws of the markets
(pp. 1–57). Oxford: Blackwell.
Callon, M. (1998b). An essay on framing and over?owing:
Economic externalities revisited by sociology. In M. Callon
(Ed.), The laws of the markets (pp. 244–269). Oxford:
Blackwell.
530 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
Callon, M. (1999). Actor network theory: The market test. In J.
Law & J. Hassard (Eds.), Actor network theory and after
(pp. 181–195). Oxford: Blackwell.
Callon, M. (2005). Why virtualism paves the way to political
impotence. Callon replies to Miller. Economic Sociology
European Electronic Newsletter, 6(2), 3–20.
Carruthers, B. G., & Espeland, W. (1991). Accounting for
rationality: Double-entry bookkeeping and emergence of
economic rationality. American Journal of Sociology,
97.
Central and Eastern Europe Bankwatch (CEE) (2005). An
analysis of additionality. The PCF’s JI project in the
Czech Republic: Sixteen small hydropower plants. Prague:
CEE.
Channel 4 (UK). (2007). Dispatches: The great carbon smoke-
screen. London.
Clark, J., Burgess, J., & Harrison, C. M. (2000). ‘I struggled with
this money business’: Respondents’ perspectives on contin-
gent valuation. Ecological Economics, 33(1), 45–62.
Cline Cohen, P. (1982). A calculating people: The spread of
numeracy in early America. New York: Routledge.
Coase, R. H. (1960). The problem of social cost. Journal of Law
and Economics, 3, 1–44.
Coase, R. H. (1988). The ?rm, the market and the law. Chicago:
University of Chicago Press.
Cole, D. (2007). The stern review and its critics: Implications for
the theory and practice of bene?t–cost analysis. Paper pre-
sented at bene?t–cost analysis conference at the University of
Washington.
Collins, H. M. (1992). Changing order: Replication and induction
in scienti?c practice. Chicago: University of Chicago Press.
Comaro?, J., & Comaro?, J. (1991). Ethnography and the
historical imagination. Boulder: Westview Press.
Common, M. S. (1988). Longman.
Common, M. S. (1992). Letter to the editor. The Ecologist, 22(1),
39.
Cronon, W. J. (1991). Nature’s metropolis: Chicago and the Great
West. New York: Norton.
Dales, J. H. (1968). Land, water and ownership. Canadian
Journal of Economics.
Davidson, D. (1984). Essays on truth and interpretation. Oxford:
Oxford University Press.
Desrosieres, A. (1996). The politics of large numbers. Princeton:
Princeton University Press.
Desvousge, W. H. (1993). Measuring natural resource damages
with contingent valuation: Tests of validity and reliability. In
J. A. Hausman (Ed.), Contingent valuation: A critical assess-
ment. Amsterdam: North-Holland.
Dorfman, R. (1965). Measuring bene?ts of government invest-
ments. Washington: Brookings Institution.
Dove, M. (1999). Academic relations of production and cost–
bene?t analysis. Paper presented to the conference on cost–
bene?t analysis: Strategies and alternatives, Yale University,
October.
Driesen, D. M. (1998). Is emissions trading an economic incentive
program? Replacing the command and control/economic
incentive dichotomy. Washington and Lee Law Review, 55,
289–338.
Driesen, D. M. (2003a). The economic dynamics of environmental
law. Cambridge, MA: MIT Press.
Driesen, D. M. (2003b). Markets are not magic. The environ-
mental forum, November/December, (pp. 18–27).
Driesen, D. M. (2003c). Does emissions trading encourage
innovation?. Environmental Law Reporter News and Analysis
33, 10094–10108.
Drury, R.T., Belliveau, M. E., Kuhn, J. S. & Bansal, S. (1999).
Pollution trading and environmental injustice: Los Angeles’
failed experiment in air quality policy. Duke Environmental
Law and Policy Forum 9.
Dutschke, M. (2002). Fractions of permanence – Squaring the
cycle of sink carbon accounting. Mitigation and Adaptation
Strategies for Global Change, 7, 381–402.
Espeland, W. (1998). The struggle for water: Politics rationality
and idenitity in the American Southwest. Chicago: University
of Chicago Press.
Espeland, W. (1999).Value-matters. Paper presented to the
conference on cost–bene?t analysis: Strategies and alterna-
tives, Yale University, October.
Espeland, W., & Stevens, M. L. (1998). Commensuration as a
social process. Annual Review of Sociology, 24.
Evangelista, R. (2007). Entrevistas: Larry Lohmann. ComCien-
cia. Revista Eletronica de Jornalismo Cienti?co. Sao
Paulo, March. .
Fankhauser, S. (1995). Letter to the editor. The Ecologist, 25(4),
167.
Farber, D. A., & Hemmersbaugh, P. A. (1993). The shadow of
the future: Discount rates, later generations, and the envi-
ronment. Vanderbilt Law Review, 49, 267–304.
FASE et al. (2003). Open letter to executives and investors in the
Prototype Carbon Fund. Espirito Santo, 23 May.
Ferguson, J. (1994). The anti-politics machine: Development,
depoliticization and bureaucratic power in Lesotho. University
of Minnesota Press: Minneapolis.
Figueres, C. (2007). The CDM and sustainable development.
Environmental Finance, S50–S51.
Financial Times, 16 Feburary 2005.
Fischer, C. (2005). Project-based mechanisms for emissions
reductions: Balancing trade-o?s with baselines. Energy Policy,
33(14), 1807–1823.
Fischo?, B. (1991). Value elicitation: Is there anything in there?.
American Psychologist 46(8), 835–847.
Frank, R. H. (2001). Why is cost–bene?t analysis so controver-
sial?. In M. D. Adler & E. A. Posner (Eds.) Cost–bene?t
analysis: Legal, economic and philosophical perspectives
(pp. 77–94). Chicago: University of Chicago Press.
Geertz, C. (1973). Local knowledge: Further essays in interpretive
anthropology. New York: Harper and Row.
Ghosh, S., & Kill, J. (in press). The carbon market in India.
Kolkata: National Forum of Forest People and Forest
Workers.
Gneezy, U., & Rustichini, A. (2000). A ?ne is a price. Journal of
Legal Studies, 29(1), 1–17.
Gri?ths, T. (2007). Seeing REDD. ‘Avoided deforestation’ and the
rights of indigenous peoples and local communities. Oxford:
Forest Peoples Programme. .
Grubb, M., Vrolijk, P., et al. (1999). The Kyoto Protocol: A guide
and assessment. London: Royal Institute for International
A?airs.
Grubler, A., & Nakicenovic, N. (2001). Identifying dangers in an
uncertain climate. Nature, 412(6842), 15.
Guardian (London), 15 July 2007.
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 531
Hacking, I. (1995). The looping e?ect of human kinds. In D.
Sperber, D. Premack, & A. J. Premack, et al. (Eds.), Causal
cognition: An interdisciplinary approach. Oxford: Oxford
University Press.
Hahn, R. W. (2001). State and federal regulatory reform: A
comparative analysis. In M. D. Adler & E. A. Posner (Eds.),
Cost–bene?t analysis: Legal economic and philosophical per-
spectives (pp. 37–76). Chicago: University of Chicago Press.
Hall, J., Fu, G., & Lawry, J. (2007). Imprecise probabilities of
climate change: Aggregation of fuzzy scenarios and model
uncertainties. Climatic Change, 81, 265–281.
Hammond, R. J. (1960). Bene?t–cost analysis and water-pollution
control. Stanford: Food Research Institute.
Hamilton, K., Bayon, R., Turner, G., & Higgins, D. (2007). State
of the voluntary carbon market 2007: Picking up steam.
Washington and London: Ecosystem Marketplace and New
Carbon Finance.
Hanemann, W. M. (1994). Valuing the environment through
contingent valuation. Journal of Economic Perspectives, 8(4).
Hanke, S. H., & Walker, R. A. (1974). Bene?t–cost analysis
reconsidered: an evaluation of the mid-state project. Water
Resources Research, 10(5), 898–908.
Hanley, N. in association with ECOTEC. (1991). The valuation of
environmental e?ects: Stage II. Edinburgh: Scottish O?ce
Industry Department/Scottish Enterprise.
Hansen, J. E. (2007). Scienti?c reticence and sea level rise.
Environmental Research Letters, .
Haraway, D. J. (1995). Otherworldly conversations, terran topics,
local terms. In V. Shiva & I. Moser (Eds.), Biopolitics: A
feminist and ecological reader on biotechnology (pp. 69–92).
London: Zed Books.
Harremoe¨s, P. et al. (2002). Precautionary principle in the 20th
century: Late lessons from early warnings. London: Earthscan.
Harsanyi, J. G. (1955). Cardinal welfare, individualistic ethics,
and interpersonal comparisons of utility. Journal of Political
Economy, 63.
Harvard Law Review (1992). ‘Ask a silly question . . .’: Contin-
gent valuation of natural resource damages. Harvard Law
Review, 105, 1990.
Harvey, D. (1989). The condition of postmodernity. Oxford:
Oxford University Press.
Haya, B. (2007). Failed mechanism: How the CDM is subsidizing
hydro developers and harming the kyoto protocol. Berkeley:
International Rivers. .
Heinzerling, L. (2000). The rights of statistical people. Harvard
Environmental Law Review, 24, 189–220.
Heinzerling, L. (2002). Markets for arsenic. Georgetown Law
Review, 90, 2311–2339.
Helm, D., & Pearce, D. (Eds.). (1991). Economic policy towards
the environment. London: Croom Helm.
Herbst, S. (1993). Numbered voices: How opinion polling has
shaped American politics. Chicago: University of Chicago
Press.
Hitchens, C. (1994). For the sake of argument. London: Verso.
Holm, P. (2007). Which way is up on Callon? A review of a
review: Daniel Miller’s ‘Turning Callon the right way up’. In
D. MacKenzie, F. Muniesa, & L. Siu (Eds.), Do economists
make markets? On the performativity of economics. Princeton,
NJ: Princeton University Press.
Independent (London). (2007). Interview with Al Gore. 7 July.
Indian Express. (2005). UN environment scheme ?awed: CSE. 16
November.
Isenberg, A. (1949). Critical communication. Philosophical
Review, 58(4).
Jacobs, M. (1997). Valuation, democracy and decision-making.
In J. Foster (Ed.), Valuing nature? Economics ethics and
environment. London: Routledge.
Knight, F. (1921). Risk, uncertainty and pro?t. New York: Harper
and Row.
Kornhauser, L. A. (2001). On justifying cost–bene?t analysis. In
M. D. Adler & E. A. Posner (Eds.), Cost–bene?t analysis:
Legal, economic and philosophical perspectives (pp. 201–221).
Chicago: University of Chicago Press.
Krutilla, J. V., & Eckstein, O. (1958). Multiple-purpose river
development: Studies in applied economic analysis. Baltimore:
Johns Hopkins University Press.
Latour, B. (1994). We have never been modern. Cambridge, MA:
Harvard University Press.
Lazarus, M. (2003). The CDM quanti?cation challenge: Time for
a more standardised approach. Presentation at a World
Resources Institute/World Business Council on sustainable
development side event at the ninth conference of the parties
to the UNFCCC, Milan, 10 December.
Lempert, R. J., & Schlesinger, M. E. (2001). Climate-change
strategy needs to be robust. Nature, 412, 375.
Levy, D. S., Hammitt, J. K., Duan, N., Downes-LeGuin, T., &
Friedman, D. (1984). Conceptual and statistical issues in
contingent valuation: Estimating the value of altered visibility in
the Grand Canyon. Santa Monica: RAND.
Lewontin, R. C. (1995). Sex, lies and social science: An exchange.
New York Review of Books, 25, 43–44, May.
Lindblom, C. E. (1959). The science of ‘muddling through’.
Public Administration Review, 79–88.
Liro?, R. A. (1986). Reforming air pollution regulation: The toil
and trouble of EPA’s bubble. Washington: Conservation
Foundation.
Lohmann, L. (1991). Who defends biological diversity? Conser-
vation strategies and the case of Thailand. In V. Shiva, P.
Anderson, H. Schuecking, A. Gray, L. Lohmann, & D.
Cooper (Eds.), Biodiversity: Social and ecological perspectives
(pp. 77–104). World Rainforest Movement: Penang.
Lohmann, L. (1995). Pulp, paper and power: How an industry
reshapes its social environment. Sturminster Newton: The
Corner House. .
Lohmann, L. (1997). Whose voice is speaking? How opinion
polling and cost–bene?t analysis synthesize new ‘publics’.
Sturminster Newton: The Corner House. .
Lohmann, L. (2001). Democracy or carbocracy? Carbon trading
and the future of the climate debate. Sturminster Newton: The
Corner House.
Lohmann, L. (2005). Marketing and making carbon dumps:
Commodi?cation, calculation and counterfactuals in climate
change mitigation. Science as Culture, 14(3), 203–235.
Lohmann, L. (Ed.). (2006). Carbon trading: A critical conversa-
tion on climate change privatisation and power. Uppsala: Dag
Hammarskjold Foundation.
Lovbrand, E. (2004). Bridging political expectations and scienti?c
limitations in climate risk management – On the uncertain
e?ects of international carbon sink policies. Climatic Change,
67(2–3), 451–469.
532 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534
Lovell, J. (2007). Carbon price is poor weapon against climate
change. Reuters, 24 September.
MacKenzie, D. (2007). Is economics performative? Option theory
and the construction of derivatives markets. In D. MacKen-
zie, F. Muniesa, & L. Siu (Eds.), Do economists make markets?
On the performativity of economics. Princeton: Princeton
University Press.
MacKenzie, D. (2009). Making things the same: Gases, emissions
rights and the politics of carbon markets. Accounting,
Organizations and Society, 34(3–4), 440–455.
McCloskey, D. N. (1998). The rhetoric of economics. Madison:
University of Wisconsin Press.
Malloy, R. P. (2000). Law and market economy. Cambridge:
Cambridge University Press.
Marland, G. et al. (2001). Accounting for sequestered carbon:
The question of permanence. Environmental Science and
Policy, 4, 259–268.
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.
Michaelowa, A. (2007). Avoiding the carbon hangover. Carbon
Markets, 32–34.
Miller, P. (1999). In M. Callon (Ed.), The laws of the markets.
Oxford: Blackwell.
Miller, P. (2001). Governing by numbers: Why calculative
practices matter. Social Research, 68(2), 379–396.
Mitchell, T. (2002). Rule of experts: Egypt, techno-politics,
modernity. Berkeley: University of California Press.
Monbiot, G. (2006). Costing climate change. New International-
ist, 396, 30–31.
Nicholls, S. (2007). Interview with Hans-Juergen Stehr. Environ-
mental Finance, S42.
Nussbaum, M. (1986). The fragility of goodness: Luck and ethics
in greek tragedy and philosophy. Cambridge: Cambridge
University Press.
Nussbaum, M. (1990). Love’s knowledge: Essays on philosophy
and literature. Oxford: Oxford University Press.
Nussbaum, M. (1997). Flawed foundations: the philosophical
critique of (a particular type of) economics. University of
Chicago Law Review, 64(4), 1197–1214.
Nussbaum, M. (2001). The costs of tragedy: Some moral limits of
cost-bene?t analysis. In M. D. Adler & E. A. Posner (Eds.),
New foundations of cost-bene?t analysis. Cambridge, MA:
Harvard University Press.
Oakeshott, M. (1962). Rational conduct. In M. Oakeshott (Ed.),
Rationalism in politics. New York: Basic Books.
Olsen, K. H. (2007). The clean development mechanism’s
contribution to sustainable development: A review of the
literature. Climatic Change, 84, 59–73.
O’Neill, J. (1995). In partial praise of a positivist: The work of
Otto Neurath. Radical Philosophy, 74, 29–38.
O’Neill, J. (1997). Managing without prices: The monetary
valuation of biodiversity. Ambio, 26(8).
O’Neill, J. (2007). Markets, deliberation and environment. Lon-
don: Routledge.
Ott, H. & Sachs, W. (2000). Ethical aspects of emissions trading.
Wuppertal Papers No. 110. Wuppertal: Wuppertal Institut fu¨ r
Klima, Umwelt, Energie.
Pearce, D. (Ed.). (1991). Blueprint 2: Greening the world economy.
London: Earthscan.
Pearce, F. (2006). The last generation: How nature will take her
revenge for climate change. London: Random House.
Pearce, D., Moran, D. (1994). Letter to the editor. New Scientist.
Petersen, W. (1987). Politics and the measurement of ethnicity. In
W. Alonso & P. Starr (Eds.), The politics of numbers
(pp. 187–233). New York: Russell Sage Foundation.
Pielke, Jr., R. A., & Sarewitz, D. (2000). Winning and losing the
global warming debate. Earth A?airs.
Pittock, A. B., Jones, R. N., & Mitchell, C. D. (2001).
Probabilities will help us plan for climate change – without
estimates, engineers and planners will have to delay decisions
or take a gamble. Nature, 413(6853), 249.
Plott, C. R. (1993). Contingent valuation: A view of the
conference and associated research. In J. A. Hausman (Ed.),
Contingent valuation: A critical assessment. Amsterdam:
North-Holland.
Point Carbon. (2006). Executive board warns against unlevel
CDM playing ?eld. 16 May.
Polanyi, K. (2002). The great transformation . Boston: Beacon
Press.
Porter, T. M. (1995). Trust in numbers: The pursuit of objectivity
in science and public life. Princeton: Princeton University
Press.
Posner, R. (2001). Cost–bene?t analysis: De?nitions, justi?cation
and comment on conference papers. In M. D. Adler & E. A.
Posner (Eds.), New foundations of cost–bene?t analysis.
Cambridge, MA: Harvard University Press.
Power, M. (1988). The audit society. In A. G. Hopwood & P.
Miller (Eds.), Accounting as social and institutional practice.
Cambridge: Cambridge University Press.
Prins, G., & Rayner, S. (2007). Time to ditch Kyoto. Nature, 449,
973–975.
Radin, M. J. (1996). Contested commodities: The trouble with
trade in sex, children body, parts and other things. Cambridge:
Harvard University Press.
Raz, J. (1986). The morality of freedom. Oxford: Oxford
University Press.
Reklev, S. (2007). Cowboys or cavalry?. Carbon Markets 27–28.
Richardson, H. (1994). Practical reasoning about ?nal ends.
Cambridge: Cambridge University Press.
Richardson, H. (2001). The stupidity of the cost–bene?t stan-
dard. In M. D. Adler & E. A. Posner (Eds.), Cost–bene?t
analysis: Legal, economic and philosophical perspectives
(pp. 135–167). Chicago: University of Chicago Press.
Rorty, R. (1979). Philosophy and the mirror of nature. Princeton:
Princeton University Press.
Rorty, R. (1991). Objectivity, relativism and truth. Cambridge:
Cambridge University Press.
Rose, N. (1999). Powers of freedom: Reframing political thought.
Cambridge: Cambridge University Press.
Rosenthal, D. H., & Nelson, R. H. (1992). Why existence values
should not be used in cost–bene?t analysis. Journal of Policy
Analysis and Management, 11, 116.
Ross, A. (1998). Real love: In pursuit of cultural justice. New
York: New York University Press.
Rowe, R. D., D’Arge, R. C., & Brookshire, D. (1980). An
experiment on the economic value of visibility. Journal of
Environmental Economics and Management, 7.
Sago?, M. (1988). The economy of the earth. Cambridge:
Cambridge University Press.
Sarewitz, D. (1996). Frontiers of illusion: Science, technology and
the politics of progress. Philadelphia: Temple University Press.
L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534 533
Sarewitz, D. & Pielke, R. (2007). The steps not yet taken. Ms.
Schlup, M. (2005). The gold standard, linking the CDM to
development and poverty reduction. In Presentation at
conference on climate or development. Hamburg Institute of
International Economics, 28–29 October.
Schneider, S. H. (2001). What is ‘dangerous’ climate change?.
Nature 411(6833), 17–19.
Schneider, S. H. (2002). Can we estimate the likelihood of
climatic changes at 2100?. Climatic Change 52(4), 441–451.
Scott, J. C. (1976). The moral economy of the peasant. New
Haven: Yale University Press.
Scott, J. C. (1999). Seeing like a state. New Haven: Yale
University Press.
Self, P. (1975). Econocrats and the policy process. Boulder:
Westview.
Sen, A. K. (1977). Rational fools. Philosophy and Public A?airs,
6, 317–339.
Sen, A. K. (2001). The discipline of cost–bene?t analysis. In M.
D. Adler & E. A. Posner (Eds.), Cost–bene?t analysis: Legal,
economic and philosophical perspectives (pp. 95–116). Chicago:
University of Chicago Press.
Sen, A. K., Das Gupta, P., & Marglin, S. (1972). Guidelines for
project evaluation. New York: United Nations Industrial
Development Organization.
Shackley, S., Young, P., Parkinson, S., & Wynne, B. (1998).
Uncertainty, complexity and concepts of good science in
climate change modelling: Are GCMs the best tools?. Climatic
Change 38(2), 159–205.
Shanks, B. (1974). The American Indian and Missouri river
water developments. Water Resources Bulletin, 10(3),
559–573.
Star, L., & Griesemer, J. (1989). Institutional ecology, ‘transla-
tions’ and boundary objects: amateurs and professionals in
Berkeley’s Museum of Vertebrate Zoology, 1907–39. Social
Studies of Science, 19, 387–420.
Stern, N. (2007). The economics of climate change: The stern
review. Cambridge: Cambridge University Press. .
Stirling, A. (1992). Regulating the electricity supply industry by
valuing environmental e?ects. Futures, 24(10).
Sunstein, C. R. (1993). Endogenous preferences, environmental
law. Journal of Legal Studies, 22, 217–231.
Sunstein, C. R. (2005). Laws of fear: Beyond the precautionary
principle. Cambridge: Cambridge University Press.
Suptitz, A. P. L. et al. (2004). Open letter to the Clean
Development Mechanism Executive Board, Minas Gerais,
June.
Sutter, C., & Parren˜ o, J. C. (2007). Does the current clean
development mechanism (CDM) deliver its sustainable devel-
opment claim? An analysis of o?cially registered CDM
projects. Climatic Change, 84, 75–90.
Taylor, M. et al. (2005). Regulation as the mother of invention:
The case of SO
2
control. Law and Policy, 27, 348–378.
Thompson, E. P. (1990). Customs in common. New York: Free
Press.
Titmuss, R. M. (1996). The gift relationship: From human blood to
social policy. London: London School of Economics Press.
Trexler, M. C., Broekho?, D. J., & Koslo?, L. H. (2006). A
statistically-driven approach to o?set-based GHG addition-
ality determinations: What can we learn?. Sustainable Devel-
opment Law and Policy 6(2), 30–40.
Tsing, A. L. (2007). Friction: An ethnography of global connection.
Berkeley: University of California Press.
Turner, R. K. (1991). Environment, economics and ethics. In D.
Pearce (Ed.), Blueprint 2: Greening the world economy.
London: Earthscan.
Ungphakorn, P. M. (1988). The cost of the Nam Choan dam. The
Nation [Bangkok], 16, March.
United Nations Framework Convention on Climate Change.
Executive Board. (n.d.). EB 36 Report, Annex 13: Method-
ological tool: Tool for the demonstration and assessment of
additionality (version 04).
Unruh, G. C. (2000). Understanding carbon lock-in. Energy
Policy, 28, 817–830.
Vadnjal, D., & O’Connor, M. (1994). What is the value of
Rangitoto Island?. Environmental Values 3.
Van Vliet, O. P. R., Faaij, A. P. C., & Dieperink, C. (2003).
Forestry projects under the clean development mechanism.
Climatic Change, 61(1–2), 123–156.
Vatn, A., & Bromley, D. W. (1994). Choices without prices
without apologies. Journal of Environmental Economics and
Management, 26, 129–148.
Verchick, R. R. M. (2005). The case against cost bene?t analysis.
Available at SSRN: .
Viscusi, W. K., Magat, W. A., & Huber, J. (1987). An
investigation of the rationality of consumer valuations of
multiple health risks. Rand Journal of Economics, 18(4),
465–479.
Viscusi, W. K., & Zeckhauser, R. J. (2003). Sacri?cing civil
liberties to reduce terrorism risks. Journal of Risk and
Uncertainty, 26, 99–109.
Walzer, M. (1983). Spheres of justice. New York: Basic Books.
Wara, M. (2007). Is the global carbon market working?. Nature
445, 595–596.
Welch, D. (2007). A buyer’s guide to o?sets. Ethical Consumer,
106.
Wiggins, D. (1987). Needs, values, truth. Oxford: Oxford
University Press.
Williams, R. (1973). The country and the city. Oxford: Oxford
University Press.
Wittgenstein, L. (1953). Philosophical investigations. Cambridge:
Cambridge University Press.
World Bank (1992). World development report 1992: Development
and the environment. New York: Oxford University Press.
World Bank (2007). State and trends of the carbon market 2007.
Washington, DC: World Bank.
Zelizer, V. (1985). Pricing the priceless child: The changing social
value of children. New York: Basic Books.
Zelizer, V. (1997). The social meaning of money: Pin money, pay
checks, poor relief and other currencies. Princeton: Princeton
University Press.
534 L. Lohmann / Accounting, Organizations and Society 34 (2009) 499–534

doc_174947704.pdf