Order out of chaos:
Public and private rules for managing carbon
Prepared for the 2011 Meeting of the American Political Science Association
1-4 September, Seattle, WA
Jessica F. Green1
In their book Local Commons and Global Interdependence, Keohane and Ostrom postulate that
the number of actors as well as their heterogeneity are key factors in explaining the likelihood of
collective action – both at the global and local levels. Their original theoretical model, which
considers the prospects for cooperation at a given point in time, treats the key focal variables as
exogenous.2 Thus, the heterogeneity of actors is taken as given and characterized in specific
ways; their interest is in the effects of this heterogeneity on their dependent variable, collective
In this paper, I seek to build upon and broaden the authors‘ work by further probing one of their
independent variables, actor heterogeneity. I expand the notion of actor heterogeneity to include
private actors—NGOs, firms and transnational networks, among others. I look not only at these
private rule-makers, but also at the interactions among privately-created and public rules. In this
paper, actor heterogeneity comprises not only the interactions among state actors, but also among
private actors, and between public and private actors. Broadening the concept of actor
heterogeneity to include private actors, produces new effects on collective action, unexplored in
Keohane and Ostrom‘s earlier work. 3 This paper uses the issue of the management of
greenhouse gases (GHGs) to understand the effects of actor heterogeneity on the prospects for
collective action to manage and mitigate GHG emissions.
I argue that uncertainty—about the quality of other privately-created rules, the fluid nature of
carbon governance and the status of future regulation—is a key factor in the interactions among
heterogeneous actors. Uncertainty about the evolution of carbon markets and the quality of other
existing rules has led private actors to hedge through the recognition of multiple standards, and
in particular the rules about carbon offsets created by the Clean Development Mechanism of the
Kyoto Protocol. Uncertainty about the quality of other standards has led many private standards
to associate themselves with those known to be of high quality. Each standard aims to stake out
a niche in the emerging regulatory environment, but also wishes to ensure their continued
relevance in the face of future changes. Both types of uncertainty promote hedging behavior:
1 Assistant Professor, Case Western Reserve University. Comments are welcome at firstname.lastname@example.org. This
paper has benefitted from comments on earlier versions by Josh Busby, Bob Keohane, Ron Mitchell and participants
at the 2011 Princeton Conference on International Relations and Climate Change and at the Duke University
seminar on Global Governance and Democracy.
2 Keohane and Ostrom 1995, p. 16.
3 Mitchell 1995 also includes non-state actors in his analysis, though as the targets of regulation rather than as the
initiators of it.
carbon standards seek to maximize their flexibility through recognition of other standards—both
public and private.
When private actors are included in the mix, uncertainty about the future and others‘ behavior
leads to complex interactions among public and private actors, which produce surprising results.
Actor heterogeneity has two unanticipated effects that serve to promote collective action. First,
the diversity of actors has given rise to healthy competition, producing robust sets of rules in the
landscape of climate governance. Despite concerns about the ―wild west‖ of carbon markets, the
dynamic effects of interaction over time has produced clear leaders in the market. Second, the
presence of many private standards concerned about future regulation has produced a degree of
convergence among the myriad rules and standards. This convergence has reaffirmed the
centrality of public authority, in spite of the undetermined future of intergovernmental rules.
Even if Kyoto fails, the rules of the CDM will almost certainly live on through private standards.
The effect of actor heterogeneity, then, is to reinforce the stature of public authority. These
findings suggest that under conditions of uncertainty, private actors can contribute to a more
cohesive set of rules and practices with respect to the measurement and reduction of GHG
emissions—potentially a positive outcome for collective action.
The paper proceeds as follows. I first define actor heterogeneity and relate it to a broader
conception of institutional heterogeneity. I then review two bodies of literature that provide
accounts of institutional heterogeneity: rational design and regime complexity. Based on these
literatures, I develop expectations about the relationship between uncertainty and actor
heterogeneity. In the third section, I describe actor heterogeneity, and present data on the private
rules related to GHG measurement and management. The fourth section evaluates the
hypotheses using network analysis and qualitative data collected through interviews.
I. Defining heterogeneity
In their discussion of heterogeneity, Keohane and Ostrom focus primarily on heterogeneity
among actors as determined by differences in capabilities, preferences and information.4 This
paper further specifies their definition in one important way. I consider an additional dimension
of actor heterogeneity: whether those involved are primarily states or non-state actors.5 Whereas
Keohane and Ostrom are agnostic about whether the actors involved are states, sub-national
government actors or non-state actors, I specify the degree of ―publicness‖ as an important
dimension of heterogeneity. I distinguish between traditional forms of public authority at the
international level which include inter alia multilateral treaties, international organizations and
bilateral initiatives and private authority, which includes rules and standards created by non-state
actors. Thus, when I refer to actor heterogeneity, I consider the panoply of public, private and
4 Keohane and Ostrom 1995, pp. 6-10.
5 A number of authors have pointed out that public is no longer a dichotomous variable, since many institutions are
combinations of different kinds of public and private actors. See, for example, Andonova et al 2009, Fransen 2011.
I agree wholeheartedly with this critique; as such, I use the public/private distinction as a rough metric,
acknowledging that it is a continuous rather than dichotomous variable. As is convention, I consider international
organizations to be a type of public actor.
―hybrid‖ units dealing with emissions management.6 Keohane and Ostrom do not exclude the
possibility of private or hybrid actors, yet they do not explicitly consider this form of variation.7
The empirical focus is on the variety of institutions created to measure, report and offset carbon
emissions. These institutions vary from wholly public to wholly private (i.e. not created by
governments). I consider institutional heterogeneity the logical outgrowth of actor
heterogeneity: with different kinds of actors come different kinds of institutions, which can be
understood as ―sets of rules that meant to govern…behavior.‖8 They will be described in greater
detail in section three.
II. Literature Review
In some ways, the study of globalization can be understood as the study of heterogeneity. The
increased flows of people, goods and ideas have been accompanied by a proliferation of actors
and institutions to manage those flows. The IR literature has devoted considerable effort to
explaining both the emergence of and variation across these institutions. In this section, I focus
on two bodies of literature in particular that seek to explain institutional heterogeneity: rational
design and regime complexity.
I begin with rational design (RD) theory, which seeks to explain variation in international
institutions. Why do some institutions centralize decision-making, while others delegate it?
Why do some institutions include ―escape clauses‖ for certain situations, while others rigidly
apply the rules?9 The answer put forth by rational design theory is that institutional variation can
be explained by the purposive actions by states to solve specific problems. Institutions exist to
reduce transaction costs, and thus, their design varies with the type of problem or market
Thus, institutional design is not simply the reflection of the distribution of power among states,
but rather self-conscious strategies to achieve desired outcomes.11 The seminal work by
Koremenos et al. identifies four groups of independent variables that account for variation in
institutional design: distribution problems, enforcement problems, the number of actors and a
variety of types of uncertainty.12 Uncertainty is a key variable in the RD literature and comes in
three stripes: uncertainty about the state of the world, about others‘ behavior, and about others‘
Koremenos et al. posit that flexibility of the institution will increase with uncertainty about the
state of the world.13 Of course, their investigation is limited to the world of states – how states
design international institutions. They do not consider the possibility, as I do below, that
flexibility can also be achieved through other types of institutions, not exclusively comprised of
6 Abbott and Snidal 2009, Andonova 2011.
7 Mitchell‘s chapter in the volume does however consider differences in capacities and incentives across state and
8 Simmons and Martin 2002, p. 194
9 Rosendorff and Milner 2001.
10 Martin 1992, Koremenos et al. 2001.
11 The classic refutation of this view is Mearsheimer 1994/95.
12 Koremenos et al. 2001, pp. 773-79.
13 Ibid, p. 793.
or created by states. Nonetheless, the important point for the purposes of this investigation is the
key role of uncertainty in explaining actor, and therefore institutional, heterogeneity.
More recently, Thompson has applied RD theory to the climate change regime. He argues that
uncertainty about the regional and local impacts of climate change,14 and the costs and benefits
of different policy options led states to negotiate a flexible agreement. Specifically, ―means
flexibility‖ allows states to meet their commitments under the Kyoto Protocol through a variety
of measures including: domestic reductions, carbon offsets at home and abroad and emissions
trading.15 This flexibility, he argues, allows states the ability to adjust their policy responses in
accordance with new information. Thompson‘s arguments are again largely focused on states’
responses to uncertainty, and the type of outcome is the same: uncertainty gives rise to
heterogeneity—multiple approaches to addressing climate change.
Regime Complexity Theory
Regime complexity theory (RC) has recently evolved as a way to understand the causes and
effects of the proliferation of international institutions. Because of the increasing number and
density of institutions, Raustiala and Victor argue that one must study the ensemble of
institutions working in a given issue area in their totality, rather than examining each one
individually. This m(or)ass of institutions constitutes a regime complex: ―an array of partially
overlapping and nonhierarchical institutions governing a particular issue-area.‖16 In short, RC
theory is a way of understanding institutional heterogeneity as constituted by a group of related
institutions which are treated as a single unit. RD theory, by contrast, seeks to explain
heterogeneity across distinct institutions.
Uncertainty plays an important role in explaining the provenance of the regime complex.
Keohane and Victor argue that the degree of fragmentation or centralization in governance
arrangements can be attributed to three generic forces: distribution of interests, uncertainty and
the search for useful ways to link various issues.17 Uncertainty comes in various forms. States
may be uncertain about the costs and benefits of their actions—a calculus which is inextricably
tied to uncertainty about the behavior of others. Complete cooperation will enhance benefits,
while free-riding by others will increase costs. The response to this uncertainty, they argue, is
often the creation of clubs, where the commitments of a select few are more credible. The
unknown of future regulation also affects the cost-benefit calculus. Exposure to future regulation
may provide an impetus for action to mitigate future costs ex-ante.
The regime complex literature also speaks the effects of regime complexity. The first, and most
thoroughly examined effect, is the phenomenon of ―shifting.‖ When there are many institutions
to choose from, states and other actors can pick the one most hospitable to their aims. This can
come in the form of regime shifting, forum shopping, or the selection of one type of regulation
over another.18 There are other feedback effects, including competition among regimes, or
―reverberation‖—where activities in one forum have (usually unintended) consequences in
14 This is akin to ―uncertainty about the state of the world‖ as put forth by Koremenos et. al. 2001, p. 778.
15 Thompson 2010, p. 281.
16 Raustiala and Victor 2004, p. 279.
17 Keohane and Victor 2011, pp. 12-14.
18Helfer 2004, Hafner-Burton 2009, Kelley 2009.
another.19 This is what we will observe in the area of GHG management. This strand of the
literature treats the regime complex as the independent variable, explaining a variety of political
and institutional outcomes. By explaining and tracing the effects of institutional heterogeneity
with respect to carbon management, this paper builds on this body of work.
Though the unit of analysis is different, the two bodies of literature share a causal story:
uncertainty about the effects of certain policies and the actions of others plays a key role in the
institutional outcomes. In the case of the regime complexity literature, uncertainty relates to
costs and benefits of collective action, conditioned by the concern about the behavior of other
states. This result is an array of non-hierarchical loosely coupled institutions, which, in the case
of climate change (and many other issue areas), includes a variety of privately-created efforts.
In the rational design literature, uncertainty about the effects of climate change in the future
gives rise to an institution that promulgates a variety of policies to achieve the same outcome.
In both cases, uncertainty about the world gives rise to institutional heterogeneity—both within
and across institutions. In the following section, I describe this institutional heterogeneity with
respect to standards that measure, report and offset carbon emissions. I also derive observable
implications of the hypothesis set forth in both the RC and RD literatures relating uncertainty to
III. Actor heterogeneity: Private standards for carbon measurement
Addressing climate change involves many different types of policies and actors. Generally, the
focus has been on the complex landscape of public efforts, at the national and international
levels.20 In this section, I expand the notion of actor heterogeneity by describing privately
created standards with respect to the management of GHGs.21 Private actors include NGOs,
private firms and transnational networks. I exclude international organizations and national
governments, while acknowledging that the boundaries between public and private are not
always neatly drawn.
Mitigation efforts are premised on the ability to measure emissions. We need to know either the
amount of GHGs being emitted, or be able to calculate the amount being ―saved‖ (i.e. prevented
from being emitted) due to some activity. Just as banks need robust ways to account for gains
and losses, so too do states need credible, accepted procedures for calculating their emissions.
Moreover, since states are by no means the only entities taking action to mitigate climate change,
such needs are distributed across many types of actors and levels of governance.
The tools for counting carbon are similarly diverse. Households will calculate their emissions
differently from firms or states. Partially as a result of these different needs, different
measurement tools, or standards, have emerged. Yet there is not a one-to-one correlation
between the type of standard and its users. Instead, we see a variety of standards for
measurement at the transnational level. Some of these have been created by governments, but
there is also a strong presence of privately-created standards. I turn now to a description of this
19 Alter and Meunier 2009, pp. 19-20, and the special issue. For a critique, see Auld and Green 2011.
20 Keohane and Victor 2011.
21 There is a small, but growing literature that highlights the role of non-state actors and the variety of governance
arrangements with respect to climate change. See, e.g. Bernstein et al. 2010, Newell and Paterson 2010, Biermann
and Pattberg 2008, Okereke et al. 2008.
institutional heterogeneity. I focus first on the private standards, and then, in the following
section, turn to an analysis of the relationship among public and private standards.
The data presented here comprise all of the private transnational standards for carbon
measurement that are currently in use.22 The transnational criterion means that the standard must
function in at least two countries; in other words, standards that operated in one country are not
included in the data set.23 I exclude those created by governments or international organizations,
though these will be considered in the following section. Although there are many standards that
include greenhouse gases among other metrics of sustainability, I include only those whose
primary aim is carbon measurement and management. Those that apply more generally to
energy efficiency, sustainability or other environmental issues are not included.24
I compiled this data by triangulating among several sources, and have updated the data as new
sources have become available.25 Table 1 (at the end of the document) lists all the standards in
the dataset. The dataset presents basic information on these standards: the year they were created,
their organizational goal, and whether they require third party verification. I address each in turn.
Figure 1 shows the cumulative number of private standards by year. There is a clear and steady
increase in the creation of standards over time. The majority of standards, 73%, were created
between 2005 and 2009, after the Kyoto Protocol entered into force. The Protocol is set to expire in
2012, creating considerable uncertainty about the permanence of the current regulatory arrangements.
The spike in private standards after 2005 provides preliminary evidence that private actors responded
to the finite lifespan of the Kyoto Protocol by creating additional rules that could possibly outlast the
22 This data was originally compiled in February 2010, and updated in August 2011. By triangulating among
multiple sources, I have reasonable confidence that this list is exhaustive. According to the criteria set out in the
coding instrument I exclude: national level standards (unless they require offsets in other countries), standards
promulgated by governments, and those created and applied to a single firm‘s activities. Finally, I focus exclusively
on standards that measure, sell or abate greenhouse gas emissions. Those that apply more generally to energy
efficiency are not included, unless they have specific provisions for GHG measurement. A full list of coding rules is
available from the author upon request.
23 The ―two-country‖ criterion can be met in two ways. It can mean that the standard is used in multiple countries.
Alternatively, the two-country criterion can mean that in order to apply the standard, certification and verification
activities must take place in multiple countries. For example, the standard may be used by a firm in Canada, offsets
are generated by a project in Brazil. If the standard requires activities in both the producer and the consumer nation,
then the two country criterion is met.
24 These decisions were made by the author based on extensive research of public and private environmental
25 The sources include Abbott 2010, www.ecolabelindex.org (last accessed June 2011), Bulkeley 2010, Hale and
Roger 2011, Hamilton et. al 2009, Kollmuss et al. 2008.
Figure 1: Creation of Climate Standards, Cumulative
Figure 2 shows the different types of standards, categorized by their functions.26 Standards have
four different functions. First, offset standards provide rules for measuring avoided emissions
(and in some cases, other environmental co-benefits) of carbon mitigation projects. The emphasis
on offset standards is the methodology for calculating emissions reductions. Second, accounting
standards provide a protocol for actors to measure their emissions. The focus with accounting
standards is on total emissions, rather than reductions. Third, transparency standards provide a
centralized repository for users to report their emissions to others. Accounting and transparency
standards differ in one important respect. Accounting standards are first-party standards: they are
used by actors who wish to catalogue information on their GHG emissions, with no requirement
to make such information public. Transparency standards are third-party standards because they
require that emissions information be made public by a third-party. These are often emissions
registries, where entities calculate their emissions and make them publicly available in one
central place. The Climate Registry, for instance, provides a platform for participating entities to
publicly report their GHG emissions. Fourth and finally, there was one private standard that
created a market: the Chicago Climate Exchange established a system for quantifying and
trading carbon emissions. However, as of January 2011, the market is being phased out and
replaced with an offsets program. There is, of course, overlap among these different functions;
the goal is to provide a sense of how institutional heterogeneity manifests itself in terms of the
different types of standards.
26 I allow for standards to have multiple goals, and code them as such. Hence, N>30.
Figure 2: Private climate standards, by function
Figure 2 yields two interesting findings. First, private actors are typically not in the business of
creating carbon markets, despite the fact that we tend to associate these two. Standard arguments
for the ―retreat of the state‖ focus on the ways in which markets ―are the masters over the
governments of states.‖27 Yet as Figure 2 shows, markets are rarely being created by private
actors. Creating markets for GHGs is largely the province of public authority.
Figure 2 also shows where private actors are especially active: they are in the business of
creating offset standards. Offsets can exist without markets, but in such cases, must rely on
altruistic consumers to purchase them. Coupled with markets however, they become an
important tool for market participants to achieve compliance. As the discussion below shows, a
very large proportion of private standards are linked to markets, suggesting that altruism alone is
not the dominant driver in the growth of private standards.
Fully 80% of private standards require third-party verification. This finding shows that most
standards care about credibility and compliance, or at least wish to appear that they do. There is
some helpful variation across organizational goals. The majority of offset standards require third
party verification (89%). By contrast, only 67% of transparency standards require third-party
verification. This is unsurprising: transparency standards have the straightforward simple goal of
getting emitters to report. This is a much shallower form of cooperation, where free-riding is not
an issue. Thus, third-party verification is less important. Of course, verification of these reports is
desirable; however, if the goal is incremental—to get organizations to begin thinking about and
measuring their emissions—required verification might scare off the more reluctant participants.
In a word, transparency standards aim to ensure broad participation; thus cooperation is
27 Strange 1996, p. 4.
By contrast, since offsets are essentially creating a currency from an intangible good (the
removal of emissions), verification is a key element for credibility.28 Without verification that
offsets have been generated, the issuing of credits is akin to printing money—creating a
commodity with no physical value. Unverified projects allow for the possibility that the offset
credits grant have no correlation to the actual physical reductions. This is not only inefficient in
terms of combating climate change, but also has the effect of deflating the value of the
―currency.‖ Thus, third-party verification is an important signal in offset standards, to ensure the
value of the commodity being purchased. Put another way, since free-riding is possible (which is
not the case for transparency standards), verification is needed.29
IV. The effects of heterogeneity
The general hypothesis put forward in this paper is that uncertainty about the state of the world,
coupled with institutional heterogeneity can have positive effects on collective action. Linking
this explicitly to private standards, we should therefore see private carbon standards adopt
strategies to help them cope with uncertainty about states‘ actions, and the behavior of other
private actors in the institutional landscape. In turn, this should lead some rough agreement on
rules for carbon management—one mode of collective action to address climate change.
There are two main observable implications of the hypothesis. First, and following the RD
literature, we should expect these private standards to exhibit high amounts of flexibility, which I
operationalize as recognition of other standards. If actors within the regime complex are
concerned about the actions of others, they will try to ―hedge their bets‖ by maximizing
compatibility with other standards. In other words, they will leverage network effects through
recognition of other standards. Moreover, to the extent that they are concerned about
regulatory uncertainty—that is, the trajectory of the Kyoto negotiations—we should expect to
see compatibility between private and public standards. Conversely, if private actors are not
concerned about the uncertainty surrounding others‘ actions, then we should see low levels of
interconnectivity within the network. Second, if private actors are concerned about the
indeterminate the quality of other private standards, we should observe competition among them
which eventually produces leaders among the complex landscape of private standards. In fact,
the analysis below demonstrates that both effects are observed.
To analyze the effects of heterogeneity, I draw on two sources of data. First, I map the
connections among standards through network analysis to show the degree of flexibility of
individual standards—their response to uncertainty. Second, I augment the network analysis
with data gathered from the standards organizations through interviews. These semi-structured
interviews asked basic questions about motivations for the creation of the standard, and
considerations about compatibility.
Figure 3 shows the network of carbon standards (on page 17). The network is based on the
original 30 private standards, presented in the previous section; it then adds all of the other
28 Victor and House 2004.
29 Of course, this does not preclude the possibility that verified offsets fall victim to the same problem: currency is
created without corresponding (or sufficient) reductions. The incentive for verifiers to sign off on questionable
credits has been point out a number of critics including Green 2008 and Wara 2007.
standards that these original 30 recognize. Whereas there are 30 private transnational standards
measuring carbon, there are 41 standards in the network. These include all of the standards which
the 30 above state that they recognize, including public standards.30
Recognition operates in two ways. A standard may accept another standard. For example, the
Carbon Neutral standard certifies entities that have offset 100% of their emissions. It accepts offsets
from a variety of other private standards including the Gold Standard, the Chicago Climate Exchange
and the Clean Development Mechanism, among others. Alternatively, Standard A may appropriate
the rules and practices of Standard B.31 The Climate, Community and Biodiversity standard can be
combined with a number of other offset standards to ensure that benefits beyond emissions reduction
are achieved. Both acceptance of other standards, and appropriation of parts of other standards
constitute forms of recognition. 8t is important to note that this operationalization does not require
mutual recognition. Standard A may recognize standard B in one of the two specified ways, but
Standard B may not recognize Standard A. Nonetheless, this one-way relationship is still coded as
recognition. In the parlance of network analysis, this is a one-mode directed network.
Uncertainty in the regulatory market
Descriptive analysis helps to understand the degree to which standards adopt a strategy of
maximizing flexibility. If the hypothesis is correct, we should expect to see a dense network,
with high levels of recognition across all of the standards. In fact, the density of the network is
rather low, at 9.2%. This means that only 9.2% of all possible ties among the standards are
present. Despite the low level of overall density, there are some standards in the network that are
much better connected than others. An hypothesis test of the density against a null of zero
density reveals a significant t-score of 5.53.32 Thus, the distribution of ties to specific actors is
not random, but rather suggests some conscious strategy by private actors to connect to others in
the network. This provides preliminary evidence that although there is an explosion in the
number of private standards post-Kyoto, there are clearly some who have emerged as more
important—and indeed more credible—than others.
To identify the important actors in the network, I turn to the centrality of individual standards.
Because this is a directed network, we can examine who is being chosen by others, a measure of
prestige, as well as who is doing the choosing. Choosers are important for diffusing regulatory
practices, whereas those being chosen are important for the rules they create. I turn first to the
prestigious actors—who is being chosen—and their characteristics. Table 2 (on page 18) shows
normalized values for both indegree centrality (prestige) and outdegree centrality (diffusion).
Figure 3 shows that the Clean Development Mechanism is clearly a key node in the network.
The CDM is one of the three market mechanisms created by the Kyoto Protocol, allowing
developed countries to fund offset activities in the developing world to help meet their emissions
30 The standards named by the original 30 constitute the ―cutpoint‖ for the network. I do not include the standards
that are recognized by the ―second-level‖ standards. This decision was made for manageability of the data rather
than any research-based rationale.
31 I operationalize recognition by examining the document(s) which outline the details of the standard. If this
document specifically invokes the rules or practices of other standards or their verifiers, I code this as recognition.
The full coding instrument is available from the author upon request.
32 Because observations are not independent, I use the ―bootstrap method‖ in UCINET. See
http://faculty.ucr.edu/~hanneman/nettext/ for an explanation of the different approaches.
targets. Quantitatively, it also has the highest level of prestige, as measured by its degree
centrality. 37.5% of all standards have chosen to recognize the CDM‘s standards. Moreover, if
the population is restricted to only those standards that include offsetting as an organizational
goal the proportion rises to 79%. The CDM is prestigious in the sense that many other standards
are choosing it.
The high level of recognition of this public standard can be understood as a response to
regulatory uncertainty. Since it is unclear whether the CDM will remain in effect after the expiry
of the Kyoto Protocol, maintaining compatibility with its standards is a reasonable strategy to
ensure the continued relevance of private standards in the event that the CDM continues to
Another interesting potential effect of the CDM‘s prestige is the likelihood that the standards set
by the CDM will persist, even if the institution itself does not. It appears that the CDM‘s rules
will be perpetuated through privately-created offset standards. This suggests the possibility of
long term residual effects of public rules within a heterogeneous set of actors and institutions.
Even if the intergovernmental process continues to deteriorate, the strength of publicly-created
rules is being reaffirmed and embedded through private actors.
Interestingly, although the CDM is the most prestigious of all the standards in the network, it is
the anomaly among the public standards in the network. The dark circles in Figure 3 are
national and international public carbon standards. There are only five named by the original 30
private standards: the Clean Development Mechanism, the EU Emissions Trading Scheme,
Climate Leaders (a voluntary program created by the US EPA), the Regional Greenhouse Gas
Initiative (an emissions trading system in the Northeast of the US), and the Intergovernmental
Panel on Climate Change.33 All except for Climate Leaders have prestige values below the
Since there is little uncertainty associated with these public standards, the low level of prestige of
the other public standards provides additional support for the regulatory uncertainty explanation.
There is no need to hedge when the trajectory of these standards is clearly established. Put
another way, there appears to be no demand for private authority in these instances. The EU-
ETS, despite implementation challenges, is firmly established, as are the myriad accounting
procedures established by the IPCC.34 Two other public standards are on their way out: Climate
Leaders is being phased out and the Chicago Climate Exchange is transitioning from a market to
an offsets program. The Regional Greenhouse Gas Initiative was created in spite of the
intransigence of the US federal government. Although it is pursuing links to other regional
programs in the US, this process is still in its preliminary stages. Thus, it has yet to establish
itself as a well-functioning and prominent set of rules. Hedging through recognition would be
premature at this point.
33 Among other duties, the IPCC has created numerous methodologies for carbon measurement.
34 The IPCC accounting practices present a slightly different story. They are frequently used for basic calculations
such as emissions factors. Because these are widely accepted, but often used in small pieces, the IPCC is not
frequently named by other standards.
Uncertainty about the quality of other standards
Clearly, private standards are seeking to maximize their flexibility with respect to public rules
under the CDM. But this fact does not explain recognition among private rules. Surely, these
private standards are unlikely to become public law. This raises the question: why recognize
other private standards? The reason for recognition, I argue, stems from a different concern
about uncertainty—about the credibility of other private rules.
The explosion of private standards following the entry into force of the Kyoto Protocol created
great variability among the quality of standards. Higher quality standards, concerned about
establishing and maintaining their reputation, sought to associate themselves with other high
quality standards—creating a club of standards that lead the pack in terms of prestige. An
examination of the network shows that not all standards are created equal. Despite an overall
low density of the network, there are there are several clear leaders among private standards
which enjoy high levels of prestige; they provide evidence for the concern about credibility.
Interestingly, these leaders tend to be older standards; there is a small, but positive correlation
between the age of a standard and its level of prestige (0.11).
Along with the CDM, there are five other particularly prestigious standards (greater than one SD
above the mean), as demonstrated by Table 2 (on page 18). Three ISO standards (for simplicity,
referred to as the 14064 series) are very frequently recognized by others. Two of these are
accounting standards (1&3), and the third is used for measuring offsets (2). The GHG Protocol
was the basis for ISO 14064-1; hence many other standards recognize both ISO 10464-1 and the
GHG Protocol, which are substantively very similar.35 Finally, the Verified Carbon Standard,
which is an offset standard that emphasizes environmental co-benefits to carbon offset projects,
is also frequently selected by others—22.5% of all possible incoming ties are present. Together,
these seven standards represent 40% of the total ties in the network.
Credibility of one‘s own standard boosts consumer confidence, which is certainly important for
the continued viability of the organization and its work.36 Indeed, one organization noted that
firms who buy offsets are very concerned about the credibility of the standard and its work:
―firms are motivated to be savvy consumers in the voluntary market because one false step
would sink them.‖37 Credibility may also be useful in cultivating long-term relevance. As
discussed below, some of the more prominent private standards are now recognized by some
governments. In other cases, offset purchasers want to enhance the likelihood that credits
purchased in the voluntary market could be used to meet future regulatory requirements. Indeed,
one interviewee noted that big emitters are coming to them to purchase lots of offset credits,
because of the concern that once the Kyoto Protocol expires, they will need offsets to meet
reduction requirements under other regulatory systems (such as the EU trading scheme), but will
not have may Kyoto credits as an option.38 To the extent that emitters are taking this view, they
are trying to ―pick winners‖ – by purchasing those credits that they believe will be compatible
35 For more on the creation of the Greenhouse Gas Protocol, see Green 2010.
36 Personal communication, XX, Gold Trust Standard, 30 June 2011.
37 Personal communication, Brennan Duty, Sustainable Carbon, 29 June 2011.
with regulatory requirements in the future. Another interviewee added that the expiry of Kyoto
could be a positive development for private standard setters.39
Finally, many of those interviewed emphasized the importance of credibility, both for their own
standards, and for those that they recognize. This translates most frequently into openness,
transparency, widespread consultation with interested parties, and periodic review of the rules
and their application.40 In the words of one interviewee, credibility and robustness stem from
―how many eyes have looked at [the standard]‖, as well as the caliber of those reviewers.41
Thus, the rapid growth of carbon standards coupled with their individual concerns about
maintaining credibility gave rise to an important effect: clear leaders have emerged among the
patchwork of private carbon standards. Actor heterogeneity has produced some consensus about
which carbon standards should be used: those that are transparent, have had an open drafting
process, and have been (and continue to be) the subject of rigorous review. As one interviewee
succinctly noted: ―[M]ore standards – better quality. The competition [makes] the market and
[as a result], users win.‖42 High quality standards have emerged, and are at the center of the
Diffusion of leading practices
High quality standards do not simply appear, they are made. One way in which they are made is
through recognition by other standards. I turn now to the ―choosers‖ of standards, as indicated
by their outdegree centrality. These are important consumers of standards, enhancing the
network effects of those with high prestige.43 Interestingly, the most outdegree central standards
are all private: Certified CarbonFree, Carbon Neutral and Planet Positive. 44 Certified
CarbonFree, for example, recognizes 30% of all standards in the network. This provides
additional support for the hypothesis: private standards are hedging by maximizing compatibility
with other standards, both public and private. One effect of hedging is to reinforce the
importance of public rules. A second effect, as indicated in this section, is to create leaders
among private standards .
Indeed an analysis of the key players in the network shows that two private standards are the best
connected: VCS and ISO 14064-1 reach 98% of the network via a maximum of one
intermediary.45 The high prestige (as indicated in the previous section) and the impressive reach
of these two standards provide additional evidence that quality standards occupy key positions in
the network. VCS has also received an additional affirmation of its quality: it is now recognized
by the Australian Government in its voluntary National Carbon Offset Standard.46 Other
standards are also being recognized by public rules. The Australian government also recognizes
39 Personal communication, Nevena Pingarova, TUEV Sud, 13 July 2011.
40 Personal communication, Jonathan Shopley, Carbon Neutral. 5 July 2011.
41 Personal communication, Brennan Duty, Sustainable Carbon, 29 June 2011.
42 Personal communication, Nevena Pingarova, TUEV Sud, 13 July 2011. Similar sentiments were expressed by
representatives of the American Carbon Registry and the Chicago Climate Exchange.
43 On network effects, see Katz and Shapiro 1994.
44 These are the only three standards with outdegree centrality greater than one SD above the mean.
45 As computed by KeyPlayer.
46 Australian Department of Climate Change and Energy Efficiency 2010, p. 8.
the Gold Standard. The California cap and trade system, though still being drafted, is basing its
offset standards on those created by the Climate Action Reserve.
By contrast, four of the five public standards have an outdegree centrality of 0. The CDM,
RGGI, Climate Leaders and the IPCC have not chosen to recognize any other standards. The
one exception is the EU-ETS, which accepts offsets generated under CDM standards, as codified
by the ―Linking Directive‖.47 Again, this underscores the finding that private rules, rather than
public ones, are serving as diffusers of authority. This is further evidenced by the correlation
between ―privateness‖ and outdegree centrality, illustrated in Table 3. As actors go from public
to private (coded 0 and 1 respectively), outdegree centrality increases, and indegree centrality
In sum, the analysis of network data points to three main effects of uncertainty on private
standards‘ related to carbon measurement. First, uncertainty about the future international
regulations has led almost all offset standards to recognize the rules created under the CDM.
Second, uncertainty about the quality of different standards has given rise to a set of leading
private standards, which are more prestigious (in the sense of network centrality) than other
private standards in the network. The former ―wild west‖ of carbon markets is consolidating
around a clear set of leaders. Third and finally, the fact that private standards tend to have higher
outdegree centrality, shows their important role as ―diffusers‖ of leading standards—both public
and private. By contrast, public standards virtually never choose to link themselves to other
standards, with the one exception of the EU-ETS and the CDM. The interaction among public
and private actors shows that private actors, not public ones, are reacting to uncertainty about the
future of the regulatory process. This has had the unanticipated effect of producing some
convergence around what constitutes a ―good standard‖ for measuring and managing carbon.
Global interdependence, as Keohane and Ostrom point out, requires cooperation through
reciprocity rather than hierarchy. Traditional approaches to international cooperation focus on
reciprocity among states, through intergovernmental agreements and international institutions.
However, global interdependence has also resulted in the expansion of relevant actors in world
politics, and importantly for this paper, in their modes of interaction.
This paper demonstrates that the heterogeneity of actors and their accompanying institutions in
the issue area of GHG management has provided new spaces for interaction and unexpected
effects. The presence of private actors, coupled with uncertainty about the future of public
regulation and the quality of private regulation, has reaffirmed the importance of the CDM rules
and produced clear leaders among private rules. Bumpus and Livermore have argued that the
private standards are ―a network of often small private organizations and NGOs [which exist]
without reference to national or supranational bodies or ‗higher levels‘ of administration.‖48 The
evidence here suggests that this is not the case. Rather, a distinct order—and even a hierarchy—
has emerged out of the messy landscape of carbon standards. Heterogeneity among actors has
improved the quality of standards and the level of convergence among them.
47 European Parliament 2004.
48Bumpus and Livermore 2008, p. 141.
The broader implications of these findings are threefold. First, despite harsh criticism and
general pessimism about the Kyoto Protocol, there may yet be a silver lining. The offset
standards created by the CDM are now well-embedded in private standards operating in the
voluntary market. Some of these standards are even recognized by governments. It is very
likely, then, that these rules are here to stay. With private standards as faithful users and
diffusers (and in some cases, improvers), CDM rules will provide a common baseline for future
offsets. Second, and related, the case of carbon standards is redefining the nature of ―public.‖49
The interactions between public and private show the ways in which public goods—emissions
reductions—are being created and provided through a complex network structure and
heterogeneous actors. As politicians and scholars wrestle with ways to provide global public
goods, they must keep this definitional shift in mind. Finally, studies of regulatory dynamics
suggest conditions under which we can expect to upward or downward harmonization.50 This
paper shows that we should include private regulation in the mix as one of the potential
49 Cerny 2010.
50 Kagan and Vogel 2004; Vogel 1997.
Table 1: Private carbon standards
Cleaner and Greener Certification
Climate Action Reserve
Greenhouse Gas Protocol Corporate Standard
Chicago Climate Exchange (and offset std)
Carbon Disclosure Project
Climate, Community and Biodiversity Standard
Green Tick Carbon Neutral
Greenhouse Gas Protocol Project Standard
Voluntary Carbon Standard
Climate Change Action
The Climate Registry
Carbon Trust Standard
American Carbon Registry
Figure 3: Network of carbon management standards
Table 2: Centrality of all standards in the network
Cleaner and Greener
Green-e renewable energy
Cleaner and Greener
Table 3: Similarity Matrix for Network Attributes
1 2 3 4 5 6 7 8
Year Age 3rd pa Privat OutDeg InDegr NrmOut NrmInD
------ ------ ------ ------ ------ ------ ------ ------
1 Year 1.000 -0.656 0.091 0.221 0.292 -0.175 0.292 -0.175
2 Age -0.656 1.000 -0.220 -0.183 -0.299 0.101 -0.299 0.101
3 3rd party certifed? 0.091 -0.220 1.000 0.111 0.423 -0.126 0.423 -0.126
4 Private? 0.221 -0.183 0.111 1.000 0.343 -0.245 0.343 -0.245
5 OutDegree 0.292 -0.299 0.423 0.343 1.000 -0.229 1.000 -0.229
6 InDegree -0.175 0.101 -0.126 -0.245 -0.229 1.000 -0.229 1.000
7 NrmOutDeg 0.292 -0.299 0.423 0.343 1.000 -0.229 1.000 -0.229
8 NrmInDeg -0.175 0.101 -0.126 -0.245 -0.229 1.000 -0.229 1.00
Abbott, Kenneth W. 2010. ―Interactions within a Governance Complex.‖ Memo Prepared for Dynamics
of Interaction in Transnational Business Governance Regimes. 22 October 2010. Toronto, CA.
On file with author.
Abbott, Kenneth W, and Duncan Snidal. 2009. ―The Governance Triangle: Regulatory Standards
Institutions and the Shadow of the State.‖ In The Politics of Global Regulation, eds. Walter Mattli
and Ngaire Woods. Princeton, NJ: Princeton University Press, p. 44-88.
Alter, Karen J., and Sophie Meunier. 2009. ―The Politics of International Regime Complexity.‖
Perspectives on Politics 7(1): 13-24.
Andonova, Liliana B. 2011. ―Public-Private Partnerships for the Earth: Politics and Patterns of Hybrid
Authority in the Multilateral System.‖ Global Environmental Politics 10(2): 25-53.
Andonova, Liliana B., Michele M. Betsill, and Harriet Bulkeley. 2009. ―Transnational Climate
Governance.‖ Global Environmental Politics 9(2): 52-73.
Australian Department of Climate Change and Energy Efficiency. 2010. ―National Carbon Offset
Standard‖ Accessed at
Bernstein, Steven, Michele Betsill, Matthew Hoffmann, and Matthew Paterson. 2010. ―A Tale of Two
Copenhagens: Carbon Markets and Climate Governance.‖ Millennium - Journal of International
Studies 39(1): 161 -173.
Biermann, Frank, and Philipp Pattberg. 2008. ―Global Environmental Governance: Taking Stock, Moving
Forward.‖ Annual Review of Environment and Resources 33(1): 277-294.
Borgatti, S. 2003. KeyPlayer. Analytic Technologies: Boston.
Bulkeley, Harriet et al. Transnational Climate Change Governance Database. Unpublished manuscript.
Bumpus, Adam G., and Diana Liverman. 2008. ―Accumulation by Decarbonization and the Governance
of Carbon Offsets.‖ Economic Geography 84(2): 127-155.
Cerny, Philip G. 2010. Rethinking world politics: a theory of transnational neopluralism. Oxford: Oxford
European Parliament. 2004. Directive 2004/101/EC. Amending Directive 2003/87/EC establishing a
scheme for greenhouse gas emission allowance trading within the Community, in respect of the
Kyoto Protocol's project mechanisms.
Fransen, Luc. 2011. ―Why Do Private Governance Organizations Not Converge? A Political–Institutional
Analysis of Transnational Labor Standards Regulation.‖ Governance 24(2): 359-387.
Green, Jessica F. 2008. ―Delegation and Accountability in the Clean Development Mechanism: The New
Authority of Non-State Actors.‖ Journal of International Law and International Relations 4(2):
Green, Jessica F. 2010. ―Private Standards in the Climate Regime: The Greenhouse Gas Protocol.‖
Business and Politics 12(3).
Hale, Thomas N. and Charles Roger. 2011. ―Orchestration and Transnational Climate Governance.‖
Hamilton, Katherine, Milo Sjardin, Allison Shapiro, and Thomas Marcello. 2009. Fortifying the
Foundation: State of the Voluntary Carbon Market 2009. New York: Ecosystem Marketplace and
New Carbon Finance. Accessed at
Katz, Michael L., and Carl Shapiro. 1994. ―Systems Competition and Network Effects.‖ The Journal of
Economic Perspectives 8(2): 93-115.
Keohane, Robert, and Elinor Ostrom. 1995. Local Commons and Global Interdependence. Sage
Keohane, Robert O., and David G. Victor. 2011. ―The Regime Complex for Climate Change.‖
Perspectives on Politics 9(1): 7-23.
Kollmuss, Anja, Helge Zink, and Clifford Polycarp. 2008. 117 Making Sense of the Voluntary Carbon
Market A Comparison of Carbon Offset Standards. Stockholm Environment Institute and
Koremenos, B., C. Lipson, and D. Snidal. 2001. ―The rational design of international institutions.‖
International Organization 55(04): 761–799.
Mearsheimer, John. 1994. ―The False Promise of International Institutions.‖ International Security 19(3):
Mitchell, Ronald B. 1995. ―Heterogeneities at Two Levels: States, Non-state Actors and Intentional Oil
Pollution.‖ In Local Commons and Global Interdependence, Sage Publications Ltd, p. 223-252.
Newell, Peter, and Matthew Paterson. 2010. Climate Capitalism: Global Warming and the
Transformation of the Global Economy. Cambridge University Press.
Okereke, Chukwumerije, Harriet Bulkeley, and Heike Schroeder. 2008. ―Conceptualizing Climate
Governance Beyond the International Regime.‖ Global Environmental Politics 9(1): 58-78.
Raustiala, Kal, and David G Victor. 2004. ―The Regime Complex for Plant Genetic Resources.‖
International Organization 58: 277-309.
Rosendorff, B. Peter, and Helen V Milner. 2001. ―The Optimal Design of International Trade Institutions:
Uncertainty and Escape.‖ International Organization 55(4): 829-857.
Simmons, Beth A, and Lisa L Martin. 2002. ―International Organizations and Institutions.‖ In Handbook
of international relations, eds. Walter Carlsnaes, Thomas Risse, and Beth A. Simmons. SAGE, p.
Strange, Susan. 1996. The Retreat of the State. Cambridge: Cambridge University Press.
Thompson, Alexander. 2010. ―Rational design in motion: Uncertainty and flexibility in the global climate
regime.‖ European Journal of International Relations 16(2): 269 -296.
Victor, David G, and Joshua C House. 2004. ―A New Currency: Climate Change and Carbon Credits.‖
Harvard International Review (SUmmer): 56-59.
Vogel, David. 1997. ―Trading up and governing across: transnational governance and environmental
protection.‖ Journal of European Public Policy 4(4): 556.
Vogel, David, and Robert A Kagan, eds. 2004. The Dynamics of Regulatory Change: How Globalization
Affects National Regulatory Policies. Berkeley: University of California Press.
Wara, Michael. 2007. ―Is the global carbon market working?‖ Nature 445(7128): 595-596.