The approach of ecological economics
John Gowdy and Jon D. Erickson*
This paper discusses the major tenets of ecological economics—including
value pluralism, methodological pluralism and multi-criteria policy assessment.
Ecological economics offers viable alternatives to the theoretical
foundations and policy recommendations of neoclassical welfare economics.
A revolution in neoclassical economics is currently taking place, and the core
assumptions of welfare economics are being replaced with more realistic
models of consumer and ﬁrm behaviour. This paper argues that these new
theoretical and empirical ﬁndings are largely ignored in applied work and
policy applications in environmental economics. As the only heterodox school
of economics focusing on the human economy both as a social system and as
one imbedded in the biophysical universe, and thus both holistic and
scientiﬁcally based, ecological economics is poised to play a leading role in
recasting the scope and method of economic science.
Key words: Ecological economics, Heterodox economics, General equilibrium
analysis, Climate change
JEL classiﬁcations: B40, B50, Q01
1. Introduction
Ecological economics is poised to play a leading role in the ongoing effort to reconcile
economic theory and policy with accepted knowledge from other disciplines.1
Neoclassical welfare economics2
dominates economic policy across the globe, but it
is currently in a state of crisis over the dismantling of its two fundamental pillars: (1)
Manuscript received 15 May 2002; ﬁnal version received 19 September 2003.
Address for correspondence: Professor of Economics, Department of Economics, Rensselaer Polytechnic
Institute, Troy, New York 12180, USA; email: gowdyj@rpi.edu
* Rensselaer Polytechnic Institute and University of Vermont, respectively. We should like to thank
the following people for comments on an earlier draft: Jeroen C. J. M. van den Bergh, Peter Brown,
Raluca Iorgulescu, Melinda Kane, Richard Norgaard, Frank Mu¨ller, Sarah Parrales, John Proops, Irmi
Seidl and Barry Solomon.
1
There is now a large literature on the history and approach of ecological economics. We have
especially beneﬁted from the contributions of van den Bergh (2000), Brown (2001), Christensen
(1989), Costanza et al. (1991), Daly (1977), Martinez-Alier (1987), Mayumi (2001), Norgaard (1994),
Proops (2002), So¨derbaum (2000), Spash (1999) and Turner (1999).
2
The ﬁeld of economics is changing so rapidly that the term ‘neoclassical’ no longer represents the
monolithic core it once did. We use the term ‘neoclassical welfare economics’ to refer to the Walrasian
model based on self-interested exogenous preferences and complete and costless contracting. Related to
this point, we also acknowledge that many environmentally oriented economists and ecologists are using
ecological rather than economic models. These researchers are less concerned about what is happening
in economics and are making their own important contributions.
Cambridge Journal of Economics 2005, 29, 207–222
doi:10.1093/cje/bei033
Cambridge Journal of Economics, Vol. 29, No. 2, Ó Cambridge Political Economy Society 2005; all rights reserved
the theory of human behaviour embodied in the axioms of consumer choice; and (2)
the theory of production embodied in the notions of perfect competition and the
marginal productivity theory of distribution. In spite of the great strides being made in
expanding the scope of economic theory beyond these concepts, they are necessary
foundations for neoclassical welfare economics and Pareto efﬁciency. Neoclassical
theorists have by and large abandoned economic man and perfect competition;
however, the policy recommendations of economists are still based on these outdated
representations of human behaviour and commodity production. Neoclassical welfare
economics continues to offer bad advice in dealing with some of the most pressing
environmental and social issues faced in the twenty-ﬁrst century, including growing
income disparity, global climate change and biodiversity loss.
This paper discusses the major points of contention between neoclassical welfare
economics and ecological economics. By virtue of being the only heterodox school of
economics consistently focusing on the human economy as both a social system, and
as one constrained by the biophysical world, ecological economics is poised to play
a leading role in recasting the scope and method of economic science. Ecological
economic models of economic behaviour encompass consumption and production in
the broadest sense, including their ecological, social and ethical dimensions, as well as
their market consequences. As such, it is a ﬁeld of inquiry encompassing much of
contemporary neoclassical economics and heterodox schools of thought, including
behavioural economics, evolutionary economics, institutional economics, postKeynesian
economics, radical economics and social economics.
Challenges to the mainstream are not new. The economic conception of human
behaviour has been criticised for over a hundred years. The difference today is that the
most serious challenges to the standard welfare paradigm are coming from within the
professional mainstream. We now have a cadre of Nobel Prize recipients who work
largely outside the framework of the Walrasian general equilibrium model. It is
increasingly recognised that understanding the context of economic activity requires
familiarity with the relevant ﬁndings of related social and natural sciences.1
In the following pages, we ﬁrst characterise the welfare foundations of neoclassical
economic theory. We brieﬂy present the basic neoclassical model of welfare economics
and discuss in detail the major points of contention between that approach and
ecological economics, outlining for each issue the neoclassical position and the
ecological economics alternative. We end with a call for a structuralist approach that
categorises consumption and production as a uniﬁed social and biophysical process.
2. Welfare economics
Ecological economists are continually challenged by two criticisms from neoclassical
economists. The ﬁrst is the ‘we know this already’ argument which states: ‘The
shortcomings of neoclassical economics you discuss are no longer valid. The ‘‘new’’
neoclassical economics does not rely on the old axioms of consumer choice or the
model of perfect competition.’ The second criticism is eloquently summarised and
dismissed by Bromley (1990, p. 99)
1
2002 Nobel laureate in Economics, the psychologist Daniel Kahneman, was cited by the Nobel
committee for his work demonstrating ‘how human decisions may systematically depart from those
predicted by standard economic theory.’(Press release, Royal Swedish Academy of Sciences, 9 October
2002)
208 J. Gowdy and J. D. Erickson
Indeed, one often encounters criticism because the full nature and scope of that alternative
evaluated paradigm has not been presented. The refrain is familiar—‘we (may) acknowledge
that you are correct, but it is easy to criticise without offering a superior alternative. Until
you can do that, we are free to pursue the conventional wisdom.’ This defense is
disingenuous. To know that a critical aspect of economic thought is bogus and yet to swear
fealty to it on the grounds that the burden of a new approach falls on those who expose its
fallacies is irresponsible. Indeed, it borders on scholarly malfeasance to persist in passing off
known ﬁctions on the grounds that it is the problem of those who criticise also to create.
Regarding the ‘we know this already’ defence, the ﬁeld of economics has indeed made
great strides in recent years in developing realistic models of economic behaviour and
production. Many of the challenges to neoclassical welfare economics discussed below
are drawn from the ‘mainstream’ literature. Nevertheless, economic policy recommendations
remain ﬁrmly grounded in the two fundamental theorems of welfare
economics. The ﬁrst asserts that Pareto efﬁciency is implied by maximisation of
preferences under budget constraints and maximisation of proﬁts under given
technology. The second follows that almost any Pareto efﬁcient outcome can be
supported with appropriate lump sum transfers.1
The stringent conditions for the achievement of Pareto efﬁcient outcomes are well
known,2
yet this goal still dominates economic policy. As Lockwood (1987, p. 811)
asserts about the second Pareto theorem: ‘It is no exaggeration to say that the entire
modern microeconomic theory of government policy intervention in the economy
(including cost–beneﬁt analysis) is predicated on this idea.’
The fundamental theorems of welfare economics are the foundation for the ‘market
failure’ approach to economic policy (Stiglitz, 1994, p. 7). Pareto efﬁciency cannot be
achieved if the ‘wrong’ price signals are sent, and it is a legitimate function of
governments to correct these market failures. The goal of neoclassical welfare economic
policy, whatever the speciﬁc problem to be addressed, is to create the conditions of
a competitive economy so as to achieve Pareto efﬁciency. Pareto efﬁciency may be
expanded to include goods and services not traded in markets, or to account for
unpriced externalities of consumption and production, simply by expanding the
domain of prices. Methods to determine price externalities dominate applied work in
environmental economics, where contingent valuation and other survey tools came into
being in the 1970s in response to an expansion of cost–beneﬁt analysis to environmental
valuation. But here again, the solution to any problem faced—from global warming and
biodiversity loss to terms of trade and income distribution—is a secondary outcome of
‘getting the prices right’. By providing an apparently precise measure of social welfare,
the goal of Pareto efﬁciency has stiﬂed discussion of distributional questions and social
goals other than efﬁciency in allocation.
1
Most economic textbooks overstate the Second Fundamental Theorem. It does not say that any
Pareto outcome can be obtained through the market. It is well known that competitive equilibrium is
rarely unique and frequently unstable, thus the Second Welfare Theorem does not provide a justiﬁcation
for many of the ‘free market’ approaches advocated by economists (Bryant, 1994)
2
The most complete discussion of the existence of general equilibrium was that of Arrow and Debreu
(1954) who based their proof on Kakutani’s ﬁxed point theorem. The Arrow–Debreu model is
individualistic, assumes rational expectations, all agents are price-takers, there is no asymptotic
information, and money does not appear in the model. Although the Arrow–Debreu general
equilibrium model is still cited as proving the desirability of competitive markets, Arrow has always
maintained that the model is most useful in demonstrating the inefﬁciency of real-world markets
(Geanakoplos, 1987; Stiglitz, 1994).
The approach of ecological economics 209
One of the most widely used tools in economic policy analysis is the potential
Pareto improvement criterion. A Pareto improvement occurs when at least one
person is made better off without making anyone else worse off. A potential Pareto
improvement (PPI) involves a change that helps one person and harms another.
Under the Kaldor–Hicks rule, such a change is justiﬁed if the gainers from the
change value their gains more than the losers value their losses, even if no actual
compensation is made. Kaldor (1939) maintained that efﬁciency could and should be
separated from questions about distribution—a problem best left outside the realm of
economics. Problems with the PPI concept, and the separability assumption, arise
from several sources. First, insurmountable theoretical difﬁculties exist in identifying
PPIs, since the real income base is different before and after moving from one Pareto
efﬁcient point to another. This gives rise to a number of paradoxes in applying the
Kaldor–Hicks criterion (Boadway, 1974, Samuelson, 1950, Scitovsky, 1941), whose
meaning is that one cannot compare two Pareto efﬁcient outcomes without making
interpersonal comparisons of utility (Suzumura, 1999).1
Another source of problems
for the PPI concept is a growing body of evidence contradicting the axiomatic model
of consumer choice, including the existence of endogenous preferences (Bowles,
1998), the endowment effect (Knetch, 1989) and lexicographic ordering of preferences
(Spash, 2000). The axiomatic neoclassical model makes poor predictions of
economic behaviour (Henrich et al., 2001), calling into question not only the validity
of PPI as a policy guide but also the notion of efﬁciency as conceived by welfare
economics.
The theory of the consumer has been dominated by a goal of efﬁciency in utility
maximisation, supported by the notion of economic man and the underlying axioms of
consumer choice (preferences are complete, reﬂexive, transitive and continuous, and
exhibit non-satiation and diminishing marginal rates of substitution). Similarly, the
theory of production has been dominated by a goal of efﬁciency in proﬁt maximisation,
supported by the notion of perfect competition and the associated
assumptions of ﬁrm behaviour (independence of the actions of ﬁrms, no market
power, constant returns to scale, perfect information and no uncertainty).
In recent years, however, empirical and conceptual breakthroughs in consumer
behaviour and the theory of the ﬁrm have called into question the two props of
neoclassical welfare economics. If these two foundations are given up, the neoclassical
welfare enterprise falls apart. There are, of course, disagreements among heterodox
economists as to the importance of general equilibrium and Pareto efﬁciency.
Nevertheless, the rejection of potential Pareto efﬁciency as the driver of economic
policy is the deﬁning characteristic of a number of heterodox schools of economics,
and this rejection can be the focal point for constructing a uniﬁed theory challenging
neoclassical orthodoxy.
1
The basic problem with determining potential Pareto improvements is that any particular Pareto
optimum is dependent on some initial allocation of resources. Consider two Pareto efﬁcient allocations
A and B on a production possibilities frontier. Point A is optimal, given some social welfare function
based on a particular initial allocation of resources. Point B is optimal, given a social welfare function
based on some other initial allocation of resources. The existence of these different bases with different
relative prices and different marginal rates of substitution give rise to the variety of paradoxes plaguing
welfare economics. Because of these, it is impossible to unambiguously identify PPIs in a pure barter
economy, much less in a real market economy. The problems pile up when the PPI concept is used to
judge intergenerational welfare improvements.
210 J. Gowdy and J. D. Erickson
The example of climate change highlights the consequences of this ﬁxation on
efﬁciency and the resulting disconnection with other equally valid goals. A consensus
of the international science and policy community on global climate change has
emerged in the work of the Intergovernmental Panel on Climate Change (IPCC).
IPCC scientists and policy-makers, across diverse disciplines, societies and cultures,
contend that humanity faces a signiﬁcant increase in average global temperature
during this century (IPCC, 2001A). Likely consequences include sea level rise,
accelerated biodiversity loss, increased incidence of human disease, and rapid
oscillations or ‘ﬂips’ between very distinctive climate regimes (IPCC, 2001B).
William Nordhaus’s (1992, 2001) climate change model is the economic model
most widely referred to by policy-makers and has been extremely inﬂuential in
justifying delays in greenhouse gas reductions. In the spirit of the neoclassical welfare
model, the focus is on achieving an efﬁcient outcome where marginal costs are equal to
marginal beneﬁts.1
Distributional consequences, distinction between luxury vs subsistence
emissions, or increased risk to certain population segments are ignored. Other
real-world difﬁculties such as transactions costs, pure uncertainty and unfamiliarity
with new situations (difﬁculties addressed in increasingly sophisticated ways by
mainstream economists) are not taken into account.
The Nordhaus models assume certainty, constant returns to scale in production, all
the axioms of consumer choice theory, smooth and continuous climate change
impacts, and the existence of a social welfare function. They further assume that all
human desires can be represented by a single omniscient consumer and that the
world’s economic systems can be represented by either a single global or suite of
regional well-behaved production functions. Discounting the future is deemed
appropriate with respect to estimating the value of a stable climate in the distant
future. Using society’s scarce resources to moderate climate change is only justiﬁed if
this results in a net increase in economic output. The theoretical justiﬁcation for this
policy is that it would result in a potential Pareto improvement.
Welfare economic models of climate change are much more than harmless academic
exercises. They are widely cited by policy-makers as scientiﬁc proof that aggressive
policies to combat global warming are not cost effective. Organisations lobbying
against action on global climate change, including the Cooler Heads Coalition and the
National Consumer Coalition, cite Nordhaus’ results to claim that ‘the economic
effects of global warming will be modest’ and that ‘implementing the Kyoto Protocol
will be expensive and harmful to the economic wellbeing of the American people’ (see
www.globalwarming.org). Working within the framework of welfare economics, the
only greenhouse consequences that count are those that affect GNP. And since the
Nordhaus model relies on a surprise-free scenario of mild and predictable change,
these effects are likely to be modest. This conclusion is not based on scientiﬁc
objectivity, but rather on value-laden assumptions about discounting, technical
change and abatement costs at best reﬂecting the personal views of the analyst, and
at worst the expectations of research sponsors (Chapman and Khanna, 2000). Ethical
considerations are central to the global warming debate, but they are set aside by
1
Detailed critiques of Nordhaus’ model are given by Wright and Erickson (2003A, 2003B), Spash
(2002) and Howarth (2001). An excellent analysis of the welfare theory behind neoclassical climate
change models is given by Laitner et al. (2001). A good summary of the related literature on sustainable
welfare including overlapping generations models is given by Pezzy and Toman (2002).
The approach of ecological economics 211
neoclassical welfare theorists. Efﬁciency improvement is the overriding justiﬁcation for
policy intervention. This argument ignores the fact that existing prices and production
depend on ethical factors such as income distribution, environmental policies and the
level of spending on social programmes.
3. The approach of ecological economics
The major conceptual issues emerging in the ecological economics literature are Value
Monism, the Rational Actor Model, Marginal Analysis, the Treatment of Uncertainty,
the Role of Efﬁciency in Economic Policy, and Production as a Social and Physical
Process (see Table 1). These issues are critical to neoclassical welfare theory and its
treatment of broad environmental and social goals. Below, each methodological pillar
is characterised, together with an alternative approach.
3.1 Value monism
Value Monism implies that all objects of utility have some common characteristic that
allows them to be compared. Until the middle of the twentieth century there was
a lively debate in economics about varieties of value, including use vs exchange value,
labour and energy theories of value, and so on. However, in the second half of the
twentieth century, the ﬁeld of economics adopted a narrower notion of value, limited
to value in market exchange. Theories of allocation became entrenched in a model of
human behaviour assuming strict rationality and methodological individualism, an
assumption that people make all allocation decisions independently of others. Isolated
individuals at a point in time are assumed to reveal their exogenously determined
preferences fully by weighing the costs and beneﬁts of their consumption choices.
People choose what they want, and what they want is revealed by the choices they
make.
The chain of reasoning in cost–beneﬁt analysis goes from ‘human preferences’ to
‘choices made in a market context’ to the ‘market value of these choices’. It is assumed
that preferences for things such as biological diversity can be determined and made
compatible with those for market goods. Ecological economists, in general, argue that
the links in this chain of logic are weak. If human preferences are not accurately
characterised by the axioms of consumer choice, limiting choices to those made in
a market context is unnecessarily restrictive.
Empirical results support a new direction for neoclassical welfare theory, a direction
more consistent with its classical roots and broader conception of value. For example,
contingent valuation surveys soliciting preferences for environmental goods consistently
show that preferences for many environmental features are lexicographic, that is,
not subject to trade-offs. Results from behavioural economics and psychology indicate
the existence of endowment effects (people place higher values on things they already
possess), hyperbolic discounting (people discount the near future at a higher rate than
the distant future), loss aversion (people are signiﬁcantly more averse to taking a loss
than failing to receive an equal gain), the part–whole problem (people consistently
place higher values on the sum of individual components of an object of utility than on
the whole thing itself) and many other ‘anomalies’ in consumer choice theory (Gintis,
2000).
212 J. Gowdy and J. D. Erickson
Value Monism lies behind standard cost–beneﬁt analysis (CBA) which uses the
notion of consumer surplus to judge the desirability of public policy choices. Findings
of behavioural economics also cast doubt on the validity of the model of human
behaviour underlying cost–beneﬁt approaches, including contingent valuation techniques.
The endowment effect, for example, is one reason for the disparity between
willingness-to-pay (WTP) measures of welfare changes and the typically higher
willingness to accept (WTA) measures. Although WTP measures are almost universally
used to measure environmental costs and beneﬁts, WTA is theoretically more
appropriate (Brown and Gregory, 1999). The existence of hyperbolic discounting
casts doubt on the validity of CBA as a measure of costs and beneﬁts in the mediumto-distant
future.
An emerging ecological economics alternative to CBA—and value monism in
general—is multi-criteria decision aide (MCDA). As the name implies, this method of
policy analysis takes into account a wide variety of relevant information. MCDA
methods allow for the multiple dimensions typical in many decision-making problems
Table 1. Key conceptual issues
Conceptual issue
Neoclassical
welfare economics
Ecological economic
alternative
Value Monism Reduce value to
commensurable monetary
units; utility function.
Separate value into
incommensurable categories;
multi-criteria assessment.
The Rational Actor Individual consumers and
ﬁrms at the centre of
analysis.
Analyse humans as social
actors, consumers versus
citizens.
Marginal Analysis Comparative statics of
marginal changes.
Recognises discontinuous
change and total effects
Evolutionary Change Evolution as constrained
optimisation, survival of the
ﬁttest view of market
outcomes, individual based
selection.
Importance of contingency,
historical accidents, path
dependency. Considers
altruism and group selection
as well as selﬁshness.
Uncertainty Reduce uncertainty to risk.
Market outcome focus to
decision-making.
Precautionary principle to
deal with pure uncertainty.
Process-oriented,
co-evolutionary focus to
decision-making.
Decision Criteria Efﬁciency as the sole criterion,
usually based on potential
Pareto improvements.
Equity, stability, resilience of
environmental and social
systems.
Production Process Theory of allocation of ﬁxed
resources; production
function.
Production as a biophysical
process, thermodynamics;
extended IO approach, joint
production of goods and
polluting wastes.
Discounting Straight-line discounting of
future costs and beneﬁts.
Recognises the difference
between individual and social
valuation of the future;
hyperbolic discounting.
The approach of ecological economics 213
to be chosen, evaluated and weighed. Valuation can be based on diverse criteria such
as efﬁciency, equity or sustainability, allowing for a more realistic assessment of
substitutability and complementarity between criteria. MCDA allows for ethical
considerations, incongruities and concern for the distant future in a democratic
decision-making framework. Numerous case studies (Barton, 1996; O’Neill and
Spash, 2000; Spash, 2000) employing MCDA methods also highlight the strength of
incorporating qualitative information into an economic valuation framework. The
approach is most developed in Europe (Spash and Carter, 2001), but is gaining
ground in applied analysis in the US.
3.2 The Rational actor model
The starting point for economic analysis is a rational actor who makes decisions
without social or environmental context. Game theory experiments and laboratory
results involving actual human behaviour have cast doubt on the general validity of the
neoclassical rational actor. These ﬁndings indicate that preferences are endogenous,
that is, they depend on social context, individual histories, and conscious preference
development (Albert and Hahnel, 1990, Bowles, 1998). Although heroic attempts
have been made to reconcile endogenous preferences with the Rational Actor Model,
these attempts require such restrictive and unrealistic assumptions as to make them of
little practical use.
A growing body of empirical evidence shows that people bring a sense of fairness
into allocation decisions. The social framework is an important factor in economic
decision-making, and people make different decisions as members of a social group
than they do as isolated individuals. For instance, people frequently cooperate with
complete strangers, even if they know they will never meet again (Bowles and Gintis,
2002). Fehr and Ga¨chter (2002) present evidence for ‘altruistic punishment’, when
people will punish free-riders at signiﬁcant cost to themselves. Results from the
ultimatum and dictator games show clearly that ‘fairness’ is just as important as
‘selﬁshness’ in predicting human behaviour (Fehr and Ga¨chter, 2000; Gowdy et al.,
2003; Henrich et al., 2001). Results from these and other games show that the
Rational Actor Model is not a good predictor of human behaviour. Humans have more
complex social behaviour than other mammals, and altruism (or any other trait) can
be imposed by social sanctions. Cooperative action such as consensus building or
collective decision-making cannot emerge in a decision framework where only
individual preferences count. Human wants are socially contingent, not atomistic.
Public policy should take into account market-derived costs and beneﬁts, but it should
also recognise that social welfare involves much more than market values.
The Rational Actor Model also relates to controversy surrounding discounting the
future. Much has been written about the limitations of using a market discount rate to
place a value on future social and environmental gains or losses (Georgescu-Roegen,
1976; Price, 1993). First, a reasonable decision for an individual acting in a market
setting at a particular point in time might be inappropriate for society as a whole. A
good case can be made for not discounting the value of natural features essential for
long-term human survival, such as climate stability, biodiversity and uncontaminated
soil, water and atmospheric systems. Second, even within the neoclassical ‘choice’
context, a growing body of evidence suggests that a straight-line discount rate does not
accurately reﬂect how individual humans actually view the future. Studies by
214 J. Gowdy and J. D. Erickson
economists and psychologists have shown that people exhibit hyperbolic discounting,
where higher value is placed on beneﬁts delivered in the near term, followed by a sharp
drop and ﬂattening in the medium term, so that the value of something stays fairly
constant out into the distant future (Laibson, 1997). A hyperbolic discount rate would
have a dramatic effect on cost–beneﬁt calculations of the future beneﬁts of global
climate stability or biodiversity. But taking into account the well-being of future
generations involves much more than choosing the ‘socially correct’ discount rate.
In new game theoretic and behavioural models of consumer behaviour Homo
economicus is being out-competed by other species of economic actors (Fehr and
Ga¨chter, 2000; Gintis, 2000; Kirman, 1989). Models of decision-making such as
‘prospect theory’ (Tversky and Kahneman, 1981) and ‘biased cultural transmission’
(Henrich, 2004) have proved to be better predictors of economic behaviour than the
axiomatic rational choice model. Likewise, the neoclassical theory of the ﬁrm as
independent of historical time, space and the behaviour of other ﬁrms is being
replaced by more realistic models. Radner (1968) found that ﬁrm managers also have
a sense of ‘fairness’ when it comes to hiring and ﬁring workers. The heart of the
neoclassical theory of the ﬁrm, proﬁt maximisation, has failed the predictability test
and is being replaced by more sophisticated models of cooperation, altruistic
punishment and other forms of strategic behaviour. The barriers to adopting
alternative views stem from their lack of congruity with the Walrasian welfare model.
3.3 Marginal analysis
The notion of the margin embodies the basic assumptions of neoclassical economics,
including substitution, value monism, opportunity cost and equilibrium. The limits to
Marginal Analysis are apparent in the case of ecosystem valuation (Gowdy, 1997).
Removing or adding one species to an ecosystem, for example, will affect other species
and the general integrity of the system in unpredictable ways. Furthermore, the effects
are likely to be different each time a change is made. For some species the changes may
be small. Removing or adding other, key species may ﬂip the entire ecosystem to
another state. Since biodiversity is characterised by ‘functional transparency’ (Vatn
and Bromley, 1994), the contribution of one feature of the ecosystem cannot be
known until it is added or subtracted from the system. In addition, even in a market
context, people consider some goods and services to be essential and not subject to
trade-offs at the margin (lexicographic preferences).
Ingrained in the notion of the margin is a view of economic change as gradual,
continuous and progressive. This is an incomplete view of evolutionary processes.
Random, non-marginal shocks are a driving force in evolutionary change in economic
as well as biological systems. The differences in the distribution of animals 64 and 66
million years ago cannot be understood without knowing that a meteor hit the earth 65
million years ago, drastically changing the climate. The current economic situation of
the insurance and airline industries cannot be understood without taking into account
the attack on the World Trade Center on 11 September 2001. Evolutionary change is
characterised by hierarchies of selection, historical contingency and random events
(O’Neill et al., 1986). In evolutionary systems it is impossible to change one thing and
hold everything else constant. The existence of qualitative and non-marginal change is
a powerful argument for rejecting microeconomic theory as a foundation for
macroeconomic analysis (van den Bergh and Gowdy, 2003).
The approach of ecological economics 215
One ecological economics alternative to Marginal Analysis is complex adaptive
systems analysis (Gunderson et al., 1995; Holling et al., 2000; Scheffer et al., 2001).
The current state of theory is summarised in the ﬁnal report of the Resilience Project,
designed around two general observations on socio-ecological biocomplexity. First, in
response to problems and crises, current development policies may succeed in the near
term, but over time they lead to rigidity and myopia. Economic sectors become
dependent, ecosystems become more fragile, and the public loses trust in governance.
Second, complexity, diversity and opportunity in complex regional systems emerge
from a handful of critical variables and processes operating over distinctly different
scales in space and time. One of the most powerful conclusions from the study of
complex adaptive systems is that stocks and ﬂows must be adequately differentiated.
Recognition of natural capital depreciation as a negative ﬂow would favour management
regimes focused on natural capital resilience and the maintenance of broad
systems boundaries capable of stability. This ecological economic vision is that
resilience is a key feature of healthy ecosystems. The ability to adapt to unforeseen and
unforeseeable changes is an important aspect of ﬁrm behaviour and macroeconomic
stability.
Another alternative to Marginal Analysis is the use of an expanded input–output
(IO) model to examine the direct and indirect effects of large changes in economic
structure. The IO Social Accounting framework allows for a much more ﬂexible
deﬁnition of economic activity than the commonly used general equilibrium concept
found in most economic impact studies. Social accounting provides a systematic way
to organise quantitative as well as qualitative information and serves as the foundation
on which to analyse and evaluate alternative non-marginal policy options and consider
their impact in an ecological, economic and social context (see Section 3.6 for further
discussion).
3.4 The Treatment of uncertainty
The Treatment of Uncertainty is a major issue dividing neo-liberal and heterodox
economists. In the neoclassical synthesis following World War II, the insights of
Keynes’s General Theory regarding the importance of uncertainty and ‘animal spirits’
were discarded in favour of deterministic, micro-based theories of macroeconomic
behaviour. Neoclassical welfare economists also reduce uncertainty to risk to keep
their general equilibrium models tractable. In a general equilibrium framework
realism is sacriﬁced for formalism. In contrast, many ecological economists call for
a structural approach where technical descriptions of particular economies are used
for scenario analysis (Duchin, 1998). In terms of policy, one ecological economics
alternative to assuming that uncertainty can be reduced to risk is the precautionary
principle (Ciriacy-Wantrup, 1952), which suggests that we should err on the side of
caution in the face of uncertainty.
Embracing uncertainty leads to a focus on process over outcome driven decisions.
The majority of decision models in economics are built around the notion of
directing a system toward an expected optimal outcome, with little attention to the
path to optimality. For instance, the common practice in dynamic allocation
problems is to assume away uncertainty, calculate a steady-state optimum and
specify a most-rapid-approach path (MRAP) to this carefully constructed goal. In
contrast, ecological economics has supported a co-evolutionary, systems view of
216 J. Gowdy and J. D. Erickson
uncertainty and dynamic change. Co-evolution is a model that by its nature cannot
predict or be operationalised as conveniently as a utility maximisation problem. Coevolution
offers a view of the complex social–natural–physical system, providing
insight into how to structure our individual and societal choices. For example,
Norgaard (1994) has investigated agricultural development in the Amazon for
lessons from a co-evolutionary perspective. He argues that applying the presumed
universal concepts of Western agriculture and a global market to a tropical ecosystem
and culture has been a ‘resounding failure’, while traditional knowledge and
cultures, which co-evolved with this speciﬁc ecosystem, have ‘repeatedly proven
more reliable’ (1994, p. 121). Recognising an underlying co-evolutionary process
between social and natural systems can be helpful in formulating lessons to help
guide social processes (Erickson, 1999). Some lessons include: (1) experiment on
a small-scale and monitor the evolutionary chain of events; (2) experiments with
long-time commitments should be avoided; (3) diversity in co-evolving systems is
inherently good; without it there is likely to be stagnation; (4) emphasise evolutionary
processes rather than mechanical ﬁxes.
3.5 Efﬁciency, stability and equity: market preferences and social values
Neoclassical policy is concerned almost exclusively with efﬁciency in allocation.
Production is separated from consumption so that efﬁciency in production is equivalent
to maximising social welfare. The goal of the ﬁrm of reducing costs has been expanded
to become the goal of the macroeconomy and human society as a whole. As Bromley
(1990) argues, this goal is ideological not scientiﬁc, that is, it is a value judgment that
colours economic analysis. As formulated by standard welfare theory, the goal of
economic efﬁciency has no logical claim to objectivity. Without interpersonal comparisons
of utility, neoclassical welfare economists have little to say about the general
desirability of alternative social states (Suzumura, 1999). When Marshall and Pigou
wrote about the general well-being of people, they were free to argue for policies that
promoted certain social states. Sometime after World War II, ‘economics ceased to be
about people and their relationships to one another as it had been before, and began to be
about commodities’ (Bromley, 1990, p. 91). Narrowing the choice of economic policy to
the realm of efﬁciency leads to the conclusion that the market mechanism should drive
social choices. Yet the market is not well suited to making social choices. Market
outcomes only reﬂect the consumption decisions of millions of isolated individuals, not
human choice as a social phenomenon. Again, the current literature on endogenous
preferences calls into question the rational choice model.
Ecological economics distinguishes between individual values and social choices. As
Vatn and Bromley (1994, p. 142) write, ‘Just as preferences count for consumer choice
within constraints, judgments can be used as the driving concept for citisens choosing
basic norms or modifying existing constraints.’ This is where the importance of
institutions manifests itself. Mary Douglas (1986, p. 124) concludes, ‘The most
profound decisions about justice are not made by individuals as such, but by
individuals within and on behalf of institutions.’
3.6 The physical nature of production
Neoclassical ‘production theory’ is not a theory of production, but rather a theory of
allocation of a ﬁxed amount and given distribution of production inputs. Likewise,
The approach of ecological economics 217
neoclassical ‘growth theory’ is not a theory of growth, but rather a theory of the
optimal allocation of input growth rates. Pasinetti (1977) wrote of the neoclassical
theory of production:
The model clearly has nothing whatsoever to do with the phenomenon of production. The
problem it deals with is optimal allocation, through exchange, of a certain initial endowment
and distribution of resources.
Ecological economics began with the insight that the economy must be in
a materials balance between raw materials entering the process and waste leaving
(Ayres and Kneese, 1969; Boulding, 1966; Daly, 1977; Faber et al., 1996;
Georgescu-Roegen, 1976; Mayumi, 2001). A well-developed alternative to the
neoclassical production function is IO analysis. Dynamic IO models may be cast in
an equilibrium framework but the IO and computable general equilibrium (CGE)
approaches are distinctly different. An IO table can be seen as snapshot of
a particular economy at a particular point in time. It need not be interpreted as an
equilibrium model in the sense of optimisation, stability or having a tendency to
return to equilibrium if disturbed. IO analysis has been criticised for its ﬁxed
coefﬁcient assumption, but a growing body of evidence suggests that this is a more
accurate representation of actual production than the twice differentiable isoquant
of neoclassical theory.1
Ecological economists have made progress in describing the
relationship between economic activity, social institutions and environmental
features using IO analysis and systems of social accounts. The extended version of
IO analysis, the social accounting matrix (SAM), gives a concise view of economic
activity and the interconnections between economic sectors, household characteristics
and social institutions. A further extension with natural resource accounts
(NRAs) provides for a supporting environmental/natural resource base in terms of
inputs and outputs. Figure 1 highlights matrices of interdependencies in a SAM–
NRA model, corresponding to the common ecological economics framework of the
economy supported by a social framework further supported by an ecosystem base.
Economic, social and environmental transactions are captured by IO, SAM and NR
accounts, respectively. With a quantitative description of these ﬂows, a SAM–NRA
model can be used to analyse complex scenarios of economic, social and environmental
change.
4. Conclusion
It is often argued that economists must follow the narrow path of neoclassicism
because ‘there is no well-developed alternative’. As outlined above, however, the
alternative is there, but it requires abandoning the ﬂawed ‘grand uniﬁcation theory’ of
neoclassical welfare economics. Rather than a Theory of Everything, we appear to
need Theories of Theories of Things. Understanding the human economy requires an
appreciation of the importance of hierarchies, contingency and self-organisation, and
1
See Miller (2000) for a concise summary of the evidence for horizontal average variable cost curves
and other evidence that ﬁrms use the services of capital and labour in ﬁxed proportions even thought
capital stock is ﬁxed. Recent US studies done by the Federal Reserve Banks and the Bureau of the
Census document the ubiquity of ﬁxed proportions in manufacturing (Corrado and Mattey, 1997).
These studies found that plants typically change production levels by increasing or decreasing all inputs
together by shutting down or re-opening entire plants or portions of plants, not by changing the number
of hours worked.
218 J. Gowdy and J. D. Erickson
recognition of the fragility of market economies in biophysical space and cultural
speciﬁcity.
The policy dilemma is this: given the fact that a signiﬁcant part of our well-being is
derived from money ﬂows from the market economy—an economy isolated from
direct inﬂuences of the natural world—how do we create policies to preserve the life
support systems of the planet? Ecological economics is still struggling with this
question, and there are no satisfactory answers yet. But of all the conventional and
heterodox schools of economic thought, ecological economics is the only one poised to
address the problems of human survival in the coming centuries. It is the school of
thought that explicitly recognises the interconnections and interdependence of the
economic, biophysical and social worlds. We offer no grand theory, but rather
a ﬂexible approach recognising the uniqueness of speciﬁc cultures and ecosystems.
Norgaard (1989) called for a pluralism of approaches to economic theory and policy in
the ﬁrst issue of Ecological Economics more than a decade ago. Ecological economics is
barely ten years old, and it has not yet coalesced into a coherent school of thought, but
it is a leading contender among heterodox schools to become a comprehensive
alternative to neoclassical orthodoxy.
Bibliography
Albert, M. and Hahnel, R. 1990. A Quiet Revolution in Welfare Economics, Princeton, N.J.,
Princeton University Press
Transactions between
Social Institutions
Transactions
between Social
Institutions and
Value Added
Transactions from
Social Institutions to
Economic Sectors
Transactions from
Factor Inputs to
Economic Sectors
Transactions
among
Economic Sectors
Transactions from
Economic Sectors to
Social Institutions
(Final Demand)
Ecosystem
Economic System
Social System
Environmental
Impact from
Economic
Sectors
Environmental
Impact from
Social
Institutions
Environmental Inputs
to Social Institutions
Environmental Inputs
to Economic Sectors
Fig. 1. An ecological economic view of nested systems of accounts.
The approach of ecological economics 219
Arrow, K. and Debreu, G. 1954. Existence of an equilibrium for a competitive economy,
Econometrica, vol. 22, 265–90
Ayres, R. U. and Kneese, A. V. 1969. Production, consumption and externalities, American
Economic Review, vol. 59, 282–97
Barton, H. 1996. The Isle of Harris Superquarry: concepts of the environment and
sustainability, Environmental Values, vol. 5, 97–122
Bergh, C. J. M. van den. 2000. Ecological economics: themes, approaches, and differences
with environmental economics, Regional Environmental Change, vol. 2, 13–23
Bergh, C. J. M. van den and Gowdy, J. 2003. The microfoundations of macroeconomics: an
evolutionary perspective, Cambridge Journal of Economics, vol. 27, 65–84.
Boadway, R. 1974. The welfare foundations of cost–beneﬁt analysis. Economic Journal, vol.
47, 926–39
Boulding, K. 1966. The economics of the coming spaceship Earth, in Environmental Quality
in a Growing Economy, Baltimore, MD, Johns Hopkins Press
Bowles, S. 1998. Endogenous preferences: the cultural consequences of markets and other
economic institutions, Journal of Economic Literature, vol. 36, 75–111
Bowles, S. and Gintis, H. 2002. Homo reciprocans, Nature, vol. 415, 125–7
Bromley, D. 1990. The ideology of efﬁciency: searching to a theory of policy analysis,
Journal of Environmental Economics and Management, vol. 19, 86–107
Brown, P. 2001. The Commonwealth of Life, Montreal, Black Rose Books
Brown, T. and Gregory, R. 1999. Why the WTA–WTP disparity matters, Ecological
Economics, vol. 28, 323–35
Bryant, W. 1994. Interpretations of the second fundamental theorem of welfare economics:
barriers to better economic education, Journal of Economic Education, vol. 25, 75–80
Chapman, D. and Khanna, N. 2000. Crying no wolf: why economists don’t worry about
climate change, and should, Climatic Change, vol. 47, 225–32
Christensen, P. 1989. Historical roots for ecological economics—biophysical versus
allocative approaches, Ecological Modeling, vol. 38, 47–73
Ciriacy-Wantrup, S. V. 1952. Resource Conservation: Economics and Policies, Berkeley, CA,
University of California Division of Agricultural Sciences
Corrado, C. and Mattey, J. 1997. Capacity utilization, Journal of Economic Perspectives, vol.
11, 151–67
Costanza, R., Daly, H. E., and Bartholomew, J. 1991. Goals, agenda, and policy
recommendations for ecological economics, in Costanza, R. (ed.), Ecological Economics:
The Science and Management of Sustainability, New York, Columbia University Press
Daly, H. E. 1977. Steady State Economics, San Francisco, CA, W. H. Freeman
Douglas, M. 1986. How Institutions Think, Syracuse, N.Y., Syracuse University Press
Duchin, F. 1998. Structural Economics, Washington, D.C., Island Press
Erickson, J. D. 1999. Ecological economics: an emerging alternative to environmental
economics, in Chapman, D. (ed.), Environmental Economics: Theory, Application, and
Policy, New York, Addison Wesley Longman HarperCollins
Faber, M., Manstetten, R. and Proops, J. 1996. Ecological Economics: Concepts and Methods,
Cheltenham, UK, Edward Elgar
Fehr, E. and Ga¨chter, S. 2000. Cooperation and punishment in public goods experiments,
American Economic Review, vol. 90, 980–94
Fehr, E. and Ga¨chter, S. 2002. Altruistic punishment in humans, Nature, vol. 415, 137–40
Geanakoplos, J. 1987. Arrow–Debreu model of general equilibrium, in Eatwell, J., Milgate,
M. and Newman, P. (eds), The New Palgrave Dictionary of Economics, Vol. 1, London,
Macmillan
Georgescu-Roegen, N. 1976. Energy and Economic Myths, Oxford, Pergamon Press
Gintis, H. 2000. Beyond Homo economicus: evidence from experimental economics,
Ecological Economics, vol. 35, 311–22
Gowdy, J. 1997. The value of biodiversity: markets, society and ecosystems, Land Economics,
vol. 73, 25–41
Gowdy, J., Iorgulescu, R. and Onyeiwu, S. 2003. Fairness and retaliation in a rural Nigerian
village, Journal of Economic Behavior and Organization, vol. 54, 469–79
220 J. Gowdy and J. D. Erickson
Gunderson, L. H., Holling, C. S. and Light, S. S. (eds) 1995. Barriers and Bridges to the
Renewal of Ecosystems and Institutions, New York, Columbia University Press
Henrich, J. 2003. Cultural group selection, coevolutionary processes, and large scale
cooperation, Journal of Economic Behavior and Organization, vol. 53, 3–35
Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H. and McElreath, R.
2001. Cooperation, reciprocity and punishment in ﬁfteen small-scale societies, American
Economic Review, vol. 91, 73–8
Holling, C. S., Folke, C., Gunderson, L. and Maler, K. G. 2000. Resilience of Ecosystems,
Economic Systems, and Institutions, Final Report submitted to the John D. and Catherine T.
MacArthur Foundation, 30 April (www.resalliance.org/reports/)
Howarth, R. 2001. Intertemporal social choice and climate stabilization, International
Journal of Environment and Pollution, vol. 15, 386–405
IPCC 2001A. Climate Change 2001: The Scientiﬁc Basis—Summary for Policymakers,
Cambridge, Cambridge University Press
IPCC, 2001B. Climate Change 2001: Impacts, Adaptation and Vulnerability—Summary
for Policymakers, Cambridge, Cambridge University Press
Kaldor, N. 1939. Welfare propositions and interpersonal comparisons of utility, Economic
Journal, vol. 49, 549–52
Kirman, A. 1989. The intrinsic limits of modern economic theory: the emperor has no
clothes, Economic Journal, vol. 99, 126–39
Knetch, J. 1989. The endowment effect and evidence of nonreversible indifference curves,
American Economic Review 79, 1277–84
Laibson, D. 1997. Golden eggs and hyperbolic discounting, Quarterly Journal of Economics
112, 443–77
Laitner, J., DeCanio, S. and Peters, I. 2001. Incorporating behavioral, social, and organizational
phenomena in the assessment of climate change mitigation options, in Jochem, E.
et al. (eds), Society, Behavior, and Climate Change Mitigation, Dordrecht, Kluwer.
Lockwood, B. 1987. Pareto efﬁciency, in Eatwell, J., Milgate, M. and Newman, P. (eds), The
New Palgrave Dictionary of Economics, Vol. 3, London, Macmillan
Martinez-Alier, J. 1987. Ecological Economics: Energy, Environment and Society, Cambridge,
MA, Blackwell
Mayumi, K. 2001. The Origins of Ecological Economics: The Bioeconomics of Georgesc–Roegen,
London, Routledge
Miller, R. 2000. Ten cheaper spades: production theory and cost curves in the short run,
Journal of Economic Education, Spring, 119–30
Nordhaus, W. 1992. An optimal transition path for controlling greenhouse gases, Science,
vol. 258, 1315–19
Nordhaus, W. 2001. Global warming economics, Science, vol. 294, 1283–4
Norgaard, R. 1989. The case for methodological pluralism, Ecological Economics, vol. 1,
37–58
Norgaard, R. 1994. Development Betrayed: The End of Progress and a Coevolutionary
Revisioning of the Future, London/New York, Routledge
O’Neill, J. and Spash, C. 2000. Conceptions of value in environmental decision-making,
Environmental Values, vol. 9, 521–35
O’Neill, R., De Angelis, D., Waide, J. and Allen, T. 1986. A Hierarchical Concept of
Ecosystems, Princeton, N.J., Princeton University Press
Pasinetti, L. 1977. Lectures on the Theory of Production, New York, Columbia University
Press
Pezzy, J. and Toman, M. 2002. ‘The Economics of Sustainability: A Review of Journal
Articles’, Resources for the Future, Discussion Paper 02-03, Washington, D.C.
Price, C. 1993. Time, Discounting and Value, Cambridge, MA, Basil Blackwell
Proops, J. 2002. Identiﬁcation of ecological economics issues, in Encyclopedia of the Life
Sciences, Paris, UNESCO
Radner, R. 1968. Competitive equilibrium under uncertainty, Econometrica, vol. 36, 31–58
Samuelson, P. 1950. Evaluation of real national income, Oxford Economic Papers, vol. 2,
1–29
The approach of ecological economics 221
Scheffer, M., Carpenter, S., Foley, J. A., Folke, C. and Walker, B. 2001. Catastrophic shifts
in ecosystems, Nature, vol. 413, 591–6
Scitovsky, T. 1941. A note on welfare propositions in economics, Review of Economic Studies,
vol. 9, 77–88
So¨derbaum, P. 2000. Ecological Economics, London, Earthscan
Spash, C. 1999. The development of environmental thinking in economics, Environmental
Values, vol. 8, 413–35
Spash, C. 2000. Ecosystems, contingent valuation and ethics: the case of wetland recreation,
Ecological Economics, vol. 34, 195–215
Spash, C. 2002. Greenhouse Economics: Values and Ethics, London, Routledge
Spash, C. and Carter, C. 2001. Environmental Valuation in Europe: Findings from the
Concerted Action, Cambridge Research for the Environment, Policy Research Brief, no.
11, University of Cambridge, UK
Stiglitz, J. 1994. Whither Socialism? Cambridge, MA, MIT Press
Suzumura, K. 1999. Paretian welfare judgements and Bergsonian social choice, Economic
Journal, vol. 109, 204–21
Turner, R. K. 1999. Environmental and ecological economics perspectives, in Bergh, J. van
den (ed.), Handbook of Environmental and Resource Economics, Cheltenham, Edward Elgar
Tversky, A. and Kahneman, D. 1981. Loss aversion in riskless choice: a referencedependent
model, Quarterly Journal of Economics, vol. 106, 1039–61.
Vatn, A. and Bromley, D. 1994. Choices, without prices, without apology, Journal of
Environmental Economics and Management, vol. 26, 129–48
Wright, E. and Erickson, J. D. 2003A. Incorporating catastrophes into integrated assessment:
science, impacts, and adaptation, Climatic Change, vol. 57, 3–35
Wright, E. and Erickson, J. D. 2003B. Climate variability, economic adaptation, and
investment timing, International Journal of Global Environmental Issues, vol. 3, 357–68
222 J. Gowdy and J. D. Erickson