1. Introduction

Impacts of environmental policies and strategies are disappointing in many areas. One reason is obviously that they are not always ambitious. Another factor is overlooked, namely, that they may increase overall complexity of the socioeconomic system, in turn contributing unintendedly to environmental pressures. This raises the question whether increases in complexity, due to a combination of economic development and finding solutions to current environmental problems, pose a challenge to long-term environmental sustainability and sustainability transitions? We approach this question here from thermodynamic and systemic perspectives, which helps to illuminate relevant mechanisms of complexity increase and potential barriers to sustainability. The ultimate goal of this article is to obtain insights about policies that may complement existing environmental strategies to assure that complexity does not sit in the way of sustainability.

When studying the relation between complexity and sustainability, it is useful to agree about the meaning and various expressions of complexity. It is good to note that there are many different specific definitions and interpretations around. However, they tend to agree that complexity is a quality that applies to objects or entities that consist of several distinguishable components or elements. More complexity tends to equate more variety in and connectivity between these, possibly involving subsystems (which may themselves be complex), feedback loops, and particular network structures of interactions or even hierarchical organization. All these features add to the system complexity. In the case of physical entities or mechanical devices, an obvious measure is the number of parts, while in social networks, the number and frequency of communication lines may define complexity. Distinguishability is another measure—in the Shannonian sense information content or distance from randomness [1]. For distinct economic interpretations and measures of complexity, see Balland et al. [2]. This paper adheres to a systemic perspective that considers complex systems as characterized by diversity of components, feedback, hierarchies, nonlinearity, and threshold effects. In turn, these will affect the nature of the system, in terms of dynamics, potential emergent patterns, self-organization, and changes in stability or resilience [3]. An increase in system complexity means that one or multiple new components or subsystems appear, and/or that new connections are created among existing components or with new components or subsystems.

In the 1970s and through the early 1990s, there was considerable attention for how the laws of thermodynamics matter to the analysis of environmental problems and their solutions. This was approached through mass balance models, production functions with energy/exergy, calculating energy return on energy investment (EROEI), and outlining entropy implications for economic growth [4–8]. Since then, attention for this topic has waned. It is argued here that thermodynamics is still relevant and could enlighten the connection between solutions to environmental problems and increased complexity of the economy.

In summary, this article addresses a topic that has received little attention despite its potential relevance to solve environmental problems, namely, the relationship between system complexity and environmental sustainability. The ensuing discussion is guided by various questions: will sustainability solutions drive additional economic complexity? Does more complexity translate into additional energy use? Are fossil fuels essential to uphold a complex economy and will decarbonization imply a simpler economy? Should a low-carbon transition be guided by policies that limit additional complexity?

The remainder of this article is organized as follows. Section 2 discusses complexity from thermodynamic and systemic perspectives. Section 3 addresses the potential conflict between sustainability and complexity. Section 4 examines how to design policy to attune complexity and sustainability. Section 5 concludes the paper.

2. Thermodynamics and Complex Systems

Thermodynamics, i.e., the physics of energy, matters for understanding the complexity of the economy because energy rules the economic world. In another paper, we argued from a variety of angles that historically economic growth has been critically dependent on energy [9]. This critical importance can be observed in daily life—e.g., recently one could witness how energy scarcity and high prices resulting from Russia’s invasion of Ukraine dominated the economy (inflation), politics (recent election outcomes in Italy), and policy (canceling of various energy taxes in many countries). At the same time, one of the most pressing problems faced by humanity, climate change, is a straightforward consequence of energy consumption continuing to be highly dependent on fossil fuels. A quick and cost-effective decarbonization transition is crucial for a stable economy, good living conditions, and possibly even long-term survival of humans in many regions. Such a transition requires replacing fossil fuels by other sources of energy. The hope of many is now focused on exploiting wind, hydroelectric, and photovoltaic power and, in line with this, to electrify the economy. But this involves complexity increases through additional industries and infrastructure which may explain to some extent disappointing gains so far in CO₂ emissions reduction.

So, the economy shows through frequent changes an almost steady increase in complexity. Likewise, Darwinian evolution in nature tends to generate increasing biological complexity and specialization over time. The phenomenology of many natural systems shows that much of the world is made up by cyclic structures, supported by autocatalytic chemical reactions, that exist considerable distance from thermodynamic equilibrium.

To understand the energy basis of such processes, thermodynamics is helpful. The second law of thermodynamics states that every process of materials transformation, such as combustion, can do productive work but is irreversible. Moreover, it can yield complex long-lived artifacts, i.e., “create order out of disorder,” in a cyclic subsystem of the overall system. The Second Law also says that while total energy in a system remains constant, the usable fraction of it (exergy) declines with every transformation process. This causes the exergy to be converted to useless energy or waste heat (anergy), driving the cycle but keeping the total energy balanced and unchanged, i.e., complying with the law of energy conservation (the first law of thermodynamics). Moreover, while the usable fraction (exergy) of energy declines, the overall level of disorder (entropy) increases.

The relationship between exergy and information deserves a brief comment. The creation and transfer of information, at the microscale, is a physical process that involves exergy dissipation. This means that information has an exergy cost (or exergy content). A flux of exergy can “self-organize” into separate stable subsystems, many of which contain smaller subsystems and are contained within other and larger ones. All these cycles are maintained by a continuous flux of exergy from external sources. They are also stores of information. Technology has made it possible to create, transmit, and store information, as bits and bytes (e.g., in “the cloud”) almost without limit, albeit not without energy (exergy) consumption (dissipation) to drive the computers. This is one of the new kinds of complexity generated by self-organization, far from equilibrium. If information is exergy, and complexity is information, complexity of an (ecological or human-made) entity is proportional to the amount of exergy going into its construction. More specifically, is a stock of potential exergy dissipation (like a reservoir of oil) equivalent to a “complexity potential” or an information reservoir? In fact, it is very tempting to equate the capital stock of the world (or any subunit) with complexity per se. It is also tempting to equate the generation of complexity as an accumulation of Gibbs free energy in the chemical domain. One can further associate complexity with ecological diversity as well as with the diversity of contents of shopping malls—which does not mean that all diversity is equally valued. Indeed, as the opposite of complexity, there are situations where simplicity is desirable.

Erwin Schrödinger was one of the fathers of quantum mechanics. In his book “What is Life?” he set himself the ambitious task of drawing together the fundamental processes of biology, physics, and chemistry [10]. He noted that life combines two opposed processes, namely, “order from order” and “order from disorder.” Ilya Prigogine and his colleagues at the Free University in Brussels completed Schrödinger’s insight into a theory of nonequilibrium thermodynamics [11]. The Second Law admits for the emergence and existence of cyclic subsystems. This ranges from simple cycles within larger cycles, contained in still larger cycles. In a biological setting, this takes the form of structures ranging from simple cells to multicellular organisms, and beyond that to social systems consisting of multiple individual organisms. Such entities can only remain far from thermodynamic equilibrium through a permanent external input of exergy. The combination of all these cyclic processes and subsystems is crucial to the existence of life on planet Earth (and to the future of humankind, including your beloved pension fund).

The Second Law, in its basic form, implies that an isolated system—defined as having no exchange of energy and materials with outside systems—can exploit a flux of exergy to self-organize into a nested structured with multilayered subsystems. Examples are a star system, a solar system, and a planet. Under certain conditions, self-organization can occur by dissipating an exergy flux, which means converting the exergy to anergy. If the right ingredients are present, it is thus feasible to maintain an “island of stability and increasing order” far from Boltzmann’s thermodynamic equilibrium [12]. Such self-organized subsystems can survive as long as there is a permanent exergy input from another part of the overall system to compensate for local exergy dissipation. Self-organization can thus enable the creation and preservation of order, both in the form of complex material structures and as useful information or knowledge [13].

The formation and stability of such subsystems come at a cost of exergy while generating an increase in complexity. This has given rise to a new science about complexity, ranging from self-organization and far-from-equilibrium dynamics (e.g., [14]). While the mechanism for spontaneous creation of autocatalysis in chemicals is only partly understood, research on this is ongoing. Human organizations and institutions are also self-organized cycles, around information storage social organization—and technology development—in much the same way as biological organisms are organized around self-reproduction [15]. What is still not well understood is if complexity is good or bad for sustainability—and unfortunately there is hardly any research on this.

Complexity in economics has received meager attention [16]. It stresses that since agents are diverse, nonrational, possess incomplete information, innovate, and influence each other over idiosyncratic and changing social networks, the economy as a whole is never in a static equilibrium but shows volatile and emergent behavior. In addition, agent features, notably social interdependence, mean that economic dynamics are characterized by path dependence (unique historical patterns) and possibly lock-in, rather than flexibility and reversibility.

Important to all complex system is that their structure was not arranged with a preconceived goal or through hierarchical or central planning and control, but the cumulative outcome of a set of unintended and unplanned processes. This holds equally true for the universe, a living organism, and the economy. The term self-organization is used in various disciplines and research areas, notably physics (especially thermodynamics), chemistry, biology, cybernetics, computer science, and the social sciences. In social systems, self-organization has been referred by some authors as “autopoiesis” [17].

Self-organization denotes that a higher-level structure arises out of a disorganized system of lower-level physical components due to spontaneous processes involving interactions between these components, hence the popular phrase “order out of chaos” [11]. The organization of such a higher-level structure is unplanned and decentralized, also referred to as “distributed control” or “self-regulation.” The macrostructure can comprise physical, chemical, biological, and economic arrangements or even coordination of behaviors and activities of agents in a larger system. Even markets have been proposed as being self-organized, namely, through the spontaneous organizing force of Adam Smith’s “invisible hand.” An entirely different example of self-organization is cellular automata that generate visual patterns or emergent structures like networks or groups—with applications to sustainability transitions [18]. Work on them illustrates that simple rules at the microlevel can produce surprising, emergent patterns at the macrolevel.

To characterize in general terms the mechanisms underlying self-organization, one should note that it is driven by some type of gradient of exergy, matter, or information which causes associated currents, flows, or streams. For example, convection cells involve a temperature current, biological cells a nutrient flow, and social systems an information stream. A more abstract definition of self-organization is as a wide variety of complex processes at the “edge-of-chaos,” characterized by a trade-off between stability and flexibility [18]. The self-organized state is in between two extreme states: (1) a rigidly structured, ordered system with few connections and extreme stability but no flexibility; and (2) a random or chaotic system with many connections and extreme flexibility but no stability.

In summary, there is a variety of viewpoints on complexity, related to notions such as energy and material use, energy gradients, energy and material transformation, exergy dissipation, entropy increase, information flows, self-organization, far-from-equilibrium processes, positive and negative feedback, bounded rationality and social interaction, lock-in and path dependency, irreversibility, and distributed control. As we will see in next sections, several of these concepts can be explored to understand the relationship between complexity and sustainability.

3. Sustainability Versus Complexity

The relationship between complexity of the economy and environmental sustainability has received little attention in the literature on both environmental science and sustainability transitions. Regarding a static or existence theory, one could raise the question whether sustainability poses limits on complexity. The notion of “small is beautiful,” the belief in local communities, and the degrowth movement suggest an affirmative answer. Economics and innovation studies give more weight to scale economies and other increasing returns to scale, which—if scale and complexity are correlated—suggests a negative answer. Energy and climate studies stress unwanted side effects like rebound and leakage. From a dynamic angle, a relevant question is if the transition to a more sustainable economy involves mainly, or even steady, increases or decreases in system complexity. Resolving this will, in effect, help to clarify a transition theory informed about thermodynamics and complexity theory.

There are different (explicit and implicit) views about how complexity and sustainability are connected. Hall et al. [19] argue that a minimum performance of energy sources is needed, in terms of what they call ERO(E)I or energy return on energy investment. Since renewables involve a complex production network and use diffuse wind, wave, or direct solar energy, they may not achieve this easily. Hence, it is uncertain if sustainability is feasible if societies are to maintain high material consumption levels. In fact, the authors claim that the only way for renewables to achieve a sufficiently high EROI is to be indirectly based on, i.e., being “subsidized by,” fossil fuels.

Historian and anthropologist Tainter [20] argues that responses to environmental and resource problems tend to increase “cultural complexity”—comprising the economy, politics, society, technology, and information. In turn, the economy requires more energy and material inputs, creating a dilemma or paradox of insolvability or even worsening of the problem. This goes somewhat against the common view of history as increasing economic complexity going hand in hand with progress. Instead, Tainter’s conclusion is that future sustainability, and hence societal stability and progress, may be compromised by increasing complexity of the economy. This issue is also of relevance to recent pleas for circularity as the main solution to unsustainability [21]. To achieve this, a sharp increase is needed in recycling and reuse which will create new loops that make production systems more complex. In addition, material-to-energy shifting will occur, due to, e.g., melting plastics or glass, collection and transport, and reuse of old products that are energy-intensive in the use phase; another possible impact is material rebound, e.g., when regulations prohibit shops to provide plastic bags to customers resulting in more plastic packaging of the products themselves, such as vegetables.

The energy-rebound effect can be seen as a special case of Tainter’s thesis, in that technological improvements (higher energy efficiency) in response to energy resource scarcity give rise to more intense use of the technology (e.g., driving a fuel-efficient care more frequently or over longer distances) or new expenditures (due to saving money along with energy). Moreover, with increased complexity, a system will have more channels available for rebound to occur, in turn contributing to a potentially higher overall rebound effect [22]. To understand why (some type of) rebound is a special case of system complexity, note the following regularities of it: improved energy efficiency relieves resource limits, in turn stimulating use of additional activities and technologies; energy-intensive general-purpose technologies—such as engines, batteries, light gear, computers, transport vehicles, or new composites and materials—will diffuse to distinct activities in production and use phases; some efficiency improvements will translate into more technological complexity, such as an extra cycle in engines and a hybrid engine (electric and fuel combustion), adding components for storing breaking energy, or providing insulation materials; and finally, there are production chain and lifecycle effects translating in what is sometimes called “embodied energy” of the final product or service. The systemic relevance of rebound was recently supported by an ambitious review of economy-wide rebound, indicating it to be higher than 50% and reaching in some cases 100% [23] as well as by an ambitious study for the US, France, Germany, Italy, and the UK, which found macro-rebound to range from 78% to 101% [24].

Tainter’s best-known work, “The Collapse of Complex Societies” (1988), examines the collapse of Maya and Chacoan civilizations, and of the Western Roman Empire. He analyzed the subject in terms of network theory, energy economics, and complexity theory. Tainter argues that sustainability or collapse of societies follows from the success or failure of problem-solving institutions and that societies collapse when their investments in social complexity and their energy subsidies reach a point of diminishing marginal returns. He recognizes collapse when a society involuntarily sheds a significant portion of its complexity. He argues that responses to environmental and resource problems increase “cultural complexity”—comprising the economy, politics, society, technology, and information. In turn, the economy requires more exergy and material inputs, creating a dilemma or paradox of insolvability or even worsening of the problem. Tainter’s conclusion is that future sustainability will depend on continued high levels of exergy consumption.

It is easy to concur with much of this. Opposing unsustainable behavior with a host of individual regulations and rules, new oversight institutions, etc. increases societal expenses without necessarily yielding effective and efficient solutions. Especially point 5 is concerning. Indeed, in the three historical cases, Tainter noted that the civilization collapsed because it ran out of energy.

In summary, it is surprising that the fundamental question whether sustainability poses limits on complexity has been largely ignored in the immense literature on environmental science and sustainability transitions. Some approaches may be seen to answer it, but only very implicitly. For instance, the notion of “small is beautiful,” the belief in local communities, and the degrowth movement suggest simplification is a good strategy. Economics and innovation studies give more weight to scale economies and other increasing returns to scale, which—if scale and complexity are correlated—may suggest complexity is not a big concern. From a dynamic angle, a relevant issue is whether the transition to a more sustainable economy involves steadily increasing or decreasing system complexity, or instead a more random fluctuation rather than a trend. Addressing all the above questions may result in the construction of a theory about the relationship between complexity and sustainability.

4. Complexity Policy for Sustainability

The lesson of the previous sections is that one should redirect or possibly limit complexity where possible as it is likely to translate into more exergy use, more rebound channels, environmental problem shifting, and generally limited understanding and control by politicians. Some policy implications are as follows: design energy transitions that limit rebound by regulating the full scope of fossil fuel energy use; favor decentralized small-scale solutions over centralized infrastructures that necessarily mean distribution losses and more complexity (such as waste water treatment plants or electricity generation facilities); promote resource reduction over resource circularity especially in terms of plastics and metals; and make producers responsible for product life cycles.

A potentially negative insight of the thermodynamic tale of complexity is that both nature and economies tend to generate complexity without any preconceived plan. This in turn suggests that it may be difficult to avoid additional exergy use and environmental impact generated through creation and maintenance of socioeconomic complexity. On the other hand, there is a positive message too, namely, that considerable moderation of these effects seems possible, through policy approaches that limit complexity, as outlined hereafter.

To begin with, a systemic approach is required, both in terms of analysis and policy. A systemic framework and method can help to unfold complexity and its change. This involves assessing the indirect effects of well-intended strategies and policies, irrespective of whether these take the form of behavioral, technological, or institutional solutions. Such indirect impacts—including in the future without a clear time horizon—are inherent to complexity increases. In terms of systemic policy, one could think of incentives that discourage undesirable resource use or emissions (e.g., greenhouse gases) for the entire system. This will avoid that energy/carbon rebound and carbon leakage will occur, which would result in ineffective policy. Two instruments which can achieve this for energy use and CO₂ emissions are cap-and-trade (emissions trading or carbon market) and taxation (levies or charges), with the first having the advantage of putting a hard limit on the system and endogenously producing a necessary price that regulates a large number of emitters (producers and consumers) so that their joint emissions remain exactly within the limit set by the cap [22]. Any complexity that adds to energy use or carbon emissions and can be avoided will then be automatically discouraged—unless reducing emissions through more complexity elsewhere is feasible and more efficient or cost-effective. In other words, one would shift between different subsystem complexities that perform differently in terms of the combination of emissions and costs—resulting in a kind of optimizing system complexity for sustainability. For other environmental impacts, no such simple policy solution exists—which suggests a policy mix to limit shifts from climate solutions to other environmental problems [26].

Generally, a systemic policy might comprise pricing other substances than merely GHGs and virgin materials to assure their optimal use and recycling from a long-term environmental perspective. There is a lot of talk about the circular economy nowadays, but it is difficult to achieve as systemic policy is lacking, and hence different aspects are insufficiently optimized for minimum environmental impact [21]. Among others, this requires incentivizing local recycling such as nutrient recovery from water and solid waste [27], which would significantly reduce the amount of fossil fuels currently used for the production of mineral fertilizers through the Haber–Bosch process as well as the extraction of phosphate, a nonrenewable and quickly depleting critical raw material.

In addition, one must try to make innovation more ambitious in the sense of considering the indirect effects of new processes, products, and services, such as on material and energy use. This can benefit from a lifecycle perspective in innovation processes that addresses the indirect—not only the direct—environmental effects of the outcomes of innovation. Possibly, approval of innovations might involve that comparative systemic environmental impact analysis is undertaken to assure that the selected option scores best in terms of total, i.e., direct and indirect, environmental impacts. Implementing a life cycle framework in policy design which includes the impacts associated with the extraction, manufacturing, use, and end-of-life stages of products (ISO 14040) will help avoid narrow, partial analysis that may overlook important variables or lead to unintended consequences. For example, electric vehicles have been found to have considerable global warming impact during production as well as indirect environmental impacts associated with material use, while in consumption, global warming impacts depend on whether the electric mix used to charge them is heavily dependent on coal [28].

Finally, more imaginative policies may be considered, for example, giving producers responsibility for product life cycles, so that complex designs which hinder recycling are discouraged. If all the stages of a life cycle are controlled by one entity, it will be easier to find an optimal solution and trade-off of distinct strategies, such as reduce, reuse, and recycling. One might also punish activities, project, investments, and innovations that increase complexity considerably with clear environmentally negative impacts. But before that can be done, deeper knowledge is needed of the connection between sustainability and complexity, in the form of both general insights and case studies.

The first three points are much debated in politics and business. Points 4 and 5 are really subdivisions of point 3 (“pay!”). Point 6 points at a weakness of political systems, namely, that there is no clear methodology for cost-accounting of regulation-and-bureaucracy. One can analyze a problem and may even assess the public cost of it, but then proceed by creating taxes, rules, and ad hoc regulations without truly assessing their indirect cost. Of course, taxes and more generally pricing instruments (including cap-and-trade) are argued by economists as a tool to decentralize solutions and select cost-effective ones, hence minimizing the burden for society of a transition to an economy with less environmental pressure. Tainter himself mentions, albeit in passing, that such accounting needs to be accurate with respect to externalized costs, i.e., costs that society or future generations will have to bear, when the snow buildup finally triggers an avalanche. Point 7, “Recalibrate or revolutionize the activity,” refers to the (rare) options where through social or technological innovation, one type of complexity can be exchanged for another, very different one, like using fossil fuel energy and machines instead of slaves, or electric motors instead of steam engines. Taking CO₂ emissions as an example, the pricing of these will—if well designed, i.e., to cover all emissions—result in everyone paying prices for goods, services, materials, and intermediate products that reflect all the indirect emissions. This means in effect that complexity increases which translate into more indirect emissions will be priced and hence discouraged. This line of thought could, of course, be extended to other substances that cause environmental pressure.

For developing countries, many of the abovementioned policies may be more effective than for developed countries. The reason is that they have a less mature economy, allowing for distinct future paths of building an economic structure that meets the goals of resource and energy efficiency. Preventing unnecessary and unsustainable complexity increases is a timely and highly relevant strategy for two reasons. First, the share of developing countries in global CO₂ emissions and other environmental pressures (e.g., pollution of plastics) is gradually increasing. Second, the proposed complexity-limiting policies prevent environmental pressures rather than first creating them, trying to solve them, and then realizing that complexity increases due to solutions hamper sustainability. Of course, this road is paved with challenges as a more complex economy may be associated with less poverty [29], general development [30], and macroeconomic stability [31]. Further research on how such correlations limit sustainability is warranted.

We agree with Tainter that management of complexity is important and often undervalued. Sustainability is not achieved through business-as-usual, by simply creating more and more complexity in the form of bureaucracy and institutions at problems. It is too early to make definite statements about whether complexity forbids sustainability, or whether it is possible to make sufficient renewable energy to operate even more complex and energy-intensive societies. But we also see risks in focusing too much on the antithesis of complexity, that is, simplification. Examples are “Small is Beautiful” (Schumacher), Jeffersonian democracy, New England town halls, anti-government ranchers who want to shoot “varmints” and those who repeat the Reagan mantra – that “government is not the solution, government is the problem”. In an ideal world, simplification would release resources (like workers, capital, and infrastructure) from maintaining an untenable past and allow them to be used for implementing a sustainable future, with renewable energy, circular economy, and growth in the quality of life rather than in waste. Libertarians effectively say “let us do away with all the old complexities, destroy them, and make room for something new.” Such “simple solutions” are what Trump, Empire Loyalists, Brexiteers, and the extreme right in Europe are thriving on. It is what convinced, after perceiving a failure of their new democratic systems in the 1920s and 1930s, the Germans, Spanish, and Italians to vote their respective fascists into power. To avoid such simplistic and socially damaging worldviews, a more convincing theory of complexity and sustainability is needed.

With regard to climate change, a seemingly noncomplex strategy is to focus on renewable energy and forget about all the rest—no worry about energy efficiency improvements or changes in production and consumption. However, renewables are unlikely to diffuse sufficiently rapidly to decarbonize the economy so as to meet the Paris 2°C target [32]. And even if one believes this is possible, putting all one’s eggs in the renewables’ basked is risky. A safe strategy involves restructuring the economy through systemic policies to guide the transition of a complex economy, with the aim of moving quickly away from fossil fuels through a combination of reducing the carbon intensity of energy sources as well as changing the technologies and composition of production and consumption. This means triggering more energy-efficient technologies in industries and households, modifying the (KLEM) input mix of production, and altering the consumption basket. The latter will go along with parallel changes in sector structure and international trade patterns. No need to plan all this—it is just too complex. Instead, likely policies are required that set limits, such as cap-and-trade systems, and then let the economic system find the feasible and cost-effective structure that matches the limits. This will possibly involve the economy still becoming more complex in certain domains, but possibly not everywhere.

In summary, there are several ideas about “complexity policy” as part of wider sustainability policy. Setting aggregate and hard physical limits for economies transitioning to sustainability seems the best candidate solution. Since the limits need to be allocated, there is a useful role to be played by cap-and-trade. This will translate additional complexity into a price for those responsible, and hence the complexity cost of decisions will be automatically accounted for by consumers, producers, investors, and innovators. As a result, overall complexity and particularly unsustainable complexity will be discouraged. Since not all environmental problems can be controlled this way, complementary measures will be needed as well, adding to complexity of the policy mix itself. The environmental (and other) costs of the latter merit attention from research as well.

5. Discussion, Conclusions, and Further Research

This article has explored the relationship between environmental sustainability and complexity of the socioeconomic system. It argued and illustrated that solutions to environmental challenges often create additional complexity in the overall socioeconomic system, at local to global levels, which might hamper the ultimate achievement of global sustainability. As solution responses to environmental problem emerge, system complexity gradually but steadily increases. Surprisingly, while this theme seems timely and relevant, it is rather ignored in the broader study of environmental sustainability and sustainability transitions. Developing more insights about it can serve as a basis to understand related topics that are currently examined in isolation—e.g., energy rebound, carbon leakage, green paradox (fossil fuel market responses to climate policy), and environmental problem shifting. More generally, additional complexity triggers indirect physical effects that contribute to particular environmental pressures. For effective solutions, it is necessary to understand the full set of indirect effects, their nature, and their magnitude. This means a serious challenge for sustainability research.

To provide a basic framework, the relationship between complexity and sustainability was examined from thermodynamic and systemic perspectives. This gave rise to a set of mechanisms of complexity increase which make meeting sustainability goals hard. It was argued that this issue is of great relevance to developing countries as their economies are likely to undergo considerable complexity increases in coming decades, when at the same time everything must be done to stay within multiple planetary boundaries. This raises the question whether countries will be able to steer their development in a sustainable direction that limits complexity of their production structure.

The exposition ended with discussing the design and implementation of “complexity policy.” We outlined several ideas in this regard. An important role can be played by cap-and-trade, which will result in complexity increases that translate into more indirect CO₂ emissions being priced and hence discouraged. This approach could, in principle, be extended to other substances that contribute to environmental pressures. However, it should be recognized that this cannot control all environmental problems and so complementary measures are needed to regulate other pressures. This suggests the need for a well-designed policy mix where one should also try to prevent unnecessary complexity of the overall policy itself—e.g., not piling up instruments—in order to avoid unintended indirect effects of policy itself. Another reason for keeping policy as simple and transparent as possible is that all major sustainability challenges are of a global nature, implying that policy coordination if not harmonization is essential for support of effective instruments at national to local levels—to avoid concerns about competitive positions and associated free-riding tendencies of countries and other jurisdictions.

Evidently, this problem-setting and framing paper does not offer definite solutions for dealing with the potential conflict between sustainability and complexity. Further research could look into various issues. First, one could quantify how much additional complexity and associated environmental pressures result due to sustainability solutions. This provides a serious challenge for research as it will involve quantifying sustainability as well as complexity. Second, one might study which alternative strategies and instruments perform relatively well in controlling complexity increases and associated environmental pressures. This can pay attention, among others, to whether centralized or decentralized solutions (e.g., PV panels on every rooftop) are most desirable or if megaprojects have a place and if so in which form and under which conditions. Third, it would be good to link the complexity-vs-sustainability question to circular economy research as the latter is all about loops and feedback, which are basic to system complexity. Fourth, one could use complexity methods to better understand the commonalities of related concepts and problems, such as energy rebound, carbon leakage, green paradox, and environmental problem shifting. Achieving more coherence in this regard may help to define effective or integrated policy responses. Fifth, it is probably useful to try to learn from how complexity and its change are studied in other disciplines, such as physics, chemistry, and biology. In particular, evolutionary thinking in biology and other areas may be useful as evolution in effect is a spontaneous, gradual change in complex systems in response to solving all kinds of decentralized problems and challenges. Sixth, research might devote attention to whether the finding for historical human societies by Joseph Tainter—that a society can be destroyed by the cost of environmental problem-solving—applies to current global society. Finally, on a more theoretical level, it is worthwhile to examine whether there exists some optimal level of complexity, due to diminishing and ultimately negative returns of additional complexity for society.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding

This research was supported by the “María de Maeztu” Programme for Units of Excellence of the Spanish Ministry of Science and Innovation with a grant awarded to ICTA-UAB (CEX2019-000940-M).