Rethinking Economics Using Complexity Theory: Dirk Helbing & Alan Kirman
In this paper we argue that if we want to find a more satisfactory approach to tackling the major socio-economic problems we are facing, we need to thoroughly rethink the basic assumptions of macroeconomics and financial theory. Making minor modifications to the standard models to remove “imperfections” is not enough, the whole framework needs to be revisited.
LE MONDE SCIENCE ET TECHNO | 04.04.2013 à 16h55 • Mis à jour le 06.04.2013 à 22h15
Dans une étude récente, deux chercheurs expliquent qu'au grand dam des analystes financiers, en quête de justifications rationnelles aux fluctuations des cours, les marchés demeurent obstinément imprévisibles.
We study fifteen months of human mobility data for one and a half million individuals and find that human mobility traces are highly unique. In fact, in a dataset where the location of an individual is specified hourly, and with a spatial resolution equal to that given by the carrier's antennas, four spatio-temporal points are enough to uniquely identify 95% of the individuals. We coarsen the data spatially and temporally to find a formula for the uniqueness of human mobility traces given their resolution and the available outside information. This formula shows that the uniqueness of mobility traces decays approximately as the 1/10 power of their resolution. Hence, even coarse datasets provide little anonymity. These findings represent fundamental constraints to an individual's privacy and have important implications for the design of frameworks and institutions dedicated to protect the privacy of individuals.
Biological competition is widely believed to result in the evolution of selfish preferences. The related concept of the ‘homo economicus’ is at the core of mainstream economics.
However, there is also experimental and empirical evidence for other-regarding preferences. Here we present a theory that explains both, self-regarding and other-regarding preferences.
Assuming conditions promoting non-cooperative behaviour, we demonstrate that intergenerational migration determines whether evolutionary competition results in a ‘homo economicus’ (showing self-regarding preferences) or a ‘homo socialis’ (having other-regarding preferences).
Our model assumes spatially interacting agents playing prisoner's dilemmas, who inherit a trait determining ‘friendliness’, but mutations tend to undermine it. Reproduction is ruled by fitness-based selection without a cultural modification of reproduction rates.
Our model calls for a complementary economic theory for ‘networked minds’ (the ‘homo socialis’) and lays the foundations for an evolutionarily grounded theory of other-regarding agents, explaining individually different utility functions as well as conditional cooperation.
Dirk Helbing is the scientific coordinator of the FuturICT initiative, elected member of the German Academy of Sciences "Leopoldina"
and of the World Economic Forums Global Agenda Council on Complex Systems, and internationally well-known for his work on crowds, traffic, and social behavior.
Complex Adaptive Systems Modeling(CASM) is a highly multidisciplinary modeling and simulation journal that serves as a unique forum for original, high-quality peer-reviewed papers with a specific interest and scope limited to agent-based and complex network-based modeling paradigms for Complex Adaptive Systems (CAS). The highly multidisciplinary scope ofCASM spans any domain of CAS. Possible areas of interest range from the Life Sciences (E.g. Biological Networks and agent-based models), Ecology (E.g. Agent-based/Individual-based models), Social Sciences (Agent-based simulation, Social Network Analysis), Scientometrics (E.g. Citation Networks) to large-scale Complex Adaptive COmmunicatiOn Networks and environmentS (CACOONS) such as Wireless Sensor Networks (WSN), Body Sensor Networks, Peer-to-Peer (P2P) networks, pervasive mobile networks, service oriented architecture, smart grid and the Internet of Things.
How and why our conventional economic thinking causes global crises (discussion paper)20-02-13
Dik Helbing (ETH Zurich)
I believe it’s no wonder that our world is in trouble. We currently lack the global systems science needed to understand our world, which is now changing more quickly than we can collect the experience required to cope with upcoming problems. We can also not trust our intuition, since the complex systems we have created behave often in surprising, counter-intuitive ways. Frequently, their properties are not determined by their components, but their interactions. Therefore, a strongly coupled world behaves fundamentally different from a weakly coupled world with independent decision-makers. Strong interactions tend to make the system uncontrollable – they create cascading effects and extreme events.
As a consequence of the transition to a more and more strongly coupled world, we need to revisit the underlying assumptions of the currently prevailing economic thinking. In the following, I will discuss 10 widespread assertions, which would work in a perfect economic world with representative agents and uncorrelated decisions, where heterogeneity, decision errors, and time scales do not matter. However, they are apparently not well enough suited to depict the strongly interdependent, diverse, and quickly changing world, we are facing, and this has important implications. Therefore, we need to ‘think out of the box’ and require a paradigm shift towards a new economic thinking characterized by a systemic, interaction-oriented perspective inspired by knowledge about complex, ecological, and social systems. As Albert Einstein noted, long-standing problems are rarely solved within the dominating paradigm. However, a new perspective on old problems may enable new mitigation strategies
1. More networking is good and reduces risks
Many human-made systems and services are based on networking. While some degree of networking is apparently good, too much connectivity may also create systemic risks and pathways for cascading effects. These may cause extreme events and global crises like the current financial crisis.
Moreover, in social dilemma situations (where unfair behavior or cheating creates individual benefits), too much networking creates a breakdown of cooperation and trust, while local or regional interactions may promote cooperation. The transformation of the financial system into a global village, where any agent can interact with any other agent, may actually have been the root cause of our current financial crisis.
Countermeasures: Limit the degree of networking to a healthy amount and/or introduce adaptive decoupling strategies to stop cascading effects and enable graceful degradation (including slow-down mechanisms in crisis situations). Support the evolution and co-existence of several weakly coupled financial systems (to reduce systemic vulnerability, stimulate competition between systems, and create backup solutions). Reduce the complexity of financial products and improve the transparency of financial interdependencies and over-the-counter transactions by creating suitable information platforms.
2. The economy drives towards an equilibrium state
Current economic thinking is based on the assumption that the economic system is in equilibrium or at least tends to develop towards a state of equilibrium. However, today’s world changes faster than many companies and policies can adapt. Therefore, the world economic system is unlikely to be in equilibrium at any point in time. It is rather expected to show a complex non-equilibrium dynamics.
Therefore, a new economic thinking inspired by complex dynamical systems, ecosystems, and social systems would be beneficial. Such a perspective would also have implications for the robustness of economic systems. Overall, beneficial properties seem to be: redundancy, variety, sparseness, decoupling (separated communities, niches), and mutually adjusted time scales (which are required for hierarchical structures to function well).
Countermeasures: Invest into new economic systems thinking. Combine the axiomatic, mathematical approach of economics with a ‘natural science approach’ based on data and experiments. Develop non-equilibrium network models capturing the self-organized dynamics of real economic systems. Pursue an interdisciplinary approach, taking on board complex, ecological and social systems thinking. Develop better concepts for systemic risk assessment, systems design, and integrated risk management.
3. Individuals and companies decide rationally
The ‘homo economicus’ is a widely used paradigm in economics. It is the basis of a large and beautiful body of mathematical proofs on idealized economic systems. However, the paradigm of a strictly optimizing, perfect egoist is a model, which is questioned by theoretical and empirical results.
Theoretically, the paradigm assumes unrealistic information storage and processing capacities (everyone would need to have a full 1:1 representation of the entire world in the own brain and an instant data processing of huge amounts of data, including the anticipation of future decisions of others). Empirically, one finds that people behave in a more cooperative and fair way than the paradigm of the ‘homo economicus’ predicts. In particular, the paradigm neglects the role of errors, emotions, other-regarding preferences, etc. This implies significant deviations of real human behaviors from theoretically predicted ones.
Countermeasure: Use a combination of interactive behavioral experiments, agent-based modeling, data mining and social supercomputing to study (aspects of) real(istic) economic systems
4. Selfish behavior optimizes the systemic performance and benefits everyone
Another pillar of conventional economic thinking is Adam Smith’s principle of the ‘invisible hand’, according to which selfish profit maximization would automatically lead to the best systemic outcome based on self-organization. It is the basis of the ideology of ‘free markets’, according to which regulation would tend to reduce the performance of economic systems.
However, models in evolutionary game theory show that self-organized coordination in markets can easily fail, even when market participants have equal power, symmetrical information etc. Moreover, even if the individually optimal behavior also maximizes system performance and if everybody behaves very close to optimal, this may still create a systemic failure (e.g. when the system optimum is unstable). Therefore, it is highly questionable whether the systemic inefficiencies resulting from individual optimization efforts can always be compensated for by greedy motivations (such as trying to get more than before or more than others).
Countermeasures: Measure the system state in real-time and respond to this information adaptively in a way that promotes coordination and cooperation with the interaction partners. Create an information and communication system supporting collective (self-)awareness of the impact of human actions on our world. Increase opportunities for social, economic and political participation
5. Financial markets are efficient
One implication of Adam Smith’s principle of the ‘invisible hand’ is the efficiency of financial markets, according to which any opportunity to make money with a probability higher than chance would immediately be used, thereby eliminating such opportunities and any related market inefficiencies.
Efficient markets should not create bubbles and crashes, and therefore one would not need contingency plans for financial crises (they could simply not occur). Financial markets would rather be in equilibrium as the conventional Dynamic Stochastic General Equilibrium Models suggest. However, many people believe that bubbles and crashes do occur. The flash crash of May 6, 2010, is a good example of a market irregularity, which has repeatedly occurred in the meantime. Also, many financial traders do not seem to believe in efficient markets, but rather in the existence of opportunities that can be used to make over-proportional profits.
Countermeasures: Develop contingency plans for financial crises. Adjust the financial architecture and identify suitable strategies (such as breaking points) to stop cascading effects in the financial system. Introduce noise into financial markets by random trading transactions to destroy bubbles before they reach a critical size that may have a disastrous systemic impact.
6. More information and financial innovations are good
One common view is that market inefficiencies result from an unequal distribution of power, which partially results from information asymmetries (“knowledge is power”). Therefore, providing more information to everyone should remove the related inefficiencies.
However, too much information creates a cognitive information overload. As a result, people tend to orient at other people’s behaviors and information sources they trust. As a consequence, people do not anymore take independent decisions, which can undermine the ‘wisdom of crowds’ and market efficiency. One example is the large and unhealthy impact that the assessments of a few rating agencies have on the global markets.
It is also believed that financial innovations will make markets more efficient by making markets more complete. However, it has been shown that complete markets are unstable. In fact, leverage effects, ‘naked’ short-selling (of assets one does not own), credit default swaps, high-frequency trading and other financial instruments may have a destabilizing effect on financial markets.
Countermeasures: Identify and pursue decentralized, pluralistic, participatory information platforms, which support the ‘wisdom of crowds’ effect. Test financial instruments (such as derivatives) for systemic impacts (e.g. by suitable experiments and computer simulations) and certify them before they are released, as this is common in other economic sectors (special safety regulations apply, for example, in the electrical, automobile, pharmacy and food sectors).
7. More liquidity is better
Another wide-spread measure to cure economic crises are cheap loans provided by central banks. While this is intended to keep the economy running and to promote investments in the real economy, most of this money seems to go into financial speculation, since business and investment banks are not sufficiently separated.
This can cause bubbles in the financial and real estate markets, where much of these cheap loans are invested. However, the high returns in the resulting ‘bull markets’ are not sustainable, since they depend on the continued availability of cheap loans. Sooner or later, the created bubbles will implode and the financial market will crash (the likelihood of which goes up when the interest rates are increased). This again forces central banks to reduce interest rates to a minimum in order to keep the economy going and promote investments and growth. In other words, too much liquidity is as much of a problem, as is too little.
Countermeasure: Separate investment from business banking and introduce suitable adaptive transaction fees.
8. All agents can be treated as if acting in the same way
The ‘representative agent approach’ is another important concept of conventional economic thinking. Assuming that everyone would behave optimally, as the paradigm of the ‘homo economicus’ predicts, in equivalent situations everybody should behave the same. This allows one to replace the interaction of an economic agent with other agents by interactions with average agents, in particularly if one assumes that everyone has access to the same information and participates in perfect markets.
However, the representative agent model cannot describe cascade effects well. These are not determined by the averagestability, but by the weakest link. The ‘representative agent approach’ also neglects effects of spatial interactions and heterogeneities in the preferences of market participants. When these are considered, the conclusions can be completely different, sometimes even opposite (e.g. there may be an ‘outbreak’ rather than a breakdown of cooperative behavior).
Finally, the representative agent approach does not allow one to understand particular effects of the interaction network structure, which may promote or obstruct cooperativeness, trust, public safety, etc. Neglecting such network effects can lead to a serious underestimation of the importance of ‘social capital’ for the creation of economic value and social well-being.
Countermeasures: Protect economic and social diversity. Allow for the existence of niche markets and for the consideration of justified local advantages. Avoid competition on one single dimension (e.g. economic value generation) and promote multi-criterion incentive systems. Develop better compasses for decision-making than GDP per capita, taking into account environmental, health, and social factors. Make social capital (such as cooperativeness, trust, public safety, …) measurable.
9. Regulation can fix the imperfections of economic systems
When the self-organization of markets does not work perfectly, one often tries to ‘fix the problem’ by regulation. However, complex systems cannot be steered ‘like a bus’, and many control attempts fail. In many cases, the information required to regulate a complex system is not available, and even if one would have a surveillance system that monitors all variables of the system, one would frequently not know what the relevant control parameters are. Besides, suitable regulatory instruments are often lacking.
A more promising way to manage complexity is to facilitate or guide favorable self-organization. This is often possible by modifying the interactions between the system components. It basically requires one to establish targeted real-time information feedbacks, suitable ‘rules of the game’, and sanctioning mechanisms. To stay consistent with the approach of self-organization, sanctioning should as far as possible be done in a decentralized, self-regulatory way (as it is characteristic for social norms or the immune systems).
Countermeasures: Pursue a cybernetic and synergetic approach, promoting favorable self-organization by small changes in the interactions between the system elements, i.e. by fixing suitable ‘rules of the game’ to avoid instabilities and suboptimal systemic states. (Symmetry, fairness, and balance may be such principles.) Introduce a global but decentralized and manipulation-resistant multi-criterion rating system, community-specific reputation system, and pluralistic recommender system encouraging rule-compatible behavior.
10. Moral behavior is always costly
Species that do not strictly optimize their benefits are often assumed to disappear eventually due to the principles of natural selection implied by the theory of evolution. As a consequence, a ‘homo economicus’ should remain, and moral decision-making, which constrains oneself to a subset of available options, would vanish.
This certainly applies, if one forces everybody to interact with everybody else on equal footing, as the concept of perfect, free markets demands. In fact, evolutionary game-theoretical models show that these are conditions under which a ‘tragedy of the commons’ tends to occur, and where cooperation, fairness and trust tend to erode. On the other hand, social systems have found mechanisms to avoid the erosion of social capital. These mechanisms include repeated interactions, reputation effects, community interactions, group competition, sanctioning of improper behavior etc. In particular, decentralized market interactions seem to support fairness.
Countermeasures: Promote value-sensitive designs of monetary systems and information and communication systems. For example, introduce two co-existing, interacting, competitive exchange systems: one for anonymous (trans)actions (as we largely have them today) and one for accountable, traceable (trans)actions (creating ‘social’ money or information). Additionally, introduce suitable transaction costs to create incentives for accountable, responsible (trans)actions and to promote ethical behavior.
In conclusion, we are now living in a strongly coupled and strongly interdependent world, which poses new challenges. While it is probably unrealistic to go back beyond the level of networking and globalization we have reached, there is a great potential to develop new management approaches for our complex world based on suitable interaction rules and adaptive concepts, using real-time measurements.
It must be underlined that our current financial and economic problems cannot be solved within the current economic mainstream paradigm(s). We need to change our perspective on the financial and economic system and pursue new policies. The following recommendations are made:
Adjust the perspective of our world to the fundamentally changed properties of the globalized, strongly interdependent techno-socio-economic-environmental system we have created and its resulting complex, emergent dynamic system behavior.
Make large-scale investments into new economic thinking, particularly multi-disciplinary research involving knowledge from sociology, ecology, complexity science, and cybernetics.
Support diversity in the system, responsible innovation, and multi-dimensional competition.
Recognize the benefits of local and regional interactions for the creation of social capital such as cooperativeness, fairness, trust, etc.
Require an advance testing of financial instruments and innovations for systemic impacts and restrict destabilizing instruments.
Identify and establish a suitable institutional framework for interactions (suitable ‘rules of the game’) in order to promote a favorable self-organization.
Implement better, value-sensitive incentive systems to foster more responsible action.
Establish a universal, global reputation system to promote fair behavior and allow ethical behavior to survive in a competitive world.
Create new compasses for political decision-making, considering environment, health, social capital, and social well-being.
Develop new tools to facilitate the assessment of likely consequences of our decisions and actions (the ‘social footprint’). These tools may, for example, include:
-a 'Planetary Nervous System' to enable collective awareness of the state of our world and society in real-time,
-a 'Living Earth Simulator' to explore side effects and opportunities of human decisions and actions,
-a 'Global Participatory Platform' to create opportunities for social, economic and political participation,
-exchange systems that support value-oriented interactions.
Why does the adherence to norms not prevent conflict? While the current literature focuses on the emergence, maintenance and impact of norms with regard to cooperation, the issue of norm-related conflict deserves more attention.
We develop a general game theoretical model of “normative conflict” and explain how transaction failures on the macrolevel can result from cooperative motives on the microlevel. We differentiate between two kinds of conflict. The first results from distinct expectations regarding the way in which general normative obligations should be fulfilled, the second from distinct expectations as to how the norm should restrain actions based on self-interest.
We demonstrate the empirical relevance of normative conflict in a version of the ultimatum game. Our data reveal widespread normative conflict among different types of actors – egoistic, equity, equality and cherry picker. Our findings demonstrate how cooperative intentions about how to divide a collectively produced good may fail to produce cooperative outcomes.
Are random trading strategies more successful than technical ones? A.E.Biondo, A.Pluchino, A.Rapisarda, D.Helbing
In this paper we explore the specific role of randomness in financial markets, inspired by the beneficial role of noise in many physical systems and in previous applications to complex socio- economic systems. After a short introduction, we study the performance of some of the most used trading strategies in predicting the dynamics of financial markets for different international stock exchange indexes, with the goal of comparing them with the performance of a completely random strategy. In this respect, historical data for FTSE-UK, FTSE-MIB, DAX, and S&P500 indexes are taken into account for a period of about 15-20 years (since their creation until today).
“True capitalists are other-regarding” – “Perhaps we have applied the wrong theory, and our economy should be run by different people” The body of economic literature will have to change, implies a groundbreaking discovery. In their computer simulations of human evolution, scientists at ETH Zurich find the emergence of the “homo socialis” with “other-regarding” preferences. The results explain some intriguing findings in experimental economics and call for a new economic theory of “networked minds”. Economics has a beautiful body of theory. But does it describe real markets? Doubts have come up not only in the wake of the financial crisis, since financial crashes should not occur according to the then established theories. Since ages, economic theory is based on concepts such as efficient markets and the “homo economicus”, i.e. the assumption of competitively optimizing individuals and firms. It was believed that any behavior deviating from this would create disadvantages and, hence, be eliminated by natural selection. But experimental evidence from behavioral economics show that, on average, people behave more fairness-oriented and other-regarding than expected. A new theory by scientists from ETH Zurich now explains why.
Agents of influenceRobert Frederick, Science Writer
Models of complex systems have become a staple of business strategy, and now they are showing early promise for improving economic forecasts.
For cargo carriers, the most direct route is not always the cheapest. In the early 2000s, Southwest Airlines adopted a new approach to shipping: Rather than switching cargo from one flight to another to minimize the distance it covered, the airline would take circuitous routes to destinations on fewer flights. The strategy seemed counterintuitive, but it has saved the company millions of dollars in storage rentals and cargo handlers’ wages.
The 2008 financial crisis has highlighted major limitations in the modelling of financial and economic systems. However, an emerging field of research at the frontiers of both physics and economics aims to provide a more fundamental understanding of economic networks, as well as practical insights for policymakers. In this Nature Physics Focus, physicists and economists consider the state-of-the-art in the application of network science to finance.
Net gains -p119
Physics — and physicists — have had much to contribute to economic and finance. Now the science of complex networks sets a way forward to understanding and managing the complex financial networks of the world's markets.
PDF (138KB)- Net gains
Network opportunity -pp121 – 122
Michele Catanzaro and Mark Buchanan
Our developing scientific understanding of complex networks is being usefully applied in a wide set of financial systems. What we've learned from the 2008 crisis could be the basis of better management of the economy — and a means to avert future disaster.
Complex derivatives -pp123 – 125
Stefano Battiston, Guido Caldarelli, Co-Pierre Georg, Robert May and Joseph Stiglitz
The intrinsic complexity of the financial derivatives market has emerged as both an incentive to engage in it, and a key source of its inherent instability. Regulators now faced with the challenge of taming this beast may find inspiration in the budding science of complex systems.
PDF (152KB)- Complex derivatives
Reconstructing a credit network -pp125 – 126
Guido Caldarelli, Alessandro Chessa, Andrea Gabrielli, Fabio Pammolli and Michelangelo Puliga
The science of complex networks can be usefully applied in finance, although there is limited data available with which to develop our understanding. All is not lost, however: ideas from statistical physics make it possible to reconstruct details of a financial network from partial sets of information.
Reconstructing a credit network
The power to control -pp126 – 128
Marco Galbiati, Danilo Delpini & Stefano Battiston
Understanding something of the complexity of a financial network is one thing, influencing the behaviour of that system is another. But new tools from network science define a notion of 'controllability' that, coupled with 'centrality', could prove useful to economists and financial regulators.