New Economic Thinking

New economic thinking for me involves the study of economic phenomena from a perspective which sees economic systems as being non-linear and dynamic. This approach is new because it models the interactions among agents in a more complex and realistic way than in much of the standard economics. The complexity approach enables us to gain an alternative understanding of how aggregate-level properties emerge from micro-level behaviours.

In my own research, on cooperation in agricultural collectives, researches have argued that households in collectives tend to shirk collectively because shirking is a rational choice for individual households, and as a consequence mutual shirking (i.e. non-cooperation) results in Nash equilibrium. Such logic is widely used to explain the failure of agricultural collectivization. However, this argument fails to explain the existence of successful agricultural collectives, low efficient agricultural collectives (e.g. People’s communes in China) that are sustained for long periods of time and the emergence of private farms (e.g. household responsibility system in China) from strict collectives. I believe this is because using standard economics approach to model agricultural collectives it is not possible to model all the non-linear dynamics that can be found in real agricultural collectives. A complexity approach makes it possible to model the complex interactions between households, and between households and government. It is also possible to include aggregate-level features, like social cognition and trust that emerge as a consequence of the long-term interactions in collectives, which can impact on individual level decision making processes. These interactions shape and reshape the way households behave in various ways and at different times, and being able to include them in economic models will better enable us to understand economic phenomena.

In order to model economic phenomena as complex and non-linear systems it is possible to use agent-based simulation, which is a more flexible means of modelling than equation-based modelling. Using agent-based models it is possible to create heterogeneous agents (e.g. households, collectives) that have multiple attributes (e.g. marginal productivity of effort) and preferences (e.g. preference for risk), as well as being able to conduct bottom-up analysis, test deviations from rational choice theory, and include multiple ideas from across the social sciences.

Complexity economics is able to improve economic thinking in a number of different aspects. The demand for new economic thinking comes from a number of arenas.

1. The research objects, economic phenomena, economics faces are complex. The complexity keeps growing as the increase of communication and interaction amongst economic actors. The fact requires economist to improve the way they deal with the complexity in economic systems.
2. The public, as the final consumer of economic analysis, have been let down by poor economic predictions. Much of the public has little faith in the ability of economists to provide accurate information about the economy since the 2008 financial crisis. It may be possible to regains their trust by Taking the complexity of economic systems into consideration is necessary to improve economist’s work.
3. Complexity economics offers a new paradigm of examining economic phenomena. This paradigm, different from reductionism that standard economics applies, emphasizes non-linear dynamics of economic systems, and as a consequence deal with economics phenomena in a more realistic way. By combining with modern (computational) analysing tools, complexity economics is expected to compensate several disadvantages of standard economic research both theoretically and methodologically.

It’s worth mentioning that I believe complexity economics is complementary, rather than substitutive of, standard economics. Each of them bears its own strengths and weaknesses. For example, complexity economics is be able treat phenomena more realistically, but it is difficult to find a rule of modelling to follow, which can confuse researchers. Standard economics can present its ideas through clear logical deduction (with the aid of mathematic formulas), but it relies too much on strong assumptions, which undermines its realizability. Therefore, it is best to cooperate rather than compete with each other. This is especially essential for complexity economics, which at its stage of coming into maturity.