8 More Concepts of Complexity Economics: Complexity Economics Series Part 4

Published on INFORMS Analytics Magazine (Joseph Byrum)

Author’s note: This series Seeing Economic Collapse and Recovery Through the Lens of Complexity Economics will look at the pandemic-related economic problems we currently face and how we might apply important concepts of complexity economics to better understand how to move forward. Part Three introduces some foundational complexity economics concepts.

The previous installment of this blog considered the four primary concepts of complexity economics. Following are eight more concepts emerging from a review of complexity literature, along with related research and recommendations for each.

1. Animal spirits

A description of economic behavior driven more by impulse than by reason; often associated with herd behavior. This concept, which originated with English philosopher David Hume and was popularized by British economist John Maynard Keynes, predates complexity science but is compatible with it.

Research: In 1987, reflecting on the slow economic recovery from the Great Depression, Robert Higgs found that evidence from public opinion polls and corporate bond markets showed that the New Deal policies “prevented a robust recovery of long-term private investment by significantly reducing investors’ confidence in the durability of private property rights” [1]. In a 2012 reflection on this essay, J. Robert Subrick of James Madison University found that “Animal spirits have unsound foundations. Once they are cracked, it takes time to repair them. Simply having the public sector ‘do something’ does not always fix the problem. Public policy’s smothering of the urge to act may make matters worse by stifling spontaneous optimism. It sometimes fails to create ‘a political and social atmosphere which is congenial to the average businessman’ and thus gives rise to a hostile business environment” [2].

Recommendation: As businesses large and small absorb the new policies by federal governments to spend money on recovery, they should be on guard against the unreasonable fear that growing government power will remove their future choices. Business leaders must still tap into the positive side of their spirits.

2. Butterfly effect

Situations in which small fluctuations can drive a system to a completely different state. It is called the butterfly effect because it was discovered by Edward Lorenz, a meteorologist at MIT, who found small changes in climate could bring about large changes in weather. In 1982, he asked, if a butterfly flaps its wings in Brazil, can it produce a tornado in Texas? [3]. This phenomenon in complexity literature (referring to the chain of events that the flight of a butterfly can trigger) is “pretty common, for example in complex biological systems, where the action of even a single entity (e.g., a virus, or a malignant cell) can affect the whole dynamic,” says Rand, the U.S. thinktank. “On the contrary, dependencies like these are difficult to describe by ODE systems because they involve the accurate tracking of fast-growing instabilities out of tiny perturbations.”

Research: Looking back on a number of crises, one can see that a small event can cause a large crisis. The COVID-19 world pandemic has been traced to the sale of live bats in a market in Wuhan, China. For numerous examples, see “The Butterfly Effect in Competitive Markets,” by Rajagopal [4].

Recommendation: Consider the “butterfly theory of crisis management,” a phrase coined by researcher Stuart C. Strother, which says that leaders should “anticipate, and prepare to respond to, small and low probability events that have the potential to result in major harmful crises” [5].

3. Clustered volatility

The appearance of random periods of low activity followed by periods of high activity [6]. They are evident in actual financial market data, where they are called GARCH behavior (for generalized autoregressive conditional heteroskedasticity, a term from statistics). Clustered volatility is a key characteristic of complex systems [7].

Research: A seminal paper in this regard is a 2013 study of the potential “causes of clustered and asymmetric volatility” in 27 financial markets witnessed before and during the global financial crisis, and during the recession period (i.e., post crisis). The paper points out the “effect of overconfidence bias on the deep economic recession” [8]. One of the effects is that following overconfidence, investors’ views may “turn to be too pessimistic (i.e., underconfident), which may let the regulatory reforms poorly performing and prolongs the economic recession” [9].

Recommendation: Given the empirical finding that before and after economic crashes, market changes can take on a life of their own, becoming too confident or underconfident, business leaders should be sure to consider a middle ground in financial decision-making and not just the extremes.

4. Entropy

A measure of the randomness or disorder of a system. Entropy increases when a system goes from an ordered state to a disordered state – for example, when a solid changes to a liquid. In thermodynamics, entropy is a measure of unavailable energy and is also defined as the probability of the thermodynamic state. “All natural processes tend to proceed from less probable state to more probable state. Hence in a natural process, entropy always increases” [10].

Research: In the “Principle of Trading Economics,” Shen Wuang, in the economic cycle, explains that the maximum entropy appears in the early stages of economic recovery and the later stages of economic depressions [11]. At such times, “situations controlled by pessimism” begin to diverge and the “expectations for economic recovery” begin to grow. The mechanism of entropy, explains Wuang, comes from differences in risk cognition and risk tolerance, which animate different behaviors from economic agents.

Recommendation: By definition, entropy is something we humans cannot influence; it is a fact of economic life. The recommendation here is simply that as we plan for recovery, we bear in mind the power of expectations to cause both disorder and reordering in systems.

5. Feedback

In electronics, “feedback is defined as the process of returning part of the signal output from a circuit or device back to the input of that circuit or device” [12]. Feedback is a closed-loop system that can operate in a positive or negative way. In positive feedback, the output is in phase with the input, which amplifies an electronic signal. In negative feedback, the output is out of phase with the input, which reduces the signal, making the output smaller. Feedback is an important concept in complexity economics. As science writer Philip Ball summarizes, “price fluctuations are best explained not as the aggregate of many random, independent decisions, ruffled by external shocks to the system – the standard model – but as largely the outcome of the internal, ever-active dynamics of the market, in which feedback makes decisions interdependent” [13].

Research: The principle of positive feedback is easy to see in the current situation. As stated by Forbes, “a dangerous outcome is mass business closures leading to rising unemployment, creating a self-reinforcing feedback loop that locks revenue-starved companies and salary-starved householders into a downward spiral” [14].

Recommendation: Understanding the principle of feedback loops can help us assess the gravity of our situation. In some cases, active intervention – by government and/or the private sector – must occur to prevent a spiraling due to positive feedback loops.

6. Game theory

The standard quantitative tool for analyzing the interactions of multiple decision-makers. As noted in notes for a course offered by the Santa Fe Institute, “many complex systems involve multiple decision-makers and thus a full analysis of such systems necessitates the tools of game theory” [15]. The “games” in question are problems expressed in computer programs that run and obtain results. Complexity uses computer programming as well but adds a twist. “Classical game theory asks: What strategies, moves or allocations are consistent with – would be the best course of action for an agent (under some criterion) – given the strategies, moves, allocations his rivals might choose? Complexity economics by contrast asks how actions, strategies, or expectations might endogenously change with the patterns they create. It is a nonequilibrium approach.”

Research: In their 2012 article on “The Theory of Planning a Successful Economic Recovery,” Sakovics and Steiner highlight the importance of game theory in a way that proves prescient for the year 2020. They note that “interactions among several agents are referred to as a game,” and that this game “becomes a coordination game when multiple stable outcomes – or equilibria – exist and the agents can coordinate equally well in any of them.” As an example, they say, is when government creates stimulus packages for economic recovery – as in 2009 (and in our current situation, with three rescue packages passed and a fourth on the way). The current outcomes will depend on the behavior of the individuals and organizations, otherwise known as “agents,” that were expected to receive the subsidies. Game theory can show different outcomes from different allocations of funding. In their research on the recovery from the last economic crisis (2008), using game theory, they found evidence the “ideal candidates for the subsidy need to satisfy only two criteria: (i) their investment has a relatively large direct impact on the incentives of others, but (ii) they are relatively insensitive to the investment of others.”

Recommendation: In allocating post-crash funding, prioritize investments that will trigger recoveries in other areas without being dependent on them. (Sakovics and Steiner provide an example of a mall that gives a subsidy to a brand-name store that is expected to draw shoppers to other stores, without being directly affected by the success or failure of those other stores.)

7. Propagation

A property of continuation and even sudden increase in a network beyond the period of usefulness [16]. This is one of the features of a complex system according to complexity economics: “When a transmissible change happens somewhere in a network, if the network is sparsely connected the change will sooner or later peter out for lack of onward connections. If the network is densely connected, the change will propagate and continue to propagate.” This cascading effect can be harmful, as in the case of cascading toxicity of distressed assets for example. Propagation can also be helpful – for example, when infusion of cash into an economy starts a chain of positive events.

Research: The concept of propagation makes intuitive sense in relation to recovery, as both recovery inhibitors and accelerants can propagate (or percolate) in a network. This was the case with the long recovery from the Great Depression of the 1930s [17]. In “Understanding Economic Recovery in the 1930s Endogenous Propagation in the Great Depression,” Frank. G. Steindl offers empirical analysis of the separate contributions of the quantity of money, the credit channel, fiscal policy, and interest rates on the recovery, using propagation patterns as a model.

Recommendation: In forecasting recovery, take account of the phenomenon of propagation both negative and positive in systems. The goal of recovery is to return to self-sustaining individuals and institutions, not to encourage dependencies. Programs should have sunset clauses.

8. Self-reinforcing asset-price changes 

A situation in which prices are set by investors who randomly generate and revise their own forecasting methods based on changing conditions, causing spontaneous price bubbles and crashes that cannot be predicted. This is one of the main characteristics of economic phenomena according to complexity economics [18].

Research: Alessio De Longis finds that there are linkages between economic growth, corporate earnings, the health of balance sheets, the ability of companies to invest productively, and consumers’ desire to spend and consume [19]. He finds that all these phenomena are “self-reinforcing” and affect asset prices – positively as economies rise and negatively as they fall. He finds that asset prices are based on the direction of change in growth levels, not on the levels themselves. Bill Miller has noted that “The market typically overreacts to bad news. Not always, but mostly, and especially when the left tail contains really bad outcomes” [20].

Recommendation: As companies study the economic landscape for clues on what lies ahead, it will be important to look at the direction of indicators (going up, going down or experiencing unpredictable volatility). This sense of direction, more than the absolute level of these indicators, will matter most.

In our final installment, we will draw a conclusion about recovery from the 12 concepts of complexity economics.

References

1. Higgs, Robert, 1997, “Regime Uncertainty: Why the Great Depression Lasted So Long and Why Prosperity Resumed after the War,” The Independent Review, Vol. 1, No. 4 (Winter), https://www.independent.org/publications/tir/article.asp?id=430.
2. J. Robert Subrick, 2012, “Animal Spirits and Regime Uncertainty,” The Independent Review, Vol. 16, No. 4 (Spring), pp. 619-621, https://www.independent.org/pdf/tir/tir_16_04_09_subrick.pdf.
3. http://news.mit.edu/2008/obit-lorenz-0416
4. https://link.springer.com/content/pdf/bfm%3A978-1-137-43497-5%2F1.pdf
5. “Butterfly Theory of Crisis Management,” https://www.researchgate.net/publication/312038536_Butterfly_Theory_of_Crisis_Management.
6. W. Brian Arthur, “Complexity economics: a different framework for economic thought,” op. cit., note 38.
7. Ibid.
8. Jlassi, Noui and Mansour, “Overconfidence behavior and dynamic market volatility: evidence from international data,” https://core.ac.uk/download/pdf/82395598.pdf.
9. Ibid.
10. M.V. Sureshkumar and P. Anilkumar, 2015, “Engineering Chemistry,” Vikas Publishing House.
11. Shen Wuang, 2019, “Principle of Trading Economics,” Springer Nature.
12. Janet Heath, 2016, “What is Feedback,” Analog IC Tips (EEE), https://www.analogictips.com/faq-what-is-feedback/.
13. Philip Ball, 2016, “Describing People as Particles Isn’t Always a Bad Idea,” Nautilus, Issue 33, Feb. 11, http://nautil.us/issue/33/attraction/describing-people-as-particles-isnt-always-a-bad-idea.
14. https://www.forbes.com/sites/panel-of-economic-commentators/2020/03/23/economic-recovery-from-covid-19-and-geopolitical-ramifications/#241cb98b596e
15. https://www.complexityexplorer.org/courses/69-game-theory-i-static-games
16. W. Brian Arthur, “Complexity economics: a different framework for economic thought,” op. cit., note 38.
17. Frank G. Steindl, 2003, “Understanding Economic Recovery in the 1930s: Endogenous Propagation in the Great Depression,” University of Michigan Press, https://www.press.umich.edu/17722/understanding_economic_recovery_in_the_1930s.
18. W. Brian Arthur “Complexity economics: a different framework for economic thought,” op. cit., note 38. Ibid., p. 10.
19. “Dynamic Asset Allocation through the Business Cycle,” Oppenheimer, http://www.advisorselect.com/transcript/OppenheimerFunds/dynamic-asset-allocation-through-the-business-cycle-a-macro-regime-approach.
20. Bill Miller, op. cit., note 80.