Systems thinking and regulatory governance (3): Examples of systems thinking in regulatory scholarship

In the previous blog post, we have looked at how systems thinking may help to develop more effective and just regulatory governance and practice. We found that various trajectories of systems thinking exist and that they can be used side by side in regulatory reform. In this blog post, we will look at some examples of how scholars of regulation have applied systems thinking.

Given the wide variety of applications of systems thinking in regulatory scholarship, I limit myself here to (very briefly) discussing a small number of recent books.

Regulation as systems of stocks and flows, nonlinearity, dynamics and feedback

In his book Outbreak: Foodborne Illness and the Struggle for Food Safety (2019), Timothy Lytton applies insights from the work of systems thinkers such as Donella Meadows. He seeks to better understand the interaction government regulation, civil liability and private governance in the field of food safety regulation. The book carefully unpacks the complex system of food safety regulation in the USA.

The book opens with the account of the 2011 listeriosis outbreak. It was the worst foodborne illness outbreak in the USA since the Centre for Disease Control and Prevention began tracking it in the 1970s. The outbreak resulted in 33 deaths and a total of 147 confirmed cases. The origins of the outbreak were traced back to an individual cantaloupe farm. But, after carefully discussing the case, Lytton concludes what is so often concluded after a (regulatory) failure of this magnitude: “No one seems entirely to blame, yet everyone seems partially at fault”.[1]

In this specific case, and the others in Lytton’s book, it was the interplay of a range of individuals and organisations, laws and regulation, interpretations and customs, and various forms of (failed) stabilising and (unfortunate) reinforcing feedback loops that explained the full scale of the outbreak. Because of the complexity of the food system, the initial outbreak at the cantaloupe farm developed quickly in a nonlinear manner, ultimately affecting 28 states in the USA.

After discussing several related cases, Lytton illustrates how the concepts and language of systems thinking help to analyse these cases. More importantly, it helps Lytton to see patterns across the cases. This leads him to call for systemic regulatory reforms that aim to improve feedback mechanisms and learning within the food safety (regulatory) system. He calls for improving the infrastructure of food safety investigations, rather than hiring more inspectors.

Regulation in a functionally differentiated society

In her book Meta-regulation in practice: Beyond normative views of morality and rationality (2017), Fiona Simon applies Niklas Luhmann’s systems theory. She seeks to understand better how meta-regulation (here understood as government regulation of industry self-regulation) has played out over seventeen years in the Australian retail energy industry.

Simon challenges the assumption that meta-regulation is a “progressive policy design that works effectively with markets and promotes stakeholder inclusion in order to reach a more informed of the public interest and how the public interest can be met by business”.[2] That is a relevant question. Some argue that, at best, government can set the boundaries for industry self-regulation, simply because the industry has an internal logic, objectives, structure, culture, rewards, punishments, and even criteria to establish expertise. Therefore, so goes the argument in favour of the type of meta-regulation Simon discusses, government law and regulation will never be able to determine and directly steer industry activities because it cannot specify the form and interpretation of all these relevant aspects.

Building on Luhmann’s systems theory, Simon argues that exactly the opposite seems to have happened in the Australian retail energy industry. The three main function systems at play (the political system, the economic system, and the legal system) have difficulty communicating with each other. Or at the very least, do not speak each other’s language. For example, energy providers operate within the economic system, and to them events only have meaning within the language of the economic system—profitable/non-profitable. The values and language of the political and legal system do not resonate with them, and vice versa.

To Simon, it is therefore not too surprising that meta-regulation of the Australian energy retail industry has not resulted in positive outcomes. In her view, it asks for a legal understanding of rationality and morale on the side of industry players that are enforced by regulators who operate (often) under political pressure and worldviews. It results in too much miscommunication and misunderstanding. To Simon, that is a recipe for “messy regulatory outcomes”, and Luhmann’s systems theory provides her to map, explore, and analyse that mess.

Regulation and (contemporary) cybernetics: Algorithmic regulation

The link between cybernetics and regulation has always been obvious. Cybernetics is the study of control and communication in mechanic and living systems. A concept central to cybernetics is ‘algorithm’. “An algorithm is a technique, or a mechanism, which prescribes how to reach a fully specified goal”.[3] Another way of looking at algorithms is that they are a sequence of well-defined instructions to solve a problem. It resonates well with how regulation is sometimes conceptualised.

The growth of information technology and, particularly, big data has resulted in a renewed interest in cybernetics for regulatory governance. In 2019, renown regulatory scholars Karen Yeung and Martin Lodge published an edited volume titled Algorithmic Regulation. Algorithmic regulation is understood “both as a means of coordinating and regulating social action and decision-making, as well as the need for institutional mechanisms through which the power of algorithms and algorithmic systems might themselves be regulated”.[4] The authors are mainly interested in exploring the use of big data, information technology, and computer algorithms to support regulatory governance and practice. Various chapters, for example, discuss the opportunities and constraints of machine learning and artificial intelligence (AI) as tools for regulatory and enforcement decision-making.

Of course, the notion of algorithmic regulation could also be conceptualised at a more abstract level. As Tim O’Reilly, a central figure in open source software, argues: regulation could be “regarded in much the same way that programmers regard their code and algorithms, that is, as a constantly updated toolset to achieve the outcomes specified in the laws”.[5] In this sense, algorithmic regulation implies ongoing monitoring and modification of regulatory governance and practice through feedback on how regulation performs ‘in the now’ and is achieving its specified goal.

Yet, because the goal of regulation cannot, often, be specified in extreme detail, algorithms need to be complemented with heuristics. To speak with the systems thinker Stafford Beer: “An heuristic will take us to a goal we can specify but do not know, and perhaps cannot even recognize when we reach it. The algorithm (such as: ‘to get to the highest point, try one step in each direction, and move to the next higher position’) specifying this heuristic stipulates the eventual discovery of a strategy. The strategy says: ‘The best thing to do is to go up here for so far, round this, along that, then up the other.’ This strategy cannot be worked out in advance”.[6]

Regulation and the Soft Systems Methodology

An insightful book that illustrates how the Soft Systems Methodology (SSM) could help to improve regulatory governance and practice is Brian Wilson and Kees van Happeren’s Soft Systems Thinking, Methodology and the Management of Change (2015). It presents case studies of the SSM in areas of service delivery and risk management—all closely related to regulatory governance and practice.

SSM asks for a careful understanding and defining of the system at hand—known as ‘root definition’. Defining the system and the problem situation it seeks to address is part of improving it, and is best done with a variety of individuals and organisations within the system. In short, the definition includes the basic transformation a system seeks to achieve (T), the worldview that provides meaning to this transformation (W), system ownership (O), system operators (A), the customer or target of the system (C), and the environmental constraints of the system (E).

For a policing system, or a police force, a root definition could be: “A chief constable owned system [O], operated by appropriately skilled police officers and other agencies [A], to establish community [C] well being in terms of the security of people and property [T], by seeking to prevent unlawful and antisocial activity, together with other potential disruptions, responding to reported incidents and identifying and apprehending those who violate the law and delivering them to the appropriate judicial authorities [W], while acting within the accepted norms of behaviour and visibility requirements, but constrained by finance availability, Home Office and local policies [E]”.[7]

After establishing the root definition—again, this is done in a deliberative process with a variety of individuals and organisations within the system—conceptual models are developed to achieve it. These conceptual models have to be compared with the real-world situation to define possible and feasible changes.

In conclusion

Over the last years, we have seen several books published that apply systems thinking to regulatory problems. These books seek to map, explore and interrogate regulatory problems. Sometimes they also provide solutions to regulatory problems.

None of the books does, however, provide evidence that systems thinking applied to regulatory governance and practice will result in better regulatory outcomes. That evidence must wait a little longer to emerge.

---

[1] Lytton, T. (2019). Outbreak: Foodborne illness and the struggle for food safety. Chicago: University of Chicago Press.

[2] Simon, F. C. (2017). Meta-regulation in practice: Beyond normative views of morality and rationality. Abingdon: Routledge.

[3] Beer, S. (1995 [1972]). Brain of the firm – 2nd edition. Chichester: John Wiley & Sons.

[4] Yeung, K., & Lodge, M. (Eds.). (2019). Algorithmic regulation. Oxford: Oxford University Press.

[5] O’Reilly, T. (2013). Open data and algorithmic regulation. In B. Goldstein & L. Dyson (Eds.), Beyond transparency: Open data and the future of civic innovation (pp. 289-300). San Francisco: Code for America Press.

[6] Beer, S. (1995 [1972]). Brain of the firm – 2nd edition. Chichester: John Wiley & Sons.

[7] Wilson, B., & van Haperen, K. (2015). Soft systems thinking, methodology and the management of change. London: Palgrave.