On November 16, prior to the 2017 Peter Stuyvesant Ball, the Netherland-America Foundation (NAF) organized this year a lecture about Complexity and Economics. The lecture, at Baker McKenzie in New York, was a high level overview of complexity principles for public policy by speakers Dr. Roland Kupers, among other associate fellow at the University of Oxford, and via a pre-recorded video by Prof. Lex Hoogduin.

On forehand, I was very excited to attend this talk as I deal with the topic more or less as a developer advocate for IBM either via the IBM Data Science Experience (DSX), Machine Learning and IBM Watson (cognitive computing, having some experience in R programming, data science, machine learning, calculus, and 20+ years of software engineering. The evening was certainly not un-interesting, but I was a little disappointed by the lack of depth of the presentations, and neither of the speakers touched on machine learning, statistics, or mathematics which have my immediate interest. But nevertheless, the two talks touched upon enough topics for further inquiry to make the evening inspiring.

Kupers’ talk relates to his publication entitled ‘Complexity and the art of public policy’ (2016) in collaboration with David Colander. Kupers is a physicist by training with experience in the field of fractals, then switched to economic policy early in his career, being a consultant now, who among other advised the NWO (Dutch organization for scientific research).

The NWO (Nederlandse organizatie voor Wetenschappelijk Onderzoek) has a special program around Complexity (Grip on Complexity, Sept 2014) as part of 215 special scientific programs, for which Kupers was an advisor.

Kupers highlights five examples that should be considered in complexity and public policy:

• Hans Monderman‘s shared space, which is an urban design principle that aims to remove all segregations between road users and believes that the removal of order will cause users to develop a higher sense of alertness and self-organization being more efficient than pre-designed orders,
• Boids modeling for birds’ flocking behavior is another example of complexicity that teaches policy makers 3 simple principles to manage complexity: separation, alignment, and cohesion,
• A third example to organize complexity successfully is the technology industry, where deductive tinkering lies at the origin of many complex technologies, as elaborated by W Brian Arthur, The Nature of Technology (2009),
• Network analysis in the legal domain is a fourth example, visualizing highly complex organizations in the field of law and codex,
• (forgot the last one, sorry).

The main driver for Kupers seems to be the self-organizing principle of complexity in the absence of regulations and super-imposed orders. The answer by Kupers to a question how machine learning fits into complex public policies, was that technology had no understanding or knowledge of ethics and therefor was likely not suitable for public policy making. In my opinion, Kupers misses the state of affairs, the simplicity of human ethics and the advanced capacity of machine learning.

Hoogenduin’s talk about complexity and the risk of uncertainty was equally generic and unfortunately played from a pre-recorded video, in his sudden absence.

According to Hoogenduin, complexity relates to uncertainty and how we react to uncertainty defines our resilience, our capacity to survive surprises, how we adapt, how creative and entrepreneurial we are. Ironically, Hoogenduin also discusses only 5 principles for effective policy around complexity:

• The modesty principle or the modesty test states that policy objectives should be obtainable,
• The ‘do no harm’ principle states that an economic order is complex and policy should aim to not disturb a fit order,
• Keynes stated two philosophical thoughts that policy makers should keep in mind: (1) don’t attach too much weight to uncertain matters, and (2) in the long run we all die,
• Henry Hazlitt wrote ‘Economics in One Lesson‘ that warns to look beyond the immediate and visible impact of an action, and always expect unintended consequences,
• Last, the economist Milton Friedman argued that there are always tradeoffs to any decision and policy.