An Algorithmic Theory of the Policy Process

With a few exceptions, current theories of the policy process do not model or measure the policy process using the graphical process notations that are common within information science, business administration and many natural sciences. The reason is that in the post-war period the needs of business process analysis came to dominate social science applications of process science whilst the needs of public policy process analysis remained largely unaddressed. As a result, modern graphical process notations can encode and quantify the instrumental properties of cost and efficiency of a business process, but not the normative properties of transparency, accountability or legitimacy of the much more complex policy making process. There have been many other unfortunate consequences. Business process modelling evolved into business process reengineering and became a critical enabler of a period of unprecedented hyper-globalization commencing in the 1990’s. However, it did so by encoding and quantifying the instrumental dimensions of cost and efficiency of globalized production processes and not their normative dimensions of domestic employment and social welfare transfers. We live with the consequences to this day of the emergence of destabilizing populist national movements and rising security and defense tensions between former trading partners. However, in recent years, there have been several important new developments. Firstly, a new class of process modelling tools has emerged at the juncture of the disciplines of information science and business administration that can model much more complex governance and policy-making processes as rules based declarative process graphs instead of sequence based imperative process graphs. Secondly, information science is now introducing a capacity for normative reasoning and moral dilemma resolution into a range of technologies from multi-agent systems and artificial societies to self-driving vehicles and autonomous battle drones. This creates new opportunities for a collaboration between policy process analysis and information science to reengineer legacy policy making processes and organizations in terms of normatively driven declarative processes. Not only must these reengineered policy making processes score better against instrumental criteria of cost and efficiency but also against the normative criteria of transparency, accountability, and legitimacy. Consequently, the metrics presented in this dissertation re-connect public policy process analysis with the tools and results of decades of process research in the fields of information science, business administration and many natural sciences, and supports a new theory of the public policy process as an algorithm whose purpose is the generation of solutions to public goods allocation problems. To illustrate the principles of the techniques involved and the utility of the approach, a case study analysis and prediction of Chinese public health policy response to the COVID-19 pandemic of 2020/21 is presented.