If the measurement and evaluation of regulatory performance is to become holistic it must become grounded in the science of regulation. This science has produced self-regulating devices, robots, and space probes. Reliability and efficiency is essential.

Existing top down practices of regulating organizations cannot be either reliable or efficient. No such approach is found in living creatures. Self-regulation and self-governance in creatures is achieved through decentralized control and communication channels. The human brain has no Chief Executive Neuron. Its system of control and communication is based on network governance.

Network governance is a condition precedent for non-trivial stakeholder owned firms to survive. This is evident from the success of giant complex firms.  Examples are: the John Lewis Partnership in the UK; VISA International in the US, and the nested networks of worker cooperatives located around the town of Mondragon in Northern Spain. They all illustrate that network governance can be introduced without any changes in the law.

They also illustrate how network governance reduces the costs and intrusiveness of external regulators. This arises because stakeholder governed firms by necessity exist and prosper because they establish internal control and communication channels to protect and further many of the stakeholders that regulators are created to protect.

Network governance shares power with stakeholders for mutual benefits.  This necessitates bottom-up and outside-in control and communication channels to cross check the traditional top-down centralized systems. In this way the core regulatory domains of Compliance, Ethics, Deterrence, Accountability and Risk (CEDAR) become subsumed into the need for the firm to operate effectively. Stakeholder controlled firms simply cannot remain viable unless CEDAR is achieved for those individuals on which the firm depends for its existence.

The science of regulation outlined below fundamentally changes regulatory design. This is because it becomes based on instrumental rather than just normative criteria like permissibility, responsibility and legitimacy. The instrumental approach arises from Ashby’s law of requisite variety. It that states: “R’s capacity as a regulator cannot exceed R’s capacity as a channel of communication” (Ashby 1957: 211). One illustration of the law is provided by executives who limit their span of control to a limited number of subordinates. This requirement recognizes the inherent physiological and neurological limitation of individuals to process and communicate information. The problem is magnified if subordinates share or feel they have to conform to ideational bias or more commonly ‘groupthink’.

The instrumental limitation of humans to control, regulate or govern complexity arises from the observation of Nobel Laureate economist Oliver Williamson (1979: 233). He stated: “But for the limited ability of human agents to receive, store, retrieve, and process data, interesting economic problems vanish”. At the end of the last century it became possible to identify and measure the physiological and neurological limits for humans to receive, store, process and transmit data. The limits are described in terms of “bits” and “bytes”. These units are also used to describe the operating limitations of computers. In this way human limitations can be allowed for in designing the control and communication architecture of social organizations. 

A corollary of Ashby’s law of requisite variety has profound implications for the control and/or regulation of any organization. “The law of Requisite Variety, like the law of the Conservation of Energy, absolutely prohibits any direct and simply magnification but it does not prohibit supplementation” (Ashby 1957: 268). In other words any executive, CEO, board and/or regulator of a firm cannot control/regulate a firm reliably unless they have a requisite variety of co-regulators to provide supplementation. This makes it in the interest of all parties to introduce network governance.

Network governance introduces operating and competitive advantages for firms while improving the reliability and efficiency of their regulation. Instead of the cost of regulation being mostly paid by the government it becomes mostly privatized. 

Auditors and rating agencies could potentially provide supplementation to act as co-regulators if they represented independent channels of communication and control. But even together they would not provide a requisite variety of feedback channels to reliable regulate complex firms. Matching complexity is required. Stakeholder involvement provides a way of providing a requisite variety of supplementary co-regulators. The purpose of regulation is to protect stakeholders. Network governance provides the means for them to largely protect and further their own interests.

The role of regulators becomes attenuated to filling in any gaps. This would be achieved indirectly by introducing additional nodes of network governance. The focus of regulators would move more to macro concerns about market power and competition.

The measurement and evaluation of regulatory performance would become based on the degree that firms become self-governing to reduce the need for laws, regulations, and regulators. The ability of firms to become self-governing would depend upon the design changes regulators introduced into corporate constitutions. Regulators would then be evaluated like engineers on the reliability and efficiency of their design. Law faculties would need to establish courses on how constitutions could be designed to improve the ability of organizations to improve their self-governance.