The Governance Impossibility Theorem

Kenneth Arrow famously fashioned an Impossibility Theorem 60 years ago which changed the course of welfare economics and social choice theory: no voting system can convert the ranked preferences of individuals into a community-wide ranking, while also meeting a certain set of reasonable criteria, each of which is likely to be widely supported.

In other words, we cannot base collective decision making simply on the individual preferences of the members of the collective – we need to trade off some ‘reasonable criteria’ against other ‘reasonable criteria’, when we made a collective decision. (And, of course, this means giving more weight to the values of some members than to others).

So what? Well, in the intervening period, we have learnt to live with Arrow’s theorem and have reconciled ourselves to the necessarily limited claims that welfare economics and social choice theory can make about the social rankings of different courses of action. However, we have allowed an entirely new substructure of criteria to grow up in relation to social choice and community decision making, without noticing that a similar logic is likely to apply to them – the ‘principles of good governance’.

I don’t wish to attack the desirability of these ‘principles’. I’ve actually written quite a lot on them in the past (see, example, my article with Elke Loeffler on “Evaluating the quality of public governance: indicators, models and methodologies”, International Review of Administrative Sciences, Vol. 69 No. 3 (2003), pp. 313-328). However, even in that article, we hinted that any full set of principles of good governance might be ‘over-determined’ – it might not be possible to achieve all of them simultaneously. Not a surprise, when you consider that we were proposing principles which related to all of the following dimensions of public governance:

• Citizen engagement
• Transparency
• Accountability
• The equalities agenda and social inclusion (gender, ethnicity, age, religion, etc.)
• Ethical and honest behaviour
• Equity (fair procedures and due process)
• Ability to compete in a global environment
• Ability to work effectively in partnership
• Sustainability
• Respect for the rule of law

Now I’m much more confident that there is indeed an ‘impossibility theorem for public governance:

no decision making system in any organisation or society can conform simultaneously to all the reasonable principles of good governance, each of which is likely to be widely supported by most members of most stakeholder groups

I was most recently reminded of the importance of this when reading Amitai Etzioni’s essay “Strength in numbers” in the recent RSA Journal (Autumn 2009, pp. 24 -27). Etzioni contrasts the obligations which arise from one’s commitment to the community (or communities) in which one lives to those obligations which arise to all our fellow men and women – universal human rights. He writes "One cannot maximise either individual rights (and in their name destroy particularistic values and the communities on which they are based) or community (thus ignoring our obligations to all human beings). Comunitarians like me see the tension between the two as a given; hence, it is best to seek out how the commitments to both core values can be combined.” In other words, bad news, a trade-off is needed.

So what? Well, I think the main implication is that we now urgently need to explore the grounds on which we might be prepared to make this trade-off between different ‘good governance’ principles. As with the reactions to Arrow’s original impossibility theorem, we are likely to find that the trade-offs we consider most convincing will actually differ significantly between contexts, and over time. And, of course, between stakeholders – so that there is likely to be disagreement (if not outright conflict) between parties as to which governance principles should have highest priority at any given time.

This is hardly a surprising lesson. However, much of the governance literature still suggests that there is always a potential ‘win-win’ situation for all stakeholders, if they only buckle down, engage with each other, seek compromise and accept the principles of ‘good governance’. Not so. These governance principles are indeed a basic part of human interaction. But human interaction requires collective choices and prioritization – and some people are always likely to lose out in such collective processes. We should not pussyfoot about in trying to hide this.

A further implication is that there may well be more important tasks in governance today than the endless elaboration of each of the ‘good governance’ principles. Of course, it is interesting and, in the long term, it may be valuable to make ever more sophisticated our understanding of what is meant by ‘achieving transparency’ or ‘respecting diversity’. However, many of these principles are likely to be only partly met in practice, so that finding better social and political methods for achieving an acceptable trade-off between these principles may bring far greater benefits to our citizens than pushing our definitions of the principles to their logical limits.

This is not to say that governance principles do not matter. Far from it. It is because they matter so much that we need to be become more adept at choosing which ones matter MOST to us, in the communities in which we live, at this particular time.