Encompassing organisations and the problem of party membership.

Working in education, I don’t get invited to many dinner parties. Call it the ‘Ken Robinson effect’. However, if I did I would like to think there would be lively discussions about the politicization of our population, and input and insights from the cutting edge of fields such as behavioral economics. Don’t worry. These are fictional dinner parties. There’s no chance you might attend by accident.

Imagining myself at such a dinner party. I can already feel myself getting annoyed and reaching for the wine. There are very good reasons why listening to the voices of party members is highly sub-optimal from the point of view of everyone else. Lest this appear a lazy dig at the interested by the apathetic, and at the risk of further alienating the other guests, classical economical theory explains why.

Classical theories of voter behavior take as their first step the rational actor. Human behavior is predictable in reference to the incentives that operate on an individual at a given time, assuming man is a utilitarian value maximizing agent. From this, much flows. During an election, a voter should plump for the party closest to his economic self-interest. For parties, it is therefore always sensible to move towards the median voter because more votes are available at the center of the distribution of economic preferences than at the tails. It also implies that voter turnout will always be a problem. It is rational to free ride on the expected votes of others.

There is ample evidence that, when it comes to individual voter behavior, institutional incentives assuming rational choice are more effective than ‘nudge’ approaches at both explaining and predicting behavior[1]. Therefore, just as understanding the incentives that operate on individuals is important to understand their voting patterns, understanding the divergent incentives that operate on large and small groups helps to understand the probability that such groups will act in the broader national interest.

A key difference between large and small groups is the extent to which they are classified as ‘encompassing’ or ‘non- encompassing’ organisations. As Mancor Olson outlined, where organisations are encompassing their membership includes a significant proportion of a wider population. Where organisations are non-encompassing, they do not. This affects the probability that a given organisation will spend time and resources campaigning for wider public goods. It is rational for an organisation which includes 50.1% of the population to push for better provision of broadly available public goods, because their membership will directly benefit from them doing so. It is not rational for a much smaller organisation to do so. If they promote broad public goods, then the benefits will be spread across the entire population and so necessarily diluted if the membership is only a very small proportion of that population. In this instance, it is rational for the small group to pursue factional interests which are of benefit to them but at a cost to the wider society. This is because the previous logic is inverted, the wider societal costs are diluted but the benefits are concentrated among a small number.

It is therefore highly unlikely that the membership of political parties will lobby and campaign for a broadly available pure public good[2], for instance good governance, because there is little incentive for them to do so. The psychological benefit of being able to support a leader with whom an activist sympathizes with and agrees with is entirely personal. The benefit to the country of an opposition party able to fill the front bench is diluted across all residents in the UK. With the obvious, and important, exception of those at the sharp end of government action the consequences of poor public policy, higher taxes, more borrowing, slower growth, sub optimal public services, are far more broadly felt. Therefore, good governance will necessarily be under supplied to the extent that supply itself is contingent on a non-encompassing organisation.

This might seem to be a pseudo-intellectual dig at the Labour party membership. This is not my intent. This is a broader problem. While the supply of pure public goods are textbook examples of market failures, the exact nature of that failure depends on the good, and I would argue circumstance, involved. An over-reliance of political parties on small groups can be both a criticism of the role of donors and of members. David Cameron showed reasonable scepticism about the value of his own membership, his allies allegedly went further, but as his resignation honours made clear the debt owed to a small number of donors trumped the integrity of the House of Lords. It is rational for donors to value a peerage over the effective functioning of the Lords because two people can’t share one ermine gown. Nor can two people share the same directorship.

Again, this is not a personal attack on those committed to the Labour party or to the Conservative party. Nor is it intended to slight Bernie Sanders or Donald Trump supporters, against whom the same argument can be made. It is simply recognition of what is a relatively uncontroversial point; the rational self-interest of individuals can lead to sub-optimal outcomes for larger groups. Just ask ice hockey players or elk. In other instances we recognize this conflict and governments either implement selection incentives to ensure wider participation in the provision of goods, such as mandating a TV licence, or by recognizing small groups of self-interested individuals as the lobby groups that they are and seeking to reduce their influence over the nature and direction of pure public goods.

Classical economic theory points to a fundamental problem with the involvement of members in political parties, point blank. If it is the role of political parties, both in power and opposition, to supply good governance then any input of party members into that process will necessarily limit their ability to do so when party membership make up a tiny proportion of an overall population. Of course, there are instances of political organisations that can claim to be ‘encompassing’. Olson cites Scandinavian unions as an example, and cites their involvement in public life approvingly. However, unless parties can begin to claim at the bare minimum a double digit percentage of the overall population as members, those members will never have the requisite incentives to act in the interests of wider society.

Frustratingly for those still at the table and waiting for the punchline, I cannot see a solution. However, to begin to approach one we need to recognize the problem. It is an entirely noble impulse to join a political party in order to ensure your voice is heard. It does not mean that society as a whole is better off listening to those voices. The personal benefit from having your voice listened to are yours and yours alone. The consequences are ours to share.

I am open for dinner invites. November is looking quite free.

[1] I mean, obviously it’s more complex than this. Get out the vote operations using ‘nudge’ methodology have been shown to increase turnout by between 3 and 4%. Institutional reforms have effect sizes of 13% for compulsory voting and 12% in a move from FPTP to a PR system.

[2] Defined as a good where excluding participation is impossible, and an additional participant makes no difference. Street lights are good examples, as is clean air. I accept governance is more controversial. But run with it. Contrary to what most people believe, education is not a public good in the pure sense. I can exclude you from a school. That’s literally what we’d call it.