Poor political governance in Solomon Islands – what use rational choice explanations?

What are the causes of poor political governance in Solomon Islands?

In a previous post I pointed out that this is an important development question. And, prompted by two blog posts (here and here) and a Discussion Paper [PDF] written by Tobias Haque, I started to examine potential causes of poor political governance. In particular, broadly in agreement with Tobias, I argued that the central problems of political governance in Solomon Islands do not appear to be easily explained by simple cultural hypotheses. In this post I will examine Tobias’ alternative explanation. In a third post I will look at what, if anything, might be done about problems of political governance in Solomon Islands.

In his posts and papers, Tobias argues that the problems of politics in Solomons are best explained by a model of rational actors following incentives that stem from the nature of the Solomons’ state.

He argues that, because the state is weak, because most Solomon Islanders live in rural areas where its reach is limited, and because their needs are pressing, Solomon Islanders vote for the candidate who they think is most likely to provide them with immediate local assistance. Moreover, Solomons voters are caught in a collective action dilemma. In an electoral system where voters only vote for one MP out of 50, and in the absence of strong political parties or something similar which might provide confidence that others nationwide are voting for national reform, it is effectively impossible to vote for national change. Tobias argues that the rational thing for voters to do is to choose the candidate most likely to help them directly — and that this is what they do. Unfortunately, sensible local voting doesn’t aggregate to good national outcomes. Because voters vote and punish MPs on their local performance MPs don’t have an incentive to run the country well, and so they don’t.

Tobias’ explanation is a good one and there is a strong tradition in political science of rational choice models of politics and institutions. And while these models may hinge on problematic assumptions about human behaviour, as I explain here, it would be a mistake to dismiss them out of hand.

The rational choice model that Tobias proposes also resonates with much of the qualitative data that I gathered in my fieldwork. In particular, people often told me that voters voted first and foremost in search of local assistance; and voters could usually offer very reasonable, if not perfectly rational, explanations of how people figured out which candidate would be most likely to offer assistance. Yet, rational choice explanations don’t fit with all my interview data, nor are they completely commensurate with election results.

The first problem for such explanations is that if you posit a world of atomised rational voters, you find yourself back where we were early in last week’s post needing to explain why election results in Solomons confound Duverger’s Law, the prediction that first-past-the-post electoral races will tend towards two party competitions (or two candidate competitions in the absence of parties). This should happen in a world of rational voters because voters ought to abandon favourite candidates for next-best candidates, who are more likely to win. After all, in a first-past-the-post political contest, voting for a no hope candidate is effectively wasting your vote. This tendency, or something similar to it, should be present not just in countries with ideological cleavages but also clientelist polities such as Solomons, where voting for a candidate that is likely to loose means there is little chance of getting anything from the victorious MP who will focus resources on his or her supporters.

The second problem is that in some constituencies, within-constituency vote patterns vary in a way that seems at odds with a model that posits a world of equally rational actors everywhere. Take for example the case of West ‘Are ‘Are. The chart below shows 2010 election results broken down by polling station. Each line is a polling station. Each colour is a candidate. Clearly, the two different halves of the constituency have very different results patterns. In one half, large blocks of votes go primarily to one of two candidates. In the other half, votes are spread among many different candidates. Why? It is hard to explain such variation with a simple rational choice model.

The third problem for rational choice explanations is that, while they were clear about voters seeking localised material benefits, when asked why people choose to vote for the candidates that they voted for, many of my interview respondents did make reference to social features that could broadly be described as cultural. People frequently told me that in their village voters chose Wantoks, or relatives, or members of the same clan.

So what is going on? Is it culture after all?

I think the answer is that in Solomons, for many voters, cultural features do play a role in the decisions they make about voting, but that culture is neither blinding nor binding. Importantly, when interviewees would tell me about Wantok voting they could also provide a reasonable, broadly rational, explanation for why it took place. Everything else being equal, people would vote for Wantoks because, owing to the presence of norms of reciprocity within kin groups, people were more likely to obtain assistance from Wantoks if they won. This wasn’t blind loyalty, though. What is happening is simply that voters, who find themselves caught in a principal-agent type dilemma are using a reasonable heuristic to determine who is likely to help them. Significantly, when provided with other information about candidates’ propensity to help, such as past assistance, people will vote for candidates to whom they are not related (this is what explains the success of the Vietnamese candidate in West Honiara). Likewise, most people won’t vote for Wantoks if they’ve proven in the past to be unhelpful. What we have here is cultural features interacting with something akin to rational choices being made by voters in search of assistance.

Another interaction that occurs is between culture, power, and voter choices. In the case of the West ‘Are ‘Are result patterns, the answer stems from the fact that half of the constituency villages are led by powerful hereditary chiefs, who are able to tell voters who to vote for. In the other half of the constituency, village leaders have less power, and voters are free to choose themselves. The half with strong leaders is, of course, the half with simpler result patterns.

Here, culture is playing an important role in determining election outcomes by determining who has the power to tell others how to vote. While the West ‘Are ‘Are situation may not be that common across the rest of Solomons, on a smaller scale, culture and power often do work together to constrain voter choice within families. Specifically, women are often not free to vote as they choose, and are expected to vote as male family heads tell them to.

Finally, we have the lack of electoral convergence (Duverger’s law) to explain. Here I am less certain, but I think the most likely explanation is that political convergence doesn’t occur often in Solomons electorates because Solomons constituencies typically lack the social features (like cross cutting civil society movements) that allow people to interact in a way that fosters political trust and allows large-scale, ongoing collective action. A good example of how such social features work is the way that trades unions brought labour related parties into being in much of Western Europe and how this subsequently shaped party politics thereafter. Absent repeated interactions mediated by the rules of cross cutting entities, and it is very hard for consistent patterns of collective political interaction to form. There is too much uncertainty and room for double dealing.

Where does this leave us then? With several observations. First, simple cultural models of politics that tie voter choices and politics to blind Wantok loyalties, or the legacies of Big Man culture, do not satisfactorily explain why Solomons experiences poor political governance. Tobias’ rational choice explanation is an improvement. But the last thing we should do is conclude, as Margaret Thatcher once did, that “there is no such thing as society”. Through a variety of means, the social and cultural features of Solomon Islands do play a role in the politics the country experiences. This fact doesn’t render rational choice models of politics invalid, but it does mean we need to use them in a way that is careful and cognisant of the way people’s choices are shaped by social structures.

Quite possibly Tobias would agree with me in all this: in his paper he concede some role for culture. Yet he offers little detail on what this role might be. What I hope I’ve done here is fill in some of this detail and, in doing so, contribute to an ongoing discussion.

In my next post I will look at what, if anything, might be done to improve political governance in Solomon Islands. This is where the biggest differences between my view and Tobias’ lie.

This blog is a part of a series on political governance in the Solomon Islands. Other blogs in this series can be found here.

Terence Wood is a PhD student at ANU. Prior to commencing study he worked for the New Zealand government aid programme.

image_pdfDownload PDF

Terence Wood

Terence Wood is a research fellow at the Development Policy Centre. His research focuses on political governance in Western Melanesia, and Australian and New Zealand aid.


  • My personal thanks to Tobias and Terence for their excellent contributions to these key questions of development effectiveness in the Solomons. The knowledge base is sparse, despite the avalanche of RAMSI / AusAID money spent in the last decade, and all well-informed public dialogue like this will add value to our joint enterprise.

    It seems we are focusing on the subject that has been variably described as political economy, behavioural economics, or even game theory. It’s a fascinating topic and for me it goes to the heart of human development. We are looking at political governance in the Solomons and seeking to understand WHY Solomon Islands seems so resistant to political and economic reform and change at the national level, yet so resilient at the village level. I don’t think Solomon Islanders are a different species to the rest of humanity so I will proceed on the basis that they process their human experience just as any other homosapien would. Of course there is individual variance, and the individual “experience” is inherently unique, but I would posit that using a representative sample we should expect to see the same patterns that occur everywhere else in the world since we started walking upright.

    I have gone back to this blog topic many times since Terence wrote it some months ago, and used many of the links to further my understanding. I’ve found some interesting patterns I’d like to share.

    I was reading an article by the Australian economic journalist Ross Gittins. He was discussing the work of moral psychologist Jonathan Haidt and his recent book, “The Righteous Mind: Why good people are divided by politics and religion”. Gittins makes the following observations about Haidt’s work.

    “Whereas the old view was that natural selection had caused us to evolve into self-seeking competitors, Haidt argues we’re more accurately thought of as ‘homo duplex’ – a creature who exists at two levels: as an individual and as part of the larger society.

    In Haidt’s view, ‘We evolved to live in groups. Our minds were designed not only to help us win the competition within our groups, but also to help us unite with those in our group to win competitions across groups,” he says. ‘We are not saints, but we are sometimes good team players.'”

    I really liked the term ‘homo duplex’. We are so much more complex than the abstract concept ‘homo economicus’. If not we’d be most of the way towards dealing conclusively with the over-emission of greenhouse gases (amongst the plethora of other intractable human problems).

    I think Game Theory holds promise for further enlightenment in respect of trying to understand how we behave and respond to our environment. Gittins in another article discusses the recent work of US Game Theorist / Economist and 2012 Nobel Prize winner Al Roth. Gittins has this to write about the game theory-based ‘market design’ work of Al Roth (http://kuznets.fas.harvard.edu/~aroth/alroth.html):

    “Unlike conventional analysis, game theory allows the possibility of ‘multiple equilibria’ – more than one possible outcome the participants regard as satisfactory. And it studies ‘decision-making under uncertainty’ – having to make decisions without knowing what the future holds.”

    And further:

    “He’s expert in ‘market design’ – changing the rules in markets so they work more efficiently in producing the best outcomes for people. He says a ‘free market’ is one that moves freely in achieving efficient outcomes, not necessarily a market with no intervention by the government. In the United States, Roth has helped make changes that improve the pairing of medical interns with hospitals, the pairing of students with high schools and the matching of kidney donors with recipients. He’s also done the last one in Australia, actually improving some people’s lives.”

    Personally I am salivating just thinking about having Al Roth work on a project to re-design the Solomons electoral system, or the way that aid donors fund public service delivery. Not that he’d provide ‘the definitive answer’ (mainly because ‘it’ doesn’t exist in this dimension), but this kind of thinking would represent a whole universe of improvement on the status quo.

    To ruminate on the fact that it isn’t just Solomon Islanders that vote partly because of ‘cultural’ reasons, the journalist Guy Rundle writes in Crikey about the recent US elections, and how voting patterns (in one of the most economically developed and richest countries on earth) refuse to align neatly with notions of rational self-interest. Rundle observes:

    “Mississippi desperately needs some basic public services, but it will retain its steadfast self-defeating addiction to poverty, stupidity and the Republican Party no matter what is on offer. The fact that voting is now cultural, rather than economic or interest-derived, plays havoc with an electoral-college federalist system – and no-one, neither the Beltway professionals nor the Nate Silver-esque nerds understand this.”

    And the venerable Economist published an article recently that pondered the question, ‘Isn’t the act of voting an irrational act for an individual?’ Our correspondent wrote,
    “The time and hassle involved in getting to polling stations, registering to vote or even deciding between the candidates mean that rational individuals should take a free ride on the efforts of others and stay at home.”

    And further:

    “In the mature democracies of the rich world there may be a simpler explanation: like agonising over loose change, the costs of voting are so tiddly that it is easier just to go and do it than to fret about whether it is rational or not. It would be odd if something as important as democracy were safeguarded by a lack of thought.”

    Personally I have often wondered why magazines that target the business community and the international jetset have so many advertisements for watches that cost more than $1,000, when a digital watch that lasts reliably for 5-10 years can be had for $20? Is the readership being taken advantage of because it’s dumb and ignorant? Or are there other forces at play?

    Thanks Terence and Tobias for your stimulating and good work.

    • Thanks Marcus,

      I’m still chuckling at the thought that in a world of rational consumers the Economist probably wouldn’t exist 🙂

      I offer a qualified defence of rational choice modelling here.

      To summarise very briefly I don’t believe humans are rational utility maximisers however I believe that enough of the time we act **as if** we are for RC models to be useful, if very carefully used.

      Useful because they kind of approximate to reality. Useful because it is very hard to use models based on more complicated assumptions (game theory, for example, would be come very tricky if based on more complicated agents). Useful because where behaviour deviates significantly from what we would expect from the models (say in Mississippi) they still illuminate interesting questions: why, Mississippi, why?

      Re voting itself – that seems like a paradox but perhaps it isn’t. As Andrew Gelman and others argue, if you believe that stakes in an election are high enough then it is still rational for you to vote, even if you know your own impact is trivial and there is inconvenience involved.

      In Solomons case, in a clientelistic polity, if you believe that candidates can figure out whether you voted and who your voted for, and if them providing future goods to you depends on you having voted for them, then it’s definitely rational to go through the hassle of voting.

      Having said all that, I have to confess to actually enjoying voting: which is clearly irrational…such are the contradictions of life as a political science nerd.

Leave a Comment