An ignorance teaser

The teaser for my upcoming public lecture was in last Thursday's Christchurch Press but unfortunately isn't available online. Here's the text. Hope to see Christchurch readers there tomorrow night!

The most surprising finding in Fairfax Media’s recent poll on voter preferences in the electoral referendum was that only a third of voters confessed inadequate knowledge to form a preference among the alternatives to MMP. If they were honest, both with themselves and with the pollster, rather more would have disclosed not only a near complete lack of knowledge of different electoral systems, but also a shocking level of ignorance even about the workings of our current system.

Political ignorance is neither unique to New Zealand nor a recent phenomenon. In a 1964 survey of Americans, only two years after NATO stood at the brink of nuclear war with the Soviet Union in the Cuban Missile Crisis, a mere 38% of Americans knew that the Soviet Union was not a member of NATO, the treaty organization formed to defend the West against the Soviets. The broad consensus of the international literature suggests voters know remarkably little about their own electoral systems, about their systems of government, about the policy positions espoused by different parties, about what the government spends money on, and about the likely effects of policy.

The causes of widespread voter ignorance are reasonably well known. Information is costly, and so people economise on it. Where the cost of an additional bit of information exceeds the benefit I expect to draw from learning it, I’ll do something else. We are all rationally ignorant about just about everything – we all could, if we wanted to, learn a lot more about everything from automobile mechanics to the history of the space programme. All it costs is time, but time is precious. And, for most people, the benefits of acquiring political information really aren’t worth the cost. We might learn enough to avoid looking silly in conversations with our friends, keeping in mind that they too are rationally ignorant.

What does this mean in New Zealand? The New Zealand Election Survey asks, at every election, a few questions establishing respondents’ basic grasp of the system. The 2011 results have yet to be released, but in 2008, results really weren’t very good.  A bare majority of respondents correctly identified the party vote as being more important than the electorate vote; a third thought they were equally important, and the rest didn’t know. About a fifth of respondents did not know that Labour was in government – they lacked even the basic knowledge necessary for supporting or opposing the incumbent.

Among those respondents indicating a preference for MMP over First Past the Post [FPP], about ten percent reckoned that a party earning forty percent of the popular vote should earn more than half of the seats in Parliament: a result more consistent with FPP. Among those supporting FPP, about forty percent said that a party earning fifteen percent of the popular vote should earn about fifteen percent of the seats: a result remarkably unlikely under FPP. A quarter of FPP supporters also favoured coalition over single-party government. It’s not wrong to favour either electoral system, but it is a bit odd to favour options that work against the outcomes you think important. It’s more than a little disappointing that, in the fifth MMP election, voters understood so little.

The academic literature has moved from establishing the basic facts of voter ignorance to debating its likely consequences: can voters make sufficiently competent decisions to ensure decent outcomes? There are a few ways this can happen. If voters who know very little cancel each other out on election day – effectively flipping coins when at the ballot box – the election then is decided by the choices of more informed voters and ignorance does little to affect outcomes. Alternatively, the kind of political knowledge embodied in answers to quiz questions may have little bearing on the choices voters need to make at the ballot box; I would fail any quiz on mechanical engineering, but that does not prevent me from making sound decisions when buying a car.

Unfortunately, there is reasonable evidence that political ignorance affects party and policy preferences, even after correcting for the kinds of demographic factors, such as education and income, that affect political knowledge. And, the effects are large. In analysis of the 2005 NZES, I found that those with less political knowledge were substantially more likely to disagree with the consensus of most economists on fairly basic economic matters.  The NZES asked respondents whether, to solve New Zealand’s Economic Problems, the government should control wages or prices by law. I have a hard time imagining a single economist who would agree that wage and price controls would solve any problem faced by New Zealand in 2005. But a third of survey respondents supported wage controls; a fifth supported generalized price controls. And lack of political knowledge was a stronger predictor of disagreement with basic tenets of economics than was a lack of education. Further, the more politically ignorant were more likely to support both spending increases and tax cuts – either of which alone can be a defensible preference, but which tend not to go well together.

We cannot do much to improve the general state of voter knowledge. The incentives for individual voters to become well informed are too weak and, for many voters, the costs are high. While many voters do take their democratic responsibilities seriously and work to cast an informed ballot, the evidence suggests a reasonably large proportion do not. What can we then do? The evidence suggests that non-voters have, on average, less political knowledge than voters; strenuous efforts to get out the vote then seem likely to reduce the average quality of the vote. Further, when considering the desirability of different electoral systems, we should put some weight on the relative cognitive demands placed on voters. A good system will make it easy for uninformed voters to support the incumbent if they like how things are going, or to replace the government otherwise. At the margin, this lends support to systems that elect single-party governments.