Downs’ work originates from the work of twentieth century rational choice theorists. Downs was looking at all areas of how rational individuals approach politic, but for this post we’ll just be looking at the ‘Downsian Model’ (also known as spatial theory and the median voter model). This assumes that voters are arranged in a normal distribution, with the bulk of voters in the centre and gradually reducing numbers of voters to the left and right of that centre. (See the diagram, where 50 is the ‘centre ground’, 0 is extreme left, 100 is extreme right and the vertical axis is the number of voters of that view) It’s important to note that this is a model of the real world, an approximation of the actual position in order to create and test theories, not a claim that this is exactly how people are organised. Downs was seeking to explain why political parties in a majoritarian system like the USA’s tended to converge ideologically upon the centre ground.
Downs assumed that a rational voter would vote for whichever party was closest to their views. For instance, a voter at point 70 would be more likely to vote for a party at point 75 than one at point 50, and a voter at point 50 would be more likely to vote for a party at point 40 than one at point 65.
The key to electoral victory – and why this is also known as the median voter model – is capturing the centre ground and the median voter at point 50. In a two-party system, whichever party best appeals to that mass of voters at the mid-point (which includes the theoretical median voter, whose views are the exact ideological centrepoint of the nation) will win a majority of the vote. As Downs assumes that parties are vote-seeking and power-seeking, this gives them a clear motivation to appeal to that median voter. As an example of how this thinking works:
If we assume that a ‘left’ party exists with an ideology at point 25 and a ‘right’ party exists with an ideology at point 75, what we would expect to see is votes splitting 50-50. The median voter (the one sitting at the ideological centre point of 50) will be equidistant between the two parties, while everyone to the left of them would be closer to the left party and everyone to the right of them closer to the right party. If the left party then moved its ideology towards the centre (say to point 35), things would change. The midpoint between the two parties would now be at point 55, and everyone to the left of that would back the left party, giving them a majority as they are now closer to the median voter than the right. The right party would then be expected to react by moving its ideology closer to the centre, and so on and so forth until both parties are right up against the centre.
It’s important to note that while this is the most commonly seen use of Downs, he didn’t say that all societies had preferences distributed in the same manner, and also looked at what might happen with different distributions of voters. For instance, in one where voters were distributed roughly equally between views, or with a number of peaks in the distribution, parties wouldn’t have the same pressure on them to move, and there would be more of an opening for multiple parties to emerge. It’s also missed by many that Downs was proposing a model, and models in political science are always simplifications. As with many rational choice theories, Downs was trying to establish a framework of how things would be if everything was fully rational, not saying that was the way it had to be. Indeed, by setting up a model of what should happen if everything was rational, we can see where things are actually irrational, which are more likely to be interesting to study. After all, where’s the fun in writing ‘everything went exactly as the theory predicted’?
That hasn’t stopped people – including many who advise, or want to advise, political leaders – of assuming that Downs was making recommendations, not theories, and since the publication of his work in the 50s, we’ve seen many people assuming the only way to assure political victory is to head to the centre. Note that this is to take all of Downs’ assumptions – including the left-right spectrum and the normal distribution of voters along it – as given, when they might not necessarily be the case.
There’s been a lot of writing that’s followed on from Downs in the decades since An Economic Theory Of Democracy was first published, and it’d be foolish to try and summate it all in a single blog post. Suffice to say, though, that there’s been plenty studied and written on every aspect of it, from the question of whether people form coherent enough political views to be able to judge which parties are closer to them to the ongoing issue of whether the left-right spectrum is the best way to look at people’s political views. So, the objections you’re already thinking of have likely been asked already, but it doesn’t mean they’ve been answered.