Jamie PowDemocracy is all about elections, right? If you add Democracy for Realists to your summer reading list, you might be left with some second thoughts.
The book was recently named ‘Book of the Year’ by the International Society of Political Psychology (ISPP). After a series of electoral upsets, populist momentum, and a growing sense of political unpredictability, the ISPP makes a timely award.
Achen and Bartels present an uncomfortable argument: democratic theorists, journalists and citizens themselves often have vastly unrealistic expectations about voters’ behaviour in elections. The authors don’t argue against democracy – or elections – but simply that we shouldn’t assume they deliver all the things we think they do.
One core expectation is that voters reward or punish governing parties based on how they perform in office. Voters are supposed to be able to keep a ‘running tally’ of how well the economy is doing.
The provocative title of Chapter 5 gives you a hint of the authors’ findings to the contrary: ‘Blind Retrospection: Electoral Responses to Droughts, Floods and Shark Attacks’.
Yes, shark attacks(!) In the summer of 1916, there were a number of freak shark attacks in the state of New Jersey. In the presidential election of November that year, President Wilson won re-election. There was very little change in his vote share across the state of New Jersey since the 1912 election, but in the beach townships affected by shark attacks, it fell by over 13 percentage points.
With some compelling statistical analysis, the authors claim that people in these towns were punishing President Wilson for random events for which he couldn’t possibly be held responsible. This is an example of ‘blind retrospection’.
That’s one eye-catching example from over a century ago, but Achen and Bartels go on to provide a more systematic analysis of the relationship between the state of the economy throughout a president’s term and whether or not his party is successful in re-election.
The finding is stark: voters don’t tend to take into account the economic picture until the last few months immediately before the election: “Voters evaluating the economy on the basis of any one or two quarters are likely to do little better than chance at capturing the potentially meaningful differences in long-term performance represented by the administration average growth rates.”
This phenomenon is termed ‘musical chairs’. It’s how the economy is performing when the president is up for re-election that seals his or her fate, regardless of the bigger picture.
A final piece of the argument is that voters’ behaviour is overwhelmingly affected by their partisanship. This driving force is a form of identity; it can be related to issue positions and ideology, but yet a very distinct factor in its own right.
Partisanship contributes to polarisation and, disconcertingly, strong biases in favour of a political party in spite of objective facts. When the budget deficit had decreased substantially under President Clinton in the 90s, Republican supporters still overwhelmingly believed that the budget deficit was increasing. Across multiple measures, partisan biases are remarkably robust.
Achen and Bartels do not argue that this means that voters are necessarily irrational, but simply that “most voters have very little real information” with which to interpret the world more objectively. Partisanship provides a handy shortcut without having to read economic reports and policy papers.
So, where does this leave the state of our conventional understanding of democracy and elections?
The short answer is: in a shambles. All the conventional defences of democratic government are at odds with demonstrable, centrally important facts of political life. One has to believe six impossible things before breakfast to take real comfort in any of them. Some of the standard defences romanticise human nature, some mathematize it, and others bowdlerise it, but they all have one thing in common: they do not portray human beings realistically, nor take honest account of our human limitations.
This all sounds fairly depressing, but, thankfully, the authors ask a final question: ‘What is good about democracy?’ They provide a list.
- “Perhaps most obviously, elections generally provide authoritative, widely accepted agreement about who shall rule.” This is crucial for legitimate decision-making.
- In effective democracies, “parties that win office are inevitably defeated at a subsequent election.” In contrast to dictatorships, power changes hands.
- Competition in elections provide “incentives for rulers at any given moment to tolerate opposition.” This allows a perpetual debate to take place over future political direction.
- Participation in democracy has benefits “for the development of human character,” enhancing people’s appreciation of their stake in society and their civic responsibilities.
- Politicians “will strive to avoid being caught violating consensual ethical norms in their society.” In other words, there are (usually) limits to what elected representatives can get away with, without risking their job security.
As Churchill quipped, “Democracy is the worst form of government except for all those other forms that have been tried from time to time…” But yet democracy is the only form of government that allows – and demands – constant, critical reflection.
This is a bold book. As we continue to wrestle with what it means to be governed by popular rule, it offers a systematic critique of some of our basic assumptions. This realist account may not convince idealists and pessimists, but it sets them a provocative challenge.
Democracy for Realists: Why Elections Do Not Produce Responsive Government (2016) is published by Princeton University Press.
Also published on Medium.
