Neuro-Politics: Rational Ignorance - Part 1

From "The Trouble With Elections: Everything We Thought We Knew About Democracy is Wrong," Chapter 8.1

I now turn to some underlying traits of the human mind that make elections, and mass-participation plebiscite direct democracy untenable. This chapter will present many of the reasons that our brains are not well suited for elections, and the next chapters will focus on why elected officials are not well suited for policy-making. These factors disqualify elections as the basic tool of democracy, regardless of voting method, or other election reforms. Some of these factors have only come to light since recent advancements in the science of psychology, but many of them have been recognized for eons. 

It is also important to recognize that there is a distinction between being a political leader and a political representative. However, in electoral systems these two roles are generally mixed together. Evolutionary development of leadership and followership in small scale societies in which humans have spent most of their existence, creates a mismatch with the role and selection of leadership in modern large scale societies.

Rational Ignorance

We’ll start by examining a serious problem, if not an “Achilles’ Heel” of mass elections — ignorance. To be clear, I am not deriding the intelligence of ordinary people, simply acknowledging that we all know a lot about some things, but little or nothing about most other things. Elections may be appropriate in small settings where everybody knows everybody else personally, but they become problematic in large communities. Here I will focus on governmental elections, but much of the analysis that follows is also applicable to large organizations. 

Most people would agree that for elections to work well, voters should have knowledge about their choices. This is true whether we are speaking about a referendum, or electing public officials. I will focus on the election of representatives, as the most common election scenario. To make an informed decision, voters should know the performance record and true character of the incumbent, as well as the opponents. This should not be based on passing impressions from 30-second TV spots or targeted social media ads designed to manipulate. Most Americans agree that voters go to the polls without being well-informed. The Rasmussen polling firm regularly asks voters whether they believe they are well-informed about policies and candidates before they vote, and also whether they believe other Americans are well-informed when they vote. Interestingly, although most voters believe they themselves are well-informed, only around 15 percent believe that American voters overall are well-informed. This has ranged between 9 percent and 21 percent in recent years. So, do respondents tend to overestimate their own level of knowledge, while fairly assessing everybody else? A hint might come from surveys that regularly indicate that a huge majority of people believe they are better than average automobile drivers. (Note that it is mathematically impossible for this to be correct). It is likely that both I – the author, and you – the reader, of this book believe we are better informed about politics than most people. Yet, many of the people we think we know more than, likely think they know more than us. It is natural to believe that our beliefs are true – that is what makes them beliefs. True intellectual humility is a rare commodity, but we should acknowledge that we may be the ones who are mistaken.

Hardly any voters are actually well-informed about all of the candidates, or even the “major” candidates. They may know a lot about one or two candidates in one or two races, but certainly not all of the offices up for election (in the U.S. we generally elect far more offices than in most other nations). In a party-based system of proportional representation, such detailed candidate information is not as important since the candidates are bound to support the party platform. But here as well, there is a dearth of actual knowledge among voters about party platforms, let alone the actual ramifications of a party’s proposed policies, nor what coalition compromises will be struck following the election. Since candidate-based elections also have policy implications, voters should know about the policy directions of the various candidates or parties. But more — they should have an understanding of the implications of those policies, rather than merely a superficial opinion. Further, there are innumerable policy areas, such that no voter’s individual knowledge can encompass very many of them.

In his book The Phantom Public, Walter Lippmann observed that

“The individual man does not have opinions on all public affairs. He does not know what is happening, why it is happening, what ought to happen… and there is not the least reason for thinking, as mystical democrats have thought, that the compounding of individual ignorances in masses of people can produce a continuous directing force in public affairs.”

But this is not a condemnation of people’s intelligence or competence. It is a recognition of the fact that humans have limited attention (what Jeremy Bentham called “active aptitude”), and other things to do, combined with the reality that one person’s vote in a mass election has almost no chance of actually changing an outcome (it would need to break or create a tie). However, if a small representative sample of citizens were presented with a policy choice, given resources, information, and time, and knew that their decision would be consequential, this dismissive story becomes very different.

With the de minimis impact of a single vote in a mass election (being essentially inconsequential), people have little rational incentive to exert much effort to become well-informed. Schumpeter believed that the typical citizen

“drops down to a lower level of mental performance as soon as he enters the political field. He argues and analyzes in a way which he would readily recognize as infantile within the sphere of his real interests.”

The larger the group, the less each individual feels obliged to pull his own weight. This is literally true in experiments measuring how hard individuals pull in a game of tug of war comparing just a few on a side to a game with dozens on each side. As Schumpeter put it, voters in a mass election have

“a reduced sense of responsibility, a lower level of energy of thought and greater sensitiveness to non-logical influence.”

The larger the group, the more likely each individual is to engage in informational free-riding. The reliance on social cues, or trusted “opinion leaders,” rather than expending the necessary energy to make independent assessments, is natural and typical behavior for most participating voters in any mass election. In order to be a well-informed voter it would take many, many days of serious research. The key observation is that most people are unwilling to expend a great deal of energy on an activity that is not intrinsically enjoyable when their effort results in no actual or perceived benefit. 

The term rational ignorance was coined by economist Anthony Downs, an economist at the Brookings Institution, in his seminal work, An Economic Theory of Democracy, in 1957.  He noted that since there is vanishingly little chance that one’s single vote will actually make any difference in the outcome of a large scale election¹ (as noted above, a vote would need to either break or make a tie for that vote to actually change the outcome), there is no rational grounds for investing any time or effort in becoming informed about candidates or issues. Economists are fond of doing cost/benefit analysis assuming humans act rationally. Voting is an example of a collective action “public goods problem.” The “marginal per capita return” on contributions to the public good is so tiny that there is essentially no reason to participate. In the realm of voting, economists expect a person to assess how much effort and time it would take to become well-informed about all of the candidates and issues in order to cast an informed vote. The person should also honestly assess the likelihood that their one vote will genuinely make a difference to the election outcome. A fair cost/benefit analysis would show that it is irrational to commit any effort in overcoming ignorance in this situation. The rational thing to do is put in little or no effort to overcome your ignorance. The benefit (return) on the voter’s investment of time is always essentially zero. Indeed, even taking the time to vote at all (without spending time educating oneself) isn’t rational from this perspective. Downs referred to the fact that anybody bothers to vote as “the paradox of voting.” If people were rational, they wouldn’t waste even that much time. 

Single-issue voters may not be ignorant about how all of the candidates stand on their key issue, but this is hardly a counter to the ignorance at the base of electoral “democracy,” since the elected officials will be dealing with hundreds of other issues as well. Within a multi-party proportional representation voting system, the ignorance question is more about knowledge of the various party manifestos (platforms), than of individual candidates. In a two-party winner-take-all system such as in the US, information needs are reduced by usually having only two, or often only one, choice on the ballot.

Many voters believe that any Democrat is better than any Republican, or conversely, that any Republican is better than any Democrat — so any effort learning about the candidates is unnecessary. But voters with objectively identical interests often hold the opposite belief about the better party, and assert that their counterpart supporting the other party is the one who is ill-informed. They may disagree with both party candidates on most issues, but feel compelled to settle for the lesser of two evils. Also, this party label shortcut disappears in the most consequential elections – primary elections within each party. These are the most consequential, because most electoral districts in the US, whether due to gerrymandering or cultural tradition, regularly elect candidates from a particular party, so that the general election is a mere formality.

In the continuation of this chapter (in the next post) I will examine why so many people do vote, despite the minuscule impact of their single ballot.

1. I should mention that in 1986 I lost an election by one vote (though after the recount it turned out I had lost by six votes). But, this was in a very small local election, where each candidate received merely hundreds of votes rather than hundreds of thousands, or millions. In very small local elections it is extremely rare but more plausible to argue that a single vote may matter. The larger the election, however, the less significant each vote becomes.