Peter Bye

In my previous piece, posted on 22nd November 2016, I argued that direct democracy, where the electorate is asked to decide or approve policy, is dangerous. I suggested that representative democracy is a better alternative. One of the reasons I advanced is that a majority of people in an electorate are remarkably ill-informed, making rational decisions next-to-impossible. In this post, I’d like to flesh out my views.

There is a great deal of evidence for just how uninformed people are: the gap between perception and reality can be wide. Here are just a couple of examples from research conducted by Ipsos MORI.

Wealth and taxation. The wealth of the top 1% is widely overestimated. In the UK in 2015, for example, the average guess was that the top 1% owned 59% of total wealth; the actual number was 23%[1]. In contrast, the taxation that business pays is underestimated. For the year 2014-15 in the UK, the average guess was 17% of total taxation; the actual was 29%, which is the largest proportion of tax paid by any group[2].

Immigration and religious minorities. The proportion of immigrants in a population is overestimated. In the UK, for instance, in 2014 the average guess was 24%; the actual was 13%[3]. The Ipsos MORI survey cited was across 14 countries. The overall view was close to the UK: 24% vs. 11%. There are even more gross overestimates of Muslim populations in Western countries. Some guess vs. reality figures are: 33% vs. 8% (France); 21% vs. 5% (UK); and 15% vs 1% (USA).

It’s easy to see how such perceptions can lead to problems. In the case of wealth and taxation, they can lead to an anti-business climate, which can be economically damaging. Concerns about immigration and Muslim populations, based on wrong information and fears of terrorism, stir up xenophobia and paranoia. As a consequence, people may vote for politicians who offer apparently simple solutions to complex and sometimes imaginary problems. The election of Donald Trump is an extreme result.

Why is it that people get the wrong answer so often, in spite of the fact that it’s not that difficult to find reliable sources of information[4]? There are many explanations advanced for the level of ignorance in a population. Daniel Kahneman and Amos Tversky performed extensive research over many years into the way people think, concluding that we are far less rational than we would like to believe. Their work has had a significant impact on economic theory and won Kahneman the 2002 Nobel Memorial Prize in Economic Sciences[5].

An alternative to the psychological explanations described by Kahneman has been advanced by Jason Brennan. In a recent book[6], he discusses the level of ignorance of much of the electorate, with some scarcely-believable examples. He explains at least some of it as a rational choice. Because one vote makes almost no difference in an election (although of course many votes do) people do not go to the trouble of finding out information, which can be a time-consuming and difficult process. They have other things to do. The result is a vote in ignorance. Further, he suggests that when some people do get more involved, they tend to adopt a position and then look for evidence to reinforce it, leading to a systematic bias.

What can be done? Brennan considers that democracy – one person, one vote – has worked better than alternatives, such as monarchy or dictatorship. However, based purely on results, he believes we could do better. He advances a case for restricting voting to those with knowledge: the epistocracy as he calls it. He describes several possible ways of implementing such an approach, including selected groups, and weighted voting, where some people have more votes than others. He suggests that trials could be run to find out what works best.

I find his views on voter ignorance compelling when it comes to referendums: they should never be used for anything important, avoiding the need for an epistocracy. For general elections, I have a major problem: who selects the epistocracy? I see this as fraught with difficulty.

The case for representative democracy is, I believe, strong. The UK is supposed to be governed this way but political parties have appeared for some time to be more concerned with saying what they believe the electorate wants to hear, rather than what they believe to be right. The influence of powerful media groups has played a significant role in the debasement of political debate in the UK and elsewhere.

Representative democracy is not just representing the public’s views to government but government’s position to the public. It should be a two-way process. Although wide consultation is desirable in forming policy, it falls to political parties to construct coherent policies, which they believe to be in the best interest of the country, and then offer them at elections. Policies may be unpopular but necessary in the light of circumstances at the time. The public can then judge on the outcome and, if not satisfied, can say so at the next election.

If representative democracy is to work effectively, we need politicians to review all the facts and argue for effective policies. It also needs the public to realise that governments cannot do everything. Politics may be a messy process but the alternatives are much worse.

Notes and sources

[1] The poll covered 33 countries, most of which had a similar bias. A few had a distortion the other way, for example Russia, where the top 1% own 70% of the wealth but the average guess was lower. For the results, see

[2] See

[3] See

[4] Most democratic governments, and think tanks, publish extensive statistics – see, for example, the ONS . Other organisations provide independent analysis of factual claims, for instance Full Fact – see And a variety of media organisations, including the BBC, The Economist, The Financial Times, The New York Times, The Washington Post, Le Monde, Frankfurter Allgemeine and many more are reliable sources.

[5] See Daniel Kahneman’s ‘Thinking, Fast and Slow’, Allen Lane, 2011 and Penguin, 2012, for a comprehensive account. An Ipsos MORI publication gives a summary of some of the reasons, which include some of Kahneman’s views – see

[6]Against Democracy’. Princeton University Press, 2016. See from which Chapter 1 can be downloaded.

Author :


  1. Hi Peter.
    Just crossed this series of your articles and wanted to witness my full respect for a guru who I had the joy and the honor to stay in touch with for few decades when at work. I am happy for the rest of the world that can now have the pleasure to read your usual intelligent documents that do not require to also have a deep tech backgound but only need engaged brains. Regards from Milan, Gilberto

Leave a Reply