Written evidence submitted by Stephen D Fisher, Associate Professor of Political Sociology, and Fellow and Tutor in Politics, Trinity College, University of Oxford (VUK 35)

Executive summary: Inequalities in turnout and registration are more important than their levels, but low levels are associated with greater inequalities. Thus low turnout is a more serious problem for political representation in Britain than registration rates, which compare well internationally. Establishing a national population database for voter registration and other government services would be the most efficient and effective way to improve registration. Short of this it would be helpful to use existing government records as much as possible. Turnout rates would most obviously and reliably be improved with enforced compulsory voting.

This document is based on my reading of the academic literature and my personal opinions informed from research and teaching on this subject over fifteen years: not the views of my employers. I address the published questions in turn.

Reasons for and impact of low voter engagement

1. What are the main factors that have contributed to low voter turnout in recent UK elections?

1.1 There are many important factors but the research is such that it is unclear what their absolute contributions or relative importance are. 

1.2 Relative to higher turnout in previous UK general elections, low turnout in recent elections is due to fewer people believing there is a substantial difference between the main political parties,[1] less belief that voting is a duty,[2] the replacement of older more civic cohorts with younger less engaged ones,[3] and economic globalization reducing the power of government.[4] Union members are more likely than other employees to vote and so decline of union membership might contribute to overall turnout decline.[5]

1.3 Patterns of change for elections to devolved institutions and local government are more mixed,[6] suggesting indifference between parties and declining power at the national level are perhaps more important than general population trends and long-run attitudinal change.  

1.4 Research comparing advanced democracies consistently shows that countries with first-past-the-post electoral systems tend to have lower turnout than those with proportional representation. Also countries with enforced compulsory voting have very high turnout. Both these factors therefore help to explain why Britain has lower turnout than many other advanced democracies. Evidence for other institutional factors that are often mooted is decidedly mixed.[7]

2. What are the main factors that affect voter registration? 

2.1 Electoral registration in Britain compares well with other advanced democracies, but it is most efficient and practically perfectly comprehensive where it is conducted by government on the basis of a national population database as in Scandinavian countries. While it is a legal requirement in Britain to return electoral registration forms, the law is rarely enforced. So registration in Britain can be thought of as essentially voluntary. Evidence from the USA suggests that within a voluntary registration system policies that make the process more accessible and easier tend to help increase registration rates and reduce inequalities in registration.[8]

3. To what extent does the public’s perception of MPs, Parliament, the Government and events such as Prime Minister’s questions affect voter registration and turnout?


3.1 While negative perceptions of politicians and politics are certainly more prevalent among non-voters than voters and they may be causally connected, but there is insufficient research evidence showing that changes in these attitudes definitely lead to changes in turnout.[9] It is worth noting that turnout between 2005 and 2010 increased despite the MPs’ expenses scandal and that the rise in turnout was no weaker in those constituencies where MPs were (most) implicated than it was in those constituencies not directly affected.[10] This suggests that perceptions of politicians do not have a substantial effect on overall levels of turnout.

4. What role does the media play in this context?

4.1 While some research shows that the media can enhance participation by informing and motivating people, other studies show that it can also put people off politics and voting. The effects depend on the nature of the source, content and context of the media coverage, and so overall there might be very little net effect on participation.[11]                           

5. What socioeconomic factors affect registration and turnout and what, if anything, can we learn from this about how to improve voter registration and turnout?

5.1 In the UK and internationally, both registration and turnout are lower among more marginal groups (e.g. those with lower education, lower incomes and immigrants) and among those who are more recently resident at their current address.[12] This last group disproportionately includes young people who are more generally less likely to vote than their elders.[13]

5.2 While it is possible to think about measures that specifically aim to improve voter registration and turnout among those groups that have relatively low registration and turnout, I argue below that it is more effective to change the registration system so that all eligible voters are highly likely to be registered and make voting compulsory so that all electors have to cast a ballot. These measures would inevitably have the biggest effect on the currently least well-registered groups and those with the lowest turnout.

6. What are the costs to society of low voter registration and turnout?

6.1 Lower registration and turnout imply more potential for and more actual turnout inequalities. While it is difficult to prove, there seems to be some evidence that political parties (in the UK and elsewhere) are more likely to design policies to attract the kinds of people who are more likely to vote at the expense of those less likely to vote. Thus there is a danger of policy making disproportionately in favour of older, richer and more educated citizens instead of younger, poorer and less educated.[14]

Improving voter turnout

7. What are the principal ways in which voter registration and turnout could be improved?

7.1 Registration would be best improved by automatic registration using information from the tax, benefits, NHS and other government records, perhaps but not necessarily involving a more general national population database as in Scandinavian countries. Exploiting existing information held by government more and relying less on individuals or households to register is efficient, effective and important for reducing inequalities between social groups in registration rates. The new Individual Electoral Registration system is an important step in this direction but it still ultimately puts unnecessary responsibility on individuals to register themselves when it could be done automatically for them.

7.2 Turnout would be best improved by making voting compulsory with enforcement, as in Australia. Compulsory voting really means compulsory ballot casting: people can cast blank or invalid votes if they wish. While it is commonly argued that compulsory voting is an unwarranted illiberal imposition, it would be a tiny burden compared with the costs of paying taxes and complying with other laws. Compulsory voting is a small price to pay to practically eliminate major inequalities in turnout and for a democracy in which everyone participates.

7.3 There are no other ways of making greater improvements in registration and turnout rates. Most other commonly proposed policies would have considerably weaker effects and some may not help at all.

8. What lessons can be learnt from other countries where voter registration and turnout is higher?

8.1 This has been covered in answers to questions above.                           

9. To what extent could electoral reform, rebuilding political parties or changes to party funding improve public engagement and voter turnout?

9.1 There is consistent research evidence suggests that electoral reform involving a change from first-past-the-post to proportional representation would increase turnout.[15] But the likely improvement in turnout would be small. For example, the experience of the New Zealand switch to proportional representation in the 1990s was not enough to offset long-term trends driving turnout decline.[16] Moreover, other implications of proportional representation for the political system are so much more important that proportional representation should not be recommended primarily on the basis that it will increase voter turnout.

9.2 Cross-national and over-time variation in public funding of political parties is not clearly correlated with turnout at elections, but there is considerable debate as to whether such party funding might have negative effects on the responsiveness of parties to voters.[17]

10. In what ways could new technologies be used to encourage people to vote?

10.1 There are good reasons to suppose that new technologies might make voting easier and more attractive for some people who are currently relatively unlikely to vote, especially young people.  My impression of the international experience thus far provides little evidence that use of new technologies have a substantial effect on turnout levels overall or even for young people.

10.2 The risks associated with new technologies are perhaps more important than any potential benefits from higher turnout.  With electronic voting of any kind outside polling stations it is impossible to ensure that all electors are casting their own votes.  Perhaps the main danger is that some electors will come under pressure from family or others to be observed voting, or even relinquish control over their vote. This is a common complaint about the current postal-vote system in the UK.

11. What would be the advantages and disadvantages of allowing voters to register on the day of an election?

11.1 The main advantage would be enabling those who have failed to register in advance to vote, and thereby increase turnout and potentially reduce turnout inequalities.

11.2 The main disadvantages would be the administrative costs of doing so quickly enough to avoid the kinds of long queues that have been seen at recent US elections, and the lack of time to conduct adequate checks.

12. How can arrangements for British citizens living abroad to register for and vote in elections in the UK be improved?

12.1 I will decline to comment on this other than to say that it could become part of a more general national population database as discussed above.

25 February 2014