CHAPTER 4
International Perspectives
This chapter examines voter registration and participation in an international setting. Much of the material described here can be found on the Internet. A list of some of the most useful web sites can be found at the end of the chapter, although users should note that addresses sometimes change and links become redundant. Where a site has provided the source for a table the web address has been provided. Much of the material on registration and turnout assembled here relates to the registration and electoral process within individual states of the United States, Europe and Australia.
Previous research has examined levels of turnout at parliamentary and presidential elections. Relatively little is known about participation in sub-national elections. This is partly a function of the difficulties in obtaining reliable data and partly in mapping what are often complex systems and irregular and inconsistent electoral cycles. Another problem with comparative analysis is that the functions carried out by sub-national bodies vary considerably, and that will almost certainly have an impact on electoral participation. The aim of this review of international experience, therefore, is to discover whether the problems of registration and electoral participation facing sub-national government abroad are comparable to our own experience.
A number of different themes are addressed and international comparisons are made with regard to:
- voter qualification and registration practices;
- the nature, timing and frequency of sub-national elections;
- voting procedures, including absentee voting, advance and electronic voting;
- trends and patterns of electoral participation.
These themes will be discussed under three broad headings. First, we examine the rules and procedures governing such matters as voter registration, the frequency of elections, and the different types of electoral system that occur at the sub-national level. The second section identifies a range of voting procedures and the methods chosen to enhance electoral participation abroad. The closing section considers electoral turnout in the U.S.A., Europe and Australia as a means of comparing our own experience of electoral participation with that of other liberal democracies.
THE RULES OF THE GAME
Voting qualification and registration practices
Across most of Europe and most other liberal democracies, the right to vote in sub-national elections is acquired at the age of 18. In some cases, for example, a number of Austrian Länder, the age qualification is above 18. In some German local authorities (within Lower Saxony) 16 year olds have the right to vote. Other countries, for example, the Netherlands, are actively debating a reduction in the voting qualification age to 16 years.European legislation states that citizens of the Community residing in a member state of which he or she is not a national shall be qualified to vote at local elections. Some other countries extend this right to nationals from outside the European Union. In Denmark, for example, any foreigner resident in the country for at least 3 years shall have the right to vote. Similar qualification periods operate in both Sweden and Norway. In the Netherlands it is 5 years before qualification is given; in Ireland it is just 15 working days before the election. The United Kingdom allows residents who are citizens of Commonwealth countries or Dependent territories to vote.
In the majority of countries it is not compulsory to vote in local elections. The few exceptions to this rule include some Australian states, Belgium, Cyprus, Greece and Luxembourg. Some countries, including the Netherlands and Italy, once had compulsory voting but no longer do so.
Registering to vote is compulsory in the vast majority of liberal democracies although the law is implemented to greater or lesser degrees. In New Zealand, for example, although voter registration is compulsory the law is not strictly enforced. This has led some commentators to conclude that turnout in New Zealand is inflated by the fact that relatively large numbers of voting-age adults simply do not appear on the electoral register. It was noted in Chapter 1 that many local authorities in Britain are also reluctant to prosecute those refusing to register to vote. In Chapter 3 one local authority identified its suspicions that individuals were avoiding registration in order to evade local taxation. Even where there is compulsion to register, therefore, the extent of registration will remain a problem.
In the United States voter registration is voluntary and remains a highly contentious political issue. Compulsory registration is perceived by some as infringing an individual's civil rights. However, the low levels of voter registration led to the passing of the National Voter Registration Act of 1993 (NVRA). The NVRA or so-called 'Motor-Voter' law sought to facilitate the process of electoral registration in a variety of ways, principally by tying it in with driving licence renewal:
The act requires states to register voters in three specified ways in addition to any other procedures the state uses for voter registration:
- Simultaneous application for driver's license and voter registration.
- Mail application for voter registration.
- Application in person at designated government agencies, including public assistance agencies and agencies that provide services to people with disabilities.
Election officials must send all applicants a notice informing them of their voter registration status.
Naturally, the more that electoral registers effectively include all voting-age adults in the population the greater the potential for electoral turnout to decline. Many of those that have been reluctant to register in the U.S., principally the poor, may also prove reluctant to participate in the electoral process. However, it should also be noted that the states that have achieved widespread registration are also those that historically have seen the highest levels of turnout. Nevertheless, the initial impact of the NVRA has proved disappointing and does not appear to have stimulated electoral participation. Some commentators believe, however, that it is still too early to consider its effects. Political parties still have to adjust their campaigning strategies and techniques to take account of the broader electorate.
Frequency of elections
The vast majority of local politicians are elected for periods ranging from 4-6 years, with four-year terms being modal. At the upper end of the range is Belgium where representatives to both the provincial authorities and municipal councils have six-year periods in office. This is also true in France and most authorities in Austria while in Ireland local councillors are elected for a period of 5 years although the timing of local elections is sometimes dependent upon the timing of general elections. Electoral cycles vary between four and five years for German state legislatures. Some countries, such as Estonia and New Zealand, continue to elect local officials for a three-year period. The frequency and timing of elections can have an impact on electoral turnout. Some believe that one of the reasons for low turnout in the United States (30-45 per cent in sub-national elections) is the multiplicity of elections, which leads to a sense of 'voter fatigue'. Conversely, if local elections are held simultaneously with parliamentary elections, for example in Sweden where turnout averages almost 80 per cent, then rates of participation will match those normal for a parliamentary election. Britain's experience in both 1979, and more recently in 1997, when local turnout rose dramatically at the synchronous local and general elections, certainly supports that observation.Types of electoral system
Simple plurality electoral systems (better known as 'first past the post' or FPTP), such as those used in Great Britain, are the exception rather than the rule. This system continues to be used at the sub-national level in the English-speaking democracies, most notably, Canada, New Zealand and the United States. District magnitude (the number of seats elected from a given constituency) varies considerably from single-member wards to 'at large' elections where the local authority as a whole comprises a single constituency. Other countries within the European Union use a variety of methods for electing local authorities. In France and elsewhere a system of proportional representation (PR) based on party lists is used for electing regional governments. State elections in Germany use another form of PR known as the Additional Member System (AMS). Yet another system in use for sub-national elections in Europe is the Single Transferable Vote (STV), which is used in both Northern Ireland and Ireland. These electoral systems vary greatly in terms of the choice given to voters, in how votes are translated into seats and whether special rules apply. Each of these elements may have an effect on levels of voter turnout.
Voting procedures
Most liberal democracies identify the level of turnout as a critical variable when assessing the legitimacy of elected authorities. It is understandable, therefore, that various measures have been taken to combat electoral apathy and increase levels of turnout.A large and increasing number of countries, including the UK, already permit either postal or proxy voting. Reliable data are largely unavailable but in those cases where information can be obtained the evidence is that voters who use this method participate in greater numbers than those who are required to attend a polling station. Significantly, a number of countries have sought to extend this practice so that a greater proportion of the electorate is eligible to vote by post. In Norway, for example, 20 per cent of votes are now cast by postal voters while in both Sweden and Finland the figure is nearer four in ten. Belgium allows proxy voting at relatively short notice when voters are unable to attend the polling station because of illness or for other reasons. Italy and other countries permit certain voters, for example, those in hospitals and nursing homes, to cast a ballot at these premises.
New Zealand too has been enthusiastic in its adoption of 'voting enhancements' (Bush 1999). In 1968, for example, local voting was spread over an 11-day period, mobile polling booths were introduced and voting by post was encouraged. By 1977 local authorities were allowed to use such methods at their own discretion. Between 1977 and 1995 the use of mobile voting booths declined, that of spread voting to a lesser extent, but postal voting became virtually universal. In 1989 postal voting was made compulsory but the law was relaxed in 1992. Interestingly, in 1992 only a single local authority reverted to polling places. Turnout in that authority fell to just 26 per cent, compared with a national average of 54 per cent. By 1995, every local authority used universal postal voting for its local elections. The evidence from Britain (see Chapters 1 and 2) and elsewhere appears to show that postal voting may prove an effective instrument in raising levels of electoral participation.
One of the most ambitious attempts to boost turnout in the United States has been made in Oregon. In 1987 the state legislature gave the option that local elections or state by-elections could be conducted entirely by postal ballot or 'vote-by-mail' as it is known. By the mid 1990s most counties in the state had chosen this method, primarily because of the reduced cost. The evidence is limited but voting by mail does appear to generate higher levels of turnout. In Oregon, turnout at conventionally run elections included 38% in a 1994 primary and 68% at that year's general election. By contrast, vote by mail elections saw turnouts of 66% for a Senate by-election and 56% in a 1996 primary. Oregon's voters appear enthusiastic supporters of the innovation. One survey (Southwell and Burchett 1997) discovered that 75% preferred voting by mail compared with just 15% that wanted to retain the traditional polling place. The chief reasons given for preferring a mail election were that it was easier and more convenient (79%), less time-consuming (28%) and allowed more time to read the ballot paper (18%). Among those preferring the tried and trusted method, most (42%) stated that voting by mail made the act of voting insignificant.
A number of other countries and states also allow a system of early or advanced voting whereby electors are not restricted to a single election day but can instead vote over a longer period. Norway has a system of early voting. Commencing at least two months before the election, voters can cast a ballot at properly designated premises. Similarly, a number of states in the U.S.A. permit early voting or advance voting in state-wide elections. Below, is a description of the procedure adopted in the state of Kansas.
Voting in Kansas is easier than ever. With Advance Voting, any registered voter can vote from home, by mail or in person before Election Tuesday.
Contact your county election officer to request an application for an advance voting ballot.
Complete the application and return it to your county election officer.
You can have your ballot mailed to you starting 20 days before the election.
You may vote in person in the county election office starting the Tuesday before election day, or up to 20 days before the election, depending on the county.
All ballots must be received in the county election office by the close of polls on election day.
Sick, disabled or illiterate voters may receive assistance in applying for and casting advance voting ballots.Source: http://www.kssos.org/adv.html
Another state, Texas, allowed voting for the 1998 elections, scheduled for November 3, to begin on Oct 19. The table below shows the extent of advance voting, both in person and by mail, by county as of October 30. The table shows that the extent of advance voting varies considerably by electoral area with the proportion in Jefferson county three times that of some other counties.
Advance Voting in Texas state elections, 1998 County Registered Voters 1998 In-Person Voters 1998 By Mail Voters Cumulative advance voters by 30 Oct % Advance Voting as of 30 Oct Jefferson 166,645 28,085 1,969 30,054 18.03 Lubbock 147,535 17,955 479 18,434 12.49 Galveston 168,481 18,343 2,190 20,533 12.19 Nueces 197,622 21,252 903 22,155 11.21 Travis 499,696 51,968 2,640 54,608 10.93 El Paso 332,020 30,895 2,490 33,385 10.06 Bexar 818,370 71,313 10,595 81,908 10.01 Montgomery 159,571 14,683 723 15,406 9.65 Collin 263,573 24,038 450 24,038 9.12 Tarrant 814,547 56,196 7,760 63,956 7.85 Dallas 1,146,673 77,066 6,817 83,883 7.32 Denton 246,655 17,634 311 17,945 7.28 Hidalgo 223,729 14,652 931 15,583 6.97 Fort Bend 175,820 11,192 415 11,607 6.60 Harris 1,755,809 85,061 21,730 106,791 6.08 Total 7,116,746 540,333 59,953 600,286 8.43 Source: adapted from http://www.sos.state.tx.us/function/elec1/early/103098.htm To facilitate voting a number of U.S. states have also introduced electronic ballot counting and even gone as far to allow electronic voting. Reproduced below, for example, is a description of the new procedures for electronic ballot counting now implemented in the state of Alaska:
The State of Alaska introduced a new ballot tabulation system for the 1998 elections. Punch cards are gone, and in their place is optical scanning. This means that Alaska voters pick up only one ballot at their polling places, and fill in ovals next to the names of candidates and issues they support. They insert the ballot into a small computer that looks like a fax machine. The computer - called an Accu-Vote - scans the ballot, tabulates the results, and deposits the ballot into a ballot box below. When the polls close at 8 p.m., precinct workers transmit results, immediately, by modem to a central accumulation site, where they are released.
Source: http://www.gov.state.ak.us/ltgov/elections/avintro.htmElectronic voting, whereby voters can vote from designated terminals situated in public places or even from their own homes over the Internet, is also being developed. In the United States, computers were first used to tabulate votes as early as the 1960s. The U.S. Federal Election Commission (FEC) called for independent studies of the administration of elections and following that inquiry various states have developed electronic voting systems still further. According to the FEC, some 7.7% of the registered voters in the United States used some type of direct recording electronic voting system at the 1996 elections. Oklahoma uses a state-wide election management system that handles not only voter registration but also changes precinct boundaries, verifies names on petitions, does polling place management, handles election accounting and budgeting, and processes absentee voters. Elsewhere in the U.S., states such as Delaware and Florida have developed similar systems while other experiments with electronic voting have been reported from Liege, Belgium for its regional elections and in New Zealand. Such developments are not surprising given that increasingly one of the main explanations for non-voting given by respondents to post-election polls is that insufficient time could be found for voting.
A significant development for electronic voting will arrive when voters are able to vote electronically from their own home. Below is a statement from one U.S. company involved in this field:
Worldwide Election Systems, Inc., announced today the development of the first Internet Voting System for use in general elections. This system would allow anyone to vote for their favorite candidates on a standard personal computer in the comfort of their home or office.
The patent-pending Internet Voting System will be sold as an integrated part of Worldwide's ELECTOR election system, which also includes ballot creation software, electronic voting booths and digital imaging absentee systems. The Internet Voting System is expected to be in first use during the 1999 general elections. A non-secure demonstration of internet voting can be found today atwww.internetvoting.com.
"Internet Voting will save money and resources, enfranchise voters and increase voter turnout" predicts Neil McClure, president of Worldwide Election Systems. "A lot of people would rather vote in a leisurely manner from their office desk or home PC than take the time to drive to their local polling place and then stand in line. Even if you don't have a PC, Internet Voting will allow the internet terminals in every library and shopping mall to be potential voting booths."
In fact, a recent study by ActivMedia showed that two thirds of Americans would like to vote on the internet. This system would also save governments considerable money over traditional methods of absentee voting, which often cost more than $5 for each vote processed.
Civil liberties groups, however, have raised doubts about the security of such voting systems.
LEVELS OF TURNOUT - INTERNATIONAL COMPARISONS
We have assembled data on sub-national turnout for a range of countries. The data largely describe turnout for the highest level of sub-national government within each country. Different electoral systems have been included for comparative purposes. However, we have not allowed for the fact that the powers and functions of sub-national institutions vary considerably and that this may have an effect on turnout. Moreover, we have not discussed the impact of a broad range of social, cultural and other factors that clearly have an effect upon overall levels of turnout in different countries.
Where possible, data have been sorted in descending order of turnout at the most recently held elections and the mean and range calculated. In some cases, Tasmania and Belgium for example, the extremely high level of turnout is a function of compulsory voting. Without exception, turnout at the regional/provincial level across Europe is higher than that found across Great Britain and the United States. The general trend, however, appears to match that encountered in national elections - overall levels of turnout recorded at recent elections appear to be in decline. This raises the wider issue of a widespread decline in civic responsibility and how those countries affected can address this problem. However, an analysis of civic responsibility lies outside the scope of this report.
Electoral participation in the United States
Although a very small number of local authorities in the United States use some form of PR the vast majority of representatives, including state legislators, are elected by FPTP. In this sense, therefore, we are controlling for the electoral system and its effects, when we compare the British experience with that of the United States. Generally, we have selected turnout data for non-Presidential election years. It should also be noted that turnout is frequently calculated on the basis of the most prominent elected office, although it is clear that levels of participation for other simultaneous elections do not vary by a significant amount.As noted earlier, the issue of electoral registration and turnout is highly important in discussions about the health of American democracy. Frequently, two sets of turnout figures are available - turnout measured against the voting age population (VAP) calculated from census data and turnout measured against those actually registered to vote. As the following extract reveals, the method for calculating turnout in America is itself a contentious issue:
'...turnout is arrived at by dividing those who voted by the population eligible to vote, NOT by those registered. For whatever the flaws (noted below) in the eligible vote (Voting Age Population) figures provided by the Census and minor anomalies in the vote count (also noted below), this gauge is by far more reliable and consistent than registration. Using registration as a denominator leaves out those who both did not register and did not vote (and, in the case of North Dakota, which has no registration, would leave out a whole state). Registration as a denominator has no consistency, since it fluctuates by changes in registration law and procedure. (A major change, such as the implementation of the National Voter Registration Act of 1993 - the so-called motor voter law - can show dramatic increases in registration and thus exaggerate decreases in turnout.) Registration lists are inflated by those who have either died or moved but remain on the lists because those lists have not been recently cleaned. This might not be a problem if each state had consistent list cleaning procedures with regard to frequency and timing, but they don't. Thus, there is no way of measuring the degree of distortion in individual states. And because of certain provisions in the motor voter law (noted below), state official registration statistics are more inaccurate than they have ever been.'
Source: Committee for the Study of the American Electorate 1998
http://www.gspm.org/csae/cgans5.htmlAs the following table makes clear, there is a large difference between the two measures. For the latest state elections, turnout was 36 per cent if all voting age adults were included but 65 per cent if registered electors only were counted. Over the period 1962-1990, turnout of registered voters has been stable but among the voting age population it has declined.
Trends in turnout (eligible and registered electorate) in US state-wide elections 1962-1998 Year % VAP % registered 1998 36.06 64.6* 1994 38.79 61.8 1990 36.53 60.0 1986 36.42 61.2 1982 40.09 60.0 1978 37.77 60.1 1974 38.78 61.6 1970 46.78 64.9 1966 48.61 66.0 1962 47.57 64.2 *this figure is an estimate only.
Source: U.S. Bureau of the Census
The data for the state of California are more fine-grained since county level turnout figures are given for the 1998 elections. Overall, turnout amongst registered voters was 62 per cent but only 47 per cent if the voting age population is used for the calculation. We should note the variation among counties in terms of voter registration. Tulare county appears to have a reasonable turnout when measured against registration. However, when the voting age population is used as a measure turnout falls to just over a third. The smallest gap between the two measures, 7.6, occurs in Mariposa county. These data show that levels of electoral registration vary widely across counties, as does turnout. That would appear to confirm the British experience where electoral registration varies across wards within individual local authorities, although the element of compulsion in our system means that the variation is not as great. It is also clear that there can be a large range in levels of turnout within a single state as we have also discovered from our own research.
State of California 1998 county election turnout by difference between registered voters and voting age population County % turnout among registered voters % turnout VAP % Difference Tulare 61.7 35.2 -26.5 Monterey 63.4 38.4 -25.0 Glenn 67.2 44.3 -22.9 Tehama 70.7 48.9 -21.8 Colusa 68.1 46.4 -21.7 San Mateo 64.7 43.4 -21.3 Riverside 57.5 36.3 -21.2 Santa Clare 59.8 39.8 -20.0 Yuba 53.7 33.8 -19.9 Shasta 65.3 45.6 -19.7 Orange 61.4 41.7 -19.7 San Bernardino 51.6 32.8 -18.8 Napa 68.1 49.4 -18.7 Sonoma 69.2 50.9 -18.3 Inyo 69.2 50.9 -18.3 Lake 63.9 45.6 -18.3 San Diego 58.1 40.4 -17.7 San Joaquin 57.0 39.4 -17.6 Imperial 54.4 36.8 -17.6 Fresno 56.3 38.8 -17.5 Madera 58.0 41.7 -16.3 Del Norte 60.2 44.1 -16.1 Marin 72.5 56.5 -16.0 Sutter 60.4 44.6 -15.8 Sacramento 61.5 46.1 -15.4 Solano 56.0 40.7 -15.3 Stanislaus 52.4 37.2 -15.2 Calaveras 69.2 54.1 -15.1 Siskiyou 66.5 51.4 -15.1 Contra Costa 63.1 48.0 -15.1 San Luis Opisbo 65.8 50.9 -14.9 Kings 53.6 39.0 -14.6 Kern 51.1 36.5 -14.6 Mendocino 63.2 48.7 -14.5 Alameda 55.6 41.5 -14.1 El Dorado 67.3 53.3 -14.0 Los Angeles 53.4 39.4 -14.0 Plumas 69.2 55.4 -13.8 Yolo 61.9 48.2 -13.7 Amador 74.3 60.8 -13.5 Mono 61.3 47.8 -13.5 Alpine 72.7 59.8 -12.9 Modoc 60.9 48.1 -12.8 Nevada 69.3 56.8 -12.5 Santa Cruz 62.8 51.0 -11.8 San Benito 56.9 45.1 -11.8 Humboldt 61.5 50.2 -11.3 Merced 48.8 37.6 -11.2 Placer 65.9 55.3 -10.6 Santa Barbera 55.8 45.5 -10.3 Sierra 79.2 69.4 -9.8 Butte 54.6 44.9 -9.7 Trinity 60.6 51.2 -9.4 Ventura 53.5 44.4 -9.1 Lassen 58.9 50.4 -8.5 Tuolumne 67.5 59.8 -7.7 San Francisco 55.9 48.3 -7.6 Mariposa 67.1 59.5 -7.6 Mean 61.8 46.5 -15.3 Source: adapted from http://Vote98.ss.ca.gov/Final/sov/v.pdf The analysis in Chapters 1 and 2 showed that problems of registration and levels of turnout appear to be related to the age profile of the local population. The table below presents data gathered by the United States Census Bureau that examines registration and voting by age. The first set of figures relate to different age groups within the population and the percentage of voting age adults that are registered to vote at each election. The second set of figures show the percentage of each age group that voted in an election. Thus, in 1994, the most recent year for which data are available, 42% of 18-24 year olds were registered to vote but the turnout for that age group was just 20%. The table clearly shows that the elderly are far more likely to register to vote. Once registered, the elderly are also more likely to cast a ballot than younger age groups, particularly newly qualified voters. The young, in particular, increasingly appear not to vote in non-Presidential election years. Such findings are in line with those found in Britain where a number of local authorities in our survey identified problems in registering students, for example, and where various initiatives have been launched to try and persuade the young to participate in elections in greater numbers.
Percentage turnout and registration by age for US elections 1964-1994 18-24 25-44 45-64 65+ % Registering 1994 42.3 57.9 71.7 76.3 1992 52.5 64.8 75.3 78.0 1990 39.9 58.4 71.4 76.5 1988 48.2 63.0 75.5 78.4 1986 42.0 61.1 74.8 76.9 1984 51.3 66.6 76.6 76.9 1982 42.4 61.5 75.6 75.2 1980 49.2 65.6 75.8 74.6 1978 40.5 60.2 74.3 72.8 1976 51.3 65.5 75.5 71.4 1974 41.3 59.9 73.6 70.2 1972 58.9 71.3 79.7 75.6 1970 40.9 65.0 77.5 73.7 1968 56.0 72.4 81.1 75.6 1966 44.1 67.6 78.9 73.5 1964 n/a n/a n/a n/a % Voting 1994 20.1 39.4 56.7 61.3 1992 42.8 58.3 70.0 70.1 1990 20.4 40.7 55.8 60.3 1988 36.2 54.0 67.9 68.8 1986 21.9 41.4 58.7 60.9 1984 40.8 58.4 69.8 67.7 1982 24.8 45.4 62.2 59.9 1980 39.9 58.7 69.3 65.1 1978 23.5 43.1 58.5 55.9 1976 42.2 58.7 68.7 62.2 1974 23.8 42.2 56.9 51.4 1972 49.6 62.7 70.8 63.5 1970 30.4 51.9 64.2 57.0 1968 50.4 66.6 74.9 65.8 1966 31.1 53.1 64.5 56.1 1964 50.9 69.0 75.9 66.3 Source: U.S. Bureau of the Census Chapter 2 showed that there has been a general decline in turnout among local authorities in England. Is that also typical of the United States? Clearly, the calculation of turnout using two different measures (proportion of registered voters or proportion of voting age population) can prove to be a source of some confusion. However, below are two tables that use the two different measures but in both cases turnout measured over the last decade appears to be in decline compared with earlier periods.
Trends in state-level turnout in Michigan Year % VAP 1998 43.3 1994 45.5 1990 38.7 1986 37.2 1982 47.8 1978 46.6 1974 44.0 1970 51.6 1966 52.2 1962 60.0 1958 50.0 1954 50.4 1950 45.4
Trends in state-level turnout in Oregon Year % registered 1998 59.0 1994 68.4 1990 76.7 1986 72.4 1982 70.2 1978 63.2 1974 69.3 1970 70.3 1966 73.0 1962 73.0 This comparison with the United States has examined elections for state legislatures. Turnout for local authorities is difficult to obtain but those data that are available suggest that turnout is lower for these types of election. The general picture, however, shows a number of interesting points. It is clear that a system of voluntary registration does not work as well as compulsory registration. Many of those who are eligible to vote, particularly blacks and the poor are omitted from the electoral register. As in Britain, the issue of low turnout is one that attracts a great deal of attention with new initiatives designed to halt what is perceived to be a general decline. It is also apparent that programmes designed to improve voter registration and to encourage turnout have met with mixed fortunes.
Electoral participation in Europe
We begin our comparison with Europe by asking whether the decline in turnout seen in both Britain and the U.S. is replicated in countries within the European Union. The table opposite shows electoral turnout across most EU countries for the last two sets of
sub-national elections. With two exceptions, Portugal and Spain, turnout has declined over the decade, suggesting that the issue of falling turnout is independent of the type of electoral system employed. Nevertheless, Britain still records the lowest local electoral turnout of all member states.
Average turnout at sub-national elections within the EU Mean (%) Change (%) before 1995 after 1995 Luxembourg 93 92 -1 Italy 85 80 -5 Austria 82 79 -3 Sweden 85 79 -6 Denmark 80 72 -8 Spain 65 72 7 Germany 72 70 -2 Portugal 60 62 2 France 68 59 -9 Ireland 60 50 -10 Netherlands 54 47 -7 Great Britain 40 35 -5 Belgium 93 - Mean 72.1 Mean 66.3 Range 53 Range 57 The data appear to confirm that turnout is generally in decline although the rate of that decline varies markedly. Variability in turnout within countries is also in evidence. Turnout does appear to be lower for urban areas, again consistent with our own findings. Sub-national turnout data for a limited number of European countries are shown below.
Austrian Länder by descending turnout at most recent election Bundesland previous election % last election % %change Vorarlberg 1994 89.3 1999 87.8 -1.5 Steiermark 1991 89.6 1995 86.9 -2.7 Burgenland 1991 85.5 1996 81.6 -3.9 Oberösterreich 1991 85.2 1997 81.0 -4.2 Kärnten 1994 85.3 1999 80.5 -4.8 Tyrol 1994 88.1 1999 79.9 -8.2 Salzburg 1994 75.4 1999 74.1 -1.3 Niederösterreich 1993 75.5 1998 72.0 -3.5 Wien (Vienna) 1991 65.4 1996 68.5 3.1 Mean 82.1 Mean 79.1 -3.0 Range 24.2 Range 19.3 Source: Adapted and up-dated from European Union, Committee of the Regions, 'Voter turnout at regional and local elections in the European Union 1990-1999.
French Regions by descending turnout at 1998 elections Region % turnout Limousin 65.41 Midi-Pyrénées 64.34 Auvergne 62.51 Languedoc-Roussillon 62.43 Franche-Comté 62.42 Picardie 61.93 Aquitaine 61.21 Nord-Pas de Calais 59.88 Bretagne 58.98 Haute-Normandie 58.70 Centre 58.50 Basse-Normandie 58.02 Provence Alpes-Côte d'Azur 57.79 Poitou-Charentes 57.73 Pays de la Loire 57.31 Bourgogne 56.95 Champagne-Ardenne 55.57 Île-de-France 54.39 Alsace 54.24 Rhône-Alpes 54.00 Mean 59.12 Range 11.41 Source: Adapted and up-dated from European Union, Committee of the Regions, 'Voter turnout at regional and local elections in the European Union 1990-1999.
Irish County and County Borough Councils by descending turnout at most recent election Authority 1991 1999 % change Leitrim 76.0 75.6 -0.4 Longford 62.0 69.8 7.8 Sligo 68.0 68.4 0.4 Roscommon 73.0 68.0 -5.0 Monaghan 65.0 65.5 0.5 Tipperary N.R. 68.0 64.6 -3.4 Mayo 68.0 63.7 -4.3 Cavan 66.0 62.8 -3.2 Kerry 66.0 62.6 -3.4 Tipperary S.R. 66.0 62.5 -3.5 Donegal 61.0 61.5 0.5 Laois 65.0 59.9 -5.1 Clare 65.0 58.5 -6.5 Waterford 63.0 57.7 -5.3 Galway 63.0 57.4 -5.6 Offaly 61.0 56.5 -4.5 Cork 59.0 56.3 -2.7 Kilkenny 64.0 55.4 -8.6 Limerick 64.0 55.1 -8.9 Westmeath 60.0 54.5 -5.5 Carlow 58.0 51.0 -7.0 Wexford 58.0 50.5 -7.5 Wicklow 57.0 50.5 -6.5 Limerick Corp* 57.0 47.6 -9.4 Cork Corp* 52.0 46.1 -5.9 Meath 51.0 46.0 -5.0 Louth 56.0 45.0 -11.0 Waterford Corp* 61.0 44.8 -16.2 Galway Corp* 52.0 44.2 -7.8 Kildare 49.0 41.7 -7.3 Dublin Fingal 49.0 38.8 -10.2 Dun Laoghaire/Rathdown 42.0 38.6 -3.4 Dublin Corp* 43.0 34.9 -8.1 South Dublin 41.0 33.8 -7.2 Mean 59.7 Mean 54.4 -5.3 Range 35.0 Range 41.8 24.0 *indicates county borough council
Source: Data for 1991 were adapted and up-dated from European Union, Committee of the Regions, 'Voter turnout at regional and local elections in the European Union 1990-1999.' Data for 1999 were adapted from the Tallyman's Guide to Local and European Elections 1999, (Dublin: Limelight Communications).
German Länder by descending turnout at most recent election Bundesland previous election % last election % % change Saarland 1990 83.2 1994 83.5 0.3 Meckl.-Vorpommern 1994 72.9 1998 79.4 6.5 Lower Saxony 1994 73.8 1998 73.8 0.0 Schleswig-Holstein 1992 71.7 1996 71.8 0.1 Sachsen-Anhalt 1994 54.8 1999 71.5 16.7 Rheinland-Pfalz 1992 73.9 1996 70.8 -3.1 Bayern 1994 67.8 1998 69.8 2.0 Hamburg 1993 69.6 1997 68.7 -0.9 Bremen 1991 72.2 1995 68.6 -3.6 Baden-Württemberg 1992 70.1 1996 67.6 -2.5 Berlin 1995 68.6 1999 65.9 -2.7 North Rhine Westphalia 1990 71.8 1995 64.0 -7.8 Hessen 1995 66.3 1999 62.0 -4.3 Saxony 1994 58.4 1999 61.1 2.7 Thüringen 1994 74.8 1999 59.9 -14.9 Brandenburg 1990 67.1 1994 56.3 -10.8 Mean 69.8 Mean 68.4 -1.4 Range 24.8 Range 27.2 Source: Adapted and up-dated from European Union, Committee of the Regions, 'Voter turnout at regional and local elections in the European Union 1990-1999.' ELECTORAL PARTICIPATION IN AUSTRALIA
This brief section is included to permit an assessment of the impact of compulsory voting on electoral turnout. Both federal and state elections in Australia involve compulsory voting. As the table for the Tasmanian Assembly elections makes clear turnout is extremely and consistently high, suggesting that the penalties for non-voting are a deterrent. However, Australian voters are allowed an 'informal vote'. This means that, should the voter not find a suitable candidate which to support then he or she can simply register the fact that they have participated in the election by casting an informal vote. This will not count towards the election of any candidates but will be accounted for in the calculation of turnout.
Tasmanian House of Assembly Elections 1992 1996 Electoral area % Turnout % informal vote % Turnout % informal vote Bass 90.2 5.1 90.2 5.5 Braddon 91.8 3.9 91.1 5.4 Denison 89.4 4.1 90.2 4.9 Franklin 91.1 4.0 91.2 5.0 Lyons 90.7 4.5 91.1 5.2 State total 90.7 4.3 90.8 5.2 Range 1.6 1.2 1.0 0.6 In Australia, local authorities below the state level employ a range of different voting systems, including STV and FPTP and a mix of voluntary/compulsory voting. At local elections in New South Wales, Queensland and Northern Territory voting is compulsory. In the state of Victoria voting is compulsory for enrolled residents but voluntary for enrolled non-residents. For the remaining states of South Australia, Western Australia and Tasmania voting is voluntary at local elections. At the last local elections for which there are data, turnout averaged 28 per cent in South Australia, 10 per cent in Western Australia and 59 per cent in Tasmania. The impact of compulsory voting, therefore, is clearly dramatic and that where there is freedom of choice levels of participation can fall dramatically.
CONCLUSIONS
This chapter has shown that there are similarities and dissimilarities between countries in the areas of electoral registration, responses to low turnout and trends and patterns in electoral participation. Whether electoral registration is voluntary or compulsory clearly makes a significant difference to the proportion of voting age adults that register. Those countries that have a compulsory system successfully register upwards of 90 per cent of eligible adults. Unfortunately, we were unable to obtain comparative data on how the registration process operates in these countries. By contrast, the United States, which operates a voluntary system of electoral registration, has been forced to introduce legislation to cope with electoral apathy. There is no firm evidence that the 1993 legislation has had a major impact in raising levels of electoral participation and indeed the contrary position may well prove to be the case. Those reluctant to register may also prove reluctant to vote. In the United States, the sections of the population least likely to register and vote are the poor and the young. This agrees with our own broad findings.
Various countries have sought to facilitate the act of voting in sub-national elections. Measures to encourage voter turnout at this level are particularly needed because there is generally less publicity given to these elections. These initiatives range from making voting compulsory (which arguably raises a different set of problems) to those designed to bring the election process nearer to the voting public. Various countries and states have introduced early or advanced voting whereby the election in effect takes place over several weeks. Others have sought to relax the rules with regard to absent voting, particularly postal voting. There is strong evidence to show, as in Britain, that electors that apply for a postal vote are more likely to vote than those that are required to vote at a polling station. The most recent initiatives have occurred in the field of electronic voting. Allowing people to vote at remote computer terminals, whether situated in public places or indeed within their own home, represents the 'cutting edge' in terms of facilitating the voting process. Thus far, such experiments are not widespread but as the flow of information across the Internet becomes more secure and as the concerns of various civil liberties' groups are addressed then the practice will be extended.
Our comparative analysis of sub-national electoral turnout has confirmed prior research into national election turnout. Turnout in sub-national elections appears to be in general decline. Few countries show an increase in participation in recent years. This, of course, raises an important issue. Any measures designed to address declining turnout in Britain need to be aware of the fact that this phenomenon is widespread. In short, the problem is not solely a function of the British context but those wider factors, associated with a reduced sense of civic responsibility, are also at work.
USEFUL WEB SITES
General
http://www.int-idea.se/Voter_turnout/index.html. The main website for International Institute for Democracy and Electoral Assistance provides details on turnout at parliamentary and presidential elections from around the world.
http://electionresources.org/europe.html
http://www.agora.stm.it/elections/election/
http://www2.dst.dk/internet/andreuk.htm Statistic links
http://www.istat.it/Astatweb/estero.html Statistic links
Country specific sites
Australia
http://ElectionResources.org/australia.html#AU
http://www.electoral.tas.gov.au/
http://www.waec.wa.gov.au/1996results.htm
Austria
http://www.noe.gv.at/POLITIK/WAHLEN/98_LTG/Index.htm
http://ebweb.tuwien.ac.at/apsm/stmk95/stmk91.html
http://ebweb.tuwien.ac.at/apsm/oow97/oow.html
http://www.tirol.gv.at/landtag/wahlergebnisse.html
http://www.land-sbg.gv.at/statistik/wahlen/ltw/index_2.htm
http://www2.vlr.gv.at/Wahl99/chart.asp?WahlID=2&VWahlID=3&chartID=Detail&Display=gesamtliste
http://www2.vlr.gv.at/Wahl99/default.asp
http://www.magwien.gv.at/ma66/wahlen/
http://www.magwien.gv.at/ma66/wahlen/igproz45.htm
http://www.oestat.gv.at/indexde.htm
Belgium
http://parlement.wallonie.be/presse/index.html
http://www.wallonie.be/Html/M3_Institutions/Institutions_1.htm
http://belgium.fgov.be/pa/nla_frame.htm
http://belgium.fgov.be/pa/nla_frame.htm
http://www.vlaanderen.be/ned/sites/statistieken/index.html
Denmark
http://www.statistikbanken.dk/
http://www.inm.dk/municipalities_counties/kap02.htm
http://www.ft.dk/baggrund/00000048/00232623.htm
http://www.dst.dk/siab.asp?o_id=665
http://www.vestamt.dk/amtsraad/1970-97/stemkom.htm
http://www.vestamt.dk/amtsraad/1970-97/andele.htm
http://www.vestamt.dk/amtsraad/1970-97/persstem.htm
http://www.vestamt.dk/amtsraad/1970-97/mandater.htm
http://dk.dir.yahoo.com/Regionalt/Lande/Danmark/Amter/
http://www.nja.dk/diverse/politik/valg97/resultat.html
http://www.kbhamt.dk/kbh_amtsraad/kbhsamtsraad_valg.html
Finland:
http://www.stat.fi/tk/he/vaalit/vaalit_en.html#The 1996 municipal elections
http://www.stat.fi/tk/he/vaalit/vaalit92/aanivp92.html
http://www.stat.fi/tk/he/vaalit/vaalit92/aanivp92.html
http://iep.univ-lyon2.fr/Ressources/Liens/election.html#France
http://www.adminet.com/world/gov/
http://www.multimania.com/psdoc/election/cadre.html
Germany
http://www.wahlrecht.kpnet.de/ergebnisse/hessen.htm
http://www.wahlrecht.kpnet.de/ergebnisse/mecklenburg.htm
http://www.wahlrecht.kpnet.de/ergebnisse/niedersachsen.htm
http://www.wahlrecht.kpnet.de/ergebnisse/nordrhein-westfalen.htm</p>
http://www.wahlrecht.kpnet.de/ergebnisse/rheinland-pfalz.htm
http://www.wahlrecht.kpnet.de/ergebnisse/saarland.htm
http://www.wahlrecht.kpnet.de/ergebnisse/sachsen.htm
http://www.wahlrecht.kpnet.de/ergebnisse/sachsen-anhalt.htm
http://www.wahlrecht.kpnet.de/ergebnisse/schleswig-holstein.htm
http://www.wahlrecht.kpnet.de/ergebnisse/thueringen.htm
http://www.wahlrecht.kpnet.de/ergebnisse/thueringen.htm
http://www.wahlrecht.kpnet.de/landtage/berlin.htm
http://www.wahlrecht.kpnet.de/ergebnisse/brandenburg.htm
http://www.wahlrecht.kpnet.de/ergebnisse/bremen.htm
http://www.wahlrecht.kpnet.de/ergebnisse/hamburg.htm
Greece
http://www.statistics.gr/en/data/index.htm
Holland
http://www.amsterdam.nl/verkiezingen/uitslagen/provincie/uitslag.html
http://geodata.prv.gelderland.nl/verkiezingen2/
http://www.xs4all.nl/~zira/kiez.html
Italy
http://www.istat.it/Astatweb/naz.html
http://statistica.regione.abruzzo.it/
http://www.regione.emilia-romagna.it/statistica/tabelle/riepilogo.htm
http://www.regione.emilia-romagna.it/statistica/tabelle/ita_ele.htm
http://www.regione.fvg.it/ele/elezioni.htm
http://212.210.110.158/e99/ballotta/listpre.htm
http://www.ars.sicilia.it/Il_Parlamento/Frame/Parlamento_FR.htm
http://www.regione.toscana.it/ita/cif/pubblica/elp96710/indic710.htm
http://www.regione.taa.it/elezioni/
http://www.regione.umbria.it/default.htm
http://www.regione.veneto.it/statistica/frame-vencifre.htm
Luxembourg
http://www.statec.lu/html_en/statistiques/luxembourg_en_chiffres/index.html
Norway
http://www.ssb.no/www-open/english/yearbook/tab/t0001010.shtml
http://www.ssb.no/www-open/english/yearbook/tab/t0001011.shtml
http://www.ssb.no/www-open/english/yearbook/tab/t0001009.shtml
http://www.ssb.no/www-open/english/yearbook/tab/t0001005.shtml
http://www.ssb.no/www-open/english/yearbook/tab/t0001007.shtml
http://www.ssb.no/www-open/english/yearbook/tab/t0001008.shtml
http://www.ssb.no/english/search/
http://www.ssb.no/www-open/english/yearbook/tab/t0001002.shtml
Portugal
http://www.infocid.pt/Infocid/0_92.htm
Spain
http://www.eleccions99.org/index.html
http://www.dipusevilla.es/frames.htm
http://www.iea.junta-andalucia.es/sima_web/
http://www.iea.junta-andalucia.es/dtbas/dtb97/ic7.htm#ic7-1
http://www.iea.junta-andalucia.es/
http://www.map.es/internet/ccaa.htm#Galicia
http://www.istac.rcanaria.es/enlace.html
http://www.ine.es/espcif/espcifin/elec98in.pdf
http://www.parlamento-cantabria.es/
http://www.parcan.rcanaria.es/elecciones99/index.htm
http://www.larioja.org/elecciones/pag2.htm
http://www.comadrid.es/cmadrid/index.htm
http://www1.euskadi.net/Emaitzak/datuak/indice_c.asp
United States
http://ElectionResources.org/america.html#US
The Office of the Deputy Prime Minister is not responsible for the contents or reliability of the linked web sites and does not necessarily endorse the views expressed within them. Listing should not be taken as endorsement of any kind. We cannot guarantee that these links will work all of the time and we have no control over the availability of the linked pages.
Published 3 May 2000
Return to Local and Regional Government Index
Return to Homepage
Web site terms