By Hannu Lahtinen, Heikki Hiilamo and Hanna Wass
Voter turnout is – increasingly – a class issue. In the 2010 British parliamentary elections, the difference in reported turnout between the working and middle classes was 19 percentage points, compared to less than just five percentage points in 1964. Nordic countries, too, have been characterised by a relatively clear class bias in electoral participation. Along with many other socio-economic measures, patterns of participation and non-participation are passed down from one generation to the next.
Voting is habitual: The first few elections are crucial in developing the habit.
This means that differences in levels of political activity persist throughout an individual’s life. If a child never witnessed her parents going to the polls, she may well never do so either.
But our recent study suggests the picture might not be as gloomy as it seems. Intra-generational social and economic mobility seems to work to reduce the socio-economic bias in turnout. This effect is called “gradient constraint”, and was originally described in social epidemiology by Mel Bartley and Ian Plewis. Applied to electoral participation, the gradient-constraint hypothesis implies that turnout among upwardly-mobile individuals is higher than the non-mobile members of the class they came from – but still lower than the class where they end up. Among those whose class status has shifted downwards, the opposite applies.
Our results show that turnout among the mobile individuals settles between the averages of their original social class or income group and those among the class where they end up who did not experience social mobility during the period covered in our analysis. An individual’s current social class and income are linked to their voting habits in a way that cannot be fully attributed to education, past voting, parental voting or their previous social class or income. Instead, the results suggest that voters have a tendency to adjust their turnout patterns to match the standards of their new social standing. It is worth pointing out that they do not vote quite as often as the average of their new group. This seems only logical, given the importance of childhood and early adulthood in forming habits. Overall, however, when it comes to turnout, intra-generational social mobility does help to narrow socio-economic inequality.
In validating the gradient-constraint hypothesis, we used a unique dataset that links individual-level voting records from the 2012 Finnish municipal elections with information on socio-economic background characteristics compiled by Statistics Finland. On the bases of personal identification codes, voters’ social class and income in 2000 and 2011 were matched to voting in the 1999 Finnish parliamentary elections in the dataset. Using actual voting records means our data are not subject to the bias characteristic of self-reported turnout, such as misreporting due to faulty memory, or over-reporting due to social desirability.
Social class and income-related representational bias in parliamentary institutions is very important, since they are directly related to an individual’s position on policy issues like welfare benefits and tax rates. If the values and interests of non-voters differ from those of voters, citizens who cast their vote become better represented. Studies conducted in the US have shown that political decision-making favours wealthier citizens. Similar results were found in a cross-country comparison.
As intra-generational social mobility can go some way to neutralising voting inequalities, achieving an open society with a higher level of mobility may narrow the socioeconomic gradient in turnout. Policy makers could, for example, invest in adult education and in vocational training especially targeted at those with fewer qualifications and in less-skilled jobs. In addition, narrowing overall income inequality is likely to lead to more intra-generational income mobility, since the absolute “distance” between income groups becomes smaller and thus easier to overcome. With the gradient-constraint mechanism, this would reduce the gap in turnout between income groups.