If a man is not a socialist by the time he is 20, he has no heart. If he is not a conservative by the time he is 40, he has no brain.
At first glance, this quote, attributed to Winston Churchill, appears to fit the evidence in Britain. A survey conducted during the 2019 general election reported in our recent book showed that 23% of respondents under the age of 30 voted Conservative and 55% voted Labour. In contrast, 59% of the over 65s voted Conservative and only 13% voted Labour.
However, a comprehensive analysis of the relationship between age and voting in Britain over a period of 55 years from 1964 to 2019 shows that growing older does not directly affect support for the Conservatives. At the same time, it does appear to influence Labour voting. As voters get older they are a bit more likely to support Labour but not the Conservatives.
The paper which presents these findings is part of a special issue of the journal Electoral Studies, being published in memory of our late colleague, political scientist Harold Clarke, who edited the journal for many years.
How is this rather counter-intuitive finding explained? The answer is that the relationship between age and voting is more complex than many people think. Political behaviour can certainly be affected by life-cycle effects – that is, people changing their politics as they grow older. But this is not the whole story. There are two additional aspects of age-related voting which need to be considered.
The first is what is described as a period effect. This refers to the fact that specific election campaigns can influence age-related voting.
For example, the 2019 election took place after three years of political turmoil following the referendum on UK membership of the European Union. In the event, Boris Johnson’s slogan “Get Brexit Done” proved very effective and the Conservatives won an 80-seat majority.
This was very different from the 2017 election in which a barnstorming campaign by the newly elected Labour leader Jeremy Corbyn, and a stumbling campaign by prime minister Theresa May resulted in a Conservative minority government. These campaign-related differences can affect the relationship between age and voting independently of life-cycle effects.
The second factor which needs to be considered are cohort effects. These arise from the fact that each new generation has different socialisation experiences which influence their political beliefs and voting behaviour.
For the most part children in their early teens do not pay much attention to politics. As they grow older, they become politically aware so that their attitudes and behaviour are formed in late adolescence and early adulthood. As this happens, they are influenced by the economic and political circumstances of the time.
This means, for example, that voters who came of age politically in the relatively affluent 1960s are likely to look at the world differently from those who came of age in the turbulent 2010s. This is different from a life-cycle effect because research shows that once acquired these attitudes and values remain relatively stable over time as people grow older, even when their social and economic circumstances change.
It turns out that separating life-cycle, period and cohort effects is a tricky exercise. But when it is done, it shows something rather surprising.
For Labour there are no discernible cohort effects but there are life-cycle and a few period effects. This means that to win elections, Labour needs to do well on the key issues such as the management of the economy and have a leader who is viewed positively by voters.
For Labour, these factors have a more powerful impact on voting than age. In addition, it helps the party to have a significant group of voters who identify themselves as loyal supporters. These measures change more rapidly over successive elections than age does, and so are more important.
For the Conservatives, there are very strong cohort effects and a small number of period effects – but no life-cycle effects. In other words, ageing alone does not account for people turning towards the party. It is more about cohort differences. And this has greatly weakened support for the party over time.
Local party infrastructure in decline
The Conservatives have traditionally relied on a strong cohort of voters who were socialised in their formative years by family, communities and social ties to be loyal supporters. They would generally support the party because they learned to do so when they became politically aware in their youth.
The problem is that this source of support has now greatly weakened, so that new cohorts, such as the one socialised during the austerity years following the 2010 election, cannot be counted on to identify with the party. In fact, they are very opposed to it.
There are a number of reasons why the socialisation mechanisms underlying Conservative support have declined in this way. It appears from the data that the era of Labour dominance between 1997 and 2010 weakened many of the normal processes of socialisation which the Conservatives relied on previously.
Far fewer people learned to support the Conservatives during this period. In that respect New Labour changed the political landscape.
A second factor is the decline in the Conservative party as a voluntary organisation in the community. In the 1950s the party had the largest grassroots membership of any party in Europe. A large, well-organised grassroots party with a significant youth movement is an excellent mechanism for socialising people into lifelong support for the party, but this has now disappeared.
With each election relying on short-term forces such as issues and voter evaluations of leaders, rather than longer-term forces anchored in family and community attachments, general elections in the future will become more volatile and unpredictable. And if the Conservatives lose the next general election on the scale of their 1997 defeat it may be a very long time before they can hope to win again.