Background Paper – preliminary and draft, comments welcome
In each one of the UK general elections that took place from the 1960s up until and including in the 1992 general election (when the Labour party under Tony Blair won a landslide victory), around 66% of young people have turned out to vote.
But, since 1992, youth turnout rates at UK general elections has been in sharp and steady decline. More precisely (and as illustrated in Figure 1 below), in the subsequent, 1997 general election, the youth turnout rate dropped by ten percent points to 56%; and then it dropped yet further by another fifteen percent points in the 2001 general election to 41%; and then it dropped yet again by another 3 percent points in the 2005 general election to 38%.
In short, over the thirteen year period from 1992 till 2005, youth turnout at UK general elections sharply and steadily declined by twenty eight percentage points, from 66% in 1992 (and before) reaching down to 38% in 2005. It increased back to a 49% rate in the 2010 election but this is still down by seventeen percentage points on turnout rates over the 1964-1992 period.
Over the thirteen year period from 1992 till 2005, youth turnout at UK general elections sharply and steadily declined by twenty eight percentage points, from 66% in 1992 (and before) reaching down to 38% in 2005
These dramatic facts beg the following fundamental question: Why has this happened? More precisely, what are the main factors that explain the steep (discontinuous) and sudden drop in the number of young people voting in the 1997 election (after a stable and reasonably high turnout rate for over 30 years), and the continued and further reductions in the 2001 and 2005 elections? Has something fundamental changed to explain this new state of affairs?
The above stated questions become even more pertinent and puzzling when, as illustrated in Figure 1 below, this has not been the case for the other, older age groups. For such groups, especially the over 55 year olds, the turnout rate at all these general elections has been around 75% with the exception in the 2001 general election when it dropped to 60%. But other than that one-off drop, it has remained stable at around 75%. So, from, and including, the 1997 election, there has been, over the intervening 15 years, a stable and robust gap of around twenty five percentage points between the turnout rates of the young and the old . Why has this been case? What explains this “political” (or “turnout”) inequality between the young and the old? Will it have implications for economic inequality?
In this briefing note, we summarize answers to the questions posed above, informed by the vast literature, academic and non-academic, on what has become today a burning issue of what appears to be a serious and persistent disengagement by the youth (and youth alone) with electoral politics, as we gear up to the 2015 UK general election.
Weighted Average Turnout of 18-34 year olds and older age groups
Before proceeding further, it is helpful to put the above facts in relation to turnout rates in other countries and ask whether this turnout inequality in the UK is also present elsewhere. As can be seen in Figure 2 below, it turns out that the UK’s “generation gap” does appear to be particularly bad versus other countries. So, perhaps surprisingly, turnout inequality appears to be UK phenomenon.
Weighted Average Turnout of 18-34 year olds and older age groups
To Vote or not to Vote? The Main Factors
Why people vote is one of the most studied questions in social science. As one of the last pieces of paperwork to go online, it’s an unusually inconvenient thing to do. It causes material disruption to a person’s day. Nobody in their social circle need know they did it, or who they voted for. And the chances of their own vote “making a difference” are vanishingly small.
It’s a question that’s both mysterious and important to powerful people as well as interested analysts, so a vast amount has been written about it. But in answer to this question of “Why did someone vote?” most of the claims can be brigaded into two categories:
- They made a choice to vote: they were decided by the costs and benefits
- They did it without thinking: it’s a habit, or some kind of a psychological reflex.
There is a lot of evidence to support both types of claim. They aren’t mutually exclusive: both can be true of the same population in different measure for different groups; both claims could even be true of the same person at different elections [and even same person at same election].
Here, we are concerned with the young and so we focus purely on why they made a particular choice on Election Day. The concept of a ‘voting habit’ is not very useful in explaining why a young person doesn’t vote and it is useless in explaining first time voting, so it’s reasonable to simplify the picture for young people. We need to understand what changed about young people’s perceptions of costs and benefits of voting from the 1990s.
Figure 3 below shows factors analysed by social scientists, with the size of the ‘weights’ roughly corresponding to what the literature has to say about their importance.
Factors that determine the decision whether or not to vote
It surprises many people to learn that the single most important thing in explaining why people vote in recent times is how they feel about democracy itself, and not – as is commonly assumed – their assessment of what they will get out of a particular party winning the election. But the fact is, many factors have material effects on the decision to vote or not vote, which is what makes it difficult to get to the bottom of what has been going on in the UK over the past two decades. The rest of this section explains more about these factors, and what is claimed about how important they are:
“It’s good to vote”
Some people will decide that voting brings them benefits, regardless of which way the election goes.
|“People will approve.” If a potential voter is surrounded by a social group (peers, colleagues, family, etc) that thinks it’s important to vote, that person is more likely to vote [Social Preference; Social Capital]||Moderate|
|“Voting makes me feel good.” A potential voter might feel a strong allegiance to a particular party or be inspired by a particular politician and enjoy expressing their support for them. They may just enjoy getting out of the house to “take part” in a civic activity. [Civic Duty]||Moderate-High|
|“I love democracy.” A potential voter is more likely to vote if they believe they derive benefits from living in a democracy and that it is important to support that system by voting. [Social Norm]||Very High|
“I might get what I want”
Some people will decide that their vote might be able to help bring about an outcome they want, and therefore they vote. This assessment is influenced by how the parties, politicians, government and closeness of the election are perceived.
|“Government is competent.” If a citizen tends to think that the system of government can deliver what is promised, they are more likely to vote||
Small but significant
|“The politicians are competent.” Otherwise known as a “valence assessment”, assessments of competence on universally desired outcomes (like a strong economy) are important to which parties people will choose to vote for on the day. If they take a dim view of the competence of all party leaders, they are less likely to vote. [Valence]||Moderate-High|
|“My vote will matter.” Landslide elections are, all else being equal, low turnout elections. A potential voter will tend to vote if the race looks close. [ Competitiveness of Election]||Moderate|
|“I see clear blue water.” If a potential voter sees a potential gap in stated policy positions that matters to them (or to people they care about), they are more likely to vote. This in turn is dependent on the actual policy positions of the party, how those are transmitted by the parties, and how they are filtered by the media and other channels. If a potential voter has strong identification with a particular party, that will also heighten their perception of the ‘difference’ from an outcome. [ Policy Differentiation]||Moderate-High|
“It’s a hassle to vote”
As we know, many millions of people do stay home on Election Day. So, unsurprisingly, it has be found that the perceived costs of voting are important factors in explaining why people do and don’t vote. There is first the time and effort cost of registering. Having incurred that cost, there is then time and effort cost of going to vote to the primary school or church hall and queuing up (or else organizing someone else to do so on one’s behalf).
The factors identified above could be called “proximate causes” in the decision to vote or not. But beneath them will lie a complex web of “drivers” that will determine how the above factors will apply to any one young person.
One of the most important drivers is the level of education: The more educated an individual is, the higher is the likelihood that they will vote. Figure 4 below illustrates this point quite starkly. It is interesting to note that over 90 per cent of older people with a degree or higher qualifications report voting in the 2010 general election, compared with just 44 per cent of young people with only GCSE or lower qualifications. The best educated older people are more than twice as likely to vote as less educated young people.
The percentage of different groups who reported voting in the 2010 general election
There are other key drivers but the three other main ones are:
- Level of affluence (income and wealth ): The more affluent the individual, the higher is the likelihood that they will vote.
- Political Knowledge: The more the individual knows about politics, the politicians/parties and their platforms, the higher is the likelihood they will vote .
- Engagement in Electoral Politics : The more engaged the individual is with elections and electoral politics, the higher the likelihood they will vote.
Explaining Decline in Youth Turnout
In short, nobody really knows why the young voted so much less during the 1990s and 2000s.
Part of the challenge is that the factors typically analysed, as listed above, don’t explain voting amongst young people as well as they explain voting amongst older people: even once you take all these factors into account, age still matters. That is, an 18 year old that looks similar in almost all the significant respects listed above to a 60 year old (similar beliefs about democracy, parties, and so on) is still about 25% less likely to vote. This means the set of factors listed above is missing something that is important to explaining voting behaviour amongst the young.
Another challenge is that it may be the case that the young and the old have been affected by a lot of the same factors, and their assessments about politics and government have changed in similar ways. But that the old formed a ‘voting habit’ at a time when the cost-benefit analysis of voting was more favourable to voting, and so now their decision to vote is “clouded” by habit. Nobody has analysed how much of the generation gap in voting can be explained by looking at the “habits” of older adults, instead of the calculations of younger adults.
Finally, there is only variable quality of evidence available for how each of these factors has changed, as particularly applied to young people, over time.
We believe we need more research that in particular addresses the following questions:
- How much of the generation gap in voting can be explained by the “voting habit” of the old?
- What are the “missing variables” that particularly affect the young that the typical voting equations are missing?
- And, for those voting drivers that we do measure and analyse, are they worse for the younger anyway? For instance, why do the young believe it’s less important to vote to support democracy?
- Is the trend we observe in the UK seen in other countries? Why, or why not?
Please send any comments to Professor Abhinay Muthoo at A.Muthoo@warwick.ac.uk