The relationship between the UK’s ageing population and Labour’s decline has received insufficient attention. Focusing on age shows the electoral terrain has transformed since Labour last dominated and helps explain the party’s descent.
In its 2001 landslide, Labour won popular majorities across almost all age groups. However, as general elections aren’t popular votes, this does little to illuminate age’s electoral role.
More relevant is how constituencies of certain age profiles voted. Of the 573 constituencies in England and Wales (seats in Northern Ireland and Scotland aren’t considered), 491 existed in 2001. Accordingly, grouping these by their median ages in 2002 (the earliest available data), reveals that in 2001 Labour possessed, from the perspective of age, an electoral edge.
Fifty-one per cent of these seats had median ages of 35-39, while another 15 per cent had median ages of
The comparative lack of older constituencies also helped. Though Labour beat the Conservatives by 17 points among voters aged 35-44, the Conservatives won 51 per cent of the 152 seats with median ages of 40-44, while Labour won only 34 per cent. And despite Labour’s popular advantage, the Conservatives averaged 6 per cent greater vote share across this range.
Similarly, that only 13 seats had median ages of 45-49 was electoral gold. Though Labour beat the Conservatives by 9 points among voters aged 45-54, the Conservatives won 10 of those seats, and averaged a vote share 31 points greater than Labour, which averaged just 15.93 per cent, and won no seats. In 2001, only one seat occupied the 50+ range, which the Conservatives won by 40 points.
Thus, though in 2001 Labour beat the Conservatives by 246 seats and with voters of almost all age groups, important patterns are discernible. Labour swept younger seats, but the Conservatives increasingly dominated past a median age of 40. Crucially, young far outnumbered older seats, handing Labour an advantage.
Since 2001, however, the UK has aged. Though, nationally, the median age has risen by under two years, ageing within constituencies has been far more pronounced. The median age has risen by at least two years in 68 per cent of constituencies in England and Wales. The median change is an increase of 2.9 years and the median increase 3.3 years, while only 12 per cent of seats’ median ages have dropped or stagnated. Simply, constituencies have aged considerably: over a third have seen their median age rise by at least four years, and a fifth by at least five. Consequently, constituencies in England and Wales are collectively much older.
Whereas, in 2002, 49 per cent of today’s constituencies had median ages of 35-39, by 2019, the proportion was just 24 per cent. Crucially, progressively older constituencies have supplied the shortfall. Between 2002 and 2010, the share of constituencies with median ages of 40-44 rose by over a third to 44 per cent, while the 35-39 category nearly halved in size. Similarly, the 40-44 range has since fallen to 33 per cent, and the 45-49 range has supplemented the drop, going from containing 2 per cent of seats in 2002 to 24 per cent in 2019: a remarkable increase.
Accordingly, the electoral landscape has transformed. In 2002, 269 more of the current seats in England and Wales occupied the 35-39 than the 45-49 range. By 2019, the margin was just three. Moreover, the number of seats with median ages of 50+ has risen from two in 2002 to 28 and will only grow as the population ages.
Resultantly, a substantial majority of seats now contain older electorates, a total reverse from 2001. In 2002, roughly two-thirds of constituencies in England and Wales had median ages of 39, and almost a third had median ages of >44.
But though the electoral terrain has transformed, the voting patterns from 2001 have, crucially, largely held true. It is therefore unsurprising that, as the proportions of older and younger constituencies have flipped, so too have parties’ electoral fortunes.
Of the five age ranges, four have voted extremely consistently since 2001. Seats with median ages of
Contrastingly, the Conservatives have dominated the three oldest groups. In 2001, 2005, and 2010, the Conservatives won 77 per cent, 73 per cent, and 74 per cent of seats in the 45-49 range. However, the Liberal Democrats captured the remainder: Labour won 0 per cent, 0 per cent, and 4 per cent.
Resultantly, the Lib Dems’ collapse in 2015 let the Conservatives sweep 87 per cent, 84 per cent, and 93 per cent of these seats in the past three elections, while Labour has remained weak, securing 10 per cent, 12 per cent, and 5 per cent. That in 2015 these seats roughly doubled in number from 2010 made this shift even more salient.
The 50+ range is also deeply blue. Since, in 2015, becoming a statistically significant grouping, the Conservatives have taken 75 per cent, 85 per cent, and 89 per cent. Labour has won only two such seats since 2001; one apiece in 2017 and 2019.
Constituencies in the 40-44 range voted with remarkable consistency across the five elections to 2017. The Conservatives captured 51-64 per cent of these seats and Labour 31-40 per cent. 2019 saw each party win anomalous proportions, but the Conservatives’ superiority continued.
Rather, only the 35-39 range has meaningfully changed behaviour since 2001. In 2001 and 2005, Labour won 73 per cent and 68 per cent. In 2010, however, the Conservatives became competitive, beating Labour by 8 points. In fact, the Conservatives have won 49-51 per cent of these seats in three of the last four elections. And though the 22-point loss in 2017 was comfortable, the margin, compared with 2001 and 2005, displays newfound staying power, attributable to the party’s overwhelming popularity among older voters.
Thus, voting trends have changed little since 2001. These trends, however, now unfold against a much-aged electoral backdrop. Whereas, in 2001, almost two-thirds of seats inhabited the younger, Labour-leaning, age ranges, only 38 per cent now do, and those largely occupy the 35-39 range.
Contrastingly, 62 per cent of seats, almost double than in 2001, now populate the Conservative-leaning age ranges, with roughly 30 per cent alone occupying the two oldest, staunchly Conservative, categories: a colossal increase on the 2001 figure of 2 per cent. Accordingly, this shift has transformed the electoral landscape, and made it far harder for Labour to win general elections.
For instance, if Labour had in 2019 won in England and Wales the same percentage of seats from each age range that it did in 2001, it would have secured only 247 seats, rather than the 357 it captured in 2001. By the same metric, the Conservatives, despite losing in England and Wales by 192 seats in 2001, would in 2019 have practically matched Labour with 244 seats.
In fact, if Labour had in 2019 equalled its best performance in each age range since 2001, it would have still been around 50 seats, or roughly all seats in Scotland, off a majority. Contrastingly, if the Conservatives had matched their worst performance in each age range since 2001, they would have won only about 30 seats fewer than that. Our new electoral reality can hardly be made any clearer.
This reality, though, is even worse for Labour than it already appears. First, Labour often struggles to compete in, never mind win, older seats. In 2019, Labour secured average vote shares of just 32 per cent, 24 per cent, and 18 per cent across, in ascending order, the oldest three age ranges. The Conservatives averaged shares 17, 32, and 40 points higher. And even in 2017, a more successful year, Labour still struggled, securing average shares 7, 24, and 31 points lower than the Conservatives. Unsurprisingly, the younger the seat, the more votes Labour generally wins, but the older it gets, the better the Conservatives typically fare.
Labour has simply long struggled to attract widespread support in older seats. The Conservatives, contrastingly, are increasingly popular.
Second, the amount a constituency’s median age changes correlates closely with voting behaviour. Rises of over two years are increasingly predictive of electing a Conservative, whilst increases of under two years, stagnation, or decreases are progressively predictive of favouring Labour. This is because really only the youngest seats have had the latter experiences, and almost exclusively older seats have aged considerably.
Accordingly, whereas many younger seats have aged, and become likelier to vote Conservative, really only young, Labour-leaning, constituencies have become younger or experienced little change. Resultantly, Labour dominates a small, shrinking pool of seats and struggles with the remainder.
This also invokes a geographic point. Practically all seats which in 2019 had median ages of
This helps explain the accusation Labour has become an urban party. Urban areas, after all, are younger, and thus contain young, Labour-leaning, constituencies. The collapse of Labour’s ‘Red Wall’ also looks explicable in this context: most ‘bricks’ lost in 2019 had median ages of >39 and none populated the
Labour’s weakness across older seats has thus left it struggling to win seats outside cities and younger suburbs, and vulnerable to the prospect of becoming, alongside the charge it is, an urban party.
Third, the ONS estimates that, by 2039, the proportion of over-64s will grow 5 per cent and the shares of those aged 4.5 per cent. Accordingly, the trends discussed will only become more extreme. In such a reality, Labour’s electoral hopes look slim-to-none.
The message is clear. To remain competitive, Labour must adapt to the UK’s new demographic character and focus on securing older seats. As alien and daunting as this may seem, demography leaves the party little choice.
Yoram Goodman is a postgraduate law student with a BSc in politics and international relations from the University of Bristol and an MSc in political theory from LSE