The ‘people’s vote’ campaign for a further referendum on our relationship with the EU has been gathering pace. Around 100,000 people joined a demonstration in London on the second anniversary of the 2016 Brexit vote. Elite support has also begun to grow, with former senior cabinet minister Justine Greening one of the latest to come out in favour.
Theresa May’s government remains resolutely opposed, and at the time of writing this appears highly unlikely to change. However, if the government does ultimately find itself holding a referendum that it had initially strongly resisted it would not be without precedent.
The 1979 referendums
The 1979 referendums on Scottish and Welsh assemblies are still two of the most significant – and controversial – referendums ever held in the UK. The failure of the devolution proposals saw the SNP withdraw support for the government, creating the conditions for the parliamentary vote of no confidence in the Callaghan government four weeks later. Meanwhile, the nature of the Scottish defeat – a 52% Yes vote being insufficient, due to a threshold requirement for 40% of the electorate to vote in favour – proved highly contentious for years afterwards.
The 1979 votes also set an important precedent that referendums should be held on major changes to the UK’s territorial constitution. Yet this was by accident rather than design, for it was not originally the Labour government’s intention to hold referendums on devolution at all.
In May 1976 Michael Foot told the House of Commons that the government had decided against putting their proposals to public votes, arguing that the 1974 general elections had offered sufficient opportunity for the public to have their say. The new Prime Minister, James Callaghan, strongly supported this position. He told Foot that he was ‘very much opposed to referenda in principle’, citing ‘the damage to Parliamentary sovereignty’.
Various schemes for referendums had been advanced from at least November 1975, albeit few resembled the eventual post-legislative Yes/No votes. Norman Buchan, a pro-devolution Scottish Labour MP, argued for a pre-legislative referendum offering a choice between devolution and independence. Edward Heath came out for a similar scheme, in expectation that it would allow Scots to express a clear view in favour of remaining in the Union. The Guardian, meanwhile, advocated a five-option referendum to ‘make sense of it all’, at which voters would have been able to choose between independence, three alternative models of devolution and the status quo. Proponents of referendums also disagreed about who should be able to vote – some MPs such as Eric Heffer thought English voters should be included.
The government’s hand was ultimately forced by an amendment tabled by Welsh MP Leo Abse to second reading of the devolution legislation, that ‘this House cannot approve the Scotland and Wales Bill unless a referendum is held before the Bill comes into effect to ascertain the wishes of the people of Scotland and Wales’. By the end of the second reading debate this had received over 150 signatures, including a range of Labour MPs extending well beyond the most vocal anti-devolutionists. A concession that referendums would be held after all became inevitable, and was announced on the last day of the debate.
It was Tam Dalyell, the most prominent anti-devolution Labour MP, who proposed to colleagues amendments establishing turnout thresholds. Bruce Douglas-Mann and George Cunningham duly tabled amendments specifying one-third and 40% respectively. Cunningham’s threshold, which he thought ‘very modest’, was agreed by 166 to 151. The fateful Cunningham amendment was supported by 35 Labour rebels, all but six representing English constituencies. Attempts were made to overturn the amendment, but to no avail. Its legacy would haunt Scottish politics for a generation, until the 1997 referendum at which legislative devolution was supported by a large majority (with no threshold requirement in place).
Lessons for today
Returning to the present day, there are several relevant insights from the events leading up to the 1979 referendums and their aftermath for the possibility of a ‘people’s vote’. The first is that especially during periods of minority government – as the Labour government was from 1976 – the House of Commons has the ability to seriously frustrate the government’s ambitions. Just two months before the Scotland and Wales Bill had its second reading Lord Hailsham delivered his much-quoted lecture characterising the UK political system as an ‘elective dictatorship’, but this was an exaggeration then and is even less true now. In the event that a parliamentary majority for a ‘people’s vote’ emerges, as it did around holding referendums in 1976, there is every chance it could force the government’s hand in a similar way. It is, for example, possible to envisage an amendment similar to Abse’s 42 years ago being tabled to second reading of the legislation required to implement a Brexit deal.
Past experience hence gives us little reason to believe that a new ‘people’s vote’ on our relationship with the EU would finally settle the issue after the 1975 and 2016 referendums were unable to.
A further insight concerns the importance of carefully thinking through the detailed rules for any referendum – a point also emphasised in the recent report of the Constitution Unit’s Independent Commission on Referendums. As was the case with supporters of referendums on devolution in 1975-76, proponents of a ‘people’s vote’ have no agreed formula for exactly how such a poll would be structured. There have even been suggestions that any future referendum should be subject to thresholds, as in 1979. The experience then suggests that mechanisms designed to help promote a particular outcome can have unintended consequences. While Cunningham’s threshold sealed victory for the anti-devolutionists in 1979, it engendered resentment that contributed to the revitalisation of the pro-devolution campaign during the 1980s and 1990s. For this reason Dalyell came to regret supporting the amendment, writing in 2011 that without it he believes there would have been a straight No vote, following which ‘the issue of a Scottish Assembly might have been put to bed for a generation’.
Although Dalyell is probably right that there would have been a straight No vote in 1979 but for the threshold, it is surely unlikely this would in fact have put the devolution issue to bed. For a final key lesson from the 1979 devolution referendums – and indeed from the UK’s experience of referendums more generally – is that referendums are rarely able to resolve an issue decisively. Even in Wales, where 80% voted No in 1979, devolution was back on the agenda before long, becoming Labour policy once again by 1992. Past experience hence gives us little reason to believe that a new ‘people’s vote’ on our relationship with the EU would finally settle the issue after the 1975 and 2016 referendums were unable to.
This article was originally published on the Bennett Institute for Public Policy blog. It draws on research for the author’s MA dissertation on ‘The Callaghan government, parliament and devolution, 1976–78’, submitted in 2016, including archival source material.