Skip to content
Skip to main content

About this free course

Download this course

Share this free course

Contemporary Wales
Contemporary Wales

Start this free course now. Just create an account and sign in. Enrol and complete the course for a free statement of participation or digital badge if available.

8.1.3 New Labour and the Welsh devolution settlement

New Labour, under the leadership of Tony Blair, promoted devolution on the grounds that ‘it will bring government closer to the people, make our politics more inclusive and put power in the hands of the people where it belongs’ (quoted in Chaney and Fevre, 2001, pp. 22–3; emphasis added). These ambitious plans to create a new inclusive politics were put to the Welsh electorate in a referendum on 18 September 1997.

Figure 17 How the vote divided Wales

A week previously, Scottish voters had voted to establish a Scottish parliament, with 74.3 per cent in favour compared to 25.7 per cent against. In Wales, only half of the Welsh electorate – 50.1 per cent to be precise – turned out to vote in the referendum. The result could not have been closer: asked whether they agreed that there should be a Welsh Assembly or not, 50.3 per cent of voters said ‘Yes’, with 49.7 per cent saying No. Figure 17 shows how support and opposition to devolution was geographically distributed across Wales; while the predominantly Welsh-speaking rural areas of north-west and south-west Wales voted in favour of devolution, the more anglicised industrial communities of north-east and south-east Wales voted to reject these proposals.

So by a mere 6,721 votes, the Welsh electorate said ‘Yes’ to devolution. This was hardly a ringing endorsement of the devolution project. The closeness of the result could be seen as a sign that the new National Assembly lacked legitimacy, given that less than half of Welsh voters voted at all, and that half of those that did didn’t want such a body to be established. This is ironic, given that devolution was meant to resolve a perceived lack of legitimacy in Welsh politics.

Activity 24

Despite the closeness of the referendum result in 1997, it was nevertheless a result which showed that attitudes towards devolution had changed in important ways since the 1979 referendum. As you read the following, note five important factors that help explain why such a change in opinion had taken place.

Extract 15 Why was 1997 different?

The timing of the Welsh referendum of 1997 could not have been better in terms of securing a ‘yes’ vote. The new Labour government was still enjoying its honeymoon period, little opportunity had existed for left-wing discontent to grow, and the Scots had already a week earlier voted resoundingly for the establishment of a parliament in Edinburgh. Furthermore, the speedy pre-legislative referendum ensured that there was little time for the deficiencies of the government’s devolution proposals to be examined. The ‘no’ campaign was a damp squib. In contrast, the political context of the referendum in 1979 was hostile indeed for the then Labour government. It had lost its majority during 1976 and was reliant on the Liberals and other smaller parties to ensure success for its legislative programme. The referendum took place shortly after the ‘winter of discontent’ and amidst the resurgence of the Conservative Party under Margaret Thatcher ... These different political contexts in 1979 and 1997 can be seen to have influenced electors’ behaviour; to have changed the pattern of support and opposition to devolution; and in the final instance, to have undermined the turnout among ‘no’ voters sufficiently for a ‘yes’ result to have crept in under the wire ...

This is not to deny that, although the increase in the proportion of the population of Wales affirming a Welsh identity between 1979 and 1997 was small, Welsh national identity increased in political salience ... [T] wo notable developments have occurred. People with a Welsh national identity have become more pro-devolution. And Plaid Cymru, the nationalist party, has become markedly more acceptable to the mass of the population in Wales ... As there was no marked social change that might account for why a Welsh identity became more politically salient, it is likely to be a political creation. Perhaps the Labour government should thank Plaid Cymru for its work in this area, for without the politicisation of Welsh identity, the swing from 1979 to 1997 would not have been enough.

Evans and Trystan, 1999, pp. 113–14


Some of the factors that you might have identified as having contributed to a ‘Yes’ vote in the referendum are:

  • the popularity of the newly elected Labour government
  • the ‘yes’ vote in Scotland a week previously
  • the speed at which the legislation for holding a referendum was passed through the Houses of Parliament
  • a badly organised ‘no’ campaign
  • the increased salience of Welsh identity.

The result of the referendum led to the creation of a National Assembly for Wales. According to the Government of Wales Act 1998, this new body would be characterised as follows:

  • Sixty Assembly Members (AMs), elected by an alternative member system (see Box 1).
  • The National Assembly would take over the powers previously exercised by the Secretary of State for Wales and the Welsh Office. These areas of responsibility include agriculture; culture; economic development; education and training; the environment; health; sport; economic development; education and student loans; the environment; health; local government and housing; sport; social services; transport and the Welsh language.
  • In these policy areas, the National Assembly would have secondary legislative powers. This meant that all laws (primary legislation) would still be made in Westminster; however, the National Assembly would be able to specify rules and regulations that adapt the legislation to the specific Welsh context.
  • The National Assembly was designed as a ‘corporate’ body; this meant that, in contrast to the distinction made in most political systems between the executive (the government) and the legislature (the parliament), the National Assembly as a whole would be responsible for proposing, passing and scrutinising policy decisions. This model of governance was designed to promote consensus and cooperation between political parties.

The first National Assembly elections were held on 6 May 1999. Table 3 provides the results from these and subsequent elections in 2003, 2007 and 2011. The difference between 1st and 2nd votes in these elections is explained in Box 1.

Box 1 The basics: first-past-the-post versus proportional representation electoral systems

In the United Kingdom, representatives to the House of Commons are elected using the ‘first-past-the-post’ (FPTP) electoral system. In each constituency, voters are presented with a list of candidates representing different political parties. The voter votes for the candidate he or she prefers, and the candidate with the most votes in the constituency wins. The ‘winner takes all’ in FPTP elections; there’s no reward for candidates that come in second or third place. So, for example, if Candidate A wins 40 per cent of the vote, and Candidate B wins 39 per cent of the vote, Candidate A will be elected because he/she has won most of the votes. FPTP electoral systems are also used in Canada, the USA and India.

‘Proportional representation’ (PR) electoral systems are different because they try to achieve as close a match as possible between a party’s share of the vote and its share of the parliamentary seats. Let’s imagine that a PR system was used to elect representatives to the House of Commons. In principle, if party A wins 40 per cent of all votes cast in the United Kingdom, it will have 40 per cent of the representatives in the House of Commons; Party B, which wins 30 per cent of all votes cast, will get 30 per cent of seats, and so on. So we can say that a party’s number of representatives is proportional to the number of votes it has. Examples of PR systems can be found in Germany and Australia (although several different forms of PR systems exist).

The electoral system used for electing representatives to the National Assembly for Wales can best be described as being semi-proportional, since it combines elements of both the FPTP and PR systems. Known as the ‘additional member system’, 40 of the National Assembly’s members are elected using FPTP, while the remaining 20 are elected using PR. This means that in National Assembly elections each voter will have two votes. The first vote is cast for a single candidate representing a specific constituency, in the same way that representatives are elected to the House of Commons (by putting an X next to the name of the candidate that he/she prefers on the ballot sheet). The second vote is used to elect the additional members that represent one of five regions within Wales. Instead of voting for an individual candidate, the vote will be cast for a political party; these additional seats will be allocated in such a way that corrects for any unfairness in the allocation of the FPTP constituency seats (there is a very complex formula that calculates how many alternative seats each political parties should be allocated!). Alternative member systems have also been used widely around the world, including in New Zealand.

Table 3 National Assembly for Wales election results, 1999–2011
Party1st vote (%)2nd vote (%)Total seats1st vote (%)2nd vote (%)Total seats1st vote (%)2nd vote (%)Total seats1st vote (%)2nd vote (%)Total seats
Liberal Democrat13.512.5614.112.7614.811.7610.685
Plaid Cymru28.430.61721.219.71222.421.01519.317.911


Wyn Jones and Scully (2004), p. 194; Scully and Elias (2008), p. 105; updated with 2011 results
Jeff Morgan/Alamy
Figure 18 Inside the National Assembly for Wales

It was always expected by prominent Labour party politicians that devolution to Wales would be a ‘process not an event’ (Davies, 1999), and that there was scope for change and adaptation over the years. However, the need for change and adaptation arose much sooner than had originally been anticipated. It didn’t take long for the drawbacks of this constitutional settlement to become apparent: within a year of its creation, there were calls for the National Assembly’s institutional set-up and powers to be revisited. The Richard Commission, chaired by the Labour peer Lord Ivor Richard, began its deliberations in 2002 and gathered extensive evidence on how different groups and interests perceived the National Assembly to be working. The Commission’s final report, delivered in March 2004, proposed a radical revision of the original devolution settlement. Some of these recommendations were included in a new Government of Wales Act 2006, which contained the following key provisions:

  • the abandonment of the idea of the ‘corporate body’, with a clearer separation being made instead between the role of the executive (i.e. the Government, responsible for proposing and implementing policy) and the legislature (the Assembly responsible for scrutinising the activities of the Government)
  • new powers for the Assembly to request the right for primary legislation to be delegated from Westminster.

Further debate about the future of devolution in Wales was initiated with the creation of an All Wales Convention in 2008 to consider the prospect of further expanding the powers of the National Assembly. The setting-up of such a convention was one of the commitments made by Labour and Plaid Cymru in June 2007 when they agreed to enter into government together. One of the main objectives of the All Wales Convention was to assess the extent of public support for moving towards full law-making powers for the National Assembly. This would give the institution primary legislative powers in the full range of devolved policy areas, although a new referendum would be required in order for this to happen. As part of its work, the All Wales Convention held a series of public meetings across Wales in order to gather the views of the Welsh public on the issue of further powers for the Welsh Assembly. The Convention’s recommendations were presented to the Welsh Assembly in November 2009. The evidence gathered suggested that if a referendum on further powers were to be held, a Yes vote was deemed to be possible although not guaranteed (All Wales Convention, 2009).

The All Wales Convention
Figure 19 A Question Time style event organised by the All Wales Convention, Cardiff 2009