The six key numbers of Welsh politics

After a decade and a half of democratic devolution, what sort of politics do we have here in Wales? To answer the question I’ve chosen six numbers that encapsulate some important, though not always very comfortable, truths about the manner in which democratic politics is conducted in Wales.

My first number is perhaps the most obvious, and should need no explanation to this audience. 6721. To jog the memory of anyone who needs reminding: the margin of the Yes victory in the 1997 referendum was 6,721 votes. Put another way, the margin between Yes and No was 0.6 per cent of the votes cast. Or to put it a third way, in a nation of 3 million people, fewer than 3,400 people changing their mind would have given victory to the No campaign.

That scenario – a No victory in 1997 – is increasingly difficult for many people to imagine. Just as university undergraduates now generally have no memory of politics before Tony Blair was Prime Minister, they also cannot recollect a time before the National Assembly. To them, the Assembly and devolution is simply part of the familiar furniture of political life. It has always been there. Indeed, to a population cohort whose numbers grow every day, the period prior to Democratic Devolution is no more part of their lived experience than the Tudors and Stuarts, Gladstone and Disraeli, food rationing, or that mysterious, mythical era before the iPhone.

But had the vote gone the other way, as it might easily have done, the political landscape of Wales would look very different. If, after the crushing humiliation of 1979, devolution had again been rejected by the Welsh people in the much more favourable circumstances of September 1997, that would almost certainly have been that: the end to the political project of winning significant autonomy for Wales.

It’s inherently impossible to know how that counter-factual would have played out. But it’s none-too-difficult to imagine London drawing the conclusion that the Welsh (or most of them, at least) didn’t really want political recognition of their nationhood, and thus winding down much of the apparatus of Administrative Devolution, and increasingly governing Wales as little more than another region of England (with perhaps a few, token nods to cultural distinctiveness). In this parallel universe, none of the broad political superstructure that develops around institutions like the Assembly would exist. Plaid Cymru, having witnessed the destruction of a key part of its raison d’etre, would probably have faced an existential crisis and might even have ceased to operate as a mainstream political party. One could go on – the consequences would range far and wide.

However that would have worked out, we can, I think, be clear on this. September 18^th 1997 was one of the most important and fateful days in the life of Wales as a nation. Whatever the inadequacies and limitations of the arrangements under which the National Assembly was created – and there were, assuredly, plenty of those – the 1997 referendum was what social scientists refer to as a Critical Juncture; or what more popular discussion sometimes terms a Tipping Point. By a vanishingly slender margin, Wales decided that it was going to be a political nation. And much has followed from that. It might very easily have decided otherwise. 6,721 votes mattered a great deal.

My second number, 67 per cent, requires some explanation. At the time of the 1997 Scottish referendum, devolution was frequently referred to – in words generally attributed to the late, lamented, John Smith – as the ‘Settled Will of the Scottish people’.

No-one could then seriously have described devolution as such in Wales. But the thing about Critical Junctures is that the path chosen has consequences, and it leads you in new directions. If you look at survey evidence gathered at the time of the referendum, or shortly afterwards, what comes through clearly is that support for democratic devolution in Wales was neither very wide nor very deep. And though few people challenged the validity of the referendum outcome, or the decision to proceed with creating a National Assembly, there were justifiable and quite widespread concerns about the public legitimacy the new body would possess.

One plausible ‘nightmare’ scenario’ was that the very existence of the Assembly, and the right of devolved Welsh institutions to wield significant governmental power, would continue to divide the people of Wales right down the middle; that the existence of devolution would be a running sore in Welsh politics.

Another ‘nightmare scenario’, at least for supporters of Welsh autonomy, was that unless the devolved institutions could be shown to deliver effectively, then the Assembly, lacking secure, majority public support, might be vulnerable to future abolition by a Conservative UK government (as had famously happened to the GLC and numerous other local authorities under Mrs Thatcher). Such fears clearly shaped the behaviour of some key actors in the early months and years of the Assembly: perhaps most obviously in Plaid Cymru’s contribution to the defenestration of Alun Michael, an action that, whatever wider merits it may have had, robbed Plaid of a substantial electoral asset…

Yet in practice these fears have largely proven groundless. After the devolutionist path was taken, an increasing proportion of the Welsh population came to accept and even support this. To be precise, the best evidence we have indicates that attitudes towards devolution and how Wales should be governed changed rather rapidly in the 5-6 years after the 1997 referendum. In the subsequent decade, the tide has continued to flow in the same direction, albeit at a much gentler speed. Opposition to devolution has come to be distinctly a minority taste – in most surveys in recent years, favoured by fewer than 20 per cent of respondents.

Yet there has been no rise in support for independence, which continues to be the preference of around about 10 per cent of voters. It is support for something in between the status quo ante and independence for Wales – devolution, Home Rule, or whatever you label it – which rose rapidly in the years after 1997, and has been the preference of the clear majority of people in Wales in all major recent surveys. In the most recent survey which we conducted, those favouring devolved government for Wales numbered 67 per cent of all respondents – even after including all the Don’t Knows and Wouldn’t Says. Devolution is now securely established as the settled will of the Welsh people, to an extent that was very difficult to imagine fifteen years ago.

Quite how that has come about is something that I’m still not sure we have a wholly satisfactory explanation for. One popular line in the period after the 1997 referendum was that support for devolution would grow as it was seen to ‘deliver for the people of Wales’. It’s none-too-clear that this holds much water. Public perceptions of the performance of devolved government have, in the main, been pretty moderate; moreover, the statistical relationship between evaluations of the impact of the devolved institutions and how people say they want Wales to be governed is generally pretty weak.

A more recent suggestion – voiced by Peter Hain in the run-up to last years’ referendum – was that support for devolution had increased in reaction to a Conservative-led government being installed in London. Given the history of Wales’ partisan preferences – something I’ll come back to – that is a very plausible idea. It’s just one which the evidence, rather inconsiderately, completely fails to support. The big changes in attitudes towards devolution happened well before Mr Cameron became Prime Minister; and, thus far, there is very little sign that the spectre of the Tories (or, indeed, of Nick Clegg) has had any significant further impact on public attitudes.

Part of the answer to the mystery comes, I think, from the very negative –occasionally hysterical – tone of devolution’s opponents in 1997. In parliamentary debates preceding the Scottish and Welsh referendums, Conservative spokesman Michael Ancram warned of the “dark, cold night” that would follow Yes votes. Well, people did vote Yes – and the sky didn’t fall in. If you issue people with blood-curdling threats about the consequences of something happening, that something does happen, and those threatened consequences don’t manifest themselves, that rather damages your arguments.

But also important in devolution becoming the settled will of the Welsh people, I think, is another factor. There’s plenty of survey and other evidence to indicate that most people in Wales retain a considerable – and for the most part, I believe, healthy – scepticism about the politicians in their devolved institutions. But they are similarly sceptical of those in power elsewhere, certainly including London. Where politicians in Cardiff Bay get at least some credit is for being more concerned with, and focussed upon, the specific problems of Wales. And since 1997, more people have warmed to the idea of having an institution devoted to addressing those problems. To put it very crudely, devolved government might be (in the evaluations of voters) a bit crap much of the time – but then so is all government. And this, at least, is our crap.

That might not seem very inspiring. But it’s valuable for us political obsessives to periodically remind ourselves that most people, most of the time, don’t find politics, politicians and government either very interesting or inspiring. The more general point is that, for whatever reason, a growing majority in Wales have come to accept the appropriateness of distinctively Welsh government institutions taking many major decisions for Wales.

There’s a final point I want to make in this section. This concerns the consequences of these changes in public attitudes. There is clear and settled majority support in Wales for the nation to enjoy substantial political autonomy within the United Kingdom. I therefore regard it as the responsibility of the political class (broadly defined) to deliver that: to produce a coherent and sustainable devolution settlement. Against a background of uncertainties elsewhere (most obviously those relating to the Scottish referendum) that may, in some respects, be far easier said than done. But it ought not to be beyond the wisdom of humanity to finally, a decade and a half on, deliver what we have manifestly yet to see. A stable, sustainable model of devolved government for Wales: a devolution settlement true to its name.

Democratic politics, though, is of course about far more than just institutional structures. Within any institutional context, the most important players in democratic politics are the parties and the voters. My remaining four numbers all relate to them.

The first of them is the number 3. It is relevant here in relation to the Three Wales Model. A brief refresher course. Prior to the work that was begun at Aberystwyth in the late-1990s, the most serious body of research into political attitudes and voting behaviour in Wales was that conducted by another Aberystwyth-based team in the 1970s and early-1980s. The element of this earlier work that is most remembered today is Denis Balsom’s Three Wales Model. This can best be understood, I think, as a simplified geographical projection of a broader and more complex interpretation of the nature of political attitudes and party choice in Wales. The Model suggested rather different patterns of politics occurring in British Wales, Welsh Wales and Y Fro Gymraeg. This built on an understanding of party allegiances as following lines of class, national identity and language.

The simple elegance of the Three Wales Model, and its apparent resonance with some subsequent events (most obviously the 1997 referendum, where all the areas identified as British Wales voted No and all those in Welsh Wales and Y Fro Gymraeg voted Yes) has helped the model to remain established wisdom on voters and parties in Wales.

In an article just published in Electoral Studies, we examine the Three Wales Model seriously for the first time. To cut a potentially very long story very short, our verdict is not favourable. Theoretically, we note that the model (and the broader body of work underpinning it) viewed party preferences and vote choice as driven by social background factors (class, national identity and language) even as electoral research was increasingly coming to recognise such social determinism to be inadequate, and developing more directly political explanations of party support. And empirically, class, national identity and language simply don’t do a very good job in accounting for how Welsh people vote: either at the time the TWM was formulated or now. Mot effective in accounting for how people vote are factors associated with what scholars term ‘valence politics’ approaches: that is, the leadership of a party and its overall impression of competence and effectiveness.

This matters because, unlike the background conditions of society, parties’ leadership and the overall impression they give of competence and effectiveness are things that parties can, relatively speedily, seek to do something about. The recognition of such directly political factors as central to driving party choice suggests much less inevitability about which voters end up voting for which parties. Sure, we all know some people who will reliably vote for the same party time after time. But there were never as many such people as the myth suggested; and all evidence shows their ranks steadily diminishing over recent decades. Parties’ electoral success is not simply down to fate, or the inevitable electoral consequences of social background factors. Things are rather more potentially malleable than that.

An immediate problem with this argument, however, would seem to be that it runs straight into the reality of long-term Labour dominance in Wales. Single-party dominance, indeed, has characterised politics in Wales for much longer than the era of Democratic Devolution; it stems back to the beginnings of democratic politics in the 19^th century. For about half a century up until World War I, Welsh politics was dominated by the Liberals. The inter-war interregnum did witness something closer to genuinely competitive three-party politics in much of Wales. But since World War II the picture has been one of almost undisturbed Labour dominance. The party have won at least a plurality of the Welsh vote, and a majority of the seats, at every one of the 18 post-war general elections. Many Welsh constituencies have been a by-word for un-competitiveness. The situation has often been rather similar in many council wards. Even in the National Assembly, where a semi-proportional voting system helps create at least some space for other political forces, Labour has won the most votes and seats at every election; and has always been in government. Indeed, while the Labour party in Scotland continues to make heavy weather of fighting the SNP, since 2010 Labour in Wales has been resurgent.

This brings me to my final three numbers: 16.0 per cent, 12.7 per cent and 1.

16.0 per cent is the average gap in all post-war general elections between the share of the vote the Conservatives have won in England, and the share they have won in Wales. This is but one of many measures I could adduce of the historic failure of Welsh Conservatism. One of the UK’s two major parties (indeed, for most of the twentieth century, the leading party), has consistently under-performed in Wales by a wide margin. Such Conservative weakness and Labour strength has been so long-standing it has come to be thought of as normal. But by any international standard this is not normal. Politics where one side pretty much always wins is really not how democracy is supposed to work.

Of course, the socio-economic reality of Wales as a relatively poor part of the UK, whose economy was dominated by highly unionised heavy industries and more recently has had a heavy public sector presence, didn’t help the Tories. But Conservative failure has gone way beyond anything that such factors could account for. (The Conservative vote share has been consistently lower among all social classes in Wales than England). Our own research has suggested that the party has been hurt by being perceived as an essentially English party in Wales; for failing to articulate an attractive, distinctively Welsh Conservatism. Recent years did see some efforts to address this problem – rewarded, to some extent in 2010, which saw the gap between the Tory vote share in England and Wales at its lowest post-war level. But it is not obvious to me that the policies of the current UK government, or the approach of the current Welsh leadership, will do much to deliver voters an appealing centre-right option.

The one occasion in recent times when Labour’s dominance of Welsh politics did appear under serious threat was the first Assembly election of 1999, when a surge of support saw Plaid Cymru capture historic Labour bastions like Rhondda and Islwyn and narrowly miss out elsewhere. But this is where my next number comes in: 12.7 per cent. This number represents the decline in Plaid Cymru’s vote share (on the list ballot) from 1999-2011: testament to a decade of pretty abject electoral failure.

We have become used, in recent years, to the SNP being a far more politically successful party than Plaid Cymru – such that many people think it is inevitable and has always been thus. It’s widely forgotten that in the first round of devolved elections in 1999, Plaid won a higher vote share than the SNP did. In fairness, the SNP weren’t fighting Alun Michael. And after Michael’s ouster, Labour recovered to do much better in the 2001 general election and the second Assembly election of 2003. Yet when the tide of public opinion turned against Labour – to the point where Labour in the 2007 Assembly election under Rhodri Morgan ended up doing worse even than they had under Alun Michael – Plaid was wholly unable to offer the credible and persuasive alternative that might have made a similar advance to what the SNP managed in Scotland.

Our research for the 2011 election shows that the hostility towards Plaid that was once prevalent across much of the Welsh population is now largely a thing of the past. There is considerable public good will towards Plaid. However, the party was stunningly ineffective at converting that good-will into electoral support.

Democracy works best with strong, credible alternatives. This has generally been lacking in Wales. But we should neither accept this as normal nor as inevitable.

This leads me to my final number: the smallest, but perhaps the most important in understanding how democracy works in Wales – 1 – invoked here in relation to the most persistently defining feature of democratic politics in Wales: one-party dominance.

To argue that this long-standing dominance is harmful is not, primarily, an anti-Labour argument. One can hardly blame Welsh Labour for being consistently successful. As I have already indicated, if any blame is to be assigned, it should be directed at those parties that have consistently failed to offer effective and credible opposition to Labour in Wales. My point is not an anti-Labour one. It is simply that sustained dominance by any single party is not a desirable state of affairs. It is not something anyone should wish for their country. Of all the nations that have experienced single-party dominance within democratic politics Sweden is probably the only one to emerge as broadly politically healthy. Studies indicate that one-partyism tends to promote phenomena like intellectual stagnation (as the dominant party is under little pressure from challengers to develop new ideas; indeed, intellectual creativity is arguably positively discouraged, as new ideas emerging onto the political agenda might disturb the hegemony of the dominant party); corruption (as the apparatus of power and that of the dominant party become inter-twined); and a sense of entitlement to rule within the dominant party. It would be brave to suggest that these phenomena are wholly foreign to the Welsh experience.

But in the case of Wales, the pathologies of one-partyism are most clearly demonstrated in something I touched on earlier: the failure to develop a sustainable devolution settlement. As Richard Wyn Jones and I have argued in our recent book Wales says Yes: Devolution and the 2011 Welsh Referendum, for several decades the key debates on whether and how devolution might occur have been internal to the Labour party; the most important divisions have been those within Labour ranks; and the party’s devolution plans have often been more about trying to avoid or bridge damaging internal divisions than providing a coherent basis for effective devolved government. Put simply, we argue that the horribly flawed process of Welsh constitution building is itself directly linked to the dominant structural feature of Wales’ political culture: single party dominance. The Italian Euro-federalist Altiero Spinelli once suggested that the great achievement of Jean Monnet had been to unite Europe; but Monnet’s great failure was that he had united it so badly. I don’t know if that was quite fair to Monnet. But a similar statement would, I think, be at least close to fair regarding the Labour party and Welsh devolution.

Any remedy to the problem of one-party domination must, of course, come primarily from the non-Labour parties: the onus is on them to make themselves more effective and serious political forces. (I have occasionally been accused of being an out-of-touch academic; I’m not so out-of-touch as to expect Labour to start fighting elections deliberately badly just because some Prof has said that one-party dominance is bad!). Quite how this is achieved is a topic for another time. What we can say now is that a healthy democratic politics in Wales must offer a wider menu of options for government in Wales than just Labour on its own or Labour in coalition; the menu must include, and occasionally provide, for a non-Labour government in Wales.

To conclude: the last decade-and-a-half have seen both considerable continuity and significant change in Welsh politics. Some of that change can be seen as broadly positive: it’s good that devolution is no longer dividing the Welsh public right down the middle. But it’s not all good: that we are on our third devolved constitution already, with at least some more change likely, is testament to the failure of Welsh constitution-making. And the dominant continuity in Welsh political life – the persistence of unhealthily lop-sided political competition – is not something I find it possible to view positively.

But whatever our views on the political life of Wales, I hope that all of us here could agree that there is a need for politics in Wales to be taken seriously: by scholars and political activists, yes, but also by civil society very broadly defined. Ireland and Iceland offer two very interesting, and on-going, examples of widespread public involvement in political reform: not just at the level of people shouting drivel on phone-in shows, but getting seriously involved in the detailed hard graft. I hope we in Wales might reflect upon and learn from these examples, because the sort of politics we have in a nation is, ultimately, everybody’s business. And in the words of the Irish essayist Hubert Butler, “We live in a small country. But its problems are complex and interesting enough for the most ambitious intelligence”.