Cross-party cooperation: lessons from Paddy Ashdown’s tenure as Lib Dem leader

For more on this topic, see 6 ways to make cross-party political deals work, The Cook-Maclennan talks: lessons in cross-party cooperation and Lessons from history on Labour / Lib Dem cooperation: Kier Hardie and Ramsay MacDonald.

Lessons from the Ashdown-Blair 'Project'

by Duncan Brack

On 9 April 1992, the Conservative Party won their fourth general election victory in a row. After an election campaign which had seen the two main parties more or less tied in the opinion polls, Labour had entertained real hopes of emerging as the largest party, and the Liberal Democrats of holding the balance of power. The result, a Tory lead of 7.5 per cent over Labour, and a Tory majority in the Commons of 21 over all other parties, came as a crushing blow to Labour; the party's leader, Neil Kinnock, resigned a week later. The Liberal Democrats too were disappointed, though at least they had succeeded in recovering from their post-merger nadir of 1988-90. (The polls' consistent over-estimation of Labour's support reflected the pollsters' failure to pick up the speed with which Britain's electorate was changing; their sampling was wrong. Plus ça change.)

Political discussion after the election focused on the seemingly never-ending Conservative hegemony. Commentators seriously considered whether Labour could ever win another election. Echoing Mark Abrams' and Richard Rose's Must Labour Lose? of 1960 (which had followed three Tory victories, from 1951 to 1959), the Fabian Society published a series of pamphlets under the heading Southern Discomfort, focusing on Labour's loss of support in southern England. And yet the combined Labour and Liberal Democrat vote in 1992 clearly exceeded the Conservative level of support, so there was also speculation about whether there might be some grounds for collaboration between the two opposition parties.

In fact, the potential for cooperation between the non-Tory parties, or a realignment of them, had been a common theme of leaders of the Liberal Party for the previous 30 years - including Jo Grimond's call for a realignment of the left in the 1960s, the formation of the Lib-Lab Pact in the 1970s, and the alliance between the Liberals and the new Social Democratic Party in the 1980s. With the sole exception of 1955, the combined Liberal and Labour vote had exceeded Conservative support in every election since 1945 (and has continued to do so in every election since 1992). It was only the division between the two progressive parties, it was argued, that had allowed the Tories to hold power for so long. The most radical version of this argument ascribed Conservative success throughout the 20th century to the disintegration of the pre-first world war Liberal Party, which had divided progressive politicians between the rising Labour Party and the surviving Liberals and had allowed the Tories to win power for most of the century on a minority vote.

Paddy Ashdown, the first leader of the Liberal Democrats, had been interested in exploring the possibility of cooperation with Labour since his election in 1988. Just five days after becoming leader he met representatives of the newly formed Labour-friendly Institute for Public Policy Research (IPPR) to talk about think-tank cooperation, and the following April he and his lieutenant Richard Holme were talking about some sort of Lib-Lab 'Programme for Britain'. The 1992 election seemed to make the case even stronger. The day after polling day he held a strategy meeting which decided that the Liberal Democrats should not attack Labour too strenuously; he recorded in his diary that 'we must make use of this opportunity to realign the left'.

Exactly a month after the election, on 9 May, Ashdown delivered what became known as the 'Chard speech', given to an audience of only 40 or 50 in a small town in his constituency; he said later that he believed it was the most important speech of his leadership. Arguing that the Liberal Democrats needed to 'work with others to assemble the ideas around which a non-socialist alternative to the Conservatives can be constructed', he called for a national electoral reform commission and for Labour to be more open to constitutional reform and a market economy, more pluralist and less a creature of the trade unions.

The speech was deeply unpopular with Liberal Democrat MPs (one took to referring to it as a 'burnt offering'), partly because at the time Labour seemed finished and about to descend, in the wake of Kinnock's resignation, into another bout of infighting. The Liberal Democrat peers were more supportive, but only last-minute wrangling prevented a strategy debate at the autumn conference placing severe restrictions on Ashdown's freedom of manoeuvre. However, there was very little response from Labour. Although some of the party's leadership, including Robin Cook and Peter Mandelson, were open to approaches from the Liberal Democrats, the new leader elected in July, John Smith, was determined to maintain his independence and believed that Labour was still capable of winning on its own. Events over the next 12 months seemed to suggest that he might be right - in September 1992, 'Black Wednesday', which saw the pound forced out of the European Exchange Rate Mechanism, destroyed the Conservatives' reputation for economic competence, and the party then started to tear itself apart over British accession to the Maastricht Treaty of European Union in 1992-93.

The Conservatives became so unpopular that between mid-1994 and the start of the 1997 election campaign the Labour share of the vote in opinion polls never dropped below 50 per cent. But many Labour politicians, and commentators, could not bring themselves to believe that the Tories would not win again; they had been too scarred by the experience of 1992. In the phrase of Neal Lawson (adviser to Gordon Brown 1992-94), Labour felt like gatecrashers at a party - they might be having a great time, but eventually someone would turn up and kick them out. So the possibility of cooperation remained alive, and it gained new impetus with the sudden death of John Smith in May 1994, and his replacement as Labour leader by Tony Blair in July.

Blair, influenced partly by Peter Mandelson, was far more open to some form of cooperation with the Liberal Democrats than Smith had been. Indeed, Ashdown had discussed the topic with him as early as July 1993 at a dinner organised by Anthony Lester (special adviser to Roy Jenkins in the 1970s, founder member of the SDP, and from October 1993 a Liberal Democrat peer). Blair, following Jenkins (whom he regarded as a mentor), recognised the strength of the argument that the historic split in the British left had handed power to the Conservatives; as he put it later, 'the 20th century had been a Tory century precisely because good and talented people who should have been together were instead in separate parties fighting each other'. Two months after his election as Labour leader, he and Ashdown began what was to become a five-year series of meetings and discussions about how to get rid of the Tories and how best to work together both before and after a potential election victory - what became known as 'the Project'.

The details of the process can be followed in Ashdown's diaries and other sources. What this essay focuses on is what in practical terms the two parties did, and what lessons can be learned from the Project for cooperation in today's very different political environment.

Assuming that the Tory vote was likely to recover before an election in 1996 or 1997, some kind of cooperation certainly seemed to make sense; and even if Labour did win, it seemed quite possible it would be with only a small majority, and might thus need additional Liberal Democrat support in the Commons. By 1993-94 the Liberal Democrats were also benefiting from the collapse in Tory support, in local elections and by-elections - in many areas, particularly south-west England, they seemed better placed than Labour to dislodge the Conservatives. In July 1995 they fought off a strong Labour challenge to win the Littleborough & Saddleworth by-election from the Conservatives, suggesting that they would not simply be swept aside by the rising tide of support for Labour.

But what kind of cooperation? One possibility was an electoral pact, where Labour and the Liberal Democrats would field common candidates. Blair suggested this in talks with Ashdown in November 1995, at least for seats in the south-west. But as Ashdown retorted, this was: 'totally impossible'. He added, 'And we would waste a lot of time dividing our parties if we tried to do it. It would also look like a grubby plan designed to gain power and votes for ourselves, instead of one based round principles and what was best for the country.' The Liberal Party had been through a process of seat allocation with the SDP, and it was a divisive, time-consuming and exhausting exercise even between two parties with very few policy differences.

What Ashdown preferred was to lay out a small number of key positions on which Labour and the Liberal Democrats agreed and make it clear that they would cooperate on them in the next government. This would enable the parties to retain their distinctiveness in other areas while at the same time promoting an atmosphere of cooperation which should encourage tactical voting amongst the electorate. Many of the discussions between Blair and Ashdown over the following 12 months explored these possibilities. This was assisted by the formal abandonment of 'equidistance' by the Liberal Democrat conference in 1995. Although many of the party's activists and MPs were highly sceptical of Ashdown's approach, and resisted any attempt to more publicly align their party with Labour, the reality was that Liberal Democrat policy positions were, with a few exceptions, much closer to Labour's than to the Tories', and the government was so unpopular by 1995 that it was ludicrous to pretend that the Liberal Democrats were indifferent, or 'equidistant', between them and the official opposition.

Discreet cooperation began to develop between the two parliamentary parties. Archy Kirkwood and Donald Dewar, the Liberal Democrat and Labour Chief Whips, began to talk more frequently, developing what Kirkwood called a 'doctrine of no surprises'. They tried to avoid major rows between their own parties and on occasion coordinated their MPs' attacks on the Tories, helping to reinforce the impression that the government was an increasingly beleaguered minority. This approach paid dividends when Kirkwood and Dewar discovered that the Conservative whips were 'pairing' the same Tory MPs against both Liberal Democrat and Labour members, effectively reducing opposition numbers. Blair and Ashdown also sometimes coordinated their attacks on the government at Prime Minister's Questions.

Talks, generally over dinners, between small groups from both parties helped each side understand each other's positions, explore the possibilities for cooperation in Parliament (short of a coalition) should the Conservatives lose power, and paved the way for cooperation over policy in the form of a series of talks on constitutional reform led by Robin Cook for Labour and Robert Maclennan for the Liberal Democrats. Starting in March 1996, a year later the group reached agreement on a package of proposals including incorporation of the European Convention on Human Rights into UK law, freedom of information legislation, devolution to Scotland and Wales (and elections by proportional representation to their parliaments), an elected authority for London, removal of the hereditary peers from the House of Lords, proportional representation for the European elections, and a referendum on voting reform for Westminster elections, comprising a choice between the existing first-past-the-post system and a proportional alternative, to be agreed by a commission on voting systems. Most of this had been Liberal Democrat policy for years (or was a watered-down version of it), but some was new for Labour. Blair saw it as part of his programme of modernising the Labour Party, and the position of the constitutional modernisers within Labour, such as Robin Cook, was thus strengthened.

The Cook-Maclennan process was public. What was discussed in secret, however, was something much more dramatic, what Ashdown called 'the big thing', an agreement to fight the election on a common platform on at least two or three major issues. Ashdown went so far as to draft successive versions of a 'Partnership for Britain's Future', covering constitutional reform along the Cook-Maclennan lines, cleaning up politics (after several examples of corruption and dishonest conduct amongst MPs), the reform of welfare systems and economic policy reform, including investing in education, awarding independence to the Bank of England, and adherence to the criteria for entry into the single European currency. From July 1996 Blair and Ashdown started to talk about Liberal Democrat participation in a Labour government; Peter Mandelson later claimed that this would have involved including two Liberal Democrat MPs, Alan Beith and Menzies Campbell, in Blair's first Cabinet.

In the end, the 'big thing' was too big a step. What worried Ashdown and his colleagues was Blair's refusal to commit firmly to the introduction of proportional representation for Westminster elections - the absolute bottom line for the Liberal Democrats, who could not be expected to tie themselves to a much bigger partner without being able to survive its eventual fall. Ashdown's diaries record in painstaking detail a long series of meetings in which Blair was first educated about what PR meant and the different systems through which it could be introduced, and then prevaricated, hinting at his own possible conversion to it (or maybe to something weaker, such as the Alternative Vote), but stressing the opposition he would face in the Parliamentary Labour Party. In the end, policy cooperation extended only as far as the Cook-Maclennan agreement.

The approach of the general election raised the question of how far the parties should cooperate - and be seen to be cooperating - during the campaign. Ashdown argued for overt collaboration, campaigning together to get rid of the Tories, even to the extent of appearing at joint rallies with Tony Blair. But polling in the party's target seats conducted by Chris Rennard, the Liberal Democrats' experienced Director of Campaigns and Elections, showed that while those who had voted Conservative in 1992 were open to the proposition that the Liberal Democrats could participate in government with Labour should the Conservatives lose, they strongly disliked the idea that the Liberal Democrats should actively campaign for a Labour-Lib Dem coalition, and any hint of this would drive them back to the Conservatives. Furthermore, the prospect of a coalition did not help with those who had voted Labour in 1992; they preferred the Liberal Democrats to the Tories anyway, and the evidence suggested that not many of them really understood tactical voting, believing that they had to vote Labour to bring down the Tories, even when their candidate was in a poor third place in their own seat. These findings were unwelcome news to Ashdown, but the polling evidence was so clear that he had to accept it; and it recalled what had happened in the last week of the 1992 campaign, when it was felt that speculation about the possibility of a hung parliament had driven voters back to the Tories. Holme also shared the figures with Mandelson.

So overt cooperation would be counterproductive. But what about covert cooperation? In January 1997 Neal Lawson and Neil Sherlock (a member of Ashdown's speech-writing team) - both supporters of cooperation - organised a dinner in the Goring Hotel in Victoria to bring together key individuals from the election campaigns on both sides, some of whom had not before been involved in the Project. Chris Rennard raised the idea of the Mirror (the main Labour-supporting tabloid) endorsing Liberal Democrat candidates where Labour had no chance. Mandelson agreed this could be done, and smoothed the way with the Mirror's staff. Originally supposed to cover just 10 seats, on the eve of poll the paper published a list of 22 seats where, if Labour voters backed the Liberal Democrats, the Conservatives would be defeated. In the event the Liberal Democrats won 20 of them.

Mandelson also requested a list of Liberal Democrat target seats so that Labour knew where to restrict its campaigning efforts. Rennard supplied a list of 28 seats to Labour deputy leader John Prescott's office, disguised (in case of leaks) as an academic paper identifying constituencies where the Liberal Democrats could win and Labour could not. It is not clear to what extent either party really held back in their campaigning - and the extent to which campaign managers can direct their activists to go to seats other than the ones they live in is probably less than the managers like to think - but it helped that in many of the targets it was obvious that the Liberal Democrats were better placed to win than Labour. The number of seats where both opposition parties were in contention with each other was very small, though there was certainly no let-up on either side in those constituencies; in the end, Labour gained two seats from the Liberal Democrats.

The parties also collaborated, in a fairly low-key way, during the 1997 election. Blair and Ashdown stayed in touch both before and during the campaign, discussed using a common language to attack the Tories, generally avoided criticising each other's parties, and focused on much the same issues in campaigning - health, education and crime (they would probably have done this anyway; these were the leading issues). The fact that the two parties' policies were different - Labour had promised to stick to Tory spending plans, the Liberal Democrats to raise income tax to invest in education - also helped, marking out a distinctive case for the Liberal Democrats and making it clear that cooperation did not mean a Labour takeover. The most obvious example of cooperation came in the decision to withdraw each party's candidate in Tatton, to give a clear run to the former BBC journalist Martin Bell, standing as an independent 'anti-sleaze' candidate, against Neil Hamilton, a Tory MP (now a Ukip's leader in the Welsh Assembly) deeply embroiled in the cash-for-questions scandal; Bell, assisted by activists from both parties, won convincingly.

The 1997 general election saw the Conservatives go down to their worst result in a century and a half, losing a quarter of their vote and half their seats. Their losses were exaggerated by tactical voting; as the psephologists John Curtice and Michael Steed concluded, 'the scale and impact of tactical voting in the 1997 election was unprecedented'. They estimated that Labour gained between 15 and 21 seats, and the Liberal Democrats between 10 and 14, as a result of tactical switching to defeat Tory MPs. Supporting one of the findings of Chris Rennard's target seat polling (see above), Liberal Democrat voters proved more willing to switch to Labour than Labour voters to the Liberal Democrats - the surge in support for Labour was so great that often the best that Liberal Democrat candidates could hope for was to keep the Labour vote at its 1992 level (which was often enough to win, given the collapse in the Conservative vote). Tactical voting was more widespread in areas where the Liberal Democrats had done well in local government over the preceding five years, which had established the party as a credible contender in the eyes of the local electorate. Effective targeting of resources throughout the Parliament also clearly helped.

In a close election, tactical voting on this scale would have made a major difference to the outcome. But the 1997 result was not close: Labour won 419 seats, for an overall majority over all other parties of 179. Nevertheless, as late as election day, Blair and Ashdown were still talking about whether they could entertain any form of cooperation. Blair declared that he was 'absolutely determined to mend the schism that occurred in the progressive forces in British politics at the start of this century'. He hinted at a coalition, though Ashdown pointed out that it would be unacceptable for Liberal Democrat ministers simply to implement a Labour programme. By the next day, however, Blair had changed his tone, talking merely of a 'framework for cooperation'. Robin Cook later confirmed that Gordon Brown and John Prescott had both made clear to Blair overnight their virulent opposition to any role for Ashdown or his colleagues in government. In any case, the size of Labour's majority destroyed any argument for it.

The rest of the story of the Project can be followed elsewhere. In place of a coalition, a Joint Consultative Committee was created between the two parties to discuss issues where there was already agreement in principle, such as devolution or first-stage reform of the House of Lords; it was later extended to European issues. It is difficult to judge what, if anything, the Committee achieved, and it was largely abandoned by Charles Kennedy in the run-up to the 2001 election. The key stumbling block to further cooperation was Blair's continued refusal to commit to proportional representation. In December 1997 the government finally announced the establishment of an independent commission on voting reform, to be chaired by Roy Jenkins - but when his report was published in October 1998, advocating an additional member system of PR, Blair's response was entirely neutral, with no commitment to a referendum. Later that day Jack Straw, the Home Secretary, rubbished it publicly, and did so again in a Commons debate the following week. Effectively this was the end of the Project, and Ashdown decided that it would also mark the end his period as Liberal Democrat leader; he announced his resignation three months later.

No comprehensive and objective assessment of the Project has yet been carried out. However, it seems clear that it made a difference in at least two respects. The Cook-Maclennan agreement had a direct impact in the shape of the constitutional reforms Blair implemented after 1997. Probably Labour would have devolved power to Scotland without any prompting from the Liberal Democrats, but Labour's attachment to issues such as Welsh devolution, proportional representation for the European elections or freedom of information was much weaker, and these policies may not have been implemented in the absence of the agreement - or their introduction may have been delayed and perhaps eventually dropped. Second, the atmosphere of cooperation that Ashdown and Blair built between them in the run-up to the 1997 election, and specific instances such as the Mirror list of winnable seats, made a difference to the election result, encouraging supporters of each party to switch to each other to defeat the Conservatives and thereby exaggerating their losses.

What lessons can we draw for the future from the experience of the Project? There are (at least) six.

First, don't talk about coalition, even if that's what you have in mind. The target seat polling from 1997 showed that those who had voted Tory in 1992 would be frightened back to their original party if they thought Labour and the Liberal Democrats were ganging up in advance of the election. This is strongly reinforced by the experience of the 2015 election, when the prospect of a Labour government in coalition with, or supported by, the SNP proved deeply unpopular (though not quite as unpopular as the Liberal Democrats' apparent willingness to join a coalition with almost anyone).

Second, it is nevertheless worth exploring the potential for pre-election cooperation over specific areas of policy. The most obvious is constitutional reform, along the lines of the Cook-Maclennan agreement or, more ambitiously, the Scottish Constitutional Convention of 1989-95. The gross distortions of the 2015 election result, where it took 34,000 votes to elect a Conservative MP, 302,000 to elect a Liberal Democrat and 3.9 million to elect a Ukip MP, makes a strong case for electoral reform in particular. Voters do not tend to get excited about these kind of issues, however, so the state of public services, particularly the NHS, could be another possible topic - the evidence suggests that voters like the idea of politicians setting aside party differences to work together (though the fact that that's what coalitions do seems to have eluded them).

Third, do everything you can to maximise tactical voting, which is a rational response to the distortions of the first-past-the-post electoral system. This will probably be more effective if it is promoted from outside the parties, as in the Mirror piece in 1997. In 2020 there should be extensive potential for arranging vote swapping via social media. Parties' campaigning resources - which, with the much greater use of telephone canvassing, are now more centralised than in 1997 - can also be deployed to avoid attacking each other, at least up to a point.

Fourth, either keep it all as quiet as you can - or be as open as you can. Some party activists are open to cooperation - and it is of course widespread in local government - but many are not. Journalists often like to look down on party members as 'tribal', but another way to put this is that they are loyal, to their beliefs and their colleagues. Most party activists will spend their entire political careers delivering leaflets, raising money and knocking on doors without ever being elected to anything, and without realising any personal financial benefit. Belief in the cause is what keeps them going, and naturally they will not be well disposed to giving up their efforts to see another party's candidate elected. So, either don't tell them, or as be as open as you can, explaining what's going on and what the benefits are. The Blair-Ashdown Project is an example of the former approach; in retrospect it is remarkable how little of what was going on leaked to the outside world. The overwhelming vote at the Liberal Democrat special conference in May 2010 to enter coalition with the Conservatives is an example of the latter, showing that reasoned argument can convince activists to work with other parties in the national interest.

Fifth, none of this will work unless the personal relationships between the key individuals involved are very good - they have to trust each other to work together and stick to their side of the deal. The Project would never have happened without the rapport that Ashdown and Blair built up over several years of dinners and discussions.

Finally, be clear what you want the outcome to be: a coalition, a confidence and supply arrangement, an agreement to reform the voting system and then call another election, or something else. And despite all the difficulties, never forget the reason: to avoid the 21st century seeing another succession of Tory victories on ever-diminishing shares of the vote.

Duncan Brack was the Liberal Democrats' first Policy Director (1988-94), chair of the party's Conference Committee 2003-10, and special adviser to Chris Huhne at the Department of Energy and Climate Change 2010-12. He is currently Vice Chair of the party's Policy Committee. Professionally he is an independent environmental policy analyst and adviser.