The Journal of Liberal History‘s special edition marking the 25th anniversary of the Liberal Democrats includes a piece from myself on the party’s campaigning history. My motivation for writing it – other than a willingness to help a 100%-volunteer Journal that Duncan Brack in particular does so much to make such a success – was two-fold.
First, to put on the record various details, some important and some less so, which every account of the party’s history I’ve seen misses. As an historian by training, I cringe every time I see what could be useful information for historians in the future fail to get recorded in ways they are likely to be able to access (and hence, for example, my post about how local parties voted on the leadership question).
Second, in doing so to debunk some myths and remedy some omissions about the party and British political campaigning, both within the party and also by political scientists. Most notably, far too many accounts of British political campaigning are written without noticing how dramatic 1997 was, not just in terms of a Labour landslide, but also in terms of the impact of constituency campaign tactics. Fail to understand what happened in 1997 and why and you not only fail to understand British political campaigning but also, closer to home, fail to understand the roots of more recent events involving Chris Rennard.
Here then is the piece.
The Liberal Democrat approach to campaigning
Started by pioneering academics such as David Denver, Gordon Hands and Justin Fisher, there is now a well-established tradition of research into the impact of local campaigning on British election results. This work, however, tends to be cross-party and to rely heavily on evidence such as officially recorded spending figures, election results, questionnaire findings and statistical analysis of all three. As a result, there is still very little written about the development of particular campaign techniques, especially where they were specific to one party for a long period of time.
This article fills part of that gap in the historical record, by considering the Liberal Democrat (and before that Liberal Party) campaign techniques, and the evolution of party strategies and structures to support them.
The Liberal Democrat inheritance from its predecessor parties
The techniques, strategies and structures were not all created afresh when the party was formed out of the merger of the Liberal Party and SDP. Rather, the merged party inherited most of its initial approach from its predecessors.
In establishing a campaign tradition for their merged successor to inherit, the Liberal Party had two big advantages over the SDP: it had been around, doing campaigning, for longer and, at least as far as local campaigning was concerned, had been seen even by many in the SDP as being the more skilled party.
Moreover, in adopting community politics in the 1970s, the Liberal Party had taken both a philosophical and a practical approach that venerated local activity and regular communication. It was about being active on the doorsteps and via letterboxes all year round and not simply (as was traditional with election campaigns) in the few weeks before a polling day. Elections were, in the old Liberal Party saying, but the punctuation marks in community politics.
The degree to which community politics should be about winning elections, and accusations that it was being dumbed down into an election-only approach, was a regular feature of Liberal Party debates over the future of the idea. As one of its seminal texts warned (and note that its authors felt the need to issue this warning): ‘Community politics is not a technique. It is an ideology, a system of ideas for social transformation. For those ideas to become a reality there is a need for a strategy of political action. For that strategy to be successful it needs to develop effective techniques of political campaigning. Those techniques are a means to an end. If they become an end in themselves, the ideas they were designed to promote will have been lost’.
By contrast, the SDP’s roots were predominantly in national politics and national issues, which did not in the same way lead naturally to a distinctive approach to local campaigning. A Labour MP who left his party for the SDP over Europe, trade union power or voting rights at a national party conference did not as a result adopt a different approach from that of other parties to grassroots politics in the same way that a Liberal attracted to community politics did.
Moreover, the relative results of the two parties in 1983 and 1987 reinforced the existing perception of the Liberal Party as being the skilled exponent of grassroots political campaigning. Despite a tortuous process to ensure a ‘fair’ allocation of seats between the two Alliance partners, at both elections the Liberals were far more effective at winning seats than the SDP (by the margin of seventeen to six in 1983, and by seventeen to five in 1987). For all the importance the SDP attached to bringing a sense of professionalism to the (as they saw it) amateurish approach of the Liberals, when it came to votes being counted it was the Liberal Party approach that consistently did better.
It is no surprise, therefore, that comparing the different campaign manuals produced for the two predecessor parties, for the Alliance and then for the Liberal Democrats reveals a strong continuity in both content and authors from the Liberal Party through to the Liberal Democrats. Indeed, in many cases large pieces of text were copied over and then updated for new editions or nominally new titles, so that even where the author’s names and titles are new, the continuity is still there.
Some of the SDP election manuals did have a distinctive approach from the contemporary Liberal Party ones. For example, the SDP’s first Local Government Election Handbook, published in January 1982, highlighted the importance of ‘Policy Formulation’, which features as early as page 2, giving it a prominence that equivalent Liberal Party publications of the time did not grant. But subsequent Liberal Democrat publications followed the Liberal Party and not the SDP choice of emphasis.
This documentary and authorial trail from the Liberal Party through to the Liberal Democrats was also reflected in the nature of the grassroots campaigning that the Liberal Democrats specialised in. The various techniques described below all had antecedents in Liberal Party activities. It is hard to identify a distinctive SDP contribution, save for perhaps one crucial one: many in the SDP viewed themselves as taking the fight for Westminster constituencies seriously, in a way that the Liberal Party, with its – in their view – undue concentration on local councils, had not.
The need for leaflets
Central to this inheritance for the Liberal Democrats was the role of leaflets, as illustrated in this extract from a film made around the turn of the century to show liberals in emerging democracies how to fight elections, showing how the Liberal Democrats won a council by-election:
If one image can sum up the approach to campaigning taken by the Liberal Democrats across twenty-five years, it would be a piece of paper on a doormat emblazoned with a bar chart and a headline screaming that ‘Only the local Liberal Democrat can beat Party X round here’.
Then Liberal Party MP David Penhaligon coined the phrase that many activists have since quoted, ‘If you believe in something, write it on a piece of paper and stick it through a letterbox’. However, it was Chris Rennard, first as the Liberal Democrats’ Director of Campaigns and Elections, and then subsequently as Chief Executive, who turned it into an effective seat-winning tactic at general elections for the party.
Both when Penhaligon first coined the phrase and during Rennard’s time in charge of the party’s campaigning, the party suffered from not only an absence of favourable newspaper owners but also a paucity of coverage on impartial radio and TV channels. As a result, the Liberals and subsequently Liberal Democrats had turned to intensive and local leafleting to get out the party’s message.
Moreover, the workings of Britain’s first past the post electoral system required that leafleting – and other campaigning – be tightly targeted geographically in order for support to be turned into seats. That is because, outside of parts of the Celtic fringe, the third party’s vote was usually fairly evenly distributed around the country, which is not the route to winning under first past the post. Only by concentrating on building up support within particular seats could votes secured be turned into actual seats won.
By the time of the Liberal Democrats, the idea of intensive leaflet campaigns and careful targeting to win local elections was well established and, in many parts of the country, successful. However, it had not achieved similar levels of success at a national level.
Part of the reason for this distinction between local level success and Westminster-level disappointments was the difficulty of applying targeting effectively at parliamentary-level elections.
At the local level, the Liberals in particular in their most successful areas had honed a very strict approach to targeting, pouring efforts into winnable wards whilst neglecting other wards – often in those doing no more than putting a candidate’s name on the ballot paper, if that. But when it came to general elections, it was far less common for candidates and their helpers to similarly abandon no-hope seats and move instead to help in winnable constituencies. (Parliamentary by-elections were an exception to this due to the paucity of other elections usually taking place on the same day – and it was no coincidence that this made it easier to encourage large volumes of help and that the party developed something of an expertise at winning such contests.)
The relative reluctance on the part of activists to move during general elections was in part because Westminster elections came with elements that made it harder for candidates to turn their back on their own seat and concentrate on helping someone who could win. Public meetings may be in a long-term decline in British politics, but most candidates felt they had to turn up for those in their own seat. Similarly, the provision by the Royal Mail of a free delivery service for one election leaflet to every voter encouraged candidates and their teams to feel that they had to ‘show the flag’ and at least get one election address out. Although the Royal Mail did the delivery of the freepost election address, its production could still require significant local campaign effort, such as to address and stuff envelopes to hold the election address, reducing the amount of time available to help in winnable constituencies.
Also, for local council elections, targeting usually meant asking people to travel less far than for targeting at parliamentary elections. Parliamentary constituencies are considerably larger, and winnable constituencies fewer. As a result, going to your nearest target constituency might even mean travelling 100 miles or more. By contrast, going to your nearest target ward usually meant only a short trip, save for the most rural of areas.
Moreover, at a national level many believed that very little could be done to influence the chances in individual seats, if indeed that was even the point of a national campaign. Thomas (Jack) Daniels, the Liberal candidate for Luton in 1966, recounted how despite his very low prospects of victory, he received an election visit from then Liberal leader Jo Grimond because Grimond simply thought it was the right thing for him to do to visit everyone who was standing as a candidate.
That sort of attitude lasted over the decades. Chris Rennard liked to tell the story of how he met with the Liberal Party’s 1987 General Election Campaign Committee and was firmly told that, ‘which seats the party won at a general election was just a matter of luck, completely out of the control of the central campaign’. That was a belief that he did not share and set out to change, and during his time in charge of the party’s campaigning it did change radically.
Table 1: Liberal, Alliance and Liberal Democrat general election performance
Election Seats won Share of the vote Seats:votes ratio
1970 6 8 0.8
1974 F 14 20 0.7
1974 O 13 19 0.7
1979 11 14 0.8
1983 23 26 0.9
1987 22 23 0.9
1992 19 18 1.1
1997 46 17 2.7
2001 52 19 2.8
2005 62 22 2.9
2010 57 23 2.5
The most notable example of the party’s improved ability to win seats under the British electoral system was 1997. At that election the party’s vote share fell, and was less than the party’s previous 1974 peak. Yet the number of seats the party won went up, to more than three times the number of seats won during the two 1974 elections.
A semi-autonomous campaign organisation
Under Chris Rennard, targeting also meant strong central control, with funds and staffing under his direction pointed towards a limited number of seats.
To be successful, this required two internal debates to be won, repeatedly. First, the ability of a well-resourced targeting operation to win seats was so important that the party should dedicate more resources to target seat campaigning. Second, that the selection of target seats should be sufficiently narrow to make it meaningful – and so only giving target seat status to those performing at a level which could bring victory.
Rennard therefore believed that effective targeting also required him to have significant power and discretion within the party organisation. It meant he argued both for an increasing share of the party’s resources to be dedicated to the key seats operation and for him and the party’s campaign staff working for him to have substantial autonomy over the key seat operation. This was only partly tempered by a need to cooperate with other parts of the party in order to bring in their resources to the operation. As a result, the selection of target seats in England in the run-up to the 1997 general election, for example, involved regional parties in helping to evaluate seats against performance criteria such as level of members and regularity of leaflet delivery. Central party staff and funding resources were only available for those seats and regions that complied with the scheme.
Supporting activity in key seats became an increasingly important funding priority for the party, with the £120,000 allocated to the Campaigns Department for key seats in 1992 growing to £1m in 1997 and continuing to grow subsequently. Controversy came from the three different sources of financial power for the 1992 and 1997 general elections. There was the mainstream federal (UK-wide) party budget. There was Rennard’s growing campaign operation. There were also specially created, autonomous structures for running the 1992 and 1997 election campaigns, under Des Wilson and Richard Holme respectively. Those separate general election structures came with their own budgetary autonomy.
All this caused some in the voluntary party’s democratic structures to complain about lack of clarity or control over what was being spent, by who and on what basis. A significant part of the party’s federal (UK-wide) funds were, for example, put into a campaign fund and a by-election fund, whose income (in particular, its sources of fundraising) and expenditure had far less detailed scrutiny from the party’s elected committees and conference than the main federal party budget. The general election budget itself received even less scrutiny from the party’s democratic processes.
There was a significant benefit for the party’s campaigning from this opacity. It protected ‘commercially confidential’ information from the prying eyes of other parties, such as over the relative amount of resources put into contesting different seats and whether or not any particular parliamentary by-election was going to see a serious campaign launched.
The party’s Medium Term Review following the 1997 campaign decided that the party’s election campaigns should in future be more accountable to the mainstream party structure and that the Campaign & Communications Committee Chair should chair the election campaigns. Tim Razzall was elected to this position following Charles Kennedy’s election as leader and he in turn and in accordance with the Medium Term Review appointed Rennard as a ‘Chief Executive of the General Election’ (he was not then party Chief Executive). This both simplified the structures and solidified Rennard’s influence, which had spread from key seats in 1992 to an increasingly important role in deciding messaging in 1997 to being the day-to-day person in charge for 2001.
Alongside this integration of budgets and structures around general elections, the Parliamentary Office of the Liberal Democrats (the operations in parliament, funded by public money and employing much of the party’s London-based staff) became more integrated with the federal HQ’s operations. Such integration brought many benefits for the party and its ability to make the best use of its resources. It also meant that Chris Rennard’s role was an increasingly powerful one – a trend strengthened by the need to fill the vacuum caused by Charles Kennedy’s lack of interest in organisational details and by his health issues at the time. Consequently, whatever benefits these changes brought, they did not assuage – and if anything increased – the concerns of some critics about the semi-autonomous nature of the campaigning control he exercised.
Some steps were taken by Chris Rennard in response to these concerns – such as the creation of a new senior management team to take on some of the powers newly concentrated at party HQ. There were also many vocal and passionate defenders of his approach, especially from the party’s organisers, agents and candidates. In part this came from a simple record of success (see Table 1) that seemed to justify this approach and which suggested that the more of the party’s funds that went on campaigning, and the more that was under Rennard’s control, the better the party did.
It also reflected the mixed reputation of the party’s Federal Executive (FE) and Federal Finance and Administration Committee (FFAC) who had control over the federal budget and were therefore sidelined when funds went elsewhere. Comments such as ‘it’s the worst committee I’ve ever served on’ were common from FE members during this time, and spanning several different FE chairs.
Moreover, there was a more subtle, cultural factor at work, arising from the way that election law requires individual ward or constituency election campaigns to be conducted. By making the candidate and their election agent legally responsible for all the campaign activities carried out on behalf of a candidate and their party, election law encourages a very centralised approach to election campaign management, which was widely reflected in the party’s general election agent manuals and similar publications for local elections. The logic was simple: if someone is going to be legally responsible, then they should be in charge and other bodies – such as the local party executive – should not have power, otherwise they would be making decisions for which someone else then has to carry the legal responsibility. Although there was no directly similar legal requirement to concentrate power at the national level for elections, the culture that elections are run by a small number of individuals who have complete control did seep over from local to the national level, especially as many of those involved at the national level in election campaigning first learnt their skills at the local level.
Finally, there was also a tradition dating back to the Liberal Party days of general election campaigns or key seat activities being run in semi-autonomous ways. When this was added to the SDP heritage, with its revulsion at the way Labour’s cumbersome committees had mismanaged elections, and a continuing tradition of Liberal Democrat leaders liking to put general election campaigns outside the party’s usual democratic structures, there was a very open door for Rennard to push at with his actions to have centralised, semi-autonomous control.
This applied first with parliamentary by-elections and then with constituency campaigning for general elections.
As with the Alliance and before them the Liberal Party, parliamentary by-elections played a large role in the party’s fortunes, especially in its early days. The Eastbourne by-election victory in 1990 was credited by many with helping to save the party after the traumas of merger, and all through the next two decades, high profile by-election victories gave the party much needed bursts of publicity and credibility.
They often also helped set the political mood, helping to shift the journalistic consensus as to what the party’s fate was likely to be. For example, after the 1997 general election gains, the media generally assumed the next general election would be about the Liberal Democrats trying to minimise the degree to which they slipped back – until the Romsey by-election victory in 2000 set the political story as being one in which the party would make further progress.
More than just leaflets
Chris Rennard’s approach to winning parliamentary constituencies was always heavily rooted in local campaigns and leaflets. When in subsequent years he recounted his role in achieving the biggest swing against the Tories in the country in 1983 with David Alton’s re-election in Liverpool Mossley Hill, it was the number of leaflets and the size of the swing that most frequently featured, even though an intensive door-knocking campaign and a growth in the volunteer base from 100 to 600 were also key parts of the campaign.
Yet there was always more to Rennard’s approach to winning elections than simply delivering leaflets. He set out his campaigning style most clearly in a guide to winning local elections, originally written for the Liberal Party but anticipating the merger. He used to offer would-be candidates a bet: he was so sure that if they did everything in the book they would win, that if they followed everything in the book and failed to win, he would refund the cover price. According to the tales, he never had to make a refund. The book covers more than just leafleting, including a key place for a regular ‘residents’ survey’, asking people what issues are important to them and what problems in the area needed fixing. Find out what the public is concerned about, then tackle it and report back regularly through repeated leaflets was the formula.
It was both a winning formula and became a controversial one for its focus was on volume of activity and populism, taking up issues that concerned people rather than preaching political philosophy at them. It was clearly effective but always risked, especially in untrained or naive hands, lapsing into crude populism without a distinctive liberal tinge. This point is explored further below (see The Rennard strategy).
Both the electoral effectiveness and ideological doubts of this approach were heighted by the heavy emphasis on appealing to tactical voters. It was an effective way of turning the disadvantages the third party faced under first past the post on their head, making the electoral system count against one of the other main parties within a particular constituency.
Where the third party was in second place (or could plausibly argue that it had moved into second place since the previous election), a strong appeal could be made to the supporters of the major party that found itself in third place, arguing they should switch their votes to the Liberals/Alliance/Lib Dems, as their favoured party could not win but by switching they would at least be able to stop the other main party winning. Moreover, in making the tactical voting case, the party was also implicitly making the case that the party could win, at least in that seat, which was also an effective tactic for appealing to voters, many of whom were willing to vote for the party as long as it would not end up being a wasted vote.
A key part of making this point effectively was to present it graphically, in the shape of a bar chart. The origins of the first bar charts on leaflets painting a contest as a two-horse between the party and its main rival, with the other party or parties labelled as unable to win are lost to history although most likely date to the late 1970s. By the early 1980s Chris Rennard was promoting tactical voting, authoring a guide for the Association of Liberal Councillors that included advice on how to use what he then called ‘block diagrams’. By the time of the Liberal Democrats the use of tactical voting appeals illustrated by bar charts was a major part of the party’s target seat campaigns.
Another distinctive feature of Liberal Democrat campaigns, less remarked on than bar chart but as with them ending up widely copied by other parties, was the use of four-page constituency newspapers. Given a non-party masthead to encourage readership, and with no prominent logo on the first page, these newspapers made clear who they were from but were written in a third-person local newspaper style and designed to attract the interest of people who would not normally give a conventional political leaflet a second glance.
The delivery of a newspapers, including one over the last weekend of the campaign containing an extremely strong tactical voting message, became a staple of the party’s electioneering for decades until reductions in printing prices, more generous expense limits and imitation from other parties made newspapers spill over into repeated appearances during a campaign.
As with other techniques, such newspapers often showed up the variation in skill levels amongst campaign teams. The newspaper produced for high-profile by-elections usually had production quality matching the independent local media. However, the format of the item showed up particularly harshly any falling short of such standards, with local campaign teams not in receipt of direct central artwork support not infrequently producing newspapers so amateurish in appearance as to undermine the concept.
Yet such newspapers continued to be produced because, as with bar charts and other parts of the successful campaign paraphernalia, in less skilled hands they became totemic concepts to copy. Copying concepts is much easier, of course, than copying quality – and with the rationale behind such paraphernalia rarely discussed widely in the party, it became easy for people to copy poorly without understanding why what they were doing was inferior.
‘Blue ink’ letters
In the search to find formats that would result in voters actually reading political literature, hand-addressed envelopes containing a reproduced handwritten letter also became common. Typically the letters used blue paper and envelopes, with the printed handwriting often in dark blue ink, giving them the name ‘blue ink letters’. The first such blue ink letters were invented by Rennard when he was organising the Liberal breakthrough in Leicester in the 1980s.
As with other campaign techniques, such as the old Liberal Party’s habit, born in 1970s Liverpool, of delivering a ‘Good Morning’ leaflet before voters had woken up on polling day, they started as a distinctive technique and ended up being copied by the other parties. Indeed, by the time of the Cheadle by-election in 2005, not only were the Conservatives closely copying the campaign tactics of previous Liberal Democrat by-elections, but many of the standard phrases and wording used in them were also being copied. It led to increasingly questions over whether the party’s campaign techniques were fresh enough and modernising at a pace to keep up with the other parties.
The Rennard strategy
Part of these criticisms was a view that Rennard-style campaigning added up to less than the sum of its parts. By fighting a series of very intensive local constituency campaigns, each shaped by the varying particular issues and concerns in the individual constituencies – so the argument went – the party was failing to build up a clear, consistent image of itself or carve out a clear core vote. Instead, it was accumulating different sets of diverse supporters in various seats, with the only possible progress coming from putting together new bespoke coalitions of supporters a few seats at a time.
This criticism can be overdone, for the party’s polling actually found rather similar results across its different key seats. The typical key-seat constituency polls with a sample size of around 450 found a consistent pattern – both around the country and across the years – of voters liking hard-working Liberal Democrats, with strong local credentials, who concentrated on the issues that the public told the pollsters were the most important to their family. These were usually health, crime and education, with issues such as constitutional reform and Europe as a result being heavily downplayed.
If anything, the problem was not so much the lack of policy consistency across different seats, for there was high consistency at Rennard’s instigation, but rather that the ‘we’re nice, hard working and concerned about the same issues as you’ message was not sufficiently distinctively liberal. As a result, the party’s pitch was one any other party could – and sometimes did – copy and match. By concentrating on promoting the virtues of individual candidates, their local connections and their local campaigns, the approach built up support for individuals who happened to be Liberal Democrats, rather than for the party in its own right. This made it hard to transfer that support to their successors as candidates. Where an incumbent MP was standing down, there was some scope for them to support the building up of their successor. However, where the party had come close but not won and the defeated candidate retired, or where an MP was defeated and a new person was seeking to regain the seat next time round, it was extremely hard to pass on the accumulated support for their predecessor to the new person.
Rennard’s response to this was partly that it was a better approach than anything that anyone else had tried or suggested, in that it did at least get Lib Dem MPs elected in far greater numbers than before. He also argued that there was a long-term strategic element to it: the aim was to build up the number of seats the party could win until it was big enough to force a hung parliament. At that point, he hoped, the party would be able to get the rules of the game changed, with a new voting system and party funding reform then letting the party fight future elections on a level playing field with the other main parties.
On that basis, the election-fighting part of the strategy almost delivered in 2010. But the mathematics of the hung parliament did not make a Labour–Lib Dem deal feasible, and so the party’s negotiating position for electoral reform was weakened. That plus the subsequent disastrous AV referendum campaign meant the strategy did not end in triumph.
Where the party did succeed in changing the rules of the electoral game – with electoral reform for Scotland, Wales, London and the European Parliament elections – this simply extended the controversy. For critics of Rennardism, the party’s failure to shine in those ballots conducted by versions of proportional representation showed its limitations. If the party could not do well in elections of the very sort it wanted, freed from the shackles of first past the post, what long-term future for the party was there?
The answer to that, for Rennardism’s defenders, was that the future required more of the same. Namely that the best results in PR elections, where votes were counted up over a much larger area than a single Westminster constituency, were garnered by concentrating activity on the strongest constituencies and wards within those larger areas. Rennard-style concentrated campaigning generated more votes overall than alternative approaches to spread the party’s activity more thinly over wider areas.
A related criticism frequently made, especially in the pages of Liberator magazine, was that by concentrating party resources on a limited number of target seats, the party failed to grow in other parts of the country and instead was left to whither. In truth, though, the number of seats being seriously fought by the party doubled across the 1997, 2001 and 2005 parliaments before Chris Rennard’s departure as Chief Executive.
Following the 2005 election, within the Campaigns Department there was a limited review of the general election result and revision of the party’s template campaign strategy for target and held seats. More generally there was debate in the party over how the general election was conducted, and whether the result constituted a missed opportunity given the political damage to Labour thanks to the fallout from the Iraq war. The party had some dramatic wins from Labour, such as in Manchester Withington and in Hornsey & Wood Green, but should there have been more given the political environment offered up by the Iraq war and resulting widespread centre-left disenchantment with Tony Blair?
Within the Campaigns Department the thinking was much more about how the party had performed against the Conservatives, and the failure to win a string of seats that on paper were promising prospects and appeared to have run the sort of campaign which had in the past resulted in victory.
After the 2001 general election, the implicit lesson drawn by the party was that a very effective key-seats operation could only get the party so far, and in addition to an effective ‘ground war’, the party needed to be better at the ‘air war’ in the national media and the like so that the ground war took place against a more welcoming backdrop. As a result, and in one of the least frequently commented on episodes during Rennard’s time at party HQ, the party’s press team was roughly doubled in size. The ground tactics changed little, for as in 1997 there was a compellingly clear and neat pattern of the seats which followed the standard Rennard campaign template winning and those that did not, losing.
For 2005, the pattern was much less clear cut, and as a result the campaign activity template that seats were expected to follow underwent significant changes, including a big increase in the volume of direct mail. This was in part a reaction to the large Conservative direct mail operation in 2005, although whilst the latter relied overwhelmingly on postage, the lower-budget Lib Dem operation was reliant on hand-delivered direct mail.
More widely in the party this still left a sense that further modernisation of campaign tactics was required. Such modernisation was part of Ming Campbell’s pitch for the party leadership in 2006 and, after he won, a campaign review headed up by Ed Davey MP was carried out. This included a study trip to Canada and the US to learn lessons from sister parties there. Before there was a chance to see the result of these changes at a general election, Chris Rennard departed as Chief Executive.
The new Chief Executive, Chris Fox, set out to organise party HQ in a very different style (more flip charts and less bar charts was the quip). His Director of Campaigns, Hilary Stephenson, nonetheless took an approach to campaigning that was very rooted in the mode of the 1997 and 2001 successes, and as a result the focus was on fighting those sorts of campaigns better and on a larger scale rather than significant changes in the party’s campaigning style.
The 2010 general election saw a huge bump in the opinion polls for the party after the first TV Leaders’ Debate, and as a result the party started putting effort into a much wider range of seats. In addition, the focus of party campaigners on targeting weakened, as more activists started staying at home to work their own seats, thinking they could win, rather than travelling to near-by more marginal seats.
As a result, when the votes came in the 2010 election was the worst for the party since 1992 when judged by the campaign machine’s ability to turn national vote share into actual seats (see Table 1). For pessimists this was the result of the Conservatives in 2005 having largely cottoned on to how to do intensive key-seat campaigning, and by 2010 Labour doing so too, leaving the party’s ability to out-perform the national picture in selected constituencies hugely reduced.
One issue that was clear is that the party called several seats wrongly in the last few days before polling day, misdirecting resources as a result. For example, a great deal of effort was directed to Oxford East on polling day, which Labour held on to by a significant margin – 4,581 votes – whereas, had the effort been directed instead to neighbouring Oxford West & Abingdon, Evan Harris would not have ended up losing by just 176 votes.
The party’s own official review was a relatively low-key affair. It made many detailed recommendations, and some significant organisational ones – particularly that the party should change its computer database software for fighting elections. In addition, the increasing emphasis in the Labour Party on the virtues of canvassing has rubbed off on the Liberal Democrats, with a switch from viewing canvassing as a data-gathering opportunity, where a virtue is made of talking to each person for as little time a possible, to an attempt to get into longer conversations about issues.
Whether such changes are the right ones and are radical enough, remains to be seen with the 2015 election yet to cast its verdict.
The Liberal Party, then the Alliance and subsequently the Liberal Democrats all set out to remake British politics. With the post-2010 coalition government not yet run its course, it would be premature for historians to cast a verdict on the extent to which this aim was achieved.
However, what is already clear is that the third party did remake British political campaigning. It was both a tribute to the success of its tactics and a frustration to its further success that other parties ended up so heavily copying its techniques. Whether it is the regular appearance of a leaflet outside of election time, heavy focusing of resources on a limited number of target seats, emphasis on the local roots and local campaign record of candidates or the widespread use of bar charts, electoral politics in Britain has followed where the third party led.
Dr Mark Pack worked at Liberal Democrat HQ from 2000 to 2009, and prior to that was frequently a volunteer member of the Parliamentary By-Election team. He is co-author of 101 Ways To Win An Election and of the party’s General Election Agents Manual.
 For example, see David Denver and Gordon Hands, Modern Constituency Electioneering: Local Campaigning in the 1992 General Election (Frank Cass, 1997) and Justin Fisher, David Denver and Gordon Hands, ‘Unsung Heroes: Constituency Election Agents in British General Elections’, British Journal of Politics and International Relations, vol. 8, no. 4, 2006.
 This gap is not unique to the Liberal Democrats or the tactics that party pioneered. Phil Cowley, now co-author of the Nuffield general election series, has made the point to the author that he and his colleagues know very little about direct mail’s impact on election results even though it is clear from his conversations with campaign organisers that they put much time and money into producing it. As a result he has posited, half in jest but also half seriously, ‘Cowley’s Law’, namely that the level of media attention given to a campaign technique is in inverse proportion to the importance given to it by campaign organisers.
 Bernard Greaves and Gordon Lishman, The Theory and Practice of Community Politics (Association of Liberal Councillors, 1980).
 The most notable of the names that provides continuity from the Liberal Party through to the Liberal Democrats is Chris Rennard. However, there are others too, such as Tony Greaves.
 This applies to some of the Liberal Democrat campaign publications which have my own name on the cover, such as the 2011 pamphlet co-authored with Shaun Roberts, Campaigning in your community. The origins of parts of the text can be traced back through to 1970s Liberal Party publications.
 ‘There were many areas of Liberal local government strength where little or no attempt was made to convert this to parliamentary elections and indeed even those with much experience of local election success had not much idea of how to adapt community politics techniques for parliamentary elections. In contrast many of the seats we did hold were not really bastions of community politics. Campaigning in Liberal-fought South Leicester and SDP-fought Watford in 1985–87, I found the former more dynamic in terms of winning council seats, but the latter more focused on the parliamentary election,’ said Iain Sharpe, now a Watford borough councillor, to the author at the Liberal Democrat autumn federal conference, September 2013.
 The same image would also cover the predecessor Liberal Party’s approach to campaigning over several decades too. The distinctive features of the party’s approach to campaigning, leaflets, bar charts and all, came overwhelmingly from the merged party’s Liberal, rather than SDP, heritage.
 Director of Campaigns and Elections 1989–2003 and then Chief Executive 2003–9. He was created a life peer in 1999.
 This reluctance to move resources and people in order to concentrate on target seats is one the other main parties have encountered too, and indeed is a reason why even as they saw the success of targeting for the third party they were not able to match it because of greater resistance to moving effort between seats from within their own organisations.
 Technology has increasingly made readying a candidate’s freepost election address a less labour-intensive operation than it used to be. The ability to print names and addresses and to automatically fold and stuff literature into envelopes has made their production increasingly something for machines rather than volunteers. The impact of this on the willingness of volunteers therefore to move to help in a different constituency is an unresearched area.
 Conversation with the author, Liberal Democrat autumn federal conference, September 2013.
 For example, the autumn 1996 federal conference in Brighton had presented to it a draft outline federal budget for 1997, a draft outline campaign development fund budget for 1997 and a by-election fund statement of income and expenditure for 1995. The 1997 federal budget envisaged expenditure of £1,487,300. The campaign development fund was at £292,000 and the previous year’s by-election fund had spent £149,909. In addition, there was a general election fund mentioned but with no details provided. Moreover, very little detail was provided for either the campaign fund or the by-election fund, each only getting one side of summary figures. The mainstream federal budget, by contrast, came with more than a dozen pages of detail on expenditure in the previous year.
 The different personalities of successive party leaders also played a role. As the Helen Morrissey report, Process and Culture within the Liberal Democrats and recommendations for change (2013) put it, ‘The relatively relaxed management style of the party leader, Charles Kennedy, compared with Paddy Ashdown who has a military background, also contributed to Chris Rennard’s power base.’
 It was also reflected in similar publications in the other main parties too.
 This changed somewhere following the Political Parties, Elections and Referendums Act 2000 with its introduction of national expenditure controls.
 The most common route to recruitment by the Campaigns & Elections Department during the Chris Rennard period was to first have been a winning election agent. Even those who did not follow this route, such as the author, usually had other local campaigning credentials.
 This was particularly a feature of Jeremy Thorpe’s leadership, during which there was secretive fundraising for campaign activities with the wider party kept out of the loop.
 For the early general elections in the Liberal Democrats’ history, a separate general election organisation was created. Integrating the general election campaign into the party’s usual operations at subsequent general elections was seen across the party as a sensible reform – and one which therefore also gave more power to Chris Rennard.
 The Eastbourne by-election was also the first by-election to make use of a relatively small sample constituency poll, of the sort that became a regular staple of key-seat campaigning under Chris Rennard. American political consultant Rick Ridder had first introduced very limited polling to the party at the end of the Richmond by-election campaign, but Eastbourne saw the first poll carried out early enough in the campaign to be used to help make decisions on messages and tactics. See Rick Ridder, ‘How Margaret Thatcher Advanced My Consulting Career’, Campaigns & Elections, May/June 2013.
 Systematic analysis over decades of the volume of leaflets delivered during election campaigns is absent from British politics. However, there are plenty of fragments of evidence that substantiate anecdotal claims about Liberal Democrat campaigns featuring a tidal wave of paper being pushed through letter boxes. For example, the Nuffield general election series publications for the 1960s and 1970s show how even in marginal seats two or three leaflets in total were the norm during a general election campaign. By the time of the Liberal Democrat election guides for the twenty-first century, campaigns were being advised to use that many leaflets simply on election day itself.
 Chris Rennard, Winning Local Elections (Association of Liberal Councillors, January 1988). The central importance of leaflets is revealed by the explanation of strategy in the book’s foreword: ‘The book explains a strategy for building a campaign. Not just what leaflets should say but when, how and why – with examples’. This quote illustrates both the key features of Rennard’s very successful approach to winning elections and also two of the features that most often attracted criticism – ‘strategy’ meaning a campaign plan for a specific election and campaigning meaning leafleting.
 As tactical voting became more talked about in the 1980s and early 1990s there was a lively academic debate on how many people voted tactically. The debate was fuelled in part by disagreements over how tactical voting should be precisely defined for the purposes of academic research. However, the party’s own polling in marginal seats consistently showed that being able to win over the supporters of the third-placed party, along with a more general message showing that the party was able to win at all, was an extremely important part of building up a winning share of the vote in such seats.
 The earliest bar chart the author has located is on a Richmond, London leaflet from 1979 (http://www.markpack.org.uk/35442/the-first-bar-chart-richmond-in-1979-possibly/). However, despite questioning several of the main campaigning experts from the 1970s, the answer to who started the use of bar charts, when or where, has not been identified. Anyone able to beat the 1979 date with an earlier example is most welcome to contact the author on email@example.com. (UPDATE: Since first writing this piece, I have further researched the first bar chart: http://www.markpack.org.uk/107519/first-tactical-voting-bar-chart-appear-political-leaflet/),
 However, it should be noted that in the first European elections by PR, in 1999, Rennard himself was a supporter of trying out some thin, blanket activity. He was instrumental in securing the funding to ensure that election addresses via the Royal Mail’s election freepost facility went out to the whole electorate. This produced little apparent benefit for the party, and in future European elections the party increasingly moved towards concentrating its funds on more and better election addresses in its stronger areas, with lower quality, cheaper and fewer election addresses elsewhere.
 This is based on the (unpublished) lists of seats circulated within the Liberal Democrat Campaigns Department during this time.
 The party’s use of ten we propose/we oppose statements during the 2005 election was a particular cause of debate. Each of the ten individual pairs had been market researched heavily ahead of the election and each was generally popular both within the party and with the public. However, many people felt that the ten individual policies did not add up to a coherent overall message or image for the party. Chris Rennard himself afterwards likened the process to being like having had all the right ingredients for a meal but not a recipe.
 Implicit in that there was not a specific post-election review which came to this conclusion. However, it was conclusion that Chris Rennard came to and communicated to others, and subsequent budget decisions flowed from this.
 This is a somewhat imprecise term as many of the press team were formally employed by the Parliamentary Office of the Liberal Democrats and reliant on state funding, which limited the range of activities they could engage in. However, as far as the outside world was concerned there was a group of press officers based in the party’s HQ and this was seen – and in practice acted as – the party’s press team.
 Nick Clegg’s commitment during his successful run for party leader to double the number of the party’s MPs added to this sense that what was needed was to fight more seats to the traditional intense template rather than to change what the template contained.
 A defence of this dissipation of effort is that 2010 appeared to offer a once in a generation opportunity to win new seats. The party may have gambled and lost by spreading its effort thin, but in the face of such an opportunity it is only hindsight that tells us going for the big prize was the wrong move.
 Some in the party blamed poor constituency polling for these misjudgements. However, given the availability of data also from other sources, simply blaming the polling is unfair on the pollsters.
 It did. After a competitive pitch process, the old EARS program was replaced not by a new product from the same team but by a product called Connect, based on NGPVAN, a US programme used by the Barack Obama presidential campaigns and by many other Democrat campaigns as well as by the Canadian Liberals. By 2014, only a small number of local parties continued to use EARS.