Skip navigation.

The challenges of opposition

Jane Green


Lessons from recent research into electoral behaviour, campaign strategy, and opposition party recovery and change.


There is a classic paradox in the study of political parties: when they are defeated, they act in ways which prevent them from regaining power.

The Labour Party of 1983, and the Conservative Party of 2001, appeared to confirm this view. Both parties were criticised for fighting ineffective campaigns. Both were described as lurching to the left, and right, respectively. They had both experienced internal pressure on policy and ideological direction. Each had unpopular leaders who were accused of eschewing polling research.

They focused their election campaigns on low salience issues which made them appear out of touch and preoccupied with internal policy commitments – rather than the issues of concern to the wider electorate. These were campaigns that history remembers as a play to the base. Rather than a ‘vote maximising’ strategy addressing a broad number of salient issues in line with the policy views of the majority of voters, both parties appeared to go after their core vote. The result, in both cases, appeared to be a self-inflicted prolonged period in opposition.

The reasons given for these campaigns in 1983 and 2001, and for both parties’ responses to defeat, are related to classic and predictable challenges of opposition. Those challenges are very similar – if not identical – to the ones the Labour Party faces today.

This article defines the classic challenges of opposition, revealing the arguments a new Labour leader will encounter. The article also presents an alternative interpretation of the 1983 Labour campaign and the Conservative campaign of 2001. It challenges the application of core vote explanations and shows that parties need not be driven off vote-maximising strategies if they understand the true constraints of their electoral context. The article concludes by highlighting more important challenges – and opportunities – facing the Labour Party in its current period of opposition.


Classic opposition problems

The following features of opposition reflect those discussed in extensive academic work, highlighting incentives which force political parties away from vote-winning appeals.


A narrow base

A defeated party will have lost a lot of votes, obviously, but its more stable voters may also have voted for other parties or stayed at home. Without these voters a party cannot build, and although more committed party identifiers are diminishing in number in Britain (and in many Western electorates), their votes are important. These voters may be the first to return when the party remains in opposition, sharing the party’s philosophy and being rooted in social ties and relevant communities.

These voters, it is argued, have different issue concerns and hold more polarised opinions: they are further to the left (or right) than more moderate swing voters. Defeated parties face particularly strong incentives, therefore, to focus on traditional issues and on policies designed to appeal to these core and pivotal voters. Remaining MPs will be those in the safest seats, and those safer seats will be populated by traditional voters in higher proportions, giving these voters a particularly strong voice in any debate regarding party renewal and change.


Weak organisation

Money and members will be in short supply, and the central party organisation will be depleted. A convincing victory sends a powerful signal about the popularity of a party. When Tony Blair took the Labour Party into government in 1997, Labour Party membership increased more after that election than it did before it. A convincing election loss has the opposite effect. A defeated party, already weakened by a period of unpopularity in government, may encounter further losses in money, members and votes. For activists and party members, the experience of being unpopular on the doorstep, their efforts being spent in vain, the loss of even more members and donors, and the experience of being relegated in importance, will all mean that the party machine requires considerable commitment.

The demand to increase membership, to energise activists, and to encourage donations can motivate parties to focus on more traditional policies. It has been argued, for example, that because the Conservative Party deepened member opportunities for policy influence between 1997 and 1999, this had the effect of pulling the party away from the issues important to the broader electorate. As I argue below, this reflects a misunderstanding of the concerns and priorities of activists, who primarily want to see the party re-elected.


Discord and disloyalty

Internal divisions, legacies and disunity can all appear when the relative discipline of government has gone. Opposition is viewed as a prime opportunity by competing factions to redirect a party towards preferred policies. The parliamentary party may be disgruntled, some of the best talent depleted, those MPs who served in government may be seeking backbench roles, others may be tainted with the blame of office, and new MPs won’t be entering parliament. Without realistic rewards, MPs have fewer immediate reasons to be loyal.

All of this means that a new leader has to provide a particularly convincing strategy, but with a weak position in the polls they will come under continual questioning and pressure. These problems characterised the leaderships of Michael Foot, William Hague and Iain Duncan Smith. The electorate sees a party not ready for government, with unclear policies and plagued by damaging disunity. The temptation to focus on a small number of issues on which policy is already clear, understood by the electorate, and where there is a clear difference between the parties is very strong – all of which may push a party onto issues of traditional philosophical commitment.


Permission to be heard

The message is difficult to define. The message of opposition – of attacking government policy decisions and spending – can all too easily backfire on the previous party in office. The defeated party’s approach and philosophy in government may have been discredited, ideas exhausted, and little attention given to a different approach.

In addition, not many people are now listening. The media are less interested, voters are less interested, and the ability to shape the debate and agenda is considerably reduced. In this context it may be tempting to reach out to old friends – media who have a bias to the party – and to provide single-issue stories which are easy to report and which fit with existing perceptions.


The distraction of introspection

Defeated parties want to understand the reasons for defeat, and while they do that, they point fingers, provide competing and inconclusive diagnoses, and waste considerable time trying to find a magic bullet solution (which does not exist) to undo the costs of government.

An inwardly focused party could spend years of a four or five year parliament analysing why it is no longer in power, then failing to mount an effective challenge at the next election when it has a chance of redressing its position. 


Some of these features of opposition will be more important than others for today’s Labour Party. New challenges face a party holding two parties in coalition to account. An election could come in five years’ time, giving the party breathing space to organise. It could also be prompted earlier, putting enormous pressure on a strategy of renewal. Much will depend on the new leader, their electoral appeal, and on the potential fall-out from cuts in public spending.

Insofar as things can be predicted, however, I offer three lessons. These lessons stem from my own recent research (see references below) into electoral behaviour, campaign strategy, and Conservative opposition party recovery and change.


Lesson one: parties are not constrained by the base 

Many assumptions about core voters have not been tested by social research. It is well-known that Labour’s working class base declined in the post-war decades and that this forced the party (and other social democratic parties in Europe) to move towards the political centre ground. It is less well-known that Labour’s previous core supporters showed remarkable loyalty when Labour adopted this shift towards the centre during the 1990s. This ideological shift was not sufficient, per se, to alienate the traditional vote of the Labour Party.

Labour’s base abstained in significant numbers in recent elections, but not between 1992 and 1997, when the most symbolic ideological changes occurred, and then only dramatically in 2005 and in 2010. Abstention began to increase when the party redirected its efforts away from its core constituencies, and towards the constituencies it needed to retain, but few Labour voters initially drifted elsewhere. Among those voters who perceived growing policy differences between themselves and the party, it was only on redistribution where this precipitated a decision to stay at home. Even by 2005 much of Labour’s base could not bring itself to change loyalties to support a different party. By 2005 and 2010 the primary reason Labour’s voters failed to turnout was due to Iraq, and due to trust – not to ideological moderation or to a range of policy performance ratings.

The base will tolerate significant and symbolic changes, I argue, as long as trust, integrity and concern for voters remain. The reasons for this are threefold: (i) the alternative party is still perceived as far worse; (ii) voters change their views alongside those of their party; and (iii) voters hold attachments for other reasons. These explanations go a considerable way to accounting for the loyalty of voters during periods of symbolic ideological and policy change. The same has been true for David Cameron’s Conservatives who, despite popular caricatures, broadly supported the party’s move away from a focus on traditional Tory concerns towards the centre ground.

If Labour wants to shore up its traditional vote base, it should dramatically distance itself from the decision to go to war in Iraq; it should renew a balanced focus on fairness and equality; and direct greater campaigning efforts at the sections of the electorate on which it has traditionally relied. This will not win the party a majority (which will still be won in middle England), but it will increase the party’s electoral mandate and its grassroots support. This, crucially, need not be at the expense of broadening the party’s electoral appeal.


Core voters do not have very different issue concerns

It is often argued that focusing on traditional voters’ concerns means campaigning on a narrow set of issues: unemployment, welfare, and the NHS (Labour); or Europe, crime and immigration (Conservatives).

These stereotypical pictures are not straightforwardly borne out in issue salience data over time. When unemployment is low in salience, it is low in salience for all parties’ voters, including the Labour base. When it becomes an important issue, it naturally becomes more important for those it affects (typically Labour voters). This means that when the NHS was the number one issue, it was the number one issue for Labour voters, Conservative voters, Liberal Democrat voters and others alike, because it was an issue which affected almost everyone. When terrorism hits the news, this is also the number one concern shared across the electorate. As immigration has risen in perceived importance, Conservatives are more likely to think this issue is a concern, but the rise in salience is very closely mirrored among Liberal Democrats, Labour voters and voters with no party affiliation. Right now, ‘it’s the economy, stupid’, for practically everyone.

The alleged trade-off between campaigning on different issues to energise various constituencies is a misnomer. It is not completely wrong, but it depends hugely on context – on what is important at a given time. A party’s own voters are also significantly more likely to follow the direction and agenda of their own party’s campaign.


The electorate is less ideologically polarised than during the 1980s

The nature of the electorate corresponds closely to changes in party ideologies. During the 1980s there were large perceived differences between the Conservatives and Labour. In 1987 just under 85 per cent of survey respondents agreed there was ‘a great deal of difference’ between the main parties. By 2005 that figure was just 23 per cent (British Election Study data). This change in perceived party polarisation – and depolarisation – has been matched by a decline in polarisation within the electorate.

There are fewer voters locating themselves at the far left or far right of various policy scales. The gap between the views of Labour and Conservative partisans has narrowed, coupled by a weakening of party attachments. This depolarisation in the electorate means there are fewer voters to constrain a party by pulling it to the far left (or right). This, furthermore, is not just a product of a declining working class – the same changes have taken place within this group. The decline is also evident among groups with different education levels, different levels of political knowledge and political engagement, and among the highest and lowest income quartiles. Fewer strongly identifying voters exist in the electorate; there are far more voters who say they support a party weakly; and there are more voters willing to support ‘others’ or to not identify with parties at all.

The reverse pattern is found in the US, where growing differences between the Republicans and Democrats have been matched by larger policy differences between partisans, as well as strengthening attachments to parties. There has also been an increase in the way voters ‘bundle’ their policy positions: a Democrat is now more likely to take Democrat positions on a range of issues, and more likely to link their attitudes to their party support. These tendencies have declined in the British electorate: voters in Britain are less likely to support a party on the basis of policy, and less likely to hold coherent ‘bundles’ of attitudes on a range of policy questions fielded in surveys. Simply put, the trends in Britain and the United States are in the opposite directions and British voters are much less clearly partisan in nature.

The causal direction between party and voter policies remains a topic for further research (although there is growing evidence of a party-led process), but the implications are already fascinating: parties in Britain need not be as constrained by more ideologically polarised voters (though US parties have more energised supporters as they jointly take contrasting ideological positions).


Lesson two: parties are constrained by issue ownership

The Labour party of 1983 and the Conservative party of 2001 had a limited set of issues which were favourable for those parties. These much derided campaigns were not appeals to the base, but they were strategies determined by the broad electoral context whereby neither party was rated positively on the salient issues of the day.

When a party’s popularity is low, the remaining issues on which the party is rated positively, competent or most trustworthy tend to be those on which the party has a longstanding reputation: these are the issues a party ‘owns’. Ownership of an issue relates to the association of a party with an issue and its perceived competence. These issues tend to be those on which a party is rated positively among voters beyond the existing base. This results in electoral dividends when those issues are high on the public agenda, providing an incentive to raise the salience of these issues by addressing them in communications and campaigns (and to diminish the salience of the best issues of an opponent). Core voters tend to rate their party positively across most issues, and so to appeal to these voters, there is little trade-off in issue appeal. To appeal to additional voters, there is a compelling argument to focus on issues of traditional strength.

The electoral incentive to focus on such issues is particularly strong in opposition. When a party is more liked and trusted, it does better on all issues – including the issues it owns. When it is least liked and trusted it does worse on all issues, but a party’s owned issues will provide a minimal but remaining advantage. Owned issues tend to be issues on which a narrative is already clear and developed, and because they are owned by the party in opposition, they may represent the only issues for which the new party in government is decisively mistrusted, offering the clearest and most potent attack lines.

The trade-off appears when a party’s few positive issues – its owned issues – are low on the public agenda. To focus on a broad set of issues risks handing other parties their best cards. Focusing on low salience issues risks making a party appear out of touch. This was the real electoral dilemma facing Michael Foot and William Hague.

Party ratings on issues tend to move up and down in common, akin to a public mood about which party is most trusted to handle the issues of the day. When a party is gaining in the polls, the palette of positive issues broadens, including new issues as the incumbent party incurs the costs of government. A party in opposition can therefore campaign on new issues to demonstrate a broad appeal, unless the government has a commanding advantage on those issues. They can focus on their owned issues when those issues are also high on the public agenda but avoid focusing only on owned issues when they lack sufficient salience. Ultimately, however, they must wait for incumbent ratings to fall and for their party to gain in popularity, with the benefit of gaining issue ratings on traditional issues and other issues alike.

The potential for this to occur for Labour looking ahead is much greater than it was for Michael Foot in 1983 or William Hague in 2001. The Labour Party of 2010 onwards can, therefore, afford to be patient.


Lesson three: policy is just a small part of the answer

There were many reasons – beyond the change to New Labour – that Tony Blair delivered a landslide victory for Labour in 1997. The greatest of these was arguably the extraordinary extent of the problems experienced by the Conservatives.

Similarly, the Conservative Party’s modernisation and change under David Cameron was a small part of that party’s success in 2010. It is difficult for political scientists to gauge these subtle and gradual electoral effects. However, it is undeniable that David Cameron’s vote share increased in the autumn of 2008, as the economic recession occurred, and diminished as the country neared the election, as Labour was again judged (by a tiny margin) the ‘best party on the economy’. As Labour’s support waned during the economic crisis, the Conservative lead on a range of criteria went from negative to slightly positive. These features suggest that much of Labour’s recovery will be due to the performance of the Coalition, rather than a process of policy renewal.

The Iraq war and a lack of trust in the government took a significant toll on Labour prior to Gordon Brown’s leadership, and this effect was sustained into 2010. However, while the Labour government did lose large numbers of voters, there were larger proportions of people who still retained a self-expressed affiliation with the party.

Voters also choose parties based on their agreement with a philosophy and specific policy approach (although less than before, as explained above). However, the preoccupation with electoral recovery via ideological moderation or polarisation – with issue-by-issue campaigns and with daily, weekly and monthly strategy – places disproportionate emphasis on a small part of the electoral solution.

Such a preoccupation holds the risk of destabilisation. When an approach doesn’t work, colleagues call for a different argument. When the party fails to make inroads, it changes its emphasis and its tone. When a leader comes from a certain ideological wing, and is popular, the policies are judged a success. If the leader is unpopular, the policies are judged a disaster. Neither outcome may have anything to do with the policies; or with the leader for that matter. The short-term demand for success and the lack thereof helped secure the Conservatives’ place in opposition. It was precisely an impatience for electoral improvement which seemed to make the party appear that it had surrendered this goal.


The leadership question

One final lesson relates to support for the party leader. It is not realistic or reasonable to expect voters to re-evaluate their perceptions of an unpopular party overnight. Furthermore, leader popularity or unpopularity is sometimes the consequence of party approval or disapproval ratings, rather than its cause.

At the beginning of William Hague’s leadership in 1997, net satisfaction ratings were the highest they were to become. They plummeted shortly after initial interest subsided in the new Conservative leader. Exactly the same trend then occurred for Iain Duncan-Smith. An initial peak was followed by the same negative trajectory. It seemed that whatever these leaders did, and whoever they were, their satisfaction ratings saw the same decline over time. The same trend happened again for Michael Howard and it even began to occur for David Cameron. His net satisfaction-dissatisfaction ratings saw a decline at first – most notably when Gordon Brown was selected as Labour leader – and then became positive only in 2008 – just two years before the election.

It is possible that until Labour becomes more popular, any leader of the Labour Party will not enjoy a consistent poll lead. The challenge for any party is to learn the difference between an unelectable leader and a party that is not yet electable under any potential Prime Minister; and to learn to be patient.

Success breeds success. Party membership will increase, money will return, activists will become inspired and loyalty will become easier when Labour’s position improves in the polls. Part of the story will come from a strong leader, part from party unity, part from a clear and coherent message. The majority will be from the costs of governing for the Coalition. Learning this lesson, and the lessons detailed above, may help Labour avoid some of the pitfalls of previous opposition parties.



Green, J. (2007) ‘When parties and voters agree: valence issues and party competition’, Political Studies 55 (3): 629-655.

Green, J. (2010) ‘Strategic recovery? The Conservatives under David Cameron’, in Geddes, A. and Tonge, J. (eds) The UK Votes: The General Election of 2010, Oxford, Oxford University Press, 85-106.

Green, J. (2009) ‘A test of core vote theories: the British Conservatives, 1997 – 2005’, working paper, available on request.

Green, J. (forthcoming) ‘Ideological moderation and permission from the base: the British Labour Party’, working paper, available on request.

Green, J, Adams, J. and Milazzo, C. (forthcoming) ‘Has the British public depolarised along with political elites? An American perspective on British public opinion’, Comparative Political Studies 45 (4).

Green, J, Adams, J and Milazzo, C. (forthcoming) ‘Which voting subconstituencies have reacted to elite depolarisation in Britain? An analysis of the British public’s policy beliefs and partisan loyalties, 1987-2001’, working paper, available on request.

Green, J. and Jennings, W. (forthcoming) ‘The dynamics of issue competence and vote choice for parties in and out of power’, working paper, available on request.

Green, J. and Jennings, W. (forthcoming) ‘Macro-competence: the measurement of mood in issue competence’, working paper, available on request.