Polling station

Home / publications / primers /

Public Opinion and Public Policy: Complexities of the Democratic Mandate

24 May 2013

This policy primer examines the relationship between public opinion and migration policy, with a particular focus on the idea of a democratic mandate for reducing the number of immigrants to the UK.


Does public opinion provide a democratic mandate for government to reduce net migration to Britain?

A key issue in migration policy is its relationship with public opinion. In particular, does widespread public opposition to immigration create a democratic mandate for government to reduce immigration or net migration to the UK? Critics have charged a democratic deficit in policy-making on migration, as migration to Britain increased substantially in the late 1990s and into the 2000s, despite consistent public preferences for decreased levels.

Polls consistently show that immigration is not only broadly unpopular in Britain, but also one of the most important issues for members of the public. In the government’s 2009-2010 Citizenship Survey, 77% of respondents who had an opinion favoured reducing immigration, including 53% who preferred reducing it “a lot”, a result firmly in line with previous polling over many years (see the briefing on ‘UK Public Opinion toward Immigration: Determinants of Attitudes‘). Immigration also consistently ranks among the most commonly-cited responses to poll questions asking the most important Issues facing Britain (see the briefing on ‘UK Public Opinion toward Immigration: Overall Attitudes and Level of Concern‘). In March 2013, for example, 29% of a representative sample of British adults listed immigration (or “race relations”) among their top three most important issues, trailing only the economy as the most frequent issue of concern, and just ahead of unemployment and the NHS (Economist/Ipsos MORI 2013).

Does public opinion therefore confer a democratic mandate to decrease the number of migrants coming to the UK? One might question whether public opinion—especially in the form of poll results rather the election results—confers democratic mandates for policy-makers on any issue. Certainly the policy process in British government is often concerned with inputs other than opinion polls. For instance, consultations commonly take into account the views of stakeholders and evidence compiled by experts, rather than polling results. At a more theoretical level, some philosophical accounts of political representation see elected officials as charged with using their own judgement and doing what is best to fulfil general goals and values, rather than simply following public opinion on issue after issue. On the other hand, issues with high levels of public concern may well affect election outcomes, and in a democracy politicians have strong motivations to address salient issues in a way that satisfies prospective voters. And, again at the theoretical level, accountability to the voters—if not necessarily to opinion poll results—is at the heart of modern representative democracy.

The present coalition government has seemed to take public opinion on immigration very seriously. It has set a goal of reducing annual net migration to “the tens of thousands rather than the hundreds of thousands”, and its assessments of its policy changes have consistently referred to public opinion or public confidence as a key motivation for taking steps in this direction. In practice, however, there are multiple obstacles standing between public opinion and policies affecting net migration, making the notion of a democratic mandate more complicated than it might appear. Specifically, 1) “immigration” does not mean the same thing to members of the public as it means in government policy; 2) government faces legal and other constraints that public opinion does not; and 3) policy-makers must take into account trade-offs with other goods while respondents to public opinion surveys rarely if ever are pushed to weigh these sorts of trade-offs.

Back to top

Immigration in policy differs from immigration in polls and surveys

Public opinion polls usually solicit attitudes toward “immigration” as a whole. In the typical survey, members of the public have little or no opportunity to express different attitudes toward different categories of migrants, nor to take constraints on policy-makers into account. But policy cannot and does not address immigration as a single undifferentiated phenomenon. Rather, it addresses particular categories or groups of migrants. As the experience of policy-making under the coalition government has shown, there is no single policy toward “immigration”, but various policies directed toward various categories of “inflows”.

For policy purposes, immigration can be divided into categories including students, workers, family members (of migrants or of British citizens), asylum seekers, and EU nationals (who may fit into several of these categories but can be considered a distinct category for these purposes, since policy decisions will reflect Britain’s inclusion in the EU (and EEA, or European Economic Area). Policies designed to reduce migration have taken different forms for each of these categories of migrants. For instance, the “cap” on immigration, perhaps the most widely discussed policy tool for reducing net migration, applies only to skilled and highly-skilled non-European labour migrants (Tiers 1 and 2 of the Points-Based System), a group that made up about 10% of non-EU immigration in 2009 (MAC 2010: p. 130; calculation excludes dependents). Family migration may be reduced by heightened English language requirements for spouses of British citizens and residents who wish to migrate to the UK, and minimum income requirements for citizens and residents wishing to sponsor their spouses’ migration to Britain. Policies proposed to limit student migration focus on reducing fraud and abuse, and also limit legal means for international students to extend their stays in Britain and to work at paid jobs while in Britain. Asylum seekers were addressed by numerous policy changes in the 2000s (see below). Meanwhile, because the EU/EEA establishes a zone of free movement across national boundaries, Britain can take few steps to immigration among EEA nationals.

For most of the history of public opinion surveys, however, “immigrants” and “immigration” have not been divided into these subcategories. Most of the evidence we have about attitudes toward immigration comes from questions about “immigrants” or “immigration” as a whole.  Furthermore, most surveys do not define what “immigration” means, leaving it to each individual survey interviewee to respond on the basis of whatever conception of immigration he or she has in mind.

But defining who counts as a migrant is a complex task, contested among specialists, and it should not be surprising that members of the public often do not define it in the same way as the government does (see the briefing on ‘Who Counts as a Migrant: Definitions and their Consequences‘). The Office of National Statistics (ONS) defines a long-term international migrant as anyone moving to another country for at least a year. In common language, on the other hand, an “immigrant” is more often thought of as someone permanently moving to another country to settle there. (The Oxford English Dictionary, for instance, defines an immigrant as “a person who migrates into a country as a settler”.) Government efforts to reduce net migration may well include immigrants with short stays in the UK, who may not fit public perceptions of who counts as a migrant. Indeed, much of the recent increase in immigration comes from migrants staying for only a few years (see the briefing on ‘Long-Term International Migration Flows to and from the UK‘).

Evidence gathered in a September 2011 survey for the Migration Observatory showed that the majority of the public, at that point in time, did not normally think about students, immediate family members, or temporary migrants when thinking about “immigrants” in the context of a public opinion survey. Majorities had in mind asylum seekers, labour migrants, and permanent migrants, among other categories.

So, even when policy-makers pursuing the popular goal of reduced migration, they may not be targeting “immigration” as a majority of citizens conceive it. Common conceptions of immigration overlap with but do not match the group of “long-term international migrants” picked out by official government statistics.

Back to top

Mismatches between public opinion and policy targets and policy tools

Furthermore, as the coalition government’s experiences are demonstrating, policies to reduce net migration inevitably act on particular categories of migrants separately. Policies can pursue the broad goal of reducing immigration in general, but the relationship between public opinion and the actual details of policy depends on more precise information about a) who the public thinks of as “immigrants” and b) which types of migration are more or less popular.

For example, net migration statistics apply to legal migration (with one key exception – some legal entrants overstay their leave to remain, becoming illegally resident after being counted in official immigration statistics). But public opinion shows concern centred on illegal rather than legal immigration, as shown in responses to a Migration Observatory/Ipsos MORI survey in September 2011. A majority (54%) of respondents who preferred less respondents also said that they wanted these reductions “only” or “mostly” focused on illegal immigration. This is consistent with surveys from 2008 to 2010, in which more than two-thirds of respondents in Britain were worried about illegal immigration, but no more than 36% worried about legal immigration (German Marshall Fund 2011).

Second, public opinion data show little preference for Eastern European nationals over others, but EU membership binds policy-makers to treat migration from all EEA countries preferentially. Respondents in one poll (Ipsos MORI 2007) were asked which groups of migrants should have priority to come to the UK to work or study and which groups should not be allowed in at all for those reasons. Eastern Europeans nations were among the most popular choices to not be allowed in at all, with 14% choosing the newer European nations such as Poland and the Czech Republic, and 16% choosing the newest European nations such as Bulgaria and Romania.  Directly fulfilling public concern at this level would involve withdrawing from the EU, a decision that would have enormous political and economic ramifications that would need to be considered as well. Of course, members of the public are not bound to consider such implications when responding to polls; indeed, it is certainly possible to wish for less immigration from Eastern Europe even while realising that EU member nations cannot enact such limits. But policy-makers certainly are expected to take such consequences into account even if the public does not (or cannot in the restrictive forum of the opinion poll).

Next, the Migration Observatory/Ipsos MORI 2011 survey showed that members of the public make distinctions among migrants by occupation and purpose for migrating. For example, solid majorities preferred less immigration among low-skilled workers, both in general (64%) and in particular cases such as construction workers (57%) and restaurant staff (59%). But only a minority of respondents wanted high-skilled immigration reduced, in general (31%) and in particular cases as well (30% for scientists and researchers, 40% for business and finance professionals).

International students and migrants who are immediate family members of British citizens were also relatively widely tolerated. On the other hand, majorities of respondents preferred fewer immigrants among extended family members and asylum seekers.

These results chime with a smattering of results from earlier surveys and ad hoc polls. For example, majorities in one survey supported increased immigration among doctors and nurses (72%) and elderly-care workers (51%) (German Marshall Fund 2011). Also, when asked which groups of migrants should be given “priority to enter” Britain, “people with skills that are needed for our economy” and “foreign students who want to come and study here” were among the three most commonly selected categories (Ipsos MORI 2007).

These priorities are difficult for policy-makers to follow, however. Labour migration, including skilled workers, is perhaps the easiest area for government to regulate the number of arrivals. Some occupations, such as medical professionals, have been exempted from limits because of shortages in the domestic labour market. But reducing net migration may create pressure to shorten the list of such exempt occupations. The Migration Advisory Committee recently recommended removal of eight occupations from the shortage list, including senior care workers, following government instructions to exclude jobs below “graduate level” (MAC 2011). International students, meanwhile, arrive in such large numbers that it is difficult to imagine government reaching its target without reducing student migration substantially. But students and highly-skilled or needed workers do not appear to be the top concerns for most of the public, and limiting their arrivals may create economic costs that bear consideration as possible trade-offs.

Back to top

Policy-making faces constraints that public opinion does not

For each category of immigrants, there are also different sets of legal constraints that bind policy-makers, but may be ignored in public opinion.

Individual rights, in particular, pose a direct constraint on the implementation of public opinion in policy-making in any area, not just migration. Liberal or constitutional democracies cannot follow public opinion blindly, but rather are legally bound to respect certain individual rights and other constraints on government power. Examples from criminal law are familiar—few would consider it legal or legitimate for a democratic government to imprison a citizen without a proper criminal trial, even if such an action would be popular. Likewise, in the case of migration policy, attempts to limit migration can run up against human rights. Of course, this notion of legal constraint is problematic in some ways. Rights, at least in the legal sense, depend heavily on governments to establish, recognize, and enforce them. Thus, rights might not be seen as independent constraints on governments. Moreover, in practice governments may fail to respect rights, and when they do so it can prove impossible for individuals to enforce their claims. And in the UK, Parliamentary sovereignty may be seen to give government more freedom to act than other constitutional democracies with written constitutions that include enumerated individual rights.

Nonetheless, even within the British constitutional configuration, there are constraints on government policy and even de facto on Parliament under the European Convention on Human Rights (ECHR), EU law, and even domestic common law. In fact, in the case of migration policy, several Labour government initiatives have been overturned, or declared incompatible with the European Convention on Human Rights, in domestic courts.

Legal wrangles concerning asylum decision-making and the treatment of asylum seekers most clearly illustrate constraints on migration policy. As a party to the 1951 Convention on Refugees, Britain is not permitted to return to their home countries any asylum applicants with “a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion…”  But almost half of the public in effect endorsed this course of action in a 2001 poll, with 44% agreeing that “Britain should not accept any more asylum seekers” (Ipsos MORI 2001). Even if that position had gained overwhelming majority support in subsequent polling, the government could not have followed it without violating commitments.

While British governments eschewed this path, other policies on asylum have led to legal challenges and objections from international institutions. For instance, the Nationality, Immigration, and Asylum Act of 2002 broadened the circumstances in which asylum applicants could be excluded from protection. The UNHCR (United Nations High Commissioner for Refugees) objected formally to this move. For another, the Labour government changed policy to withdraw housing support from asylum seekers who did not file their applications within three days of arrival in the UK (Somerville 2007). This policy was overturned in British courts as a human rights violation.

Family migration policy faces important legal constraints as well. Restrictions may come into conflict with international agreements and EU law, including the Charter of Fundamental Rights of the European Union and the European Convention on Human Rights (ECHR). On the other hand, European case law has allowed governments significant scope for limiting family migration (Kraler 2010), so the constraints may be flexible in practice. New restrictions on family migration notably require a minimum income for British citizens who wish to sponsor the immigration of a spouse with non-EEA citizenship. Thus, they affect British citizens’ as well as prospective migrants’ scope for marrying and form a family with the person of their choosing. This provision has already attracted legal challenges.

The previous Labour government also faced legal constraints on its efforts to reduce family-based immigration occurring through marriages of convenience or “sham marriages.” A 2006 High Court decision ruled against a policy requiring special applications for marriage for non-EU nationals, at a cost of £135 per application (Casciani 2006). The High Court saw this as interfering with Article 12 (ECHR) rights to marriage and also as discriminatory on the basis of nationality and religion under ECHR Article 14 (Department for Constitutional Affairs 2006).

Back to top

Attitudes may stand alone, but policy-makers must consider trade-offs

Aside from clear constraints, policy-making involves trade-offs, striking balances among competing goods. In contrast, public opinion almost always treats issues in isolation. Polls ask for people’s opinions about immigration without explicit consideration of the impact that migration policy might have on other areas. Yet such trade-offs have come to the fore as the coalition government has sought to develop policies to reduce net migration. The “cap” on non-European labour migration was implemented with a significant exemption for intracompany transfers (ICTs). Business interests argued that capping ICTs would create significant economic costs. ICTs accounted for 29,255 entry visas to non-EU nationals in 2012 (out of a total of 45,444 entry visas granted to main applicants in the skilled labour components (Tiers 1 and 2) of the Points-Based System). Similarly, in debates about reducing immigration of international students, the higher education sector has argued that international students constitute a valuable export industry, and make important intellectual contributions as well. Declines in student visas in 2012 were almost exclusively among foreign language schools and further education colleges, with universities insulated from changes. In the cases of university students as well as ICTs, policy-makers needed to weigh the potential economic and other costs against the potential benefits of satisfying public demands for less immigration.

Members of the public, meanwhile, need not make such calculations when expressing their opinions to pollsters. Public opinion on these trade-offs is unknown. Questions about policy trade-offs are rarely polled. Further, there may be good reason for this: polling on detailed policy questions is likely to provide less reliable results than more general questions that tap into basic values and attitudes. For most democratic citizens, becoming well-versed in policy matters is neither their job nor a deep and abiding interest. When polls ask people questions for which they have not formed firm opinions, research suggests that responses are often not very reliable: they are unstable, shaped by whatever happens to be “at the top of the head” at the time, and vulnerable to manipulation by changes in question wording (Zaller 1992). On the other hand, there is some evidence that public opinion can respond intelligibly to trade-offs, at least in the context of government budgets (Hansen 1998).

The issue of trade-offs looms large in immigration policy-making, as shown in the efforts to limit immigration among workers and students. One additional “trade-off” might be better posed as an enduring constraint: EU membership. In agreeing to join the EU, Britain in effect made a decision to accept higher levels of migration from European countries in order to secure the benefits of free economic activity with the EEA (as well as free movement for British nationals to EU countries). Free movement within the EU by EU nationals (including British citizens) accounted for +68,000 in net migration statistics in 2010, a significant amount of movement that is not subject to direct immigration control.

Back to top

Policy Implications

What does this mean for policy-makers? They face a widespread preference for reduced migration, but, given policy trade-offs and legal constraints, the available policy options often seem mismatched with the main targets of public concern.

In the current government’s approach, policy-makers specify and try to meet a numerical target by combining different policies that reduce various streams of migration. As others have noted, there is a risk that the goal will prove impossible to meet, because net migration depends on many events beyond government control (including EU immigration, British emigration, international events that may lead to large numbers of asylum seekers arriving from new places). Moreover, reaching this target might impose some costs that would prove unpopular as well, as shown above in the discussion of policy trade-offs.

Importantly, it is not clear that any particular number of migrants would placate public concern. Even in the 1960s, when there were many fewer immigrants, similar majorities of British people still felt there were too many. Even now there is reason to doubt the existence of a specific number that would satisfy concerns. In a 2006 poll, respondents who favoured a “strict limit” on the number of migrants coming to the UK were given a follow-up question asking what that numerical limit should be. More than 60% said they did not know what the limit should be (Bremner 2011). Public preferences for fewer immigrants seems to reflect a general attitude toward immigration (however each poll respondent defines it), rather than a belief that immigration is fine as long as it is kept below some easily-specified numerical limit.

On the other hand, increases in immigration numbers correspond with an increase in the salience of immigration as an issue. Reducing numbers might plausibly lead to less urgent public concern (Page 2009), even if majorities would prefer still further reductions. Also, fulfilling a clear numerical goal might have the added benefit of restoring some public confidence in immigration policy. Some have argued that a primary public demand on immigration policy is for competent management (Saggar 2010), and reaching a clearly stated numerical goal might well be perceived as a demonstration of competence.

Finally, for policy-makers to respond to public opinion, they need finer-grained opinion data. Given the complexities of immigration inflows and the constraints and trade-offs policy-makers face, a simple public preference for less immigration does not provide clear policy guidance for the actual decisions policy-makers face, particularly if coupled with a majority preference for maintaining immigration flows among students and high-skilled workers. More detailed opinion data will not resolve all policy dilemmas, but it may help government address public concern more precisely. Migration Observatory survey work provided a useful snapshot of attitudes toward different categories of immigrants as of September 2011, but additional work would be useful to validate initial findings and track changes over time. Until its recent cancellation, the government-commissioned Citizenship Survey had been a possible vehicle for gathering such information. Without the Citizenship Survey, commercial pollsters and academic researchers are the most plausible candidates for collecting such data, but their aims may not match the needs of policy-makers to better understand public opinion on policy-relevant categories of immigrants.

Back to top

Related Materials


  • Bremner, Pamela. “Public Opinion on Immigration: Trends and Interpretation.” Presented at the Centre on Migration, Policy and Society, University of Oxford, 17 Feburary 2011.
  • Casciani, Dominic, “Sham Marriage Law Breaches Rights,” BBC News, 10 April 2006. http://news.bbc.co.uk/1/hi/uk/4894544.stm.
  • Department for Constitutional Affairs. “Review of the Implementation of the Human Rights Act.” DCA, London, 2006.
  • German Marshall Fund of the United States. “Transatlantic Trends: Immigration, Topline Data 2010.” German Marshall Fund, Washington DC, 2011. http://trends.gmfus.org/immigration/doc/TTI2010_English_Top.pdf
  • Hansen, John Mark. “Individuals, Institutions, and Public Preferences over Public Finance.” The American Political Science Review 92 (1998): 513-531.
  • Economist/Ipsos MORI. “Economist/Ipsos MORI Issues Index.” Ipsos MORI, London, February 2011. http://www.ipsos-mori.com/researchpublications/researcharchive/2724/February-2011-EconomistIpsos-MORI-Issues-Index.aspx.
  • Ipsos MORI. “Asylum Poll.” Ipsos Mori, London, 2001. http://www.ipsos-mori.com/researchpublications/researcharchive/1322/Asylum-Poll.aspx
  • Ipsos MORI. “UK Border Agency Public Attitudes Survey.” Ipsos Mori, London, 2007. http://webarchive.nationalarchives.gov.uk/20100418065544/http:/www.ukba.homeoffice.gov.uk/sitecontent/documents/aboutus/public-attitudes-surveys/
  • Kraler, Albert. “Civic Stratification, Gender and Family Migration Policies in Europe.” Final Report, International Centre for Migration Policy Development, Vienna, 2010.
  • Migration Advisory Committee. “Limits on Migration: Limits on Tier 1 and Tier 2 for 2011/12 and Supporting Policies.” Home Office, London, 2010.
  • Migration Advisory Committee. “Analysis of the Points Based System: Revised UK Shortage Occupation List for Tier 2 Comprising Jobs Skilled to NQF Level 4 and above.” Home Office, London, 2010.
  • Page, Ben. “British Attitudes to Immigration in the 21st Century.” In Migration, Public Opinion, and Politics, edited by Bertelsmann Siftung, Migration Policy Institute. Gütersloh: Bertelsmann Siftung, 2009.
  • Saggar, Shamit. “What do voters want from leaders on immigration?” Policy Network seminar, Policy Network, London, 13 December 2010. http://www.policy-network.net/articles_detail.aspx?ID=3932.
  • Somerville, Will. Immigration under New Labour. Bristol: The Policy Press, 2007.
  • UK Border Agency. “English Language Requirement for Spouses.” Impact Assessment HO006, UK Border Agency, London, 1 October 2010.
  • YouGov. “Economic Migration” Poll prepared on behalf of Migration Watch. YouGov, London, 2010.
  • Zaller, John. The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press, 1992.



Public Opinion and Public Policy: Complexities of the Democratic Mandate

Press Contact

If you would like to make a press enquiry, please contact:

Rob McNeil

+ 44 (0)7500 970081

 Contact Us 


This Migration Observatory is kindly supported by the following organisations.

  • University of Oxford logo
  • COMPAS logo
  • Esmee Fairbairn logo
  • Barrow Cadbury Trust logo
  • Paul Hamlyn Foundation logo