Policy Papers

Redrawing the boundaries of British democracy?  Census data and the Great Reform Act, 1832-2011

Stephen Thompson |

  • RSS Feed Icon

Executive Summary

  • The UK government is currently engaged in a programme of major political and constitutional reform. The Deputy Prime Minister has drawn parallels between today's changes and those brought in by the 1832 Reform Act.
  • Current policy is based on the claim that the state has too much control over citizens, most notably as a result of laws that have increased surveillance.
  • The main purpose of the 1832 Reform Act was to make parliamentary representation more equal by transferring seats from less densely inhabited places ('rotten boroughs') to fast-growing industrial towns and counties. Reformers objected to rotten boroughs because the MPs who sat for them were accountable to the borough patron, rather than the electorate. This system helped to perpetuate the influence of the executive, at the expense of the legislature.
  • Those who drafted the 1832 Reform Act made extensive use of the population census to identify constituencies with particularly small populations. The principal advantage of the census was that it provided authoritative population totals that were in the public domain.
  • The 2011 Parliamentary Voting System and Constituencies Act is only superficially similar to the 1832 Reform Act. The present government's commitment to equal-sized constituencies would, in fact, have been an anathema to Whigs and Tories alike in 1832. Today's anti-state surveillance sentiments and, most particularly, the Coalition's anti-census rhetoric implies a fundamental misunderstanding of how early nineteenth-century reformers achieved their aims.
  • Historically at least, the growth of state information gathering in Britain has roughly paralleled increasing democratization and the expansion of the franchise. This is partly because institutions like the census can be used to make the state more accountable to its citizens, not less. For instance, looking to the future, a census in 2021 would provide the best independent check on the completeness of individual voter registration (due to be introduced in 2014).
  • Today's policymakers should think twice before replacing the census with untried alternatives that seek to share data from a mixture of public and private organisations. Although such a model may seem less expensive, it may also prove less democratically accountable.


On 19 May 2010, the new Deputy Prime Minister, Nick Clegg, gave a speech in which he promised 'to transform our politics so that the state has far less control over you, and you have far more control over the state'. Warming to his theme, Clegg declared that government was about to embark upon

the most significant programme of empowerment ... since the great reforms of the 19th century. The biggest shake up of our democracy since 1832 when the Great Reform Act redrew the boundaries of British democracy, for the first time extending the franchise beyond the landed classes. That was landmark legislation from politicians who refused to sit back and do nothing while huge swathes of the population remained helpless against vested interests; who stood up for the freedom of the many, not the privilege of the few.

Later in his speech, Clegg identified three concrete changes that would transform the state-citizen relationship. First, he promised that 'This government will end the culture of spying on its citizens': no ID card scheme, no national identity register, and a halt to second generation biometric passports. Secondly, Clegg pledged to reduce the power of political elites through a series of constitutional changes, including an elected House of Lords, fixed-term parliaments, individual voter registration, a referendum on Alternative Vote, reducing the number of MPs and creating constituencies that are more equal in size. The third undertaking was to redistribute power away from the centre 'so that you will get more control over the hospitals you use, the schools you send your children to, the homes that are built in your community'. Summing up, Clegg concluded that all these changes amounted to 'our very own Great Reform Act'.

Clegg's speech contained two related claims. The first was historical, namely that the Coalition's reform package would be the most far-reaching since 1832. Clegg's second claim was that the 'proliferation of laws that increase surveillance' - ID cards, identity registers, biometric passports, CCTV, and the DNA database - have undermined British democracy by encroaching upon citizens' hard-won liberties. To recover the political freedoms which the 1832 Reform Act granted, Clegg implied, it is necessary to curtail, or even dismantle, the state's apparatus of surveillance. This paper seeks to offer a historical perspective on the validity of both these propositions.

The purpose of the Great Reform Act

The Great Reform Act is usually remembered for extending the franchise - as Clegg's speech indicates - but this famous piece of legislation also redrew the boundaries of parliamentary representation in the most literal sense, by abolishing the so-called rotten boroughs. Rotten boroughs were parliamentary constituencies where the borough patron - usually an aristocratic landowner - was able to determine the outcome of elections by corrupting the small number of qualified voters. Perhaps the most notorious rotten borough in the early nineteenth century was Old Sarum in Wiltshire, located just north of Salisbury. No-one lived in Old Sarum but it had returned MPs to Westminster since the reign of Edward I. Like most early nineteenth-century boroughs, Old Sarum was a double-member constituency. According to one contemporary description, 'Old Sarum retains the name and immunities of a borough, it might be supposed for the purpose of ridicule only'. It had seven voters in total at the start of the nineteenth century, of whom five turned out at the 1802 general election.

In many respects, it was the campaign against electoral corruption which motivated the Whig government that drafted the Great Reform Act. Since at least the 1780s, moderate parliamentary reformers had argued for 'a more equal representation of the people'. This was not because they were committed to universal suffrage, but rather because reformers thought that corruption perpetuated the 'influence' of the executive. Many rotten boroughs were ultimately controlled by the Treasury. This meant that the executive could rely on MPs representing these boroughs to do as they were told and vote with the government. Whig reformers therefore wanted to abolish rotten boroughs and transfer their representation to the more populous counties like Yorkshire and Lancashire, and to unrepresented manufacturing towns like Birmingham, Manchester and Leeds. This would eliminate the problem of electoral corruption by making it prohibitively expensive. Bribing a dozen electors was straightforward; bribing half the freeholders of a county was quite another matter. For leading Whig reformers such as Earl Grey and Lord John Russell, electoral reform was essentially restorative: they believed that the elimination of rotten boroughs together with a wider franchise would strengthen the power of parliament against an overbearing executive.

The statistical foundations of the Great Reform Act

Earl Grey's government was appointed in November 1830 and many of its ministers had no previous experience of office. The Cabinet made parliamentary reform its top priority and a four-man committee was charged with drafting the relevant legislation. The 'Committee of Four', as it became known, needed a formula for determining which boroughs were most susceptible to corruption. Rather than undertake a detailed inquiry into the number of resident voters in each of the 262 boroughs that returned MPs to the Commons, the Committee decided to rely on existing data that was already in the public domain.

During his time in Opposition, one of the Committee's members, Lord John Russell, had raised the issue of parliamentary reform in the House of Commons on half a dozen separate occasions. As early as 1819, Russell had used demographic evidence from the 1811 census to argue that unrepresented towns of at least 15,000 inhabitants should receive the elective rights of rotten boroughs. Two years later, Russell used the published census returns to estimate that as many as 300 MPs sat for constituencies of fewer than 5,000 inhabitants. Russell's reliance on publicly-funded, official population statistics set him apart from the late eighteenth-century generation of reformers. Before the first national census of 1801, reformers had relied on private and incomplete analyses of poll books to demonstrate the statistical anomalies of the existing electoral system. Census evidence had two principal advantages over poll book data: authority and completeness. As Russell himself put it, the census returns were published in 'a well known statistical book, as free from error and doubt as could be expected.'

Once he was in government, Russell remained convinced of the merits of census data. He had little difficulty in persuading his colleagues that the amount of population constituted the 'surest proof of the necessity of disfranchisement in some cases, and an increase in the number of members in others'. The Committee of Four devised a simple formula for deciding which constituencies should be disfranchised. Any borough with fewer than 2,000 inhabitants at the time of the 1821 census would cease to return MPs altogether. In addition, any double-member borough with fewer than 4,000 inhabitants would lose one MP. Having applied this formula, the Committee found it had 168 seats available for redistribution. This was equivalent to 30 per cent of the House of Commons. Of these 168 seats, the Committee agreed to assign 34 to unrepresented English towns containing over 10,000 inhabitants, 8 seats to the metropolis, 5 to Scotland, 3 to Ireland and 1 to Wales. Another 55 seats would be shared out among counties with over 150,000 inhabitants. The remaining 62 seats would be suppressed, thereby reducing the size of the Commons by around 10 per cent from 658 seats to 596 seats.

When Earl Grey presented these proposals to William IV in February 1831, the King quickly grasped that the foundation of the plan was 'the amount of population'. He therefore advised Grey to take a new census 'to remove all possibility of objection and doubt'. The act of parliament authorising the 1831 census had been passed six months earlier, but it was not due to be taken until 30 May 1831. Grey was adamant that the Reform Bill could not wait because of the risk of 'increased agitation and the propagation of dangerous theories'. In addition, Grey argued, if it became widely known that the fate of boroughs rested on the results of the forthcoming 1831 census, 'the accuracy of the new returns might be affected'. The 1821 census, by contrast, could be used, he insisted, 'without the possibility of a suspicion that any inaccuracy can have crept into it from interested motives.'

Opposition to reform: the population principle

Grey should have listened to the King. Almost as soon as the Bill was presented to the House of Commons, opponents of reform attacked its statistical foundations. By far the most tenacious critic was the MP for Aldeburgh in Suffolk, John Wilson Croker. Aldeburgh had a population of 1,212 in 1821 and was therefore one of the 60 boroughs which would disappear under the government's plans. In the first of many interventions during the debates on the Reform Bill, Croker accused the government of cynically protecting its own interests. It was no coincidence, Croker suggested, that the Whig boroughs of Malton (pop. 4,005), Knaresborough (pop. 5,283) and Tavistock (pop. 5,483) narrowly escaped the axe.

During the Bill's committee stage, the MP for Liverpool, General Gascoyne developed a far more damaging critique of the government's reliance upon census data. He pointed out that one effect of redistributing seats on the basis of population would be to transfer seats to Ireland at England's expense. Gascoyne wondered how this could possibly be just given how little the Irish population contributed in taxes. He therefore tabled a wrecking amendment to maintain the existing number of English and Welsh MPs. This was carried by 8 votes and the government responded by requesting the dissolution of parliament from William IV.

The government won a landslide victory at the general election, securing a majority of over 130. As soon as the new parliament met, ministers introduced a second Reform Bill that was virtually identical to the first. The seat redistribution formula was still based on borough populations recorded in the 1821 census, despite Tory attempts to force the government to use the 1831 census instead. Opposition in the Commons certainly delayed the Bill, but ultimately there were not enough votes to stop it being passed.

Things were different in the House of Lords. As in the Commons, opponents of reform objected to the Bill's seat changes. However, what worried peers was not the accuracy of census data - whether to use data from 1831 instead of 1821 - but the principle of making representation dependent upon population. This principle was associated above all with the Constitution of the United States. The Founding Fathers had declared in Article 2, section 2 of the Constitution that a census should be taken every ten years for the purpose of apportioning 'Representatives and direct Taxes ... among the several States.' Rightly or wrongly, Tory peers believed that the government's decision to base seat changes on census data revealed their true radical intentions. Ministers repeatedly denied accusations that they wanted to introduce equal-sized constituencies, but they failed to persuade a sufficient number of peers to back their plans. The Lords threw out the second Reform Bill by a majority of 41 votes.

The Cabinet realised that it would have to make concessions if the measure was ever going to get through the upper House. The most important of these became known as the 'Drummond Scale', after the man who devised it, Lieutenant Thomas Drummond. The Home Secretary, Viscount Melbourne, instructed Drummond to rank the smallest 120 boroughs in order of importance. In place of a purely demographic measure of importance, Drummond substituted a new formula. He gave equal weighting to the number of houses at the time of the 1831 census, and the amount of assessed taxes paid in each borough. This compromise did just enough to get the third - and final - Reform Bill through its second reading stage in the House of Lords. Even more importantly, though, it reassured William IV. The King had initially questioned the wisdom of using the 1821 census. He was therefore gratified to learn that the third Reform Bill's redistribution schedules would be based on the more up-to-date results of the 1831 census. William's support was crucial. Without it, the third Reform Bill would have been lost too. During the 'Days of May', when the government resigned in protest at peers' blocking tactics, the King threatened to create an unlimited number of new peers to get the Reform Bill through its committee stage. This broke the deadlock in the House of Lords and the Reform Act became law on 7 June 1832.

One important legacy of 1832 is that the British census has never assumed the political importance which its US counterpart enjoys. This is because opponents of parliamentary reform were extremely effective at portraying the first and second Reform Bills as the thin end of the wedge: the first step towards an American understanding of political representation, prioritising equal electoral districts above place-based, territorial considerations.


How valid, then, are comparisons between the Coalition's current legislative programme and the reforms of 1832? Nick Clegg has called government plans to roll back state surveillance, reform parliament, and decentralise power as 'our very own Great Reform Act'. At one level, this analogy does not seem entirely far-fetched. After all, the government's flagship Parliamentary Voting System and Constituencies Act, passed last month, appears, superficially at least, to perform the same dual function as the Great Reform Act of 1832. On the one hand, it provides for a change in the franchise, subject to a 'Yes' vote in the forthcoming AV referendum, and on the other, it requires the Boundary Commissioners to redraw constituency boundaries across the United Kingdom.

At another level, however, drawing such a parallel reveals a fundamental misunderstanding of the mechanisms through which political reform was achieved in 1832. According to Clegg, the state has become locked into a 'culture of spying' on its citizens. The only way to break this culture, so the argument goes, is to dismantle the machinery of surveillance. What this might mean in practice is unclear: Clegg himself did not include the census among his list of 'laws that increase surveillance', but the Cabinet Office Minister, Francis Maude, called for 'a less intrusive, much cheaper census' while still in Opposition. More recently, Maude went much further in comments made to the Daily Telegraph. In an article entitled 'National census to be axed after 200 years', it was reported that Maude is convinced that there are alternatives sources of data - including databases held by credit checking firms, Royal Mail, councils and Government - 'which will provide better, quicker information, more frequently and cheaper'. Two quite separate arguments about the merits, or otherwise, of state surveillance have coalesced in the last year: one is concerned with individual freedoms, the other is about cost. Neither has much to do with the Great Reform Act.

In 1832 two by-products of the nineteenth-century state's 'culture of spying' - census data and tax office returns - were integral to the reform process. Without concrete, authoritative knowledge of the wide discrepancies in borough populations (or to use the later terminology, 'importance'), the Whigs would have found it impossible to undertake large-scale reform. Electoral corruption was extremely hard to prove and only a handful of boroughs - New Shoreham (1770), Cricklade (1782), Aylesbury (1804), Grampound (1821), East Retford (1828) - were ever convicted of it. To get round this problem the Cabinet decided to abolish an entirely arbitrary number of boroughs. But to justify this decision, ministers needed some kind of empirical threshold, and this was where census data proved so invaluable, albeit controversial.

So controversial indeed, that census data plays no part in determining the size of present-day parliamentary constituencies. Instead, the Boundary Commission is required to use figures taken from electoral registers - which only exist because of the Great Reform Act - when making boundary changes. As a result of the 2011 Parliamentary Voting System and Constituencies Act, we can expect a complete overhaul of constituency boundaries before the next General Election. This is because the Coalition is committed to creating fewer and more equal-sized constituencies. In 1832, by contrast, the Whigs abandoned their plan to reduce the size of the Commons by around 10 per cent following General Gascoyne's amendment to the first Reform Bill. On the issue of constituency size, the Whigs consistently and repeatedly denied Opposition claims that their real objective was the introduction of equal electoral districts. It is therefore particularly ironic that Nick Clegg should have made the Great Reform Act his historical precedent of choice.

Under the new boundary rules, introduced by the 2011 Parliamentary Voting System and Constituencies Act, commissioners have been told that that the number of registered voters - not people - in a constituency must not deviate from the UK average (the 'electoral quota') by more than plus or minus 5 per cent. This is a much more rigorous interpretation of the concept of equal constituencies than was applied under the old rules. To achieve this degree of equality, many constituencies will no longer respect existing local authority or county boundaries. This has caused considerable disquiet in parts of Cornwall, where locals are unhappy about the possibility of sharing an MP with Devon. In the Isle of Wight, meanwhile, campaigners won a last minute reprieve from the new Act's provisions and will continue to be treated separately from the rest of England - gaining an extra MP in the process.

Recent constitutional changes should not be seen as confirming the worst fears of nineteenth-century Tories, however. The US Constitution's commitment to demographic equality in the distribution of representation has not been imported by the back door. For this to happen, the UK census would have to become embedded within the boundary rules, rather than face the axe. Within a US context it would be unthinkable to propose the abolition of the national census because its results are required for the redistribution of Congressional representation every ten years. What the UK now has is the façade of equal representation. It is widely recognised that one of the chief defects of electoral register data is that there is considerable under-registration of voters. According to a report published by the Electoral Commission in March 2010, there has been a long-term decline in the completeness of Great Britain's electoral registers. The Commission can say this with confidence because it is able to compare electoral registers with census returns. It has been estimated that the completeness of electoral registers has fallen from a maximum level of completeness of 95 per cent in 1950 to only 91 per cent in 2000. In addition, it is thought that geographical variations have probably widened since the late 1990s, with metropolitan and urban areas experiencing the greatest levels of decline in completeness. Interestingly, the Electoral Commission describes this census-based method of estimation as the 'gold standard' for establishing the completeness of registers. Given that Coalition policy has made equal constituencies such a totemic part of its reforms, it is surprising that ministers should also be contemplating the abolition of the most effective independent check for verifying that equality. Moreover, with individual voter registration now expected by 2014, a census in 2021 would provide the best benchmark for measuring the effectiveness of the new system.

The British state has gathered information on its citizens long before it could legitimately be described as democratic. Universal adult suffrage was not achieved until 1928, nearly a century after the supposedly 'Great' Reform Act of 1832. Successive generations of citizens have won new political rights and freedoms not in opposition to information gathering by the state, but on the back of it. The Great Reform Act was a limited measure, which did not, and was never intended to, make Britain a democracy. Nonetheless, the Whig reformers were committed to challenging electoral corruption and vested interests. Crucially, the state itself provided the means for doing this. First in Opposition, and then in government, Lord John Russell turned published, parliamentary census data into a powerful empirical resource that he drew on repeatedly to attack the absurdities of the existing electoral system.

Today's reformers tell us, by contrast, that the state knows too much about its citizens and this knowledge is anti-democratic. Historically, however, officially generated knowledge has played an important role in helping citizens to hold their rulers accountable. This suggests that policy-makers ought to focus more on ensuring that state surveillance is subject to appropriate safeguards to protect citizens' information - such as the closure of personal census data for 100 years - than on dismantling and sub-contracting these functions. Alternatives to the census, which would most likely involve extensive data sharing between a variety of public and private organisations, may prove cheaper, but could well end up being more intrusive and less accountable. If the state decides to cut back on the provision of publicly controlled and publicly accessible data, citizens may discover that they are less able to challenge 21st-century vested interests - political or otherwise.

Further Reading

Nick Clegg, New politics (19 May 2010).

The completeness and accuracy of electoral registers in Great Britain, The Electoral Commission (March 2010).

Edward Higgs, The information state in England (Basingstoke: Palgrave Macmillan, 2004).

Christopher Hope, 'National census to be axed after 200 years', Daily Telegraph (9 July 2010).

T. H. B. Oldfield, The representative history of Great Britain and Ireland (6 vols, London, 1816).

S. J. Thompson, 'Population combined with wealth and taxation': statistics, representation and the making of the 1832 Reform Act', in Tom Crook and Glen O'Hara (eds.), Statistics and the Public Sphere: Numbers and the People in Modern Britain, c. 1800-2000 (London: Routledge, 2011).


Papers By Author

Papers by Theme


Sign up to receive announcements on events, the latest research and more!

We will never send spam and you can unsubscribe any time.

About Us

H&P is an expanding Partnership based at King's College London and the University of Cambridge, and additionally supported by the University of Bristol, the University of Edinburgh, the University of Leeds, the Open University, and the University of Sheffield.

We are the only project in the UK providing access to an international network of more than 500 historians with a broad range of expertise. H&P offers a range of resources for historians, policy makers and journalists.

Read More