This website uses cookies so that we can provide you with the best possible experience. If you continue to use this site we will assume that you are happy with this. You can find out more about how we use cookies here. If you would like to know more about cookies, or how you can delete them, click here.

What’s at stake with changing the census?

09 October 2013

Michael Harris

The Office for National Statistics (ONS) is consulting on the census and the future provision of population statistics in England and Wales. As you would expect, the RTPI will be responding. So what’s at stake?

The ONS suggests that this would be the most radical change in the production of population statistics in England and Wales since the start of the nineteenth century (the Scottish and Northern Ireland Governments have responsibility for their own censuses). Of course, the census has changed in the past, but why the proposals for change now?

The census is facing change. As part of its ‘Beyond 2011’ programme, a couple of weeks ago the ONS published a consultation document on its two suggested approaches for the future – a primarily online census, or a census using existing government and private data and compulsory annual surveys. In the first scenario, the census would continue to be taken every 10 years but would be conducted primarily online, except for people with no internet access. In the second, the ONS would use existing data held across the public sector, for example tax, welfare and school records, in which case a ‘snapshot’ of the population could be produced every year.

The ONS suggests that this would be the most radical change in the production of population statistics in England and Wales since the start of the nineteenth century (the Scottish and Northern Ireland Governments have responsibility for their own censuses). Of course, the census has changed in the past, but why the proposals for change now?

The primary reason seems to be cost. In 2011, the census cost £482 million and required 40,000 staff, most of whom were needed for just six weeks; this was 35% more in real terms than its 2001 predecessor and costs are likely to increase again by 2021. The ONS predicts that a predominantly online full census would cost about £625 million (in 2011, 17% of respondents completed the census online, and for the 2021 census the ONS expects at least 65% of households to use the internet). An annual sample is said to require £460 million over 10 years. 

But the Government has also said that compiling information from existing data would produce more accurate and timely data. Approximately 1.5 million households failed to fill in their forms in 2001. And it’s true that a census once every ten years can mean a significant time lag in critical data. Most recently, the 2011 census discovered 480,000 more people than expected by central government, the result of higher immigration and birth rates – but this could be (and has been) used as an argument for or against the present approach. 

Different countries take a range of approaches. Germany can exploit data from compulsory identity cards. Scandinavian nations can use administrative data. Canada has moved to an online survey. To critics of the proposed changes however, the fear is that scrapping the current approach would undermine the ability of policymakers, practitioners and others to identify and so address critical issues that should be reflected in public policy, for example population growth, mortality rates and housing need. The changes might also be regarded as being in tension with the UK Government’s stated commitment to evidence based policy and practice

Danny Dorling, professor of human geography at Oxford University (formerly at the University of Sheffield), has consistently argued that moving away from the existing approach would make it harder to understand and track significant social changes. Writing before the ONS released its consultation document, Dorling wrote that: “If councils don't fight now to save this important resource, they will be flailing in the dark when it comes to strategic decisions and informed budget-setting. Private organizations may be able to fill in the cracks, but they can never replace the open exchange of public data that the census allows.” 

Dorling and others argue that there are already smaller sample surveys which provide ‘inferential statistics’ (making predictions or inferences about a population from observations and analyses of a sample), but the whole point of a census is that it is detailed and that you can do with the (traditional) census that you can’t do with other datasets. Dorling provides an example in one of his articles

“The Census allows social scientists to determine in what direction the trends are going. Within a week of the 2011 results being published, Ludi Simpson and Stephen Jivraj, on behalf of the Centre on Dynamics of Ethnicity at The University of Manchester, had analysed the results and determined that every single ethnic minority group within England and Wales had become more dispersed geographically despite rising in numbers in most cases… In the week before Simpson and Jivraj’s analysis was complete, the UK press had already decided that the rising numbers of many groups of people born outside of Britain meant that there had to be ethnic polarization within Britain. They were wrong, and because we had a Census and hence data for every local authority, it was possible to show that they were wrong. Without a Census we would not know.” 

According to this view, if the census were to be replaced with the annual surveys, reliable statistics would only be produced down to local authority level, in contrast to what can be done now where data can be available for areas with a population as small as 100-125 people. This could mean that it would be impossible for authorities to plan and deliver services for local populations, especially in highly populous and diverse areas. 

There’s also the argument that the census has a stature and an authority much greater that any ‘bundle’ of other data sources is likely to have, and in the absence of this authority there could be significant recurring (and even legal) disputes over the validity of the decisions that this data is used to inform. 

There are also issues that are especially pertinent to bodies such as the RTPI – whether for example abolishing the census in its current form would undercut efforts to promote spatial thinking in policy beyond planning. It’s also worth remembering that not only does data underpin the work of planners today, but as noted here previously, it played a major part in the development of planning as a profession.

In the UK, the research undertaken by Charles Booth between 1886 and 1903 to quantify London’s social problems – considered to be the first modern surveys – helped lead to the public health and housing acts in the late nineteenth century, and relatedly the planning system and a planning profession that could ensure the development of better organised, healthier towns and cities. Nineteenth century social reformers were arguing for more data  on social conditions in the belief that shedding light on problems was the first step towards solving them; while efficiency is obviously important, there is more at stake in these changes than saving money alone. 

These are the types of issues that the RTPI will be considering as part of its response to the ONS consultation – and we welcome your views. Do tell us what you think: 

The ONS consultation document can be found here. The consultation runs from 23rd September to 13th December, and Parliament will vote on the final proposals next year. 

We’re also conducting our own census of sorts, in our RTPI Membership Survey. If you haven’t already, you can take part at: /membership/membership-survey/

Also look out for a forthcoming RTPI research briefing on ‘Planning for More People’ in the next few weeks.

About Michael Harris

Dr Michael Harris is Deputy Head of Policy and Research at the Royal Town Planning Institute, where he leads on the RTPI’s research activities. Previously he was a senior associate at the new economics foundation (nef) think tank, and Director of Public and Social Innovation at Nesta (the National Endowment for Science, Technology and the Arts). He has also worked in local government and academia. Michael has an on-going interest in localism, health and wellbeing, and community engagement.