Sixth Standing Committee on Delegated Legislation
Wednesday 2 February 2000
[Mr. Jim Cunningham in the Chair]
Draft Census Order 2000
The Economic Secretary to the Treasury (Miss Melanie Johnson): I beg to move,
That the Committee has considered the draft Census Order 2000.
The draft Order in Council gives effect to the Government's proposals for a census in England and Wales on 29 April 2001. The proposals were set out in the White Paper in March 1999.
Under the Census Act 1920, an Order in Council is necessary to prescribe the date of the census, the people to be counted, the people to be responsible for making the census returns and the information to be given in those returns. We have had a census in Britain every 10 years since 1801, except for 1941, for obvious reasons. They have usually been held in April, because that allows the necessary field work to be conducted at a time when there are sufficient daylight hours and less risk of poor weather conditions.
Following devolution, separate legislative arrangements will be made in Scotland and Northern Ireland for the censuses there. Our proposals comply with European Union guidelines for the 2001 round of censuses.
The categories of people to be counted are set out in article 4 of the draft order. Everyone will be recorded at the place where he or she is usually resident. Visitors at an address who are normally resident elsewhere in the United Kingdom will be required to supply full information at their usual residence. That is a change in procedure and will reduce the overall burden on the public, in that information about any individual person will need to be supplied only once. Students and boarding schoolchildren will be regarded as being resident at their term-time address, thus making the census more consistent with the base for the annual mid-year population estimate and, in turn, for that used for resource allocation purposes. Responsibility for making the returns is covered in article 5.
The final layout of the questions that will appear on the census form will be published in regulations to be laid before the House when the Order in Council has been made. They will set out the detailed conduct of the census and contain facsimile copies of the forms.
Under the Census Act 1920, the questions in italics in schedules 2 and 3 of the draft order can be included only if they are approved by resolution. A few questions in schedule 2 are either new to the census in England and Wales or have been significantly revised since the 1991 census. I shall go through them briefly.
Paragraph 2 of schedule 2 relates to the relationship of each person in a household of two or more people to the first person on the form. The new style of question expands the information that was collected in the 1991 census and responses will provide statistics for more detailed analysis of household and family structure to reflect the need for information on an ever-changing society.
Paragraph 9 of the schedule contains a revised question on ethnic groups. The ethnic group question was one of the main successes of the 1991 census. Information from it has been used to provide baseline figures against which the Government can monitor possible racial disadvantage. The wording of the question set out in the draft census order has been extensively researched and tested since then to meet users' requirements for additional information and to be as acceptable as possible to respondents. The form of the question also allows respondents to describe their ethnic group in their own words, using the write-in option.
Paragraph 10 allows people to assess their own health during the preceding 12 months as either good, fairly good or not good. The question is new and simple, and the information from it has been demonstrated in sample surveys to have a good predictive power for health policy and the provision of services, particularly for the elderly.
Paragraph 11 relates to the provision of unpaid personal help. The Government realise that an increasing amount of unpaid personal help is given to people with poor health and this new question will help the understanding of the variations in the need for care and the targeting of resources more effectively.
In paragraph 14, the information from a revised question on academic and vocational qualifications will be used to assist in the better provision of education and training and in the monitoring of the take-up of Government initiatives.
Paragraph 16(e) relates to the time since last paid employment. That new question will help to determine local differences in the periods of unemployment and the extent of long-term unemployment. It will also be used to assess and monitor disadvantage and exclusion in education and training planning, labour market analysis and mortality and morbidity studies.
A new question about the size of the work force in an employer's organisation or company at the place of work is contained in paragraph 17(b). It will help us to understand changing employment patterns.
Paragraph 23 seeks to collect information on the lowest floor level of a household's accommodation. That information will be new only in England and Wales--the question was included in the 1991 Scottish census.
The 2001 census White Paper suggested that a question about religion should be included for the first time in the census of England and Wales. That question would help to provide information that would supplement the output from the ethnicity question by identifying ethnic minority sub-groups, especially those whose members originate from the Indian subcontinent, who increasingly prefer to identify themselves by their religion.
As it is currently drafted, the schedule to the Census Act 1920 does not permit particulars on such topics to be collected. Consequently, the inclusion of that question in the 2001 census of England and Wales will depend on primary census legislation. To effect that change, a short Bill, the Census (Amendment) Bill, was introduced in the House of Lords on 16 December and received its Second Reading on 27 January.
Great care will be taken throughout the process to prevent the inadvertent disclosure of information about identifiable individuals in the tables that are derived from the census, especially those relating to local areas or to specific sub-groups of the population. The security of census data and the measures that will be taken to protect confidentiality in output will be subject to independent reviews, the results of which will be announced prior to the census.
I give Committee members an assurance about the confidentiality of all census information. The census office in England and Wales and its partners in Scotland and Ireland have an outstanding record for maintaining confidentiality, and they are determined to maintain it.
The main national and local results will be released, to a committed timetable, as speedily as possible over a short period of time. That timetable will be announced when all the production systems have been fully rehearsed.
The census is expected to cost an estimated £202.3 million. About three quarters of that will fall into the next two financial years. Financial provision has already been made for that expenditure.
The census will meet crucial requirements for statistical information and the burden on those who have to complete the census forms has been kept to the minimum that is consistent with those needs. The questions have been fully tested and found to be acceptable. I am confident that the proposed census package strikes the right balance between meeting the essential needs for statistics and minimising the burden on the public. I commend the Order to the Committee.
Mr. Oliver Letwin (West Dorset): I approach this order with the scepticism that the Opposition always approach such matters. The Economic Secretary will be aware that my noble Friends in another place propose to table an amendment to the Census (Amendment) Bill about the religious test, which is of considerable controversy. We believe that the test should be voluntary.
I searched through the order for signs of a similar controversy and I asked myself interesting constitutional questions about whether the order could be considered before the Census (Amendment) Bill had been enacted. However, I alighted on the helpful explanatory notes and discovered that this is an entirely uncontroversial order that makes no reference to what is to come. I was unable to find anything to object to in it, and I shall therefore terminate my remarks. I have no intention to cause problems.
Mr. Edward Davey (Kingston and Surbiton): I wonder whether Committee members are as surprised as I am by the remarks of the hon. Member for West Dorset (Mr. Letwin). The Economic Secretary and I recently served on a Standing Committee with the hon. Gentleman, and we became used to him making rather longer, but very good, speeches. However, I intend to speak for slightly longer than he did because important questions need to be asked and because we seek reassurances from the Economic Secretary.
I agree with the hon. Gentleman that this is not a controversial measure. Censuses have been efficiently conducted in this country for two centuries, and we have learned from that experience. The census office has a record that is second to none, internationally, and I should like to put on the record our appreciation of the work of its statisticians.
There are always worries in relation to orders such as this. On the issue of compulsion, there is general cross-party agreement that we must require people to respond to the census form. Otherwise, the data would not be reliable and the uses to which it is put by the Government would be questionable. However, is it right that people could face prosecution and prison sentences for not answering questions that are particularly intrusive and sensitive? Will the Economic Secretary tell us, from the experience of the 1991 census, whether such people were prosecuted and, if so, how many? The hon. Member for West Dorset referred to a Bill that is in another place in that respect.
I am also concerned about privacy, and how that relates to small area information. Commercial firms are allowed to buy some of the information that is collected from the census. In the case of the 1991 census, that sometimes related to areas containing as few as 16 households. Although the information that they receive has no named individuals, when it relates to such small numbers of people it is possible that those analysing the information could identify individuals. I understand that the statisticians who examined that issue in relation to the 1991 census concluded that there was no evidence that it had happened. However, we should be concerned about the possibility of it happening, especially if the census is to contain increased detail in respect of ethnicity. In some areas of the country, people from ethnic minorities are very much in a minority. If one is dealing with small area information that covers only 16 households, someone from an ethnic minority may be easily identifiable. Given that the 2001 census will include more questions about ethnicity, have the Government considered increasing the minimum thresholds?
Will the Economic Secretary clarify whether the order is compatible with the Human Rights Act 1999? Because it is a statutory instrument, neither the Economic Secretary nor the Home Secretary are required to certify that it is compatible. Given that the census deals with very detailed information on private individuals, it is important to consider that point. It would be wrong if the order, having been approved by the Committee, were then to be challenged in the courts and found to have transgressed the right to privacy in the Human Rights Act.