CENSUS ORDER 1970

HC Deb 17 February 1970 vol 796 cc329-58

Mr. Deputy Speaker (Mr. Sydney Irving)

Before calling on the hon. Gentleman to move the Motion. I ought to draw the attention of the House to the fact that there is both a Motion and a Prayer relating to the Census Order, 1970. I, therefore, propose to follow a recent precedent—HANSARD, 18th March, 1965; Vol. 708, c. 1617—and suggest that both should be discussed together. In that case I would allow the scope of the debate to be wide enough to cover both the Motion for the affirmative Resolution and the Prayer. If necessary, I would be prepared to allow the House to divide on each Motion separately. But if the hour of 11.30 is passed and the debate is still continuing, I will only be able to put the Motion for the affirmative Resolution, and the Prayer will fall and cannot be carried over to any subsequent day. I hope that this course will be acceptable to the House as it was on previous occasions.

9.7 p.m.

The Joint Under-Secretary of State for the Department of Health and Social Security (Dr. John Dunwoody)

Thank you for your helpful Ruling, Mr. Deputy Speaker.

I beg to move, That items 2, 7, 8, 10, the words in item 11(a) 'and the month and the year of birth of each such child', 11(b) and (c), 12, 13, 14(a)(vii) and (viii), 15, 18, and 19 in Schedule 2 to the draft of the Order in Council, entitled the Census Order 1970, which was laid before this House on 28th January, be approved. The draft order providing for the taking of a census of the population of Great Britain in 1971 has been laid before both Houses of Parliament in accordance with the Census Act, 1920. The order, as a whole, is subject to negative resolution, but the Act requires that if there are any items of information which are wanted in the census returns but not specifically mentioned in the Act, the parts of the Order relating to them shall be subject to affirmative resolution of both Houses. There are 12 such items in this draft order, and these are the subjects of my right hon. Friend's motion.

But we now have the opportunity to debate the order in full because of the Motion standing in the names of hon. Gentlemen opposite, and I propose therefore to say something about the general provisions of the draft Order before turning to particulars which require affirmative Resolution.

In accordance with the Act of 1920 the Order prescribes the date on which the census is to be held, the persons to be included in it, the persons who will be responsible for making the returns, and particulars to be given in the returns. Other matters concerned with the details of the arrangements for carrying out the census have to be prescribed in regulations, and these will be laid before Parliament at a later date; the regulations, which will be subject only to negative resolution, will contain facsimiles of the actual forms of return which it is proposed to use.

First, I should like to say some things about the date proposed for the census, which is 25th April, 1971. There have been censuses of population in Great Britain at ten-yearly intervals since 1801—with the exception of 1941, on account of the Second World War. Since 1961, however, mainly because of the greater mobility of the population—of whom about 10 per cent. now move house every year—and the consequent rapidity with which census figures for local areas become out-of-date, the ten-year interval has been accepted as too long, and, as hon. Members will recall, it was decided in 1963 to hold a mid-decade census in 1966, using a 10 per cent. sample of the population only.

The need for a further census after five years is just as great now as it was then, and we must therefore hold a census in 1971, which is, under the Census Act, 1920, the earliest year in which a census could be taken again. By holding a census in 1971 we shall maintain the 10-yearly series, and therefore be able to make direct comparisons between the changes occurring in the period from 1961 to 1971 and those occurring in previous ten-year periods.

The choice of Sunday 25th April is in line with past practice. Sunday night is the night of the week when most people are at home, and April is a time of year when most holiday movement has not yet started but when the lighter evenings, combined with better weather prospects than earlier in the year, can be expected to help the enumerators in their task. 25th April is the corresponding Sunday to those on which the 1961 and 1966 censuses were held and is well clear of the Easter weekend.

Hon. Members will next wish to know about the type of census which it is proposed to conduct. The draft order provides for a count to be made of the entire population in Great Britain and for all the relevant questions to be asked in respect of everyone. Although it was found sufficient to conduct the 1966 mid-decade census on a sample basis, the sampling procedure depended on the complete count having been taken only five years previously. It would not be practicable to repeat this procedure on the basis of a full count that will be ten years out of date. In particular the annual estimates of the population of local authority areas, based upon census figures, will by 1971 require the full correction that will be provided by a complete count of heads.

The advantages of obtaining answers to all the census questions in respect of all the people, as opposed to asking some of them on special sample questionnaires, as was done in 1961, is that it enables us to produce accurate figures relating to particular groups of the population, even if they form a small proportion of the whole. For example, information is needed on highly qualified manpower, who constitute a very small section of the total population; the information needed is not just the qualifications held by these people but also their current occupations, their ages and other items included on the form.

Another example is the members of the immigrant communities, for whom we shall be concerned to have figures showing what progress has been made in providing adequate housing and education and in assimilating them into the employment structure. But we do not just need the full picture in relation to minority groups of the population; by asking all the census questions of everyone we shall make it possible to carry out special studies of particular areas of the country. A sample census is of limited usefulness where small administrative areas are involved.

Turning now to the proposed contents of the census forms, the particulars to be asked for are listed in Schedule 2 to the draft Order. They are set out there in a manner consistent with the legal nature of the Schedule and as concisely as possible, but the actual questions will appear on the forms in more everyday language, and there will be notes to help people understand how particular questions apply to them.

Mr. Antony Buck (Colchester)

May I take it that there will be regulations following this Order and setting out these matters in greater detail?

Dr. Dunwoody

I said that there will be regulations at a later date, and there will be a facsimile of the census form.

The basic purpose of the census is, of course, to provide statistics on the size and structure of the population by areas, but the proposed questions will give information on a wide range of other topics of concern to Government Departments, local and regional planners and those engaged in many fields of social and economic research. The census is a major source of information in the fields of housing, employment and education. For the provision of these services reliable and up-to-date statistics are essential.

Planners and others look to the census for information on migration patterns and the demands for road space and transport facilities. Information on the trends in population change is a basic requirement for making population projections, on which, in turn, so much of forward planning of all kinds depends.

There have been many other items which have been suggested for inclusion in the census but which have had to be omitted. Some of these were unsuitable for an operation such as the census in which the information is obtained by means of forms filled in by householders. There have also been other matters on which it would have been desirable to include questions if room had permitted, but we must keep the form to a manageable size.

Regard must be paid to the burden which the census places on householders and it was found, in the course of testing out a proposed form of questionnaire, that the range of questions included on that form would need to be reduced. Of course, making that reduction has meant sacrificing some of the useful information which it had been hoped the census could provide. The final list, in Schedule 2 of the order, represents a set of questions which will be only slightly greater than that of the 1966 census, and which experience tells us is not an impracticable one for members of the public to cope with.

We are grateful to all those members of the public who, in various parts of the country, took part in the test censuses held in 1968 and 1969 and have, by doing so, greatly helped us to formulate the proposals for the census.

I come to the items which are subject to affirmative Resolution. The first of these relates to the relationship of each member of a household, to the head of the household or to a person's position in an institution or establishment. These are customary questions in a census but are subject to Resolution because they are not specified in these precise terms in the Census Act, which mentions only relation to head of family.

Relationship to head of household enables us to produce statistics on the composition of private households; position in an institution or establishment is required in premises other than in private dwellings for the purposes of interpreting the census forms and distinguishing between, say, the staff and the patients of a hospital.

The second and third items relate to new questions which have not been included in previous census. They will be of particular use in enabling us to produce from the census the information which is needed about progress in meeting the special needs of immigrant groups, in particular enabling us to produce basic data about the different immigrant groups: numbers, age-groups, areas of residence and occupations. This will apply to all immigrant groups, whatever their country of origin, and this data will be important both in terms of planning in the sphere of community relations and as a background against which the effect of existing policies is decided.

The proposed question on year of entry into the United Kingdom is necessary for the purpose of studying the time element in the process of assimilating immigrants to the general national standards in, for example, employment and housing. With the same purposes in view, the proposal to ask a question on the countries of birth of a person's parents will help to identify the numbers of people born in this country to immigrant parents. There are also a significant number of people born overseas to British parents who can, by means of this new question, be excluded from the count of immigrants.

The next item is concerned with people's addresses at earlier dates; that is, at a date one year and a date five years before the census. This will make it possible to measure the extent and characteristics of population movements both into Great Britain and within the country. The measurements of internal migration are essential for local population forecasts on which a wide range of planning work depends. This subject was first introduced into the 1961 census, and repeated in 1966 when the questions were virtually identical to those now proposed.

As in 1961, questions are being asked of married or formerly married women concerning the dates at which they first married, and also when their first marriage ended, if it has ended, and about the children born alive to them in marriage. This time, however, instead of asking merely for the total number of children born in marriage, women will be asked for the date of birth of each child. The need for this stems from our recognition of the fundamental importance of accurately assessing present trends in fertility, with particular emphasis on the spacing of births as well as their total number.

Mr. Eric Lubbock (Orpington)

How will the Department get this information when the question is being asked only of married women, which the Department is only entitled to do under the Act? Women who are not married are surely equally relevant in this connection.

Dr. Dunwoody

I am coming to that point. As in 1961, questions are being asked of married or formerly married women concerning the dates on which they were first married. To avoid asking about matters which some people would be unwilling to record, the question does not refer to illegitimate births. The point raised by the hon. Member is what we are interested in, the patterns of married women which can be demonstrated effectively. Also, in order to simplify the questionnaire for older people, the questions on fertility have been confined to women who will be under 60 on the census day. Neither of these limitations will seriously detract from the value of the information obtained.

The next few items are concerned with education. It is proposed to ask whether a person will be a full-time student in the ensuing academic term; this information has been obtained in previous censuses and is needed to supplement the particulars of current employment in order to identify those who are working only temporarily at the time of the census and do not therefore form part of the normally available labour force. As the census will be taken in April, there may be a substantial number of students on vacation work at that time.

A new question is proposed concerning educational attainments of the standard of G.C.E. "A" levels and its Scottish equivalents. This replaces a question asked in the 1961 census on the age at which full-time education was terminated but is intended to serve the same purpose as far as those not proceeding beyond the secondary school stage is concerned; it will provide figures for use in planning the provision of educational facilities and by cross-tabulation with data on occupations, etc. will help in estimating the return on the investment in education.

The answers to this question will supplement the information obtained under the next heading, which relates to higher educational, professional and vocational qualifications. This heading covers qualifications obtained after the age of 18, at first degree level and above, including those awarded by professional institutions; information concerning people with these qualifications is becoming much more important for planning purposes in the field of further education and in connection with the economic deployment of highly qualified manpower. The question also covers specific vocational qualifications, in particular those of teachers and nurses, because there is a continual need for information on the number of people qualified in these fields but not currently employed in them. This question on degrees and other attainments since the age of 18 is not a new one; the same information was obtained in the 1966 census, and there was a similar question in 1961 which was restricted to qualifications in science and technology. What is new in the present proposal is the inclusion of the names of the institutions which awarded the qualifications among the particulars asked for; the purpose of this addition is simply to improve the classification of qualifications to their appropriate levels.

Questions on employment have featured in many previous censuses and are specifically mentioned in the Act. Only two items among the present proposals on this subject require affirmative resolution. Firstly it is proposed to repeat the 1966 question on the main means of transport used by people travelling to their places of work. With the continuing increase in commuter traffic, there is a strong demand, particularly among authorities in the conurbations, for detailed analyses of the kind produced from the 1966 census, in which information on means of transport was correlated with that on area of residence and area of workplace.

A new item concerns a person's occupation one year prior to the census. This is a straightforward question which will give information for cross-tabulation with particulars of the current occupation and thus enable analyses to be made of movements between occupations.

The last three questions referred to in the motion are among those which will be put to heads of private households only. It is proposed that a question on housing tenure, asked in 1961 and 1966, should be repeated. This will provide information concerning the extent to which private householders own their accommodation, rent it from local councils or rent it—either furnished or unfurnished—from private landlords.

A further proposal requiring the approval of the House concerns a series of straightforward questions about the availability and sharing of certain amenities on the same lines as questions asked on the past two occasions. The amenities concerned are cookers, sinks, piped hot water, baths and showers, and lavatories. Information concerning each of these is required as an indicator of progress in meeting housing needs; analyses produced from it are important for the formulation of national and local housing policy.

The last item concerns a question on the number of private cars which was introduced in the 1966 census and which it is proposed to repeat in a slightly modified form. To those concerned with town planning and the solution of traffic problems the value of a count of the cars in private use is that this provides a measure of the demand for space on the roads which is not limited, as the question on transport to work would be, to those used for commuting purposes. In addition to this use the question will provide data which can be correlated with other household characteristics such as socio-economic group, which can be derived from the census.

This brings me to the end of the list of items referred to in the Motion. I have indicated which of them relate to questions being introduced for the first time into a census, and I would now like to refer briefly to the other new items. It is proposed to supplement the customary question on a person's occupation by asking also for a short description of the work done in that occupation, to help in classifying the job properly. There is a question on normal hours worked in a person's job which extends the question asked previously only of part-time workers to all in employment. Another proposal concerns the identification of households living in structurally separate dwellings and those whose accommodation is not self-contained; in this connection it is proposed to ask a question about the sharing of rooms, halls and landings on the form, whereas in the past it has been the practice to leave this task of classifying the accommodation entirely to the enumerator.

Hon. Members will see at the end of Schedule 2 to the draft Order that it is planned to ask questions about the speaking of the Welsh language in Wales and the Gaelic language in Scotland. Census information on these languages was last collected in 1961. On this occasion there are new items concerned with people's ability to read and write these languages.

Having mentioned these new items I would like to point out that there are a number of questions asked in the past which have been omitted from the proposals this time. For example, a person with more than one job will not be asked—as he was in 1966—for details of his second job. Questions on the age at which a person's full-time education ended, and on that of garaging of cars have also been dropped. I mention this lest hon. Members should get the impression that the increase in the length of the census form will be greater than is in fact the case.

I referred earlier to the tests which the General Register Office has carried out. All the new questions have been tried out by that means, but there is one which was tested and has been omitted but on which I have something to say. I want to say something on the question of income, on which the need for information has been strongly pressed by almost all users of census results. Local authorities, regional planners and central Government require information on the relative wealth of different areas, so that services can be planned to meet local needs and economic aid can be directed to the places where it is most necessary. At the national level income statistics would serve to identify more reliably those sections of the population in which genuine poverty exists.

It is true that the Inland Revenue compiles statistics on income distribution, but these statistics do not enable income to be related to data on the other subjects covered by the census, nor do they provide the complete picture of the incomes insufficient to attract tax, where much of the interest in income statistics is concentrated. Both the General Register Office and the Inland Revenue are debarred by the rules regarding confidentiality from collating the separate records which they hold on an individual basis.

There was thus a strong case for investigating in the test programme the practicability of including a question on income in the census questionnaire. However, in the course of trying out the questions two particular obstacles were encountered. Not unexpectedly the test results highlighted the difficulty of wording a question in a suitable manner to deal adequately with the complexities of income within the confines of the space available on a census form. There was also some evidence of public opposition to the question—less than was expected, but still sufficient to be taken into account, because of our wish to avoid asking any questions which even a minority of the public might find objectionable.

But the value of obtaining income data for cross-tabulation with other census characteristics is undoubted. It is therefore proposed to conduct a separate voluntary income survey of a comparatively small sample of households immediately after the census. Questions will be sent by post to 1 per cent. of households. There will be a separate form for each person, and the forms will be returned by those who are willing to respond, direct to the census office. Because participation in this survey will be entirely voluntary, it does not require legislation and is not therefore among the proposals to which I am now seeking the approval of the House. I have referred to the increasing demand for information from the census, but if it is to have its full value this information must be made available as soon as possible after the census has been taken. There was a marked improvement in this respect in the 1966 census as compared with previous censuses, but our aim is to reduce the time taken to produce the main results still further and to achieve the two-year target set for this. To this end, and bearing in mind that data relating to the whole population and not only to a 10 per cent. sample will have to be handled this time, the General Register Office now has a computer of very much greater capacity than that used for the last census. This increased capacity will be used not only to speed up the results and to handle the larger volume of data but also to produce some advance tabulations which will be available within six to nine months of Census day.

Mr. Lubbock

Can the Minister give some information about the computer and whether the software is all ready for the analysis to be made?

Dr. Dunwoody

I will leave the details of the computer to my hon. Friend who will be answering the debate. This of course will be in addition to the usual report containing preliminary population figures produced within three months of a census being taken. I would like to say a word about the effect of our proposals on the members of the public who will be supplying the information. We realise very well that the taking of the census involves some invasion of people's privacy. But government, central and local, is under continual pressure to develop the social and other services, to improve the planning of the economy and to make sure of the right priorities in capital investment.

How can all this be done without having available the hard facts on which to work? The necessary information in many cases can only come from the private citizen. But this reliance on individuals to give their personal information in turn imposes a responsibility on us to create complete safeguards against any breaches of confidence when the information is collected and handled. These safeguards take the following form: we have made provision in the Census Order for anyone who does not wish to give information to the head of his household to make a separate return which he can hand to the enumerator himself. And if anyone making a return does not wish to disclose it to the person appointed as enumerator he will be able to send in the form direct to the local census officer.

Once the information has been received at the General Register Office it is not released in any form in which individual persons or households can be identified to any other Department of Government or indeed to any other body. And there are of course penalties under the Census Act for disclosing or improperly using information given on the census form.

In a word, the census is a purely statistical operation. There is no danger of any breach of confidence. I think it important to emphasise this, so that people can be assured that they can safely give the particulars required. The success of the census must depend ultimately on the part played by members of the public and the care which they are willing to devote to the completion of the forms. I hope, too, that hon. Members will feel satisfied on this point and also that they will have the confidence which I have in the census plans which the draft order authorises.

Mr. Speaker

Order. May I remind the House that my Deputy in the Chair ruled earlier we are taking with this Order, which is an affirmative Order, the Prayer No. 4 on the Order Paper. If this debate lasts after 11.30 I will be unable to put the Prayer.

9.33 p.m.

Mr. Antony Buck (Colchester)

The reason why the Prayer was put down by my hon. Friends and myself was to enable us to have the free-ranging debate, which we have been having. I would like to thank the Minister for the great meticulousness with which he has dealt with the Schedule. It has taken a certain amount of time and I hope that he will forgive me if I do not follow him through the intricacies and details of Schedule 2. It is sufficient to say that we think the balance is about right.

Over the last decade or so there has been what I think might be described as a happy consensus between all three parties about this matter of the taking of a census to provide the essential and important information necessary for future estimates of development, particularly within the realms of housing, education and the public services. The need for a full census every 10 years and a mid-term one is illustrated by what the Miniser said when he repeated facts given earlier, that 10 per cent. of the population would appear to change its address each year. That came as a very surprising statistic to me when I first read it. It perhaps illustrates the changing character and fluidity of our society today and emphasises the need for a census of this character.

The House has recently had its attention drawn, through the initiative of the hon. Member for Birmingham, All Saints (Mr. Walden) to the need for privacy and of the increasing public concern at the infringement of privacy. The hon. Gentleman's initiative caused the House to reflect on this and the Government announced that there would be a Commission to go into the whole matter. There is growing public concern about infringement of privacy and I regret to say, in the context of this public concern, that it is, I fear, just possible that there may be marginally less public acceptance of the taking of a full census than on past occasions. It seems to me necessary, in order to gain public support, that the strongest and the most emphatic assurances should be given about the utter confidentiality of the information supplied. As has been said in previous debates, the success of a census depends on public goodwill and co-operation.

The assistance in past years of the broadcasting service and the Press has been sought, and I imagine that this will be done again. Perhaps the Minister who is to wind up will deal with this aspect of the matter. The widest publicity must also be given to the fact, which was mentioned in passing by the Minister, that any person employed in the taking of a census who misuses information which he gains through being involved in the census will render himself liable to a penalty under Section 8(2) of the 1920 Act. These penalties are not inconsiderable. There is a penalty of two years' imprisonment—I think in the original Act it was of penal servitude—combined with an unlimited fine or, alternatively an unlimited fine. I was wondering whether it would be correct to advocate increases in the basic penalty, but what is referred to in criminal circles as "two years inside" would appear to be sufficient. If the census is to be successful, this fact must be given publicity in order to ensure additional public confidence as to the confidentiality of the matter.

Further, it might be appropriate—and I hope it will be considered appropriate by the Minister—to give the House an exact account of what precisely will happen to the completed census returns after they are gathered up on and after Monday, 21st April, 1971. What precisely does happen to them? Where do they go? What security precautions are taken all the time to see that they are kept under lock and key? It is essential at every stage to see that the information contained within the census forms is kept utterly confidential. This is a highly important matter in view of the increasing public concern about the citizen's right to privacy.

Publicity should also be given—and this again was a matter mentioned en passant by the Minister—to the rights of the individual conferred by paragraph 5(1) of the Order—that is the right of a person to make a personal return if he chooses, rather than that the return should be made by whoever is the nominal head of the household. This is an important right and when exercised precludes anybody else from making a return about that individual. This additional safeguard should also be given maximum publicity.

I notice with some interest that, in this census, it is not intended that additional information shall be provided from a sample proportion of the population. I was surprised by this, but I think that what the hon. Gentleman has said satisfied me on the point, particularly what he said about the voluntary censuses which there are to be about additional matters. It seems right that this should be an over all census for added simplicity.

What the hon. Gentleman had to say about the length of time before the results of the census become available was a little disappointing, although one commends the fact that the results in total will be available in two years compared with a period nearer three years in the past. I am sure that the hon. Gentleman will respond to the request made to him by the hon. Member for Orpington (Mr. Lubbock) for details concerning the computer which is to be used in processing the information.

Another matter is the question of the cost of the census. Perhaps the hon. Gentleman can tell us the additional cost because of the decision not to have a sample census. I welcome the fact that there is to be additional information, particularly as it is appropriate that we should have the additional information provided for in paragraphs 7 and 8 of Schedule 2 which will require information to be given In respect of any person not born in the United Kingdom…". in that he is required to state the year of his first entry into the United Kingdom and give details of the country of birth of his father and mother. This information is especially necessary for those of us concerned particularly with matters relating to immigration.

There are some sensible Amendments in the order. It was an absurd requirement in the earlier censuses for details of the fecundity of ladies over 60. It was ridiculous to ask them to provide such information. One or two such details have rightly been eliminated this time.

We put down this Motion in order to enable a wide-ranging debate to be held. We shall view with interest in due course the pro forma of the census. Five years ago, in a debate on the subject of presentation, Mr. Geoffrey Howe, who, we hope, will be with us again before long—I am sure that he will—made substantial points concerning the format and layout and the way in which the census should be worded. I hope that we shall hear something about that and that, when the pro forma comes up, it will be in a rather more modern idiom than in the past.

We welcome the taking of the census. On a similar occasion many years ago, my right hon. and learned Friend the Member for St. Marylebone (Mr. Hogg), who was then in another place, commented on the somewhat cumbersome character of the whole procedure under the substantive Act, and, in due course, when we have returned to power, we shall examine this matter in detail and perhaps effect some alterations. Meanwhile, we welcome the proposals for a census in 1971 as provided for in the Order.

9.44 p.m.

Mr. Eric Ogden (Liverpool, West Derby)

You indicated, Mr. Speaker, that if this debate went on longer than 11.30 p.m. there might be some difficulty. I have no intention of trying to keep it going more than five minutes.

The hon. Member for Colchester (Mr. Buck) referred to something which caused me concern—the happy consensus between Government and Opposition, whichever party happened to be in office, about the usefulness, effectiveness and extensiveness of censuses. I like the idea no more because there is a happy consensus than I would if there was no such consensus.

The hon. Member referred to the importance of having the public's good will and co-operation but he managed to avoid mentioning, in a wide-ranging speech, that the actual return of the census form is not voluntary. This is the part that concerns me and certainly concerns some outside. So I join in this debate to indicate that there is one Member of the House—and possibly there are many others—who likes it none the more because it is compulsory than if it had been voluntary.

I accept what was said by my hon. Friend the Joint Parliamentary Secretary. He gave a very full and reasonable account of the reason for the changes in the procedure and the changes in the questions that have been asked. He indicated that the information which would come in from this census would be used to good purpose, to provide better services for people throughout these Islands.

My hon. Friend will accept that some of the information is already available in one form or another in Government Departments, though it is a fact that doing it in this way brings it in at this central level and for a good purpose. Good purposes are not always a good reason for doing compulsorily what might be done voluntarily. I accept that in this I am thoroughly reactionary. My hon. Friend can claim precedents going back 2,000 years, for we read in the Gospel of St. Luke, chapter 2, that: There went out a decree from Caesar Augustus that all…should be taxed…and all went…to his own city. but I do not believe that he would claim that Caesar Augustus was a very good precedent.

Mr. Speaker

Order. We are not discussing that census at the moment.

Mr. Ogden

I am simply making the point that, regardless of the reasonableness of the questions and regardless of the use of the information that comes along, something sticks in my own throat and in the throats of many outside when they are asked to give information compulsorily, not because of something like income tax or claims they are making but to give information by compulsion with the threat of prison. It may be that at some stage parts of this could be extended and only the more essential information sought compulsorily, other information being asked for volutarily.

I would like an extension of the proposal for a voluntary census. My hon. Friend might then be able to compare the results obtained from such a census with those obtained compulsorily. I am simply voicing my objection to a compulsory census while accepting in full that the intentions of my hon. Friend and those of the Opposition are entirely for the best.

9.48 p.m.

Mr. Eric Lubbock (Orpington)

I am glad that the hon. Gentleman has raised the question of the principle of voluntarism because at least on an occasion when we are discussing a census order it ought to be ventilated and we should have proper answers from the Minister who replies as to why these census returns have to be compulsory.

If one had evidence to show that a large number of people would refuse to fill in census forms and that their reluctance to do so would make the results less significant or that they would be in any way undermined by the refusal of particular categories of people, which would distort the results, I could understand the need to make return of the forms compulsory and provide for penalties in the event of a refusal.

The hon. Gentleman, however, did not argue this in his speech. Looking back over the years, as far as I am aware—though I have glanced only at the 1965 debate—it does not seem to have been a matter of any concern to the House. I agree with the hon. Member for Liverpool, West Derby (Mr. Ogden) that in spite of the fact that there are respectable precedents in this country going back to the Census Act, 1920, we should review this and find out whether we could not obtain this information by persuasion rather than compulsion. If not all of the information requested in the forms could be obtained in that way we should consider whether that would in any way lessen the value of the results to the Government and the various Departments which are to make use of the information once it comes in. I would be grateful if the hon. Gentleman who is to wind up would deal with that point.

My next point concerns the question of convincing the public that what they are doing in filling in this form has a value in providing them indirectly with the services the Government are planning. The hon. Member for Colchester (Mr. Buck), who spoke about publicity through the radio, television and the Press, was absolutely right: obviously a good deal of money must be spent in explaining to the public, as the Minister explained to the House so painstakingly tonight, why this information is required and how it will benefit the form fillers as citizens in terms of better services the Government can provide.

Apart from broadcasts and announcements in the Press, there is another thing which is of equal importance. Not all of us can manage to see every television programme or read the newspapers every day. I would like to see the sort of explanation that the Minister gave this evening included in the census form itself. I see no reason why a brief summary should not be included of the use to which the answers to the various questions will be put when they are received by Government Departments. If it is possible for the warning to be put on the form, as I believe it is, in heavy type explaining to the citizen that if he refuses to fill in the form he is liable to certain penalties, I do not see why——

Mr. Speaker

Order. The hon. Gentleman cannot amend the Order. He can, however, denounce it.

Mr. Lubbock

I am not suggesting that the order should be amended. It is a question of how the questions that are provided for in the Order are treated in the form. We shall have another opportunity of discussing the contents of the form. It is right for us to suggest at this stage that the Government should consider these matters in deciding how these questions are translated into a document which goes into households.

As to the questions, I thought that the Minister gave a very good explanation—we are grateful to him for it—of the italicised questions which have to be dealt with under affirmative Resolution. I was happy about most of the hon. Gentleman's answers, but I am not sure that questions 7 and 8 are not ultra vires. The Schedule to the Census Act, 1920, provides that Any other matters with respect to which it is desirable to obtain statistical information with a view to ascertaining the social or civil condition of the population may be dealt with in the census return.

I do not see that the social or civil condition of the population are affected by the country of birth of one's father or mother or, in respect of persons not born in the United Kingdom, the year of first entry into the United Kingdom.

I can understand that the Minister may say that it would be useful for the Government to know how many immigrants there are in a certain area and to be able to distinguish the true immigrants as it were, from those who just happen to have been born in a foreign country but whose parents were citizens of the United Kingdom. This might be useful from many points of view, but I question whether the Minister has any right to ask those questions under the terms of the Census Act, 1920.

I think that all the other italicised questions are reasonable and that they will, in the terms that the hon. Gentleman described, provide useful information for the Government. It always surprises me that, for instance, local education authorities have practically no idea of how many primary school places they will have to provide in the next few years, because no information is available about the number of children under the age of five. Such information is obviously of vital importance to local education authorities in planning the future provision of school facilities.

It must be remembered that when people are asked to give this rather personal information about their own households, personal activities, employment, and so on, they are likely to ask what use will be made of it. I am grateful to the Minister for the assurances he gave about the confidentiality of the returns. However, there is one point that causes me some anxiety on this matter, and I shall return to it.

Before leaving the questionnaire, may I mention one matter which my hon. and learned Friend the Member for Montgomery (Mr. Hooson) has asked me to raise. I have discussed it with him, and I think that what he says is reasonable. It relates to the questions on the speaking of Welsh. My hon. and learned Friend points out that a large number of Welsh people have been forced to leave their native country to seek work in other parts of the United Kingdom. In addition, many of them come here as teachers in our schools, and we are grateful to them for that.

Many of them are native Welsh speakers, and my hon. and learned Friend says that from this questionnaire as it stands we shall not get an adequate picture of the number of Welsh speakers in the United Kingdom as a whole. As the purpose of these three questions is possibly to decide such matters as grants for Welsh books, the amount of Welsh broadcasting that is necessary on television and radio, and the number of teachers of Welsh required in schools, it is necessary for the Government to have an overall picture and not one which is confined to the Principality itself. I should like the Government to consider that so that I can discuss the matter further with my hon. and learned Friend.

I said that there was one issue which caused me concern on looking at these questions, and that is the sale of information derived from the census to commercial organisations for use in market surveys which they undertake. The sample census of 1966 was sold to an organisation called B. I. A. Ltd., which paid a large sum of money for all the data derived from the census. Although the hon. Gentleman says that the people who take the information are required to observe absolute confidence about individuals—and one accepts that, because of the penalties which go with the release of information about individuals—it is still possible for these commercial organisations to buy the information concerning ward libraries, which consist of 220 houses in a census district. It is possible for individuals to be identified within that tranche of 220 houses, and for secrets of their personal lives to become known to people with commercial interests.

That causes me some anxiety, and I should like the Government to say that they are not going to sell any of the information derived from the census, which, as the hon. Gentleman explained, is primarily of importance to Government Departments, local authorities, transport planners, and so on, and not to commercial firms. It is not the function of a Government to collect information of a highly confidential nature from private citizens and then sell it to somebody for purposes of commercial gain. From the Government's point of view the sums involved are not very large, and I think that they could forgo this income without any great loss to the taxpayer.

I should like an undertaking that this practice will be discontinued with the 1971 census. If someone wants to buy a complete census referring to a county, or county boroughs, that is harmless, because the number of people concerned is far too large for any individual to be identified, but I am not happy about this practice of selling ward libraries, and I should like it to be discontinued.

One cannot object to a census being taken as such. The information is required—indeed, it is vitally necessary—to enable the Government and local authorities to plan the provision of facilities for the citizens they serve. I am not objecting to the Census Order, but I should like some reassurance on the points that I have raised before we pass the Order this evening.

9.59 p.m.

The Joint Under-Secretary of State for Scotland (Mr. Bruce Milian)

As there has been a general welcome to the Order, it is perhaps unnecessary for me to spend any time on a justification of it, but I should like to take up a number of points, some of them quite general, which have been raised during this short debate.

The hon. Member for Colchester (Mr. Buck) raised the whole question of confidentiality. It will be an important part of the pre-publicity for the census to get over to members of the public that the information they will be providing in the census returns will be treated with extreme confidentiality. I agree that there is a question of public confidence here, and that members of the public are increasingly concerned about what they rightly consider to be infringements of privacy. The Government, as witness the recent establishment of the Committee, are very concerned about this. In this atmosphere it is essential that we should persuade members of the public that the information asked for is, first, absolutely essential and, secondly, that it will be treated with great confidentiality, and that no details of any individual or any individual family will be disclosed to anybody.

The range of census questions is restricted to some extent because of the need to get public confidence. There are obviously many questions that it would be useful for the Government to ask. Questions about income were mentioned by my hon. Friend the Joint Under-Secretary of State in opening the debate, but these questions are not included in the census because there is no guarantee that they would be as widely accepted by the public as is necessary if the results of the census are to be of the accuracy and completeness that we must have, and if people are to be persuaded to co-operate. Not many questions in the proposed questionnaire are likely to be sensitive questions, although there may be questions which are perfectly innocent for most of us and yet touch a point of sensitivity for certain members of the population.

There are penalties for disclosure of census information. In addition to penalties covering those who are engaged in the taking of the census, the enumerators and so on, there will also be a regulation to forbid heads of households who acquire information relating to other people in the household, or heads of establishments who acquire information relating to people in their establishment, from disclosing that information or using it improperly.

It will be open to anyone making a return to send it to the local census officer in preference to handing it to the enumerator, and we shall try, as far as possible, to make sure that enumerators do not know personally the families they will be covering in the course of their work. The order also provides that the individual people in a household can make their own personal returns. Their returns do not have to go through the head of the household. Similarly, in institutions and so on, individual returns, as distinct from institutional returns, are perfectly possible and can be channelled in a way that allows no one to see the information that is provided.

After the various analyses have been done and the results published, the returns are passed to the Public Record Office, where they are not made available for inspection by the public until 100 years have elapsed, and I think that is a fairly generous assurance of confidentiality. I am informed that all census returns after 100 years are extremely useful for historical researchers.

As regards passing of information to other Government Departments, local authority associations and public bodies, no information is passed out from the census office except on a statistical basis in the published documents. Nothing is passed out in a way that can identify any individual, or relate individuals to the particular pieces of information that are being passed out.

The hon. Member for Orpington (Mr. Lubbock) raised the matter of sale to commercial organisations. I assure him that areas for which information is given on a statistical basis are deliberately kept large enough to avoid any risk of identifying individual people or households. If there are particular instances in which the hon. Gentleman feels that this principle has been breached I shall be glad to look into them, but that is the principle on which information is made available. It would be too restrictive to say that under no circumstances whatever is information to be available to any outside commercial organisation. After all, many of the public documents must in any case he available to commercial organisations. I do not see that there is any principle here of the kind the hon. Gentleman has in mind. But I agree with him that one must not break down the results into such fine areas that, with a little intelligent deduction, it is possible to identify particular families by information perhaps pertinent to a small number of families. We intend that that kind of thing should be avoided.

Mr. Lubbock

Perhaps I could give a hypothetical example, not an actual case. Since one of the questions relates to the number of motorcars as related to members of private households, let us say that among 220 households one family possesses five cars. It would not be difficult for a commercial organisation interested in selling motor car accessories to pick out that one household from the 220.

Mr. Millan

I am not sure w ere the 220 the hon. Gentleman has in mind comes from, but I make the general point which I have already made that the information will not be supplied cut down to such fine areas or such fine pieces of information that it will be possible to make the kind of identification the hon. Gentleman has in mind. The areas for which information might be supplied, and the kind of information, must all depend on the particular item involved. What may be sensible for one particular item of information might not be sensible for another. I have stated the principle on which it will operate and I would be happy to look into any particular instances. It is important to give the assurance that it is an absolute principle of the census that information is not made available to anybody, it does not matter whether they be commercial organisations, public bodies or anyone else, in a form to enable them to identify particular pieces of information and individual householders or members of families.

The hon. Gentleman the Member for Colchester also dealt with the question of sampling. Since he accepted the principle on which the census is based, perhaps it is not necessary for me to deal with the matter at length since my hon. Friend mentioned it in his opening remarks. We looked at this matter extremely carefully and I took some persuading that a sample would not do the job as regards certain items of information. But all the professional advice we have is that there should be 100 per cent. enumeration, although in the actual publication of figures and the use of information afterwards, having got the 100 per cent. information, certain sampling techniques would be possible and, for certain purposes, a sample will be done at that particular time. Our professional advice is that it is essential to have all the information in the census if it is to be complete, and if we are to get the kind of information available in the detail in which we require it to justify the tremendous effort that the census involves.

I was asked about the publication of results. I agree that it is important that the results should be published as quickly as possible. I do not have time to go over all the details of publication, but perhaps I may run over one or two matters. Within three months of census day we should be able to give an enumerated population by sex for the whole of the country and for regions and local authority areas. Within six to nine months there will be another preliminary report giving some additional information, sex, age, marital condition, country of birth, distinguishing between the working population and the rest for the whole of the country, for regions and for local authority areas. Within 12 months of census day, certain sample tables will be provided. Then we go on to provide more detailed information with the target in mind of having all the basic information published within two years of census day.

This will be very much better than the achievement in 1961 and also better than that of the publication of the results of the sample census in 1966. The time table in 1961 was a rather protracted one, and the main tables were not completed until more than five years after census day. The final volume did not appear until 1967. We hope to do a good deal better than that with the census of 1971. In 1966, there was an improvement over 1961. Virtually all the reports in that census had appeared by the end of 1969, but that was working with much less information since it was only a sample census. In 1971 we shall be dealing with very much more information, and the job involved will be a much greater one because we shall be dealing with a lot more information, because it is not a sample census, because we hope to speed up publication of the results, and for a number of other reasons.

The cost of the 1971 census is likely to be a good deal more than that in 1961. The gross expenditure in 1971 is likely to be about £10 million. Even allowing for changes in the value of money, that is considerably more than the census in 1961 cost. It is virtually double the cost, even in real terms. However, that is part of the price that we pay for additional information, for the methods of handling the information that we are adopting and for the quicker publication of the results. But, in terms of the value which can be obtained from the census, especially from the more speedy presentation of results, I think that this will be money well spent.

Mr. Buck

Can the hon. Gentleman break that down a little further? In actual money terms, it must be well over twice, if one takes into account the decline in the value of money. Can he give us any breakdown of those items which will cost more?

Mr. Millan

I gave the information in real terms because I thought that that was the fairest way of presenting it. I was not trying to take advantage of the fact that prices have gone up since then. The biggest item is field costs, taking the England and Wales figures, which will go up from rather less than £3 million to just under £5 million. That is an example of the kind of change in real terms between the two dates of 1961 and 1971. The computer costs are also up, the printing is slightly up, and the headquarters costs are also very considerably increased. However, there is a general increase in all the items involved.

The hon. Gentleman raised a number of other points, especially as regards the presentation and layout of the form. I have had regard to what he said. We are not discussing the layout of the form tonight. We shall be doing that later when it comes to the regulations. However, we have taken advantage of the pretests in 1968 and 1969, and we hope very much that, when the form is presented, it will be agreed that we have made a real effort to get a layout which will be intelligible to the people who have to complete it. This is not easy, but we hope that the layout will be an improvement on past censuses. We have taken a good deal of care with it.

My hon. Friend the Member for Liverpool, West Derby (Mr. Ogden) raised the question of voluntary censuses. His point had validity to this extent at least. Since the census is compulsory we have to persuade people that the information required from the census is essential. My hon. Friend, in opening, went over the items in considerable detail and produced a justification for all of them. We have not included in the order anything that we do not feel is absolutely essential. We have jettisoned particular items of information that we would dearly like, but which we felt would make the form too long or not acceptable to members of the public.

It is not practicable to do this on a voluntary basis. It is impossible to judge with accuracy what the response is likely to be, but we have experience of voluntary surveys. Certainly on that experience, including the pre-test censuses which are on a voluntary basis, we might get a response of about 80 per cent. It might be more, depending on the publicity and care with which we presented it and so on, but it would not be anything like a 100 per cent. return.

There is no guarantee that the forms returned would, as it were, be typical of the whole. I think there is almost a guarantee that they would be untypical of the whole and we could not gross the thing up and get information on an accurate basis. Therefore, if we are to have a census in any meaningful way, it means a compulsory census. We shall try our best in the publicity about the census to explain why the information is required.

The hon. Member for Orpington suggested that we might put something on the form explaining why the information on particular items is required. That would be impracticable without making the form an even more formidable document than it will look at present. Whether we could include on the form a general statement about the necessity for the information is a matter I will look into. But when we take account of the complexity of the information, to try to provide justification for it all, apart from the possibilities of arguments between the enumerators and individual members of households whether the justification was valid or not, would make the form far too complex. However, I will look into the question whether we should include a general statement on the form.

The hon. Gentleman mentioned items 7 and 8 in the Schedule and said that he thought they were ultra vires. I must dispute that. Apart from anything else, the order has, in the normal way, gone to the Select Committee on Statutory Instruments. I am sure that if it was ultra vires that Committee would have drawn it to the attention of the House.

The information for which we are asking about immigrants is extremely important and relevant in ascertaining the social condition of the population. If we are to improve our knowledge about the situation of immigrants in this country, the kind of information for which we are asking in items 7 and 8 is indispensable. I was glad that the hon. Member for Colchester, in his opening, welcomed the emphasis that had been put in the order on improving the quality of that information.

Concerning Welsh—and Gaelic is also concerned—the answer is that there are Welsh and Gaelic speakers in England as well as in Wales and Scotland, respectively. But we must have regard to the complexity of the form and the complexity of the information which we are asking people to provide. There would be a certain danger in asking people in England, the vast majority of whom would not be involved and would, as it were, be producing a nil return, to give information of this kind. That is why Gaelic is restricted to returns made in Scotland and Welsh to those made in Wales, including Monmouthshire, but the information asked for on the form about the reading and writing of Gaelic and Welsh is new, so we will have additional information compared to previous censuses.

These are virtually all the questions that I was asked. I am glad that hon. Members have generally accepted the necessity for the census and welcomed the Order and the very detailed explanation which my hon. Friend gave in moving it. I commend the Order to the House.

Question put and agreed to.

Resolved, That items 2, 7, 8, 10, the words in item 11(a) 'and the month and the year of birth of each such child', 11(b) and (c), 12, 13, 14(a)(vii) and (viii), 15, 18, and 19 in Schedule 2 to the draft of the Order in Council, entitled the Census Order 1970, which was laid before this House on 28th January, be approved.

Mr. Speaker

Does the hon. Member wish to move his Prayer?

Mr. Buck

No, Sir.