15 Jan 2003 : Column 249WH

Westminster Hall

Wednesday 15 January 2003

[Mr. Frank Cook in the Chair]

2001 Census in Manchester

Motion made, and Question proposed, That the sitting be now adjourned.—[Mr. Heppell.]

9.30 am

Mr. Keith Bradley (Manchester, Withington): We meet this morning in very sad circumstances.

Mr. Graham Stringer (Manchester, Blackley): Will my right hon. Friend give way?

Mr. Bradley : Of course.

Mr. Stringer : I thank my right hon. Friend and you, Mr. Cook, for that indulgence. I should like to take this opportunity, the first that I have had, to offer my condolences to the family of Stephen Oake, who was murdered in my constituency last night, while doing his duty as a constable in the special branch. I also hope that the other offices who were injured in the line of duty make a speedy recovery.

Mr. Bradley : I am sure that the whole House shares those sentiments. We all send our condolence to the family of Stephen Oake.

The debate this morning is to challenge the accuracy of the 2001 national census as it applies to the city of Manchester, to highlight the financial implications of the discrepancy for the people of Manchester and to seek urgent action by the Government to resolve the fundamental and crucial problems that we believe should be clearly identified.

Manchester is one of the country's foremost regional centres and it has made great strides over the past few years to strengthen its economic base. Its success is important not only for the people of Manchester but for the north-west and the nation. Government policies over recent years have recognised that fact.

The 2001 census results published on 30 September by the Office for National Statistics showed Manchester to have a population of 392,800. That is a reduction of 10.4 per cent.—nearly 46,000 people—in Manchester's population between 1991, when the previous census was held, and 2001. It is a reduction of 10.6 per cent.—nearly 47,000 people—in Manchester's population for the nine-month period from the mid-year estimate in 2000 to the 2001 census.

In Manchester, we were, to say the least, utterly bewildered by these figures. During the 1990s, our city centre population increased by nearly 10,000. Mid-year estimates throughout the 1990s indicated that the city's population had stabilised and had started to grow. So the question is; how can these wide variances be justified?

We are told by the ONS that, in Manchester and elsewhere, previously unrecorded discrepancies going back nearly 20 years have flawed the 1991 census and the

15 Jan 2003 : Column 250WH

mid-year estimates since then. The ONS has speculated that the rate of young men in particular emigrating from England was previously underestimated by some 50,000 a year; nearly 1,000 a week. However, this trend has not been substantiated or picked up by alternative sources, including the international passenger survey. Therefore, in our view, it is not credible.

The view of the ONS will be debated by statisticians for years to come, but the issue for the people of Manchester is not an academic exercise. We want to know why the administrative sources available to us and to the ONS showed that the census results from Manchester seriously underestimate Manchester's population. We also want to understand why Manchester's residents are being penalised for these fundamental weaknesses and why serious questions must be asked in Manchester as the revenue budget for 2003–04 is prepared.

Manchester city council is a responsible local authority. It has sought, since the publication of the 2001 results, to enter into a detailed dialogue with the ONS to understand the rationale of its population figures. The initial signs were positive. The chief statistician went to Manchester and met council officers; they agreed a process of what I will call engagement. However, the outcome of these discussions remains as unclear and as worrying for the residents of Manchester as when the figures were first published.

It has taken nearly three months to obtain the information necessary to undertake even partial scrutiny of the results. Critical pieces of information are still awaited. For example, the aggregate general practitioner list data are not yet available. The ONS states that it needs the permission of the NHS security and confidentiality advisory group to release the information. We ask the Government to look at this urgently to facilitate that information being made available to Manchester soon.

A new aspect of the 2001 census was an independent follow-up face-to-face census coverage survey. In order to check the representative nature of this survey and the sample that was used, we need to see whether the survey missed households. We need to know the total number of addresses or household spaces and households in each postcode unit. We need to know the total number of addresses of household spaces and households responding in each postcode. We need to know the number of persons enumerated and inputted for each of the postcodes used in that survey. The ONS says that by releasing the postcodes sampled in the census coverage survey it effectively would be releasing information about the households and individuals sampled. We dispute that and request that the Minister intervene quickly to allow that information to be released to Manchester city council so that it can check the validity of the sample survey. What information has been produced demonstrates that the number of people who live in Manchester is underestimated by the 2001 census.

No substantive reasons have been given as to why alternative administrative sources have been disregarded by the ONS, which was always seen to be part of the quality assurance process and therefore the overall census methodology. I shall return to this point shortly.

15 Jan 2003 : Column 251WH

Crucially, it has also come to light that something like 10,000 properties that the council believes to be occupied, as council tax is being paid, were not canvassed or identified by census enumerators. We strongly believe that if people are paying council tax in the city, they may actually live in the city. We believe that unless we have an accurate property base, we cannot produce an accurate population estimate, irrespective of the competence or otherwise of the ONS.

The city council has asked the ONS to undertake a match between its own property base and that identified by the census. That would at least assist in identifying the differences and therefore those properties that appear to have been missed. That seems a reasonable request and I should have thought that this exercise was quite straightforward. I am afraid that that is not the case. The city council was first told that the ONS had resource difficulties that make such an exercise very difficult in the short term. It was also told that the ONS had additional information requirements, which we believe can be quickly met but are delaying the process. When the city council suggested that an independent specialist agency should undertake the work as agent for ONS, problems of confidentiality were again raised, and these are apparently still being considered. This needs to be resolved as a matter of urgency so that we can test the validity of the property base.

I return to the issue of quality assurance and alternative data sources. The ONS has given Manchester—in confidence, obviously—the child benefit data used for the quality assurance process, which show that there are more children receiving child benefit in the data used by the ONS than were found by the one-number census or the mid-year estimate. It is clear that there has been a miscount. Moreover, the pensions data used for the quality assurance process—again, supplied in confidence—show more people in the 65 to 79 age group in receipt of pensions than were found by the one-number census or the mid-year estimate. That supports our argument that the statistics need to be validated.

It was interesting that in oral questions on Monday, my right hon. Friend the Secretary of State for Work and Pensions said, with reference to pensioner figures:

We concur with that, as it reinforces our case for a review of the data on the number of pensioners in Manchester.

The school census data used for the quality assurance process and supplied in confidence show more children aged five to nine and 10 to 14 in the school census than in the 2001 mid-year estimate, further reinforcing the point that when brought together, independent sources show a greater population in the city of Manchester than was shown by the 2001 census.

Similarly, the electoral register in Manchester shows an increase from 310,010 in 2001–02 to 313,573 in 2002–03. The census results for Manchester indicate that the city's total population aged 17 or more was 305,193. The electoral register figure was 4,817 higher

15 Jan 2003 : Column 252WH

than the estimate. If people are registering to vote in Manchester and, as I said earlier, paying council tax, that strengthens the argument that the census underestimated the numbers.

The quality assurance process was a key part of the ONS methodology, which was extensively consulted on by the ONS before the census was undertaken. The issue is not the principle of the process, but how it has been applied and, in Manchester's case, why the information has been ignored. The information given by the ONS shows as a recurring theme for Manchester that there are more people living in the city than were projected in the census results. We need to know the possible reasons why there are more children in receipt of benefit, more children in Manchester schools and more people in receipt of pensions than the census shows. The comparisons with the electoral register and the flaws identified in the ONS property base mean that the census results for Manchester have no credibility.

Mr. David Kidney (Stafford): Does my right hon. Friend intend to deal with Manchester's status as a university town? Is he aware that Manchester is not alone in being an apparently thriving area, but having a reduced census figure? Other cities with a similar complaint include Cambridge, Oxford, Bristol, Nottingham and Stafford. Although not everyone may know of Stafford's status as a university town—it is one—the others are well known. Is that a factor in my right hon. Friend's considerations?

Mr. Bradley : Yes. The large number of students in the city is an added factor; a complicating factor, because that is such a mobile population which, to be fair to the census enumerators, caused some difficulty in the count. That is why I have avoided using that as a key indicator of the reduced population, and have instead used data that are easy to substantiate from alternative sources. The student population is certainly another factor that has not been taken properly into account or counted as part of the overall population figure. The examples that my hon. Friend gave of similar situations in other towns with a high student population reinforces the argument for re-examining the census data and the extent to which the population has been underestimated.

Mr. Bob Laxton (Derby, North): I endorse my right hon. Friend's comments. Derby is also affected by a huge reduction in population—about 14,500, according to the ONS—and is also a university town, which must be factored in. My right hon. Friend may be interested to know that I have been contacted by two former employees of the ONS data centre in Widnes, who told me that gross errors were made in the processing of the data, that the optical character reading of the forms was, in many cases, gibberish and unintelligible and that there were huge pressures on the operatives in the data processing centre in Widnes to increase their hourly key strokes to a very high level. That was achieved by cheating; by people holding their fingers on keyboards and blasting through screens to try to get their key stroke figures up. Those two individuals, who have asked to remain anonymous as they are clearly whistleblowers, pointed out forcefully to me that there will be huge distortions and errors around the country

15 Jan 2003 : Column 253WH

as a result of the ineffective and inefficient way in which the data processing centre at Widnes, which was the only one handling the census details, operated.

Mr. Bradley : My hon. Friend makes a powerful point, which brings me to my next point. Whatever is said about the ability of the ONS to count, as an organisation it does not transmit an image of public accountability and responsiveness. We need to consider that carefully as a crucial issue, on which I am sure my hon. Friend the Minister will want to comment. While it prevaricates or fails to substantiate its decision-making for the results in Manchester and other cities around the country, as we hear, the decisions crucially feed into the rate support grant settlement for the coming financial year, which has been announced by the Office of the Deputy Prime Minister. We believe that in Manchester we would have lost more than £18 million worth of grant if the figures had been applied to the financial year 2002–03. I shall deal with the current year in a moment.

Such reductions are untenable in the light of what we believe to be the fundamental flaws in the census results and the grossly unfair impact that such reductions in grant will have on the people of Manchester, which is by no stretch of the imagination a wealthy community. It has many areas of poverty and deprivation that can ill afford any reduction in grant. We must address that issue urgently.

In Manchester, we all want to understand why the Government appear to be unwilling to scrutinise the results of the census and satisfy themselves that their resource allocation policies are a fair and accurate reflection of needs across the board. If the Government decide to use the ONS results, they are, by implication, telling everyone that they accept, at least for the time being, the adequacy of the information that has been supplied by the census. That is totally unacceptable. We also want to understand what steps the Government will take to make the ONS more responsible and accountable. The Government have rightly placed public service reform at the heart of the modernisation programme, but the ONS has clearly missed out in that process. The Government must urgently address that point.

We also want to urge my hon. Friend the Financial Secretary to examine the evidence that we are placing before her in this debate and to arrange an urgent meeting with Manchester Members of Parliament, the leader of the city council and other key players as soon as possible. As I have already made clear, we would lose £18 million under this year's rate support grant settlement and, under next year's settlement, we are again likely to lose millions of pounds. The settlement and the way in which it is now calculated rely heavily on the population of the city, so unless we get the figure for the population right, we will lose millions of pounds of grant. The city can ill afford that.

Unless we make an urgent reappraisal of the population of Manchester so that we can determine the true figure of grant that should go to the city from 1 April 2003, we shall perpetuate the problem in years to come. Unless we get the baseline right now, the loss of money will be rolled forward in whatever adjustments are used in mid-year estimates or future surveys. There is no sign that the Government will compensate us even if they find at a later stage that the figures are inaccurate in the way that I have suggested.

15 Jan 2003 : Column 254WH

Immense difficulties have been created for the budgeting process in Manchester. Determining what can be made available under a grant based on erroneous population figures means that services may have to be reduced to meet the reduced grant. That is totally unacceptable to a city such as Manchester that relies heavily on crucial education, social and other services. Any unjustified reduction for the city will be unacceptable to its Members of Parliament and population.

I therefore urge my hon. Friend the Financial Secretary to consider carefully the alternative data sources and to examine the validity of the arguments that my colleagues and I will make in this debate. I urge her to take quickly the view that the census figures are flawed and that they underestimate the population of Manchester. She must be certain that the rate support grant settlement from 1 April reflects the city's true population and the true needs of its residents.

Several hon. Members rose—

Mr. Frank Cook (In the Chair) : Order. Some Members may not be familiar with the conventions of this Chamber, so I remind them that we begin the first of the three wind-up speeches 30 minutes before the end of the debate. We therefore have until 10.30 am for the general debate. I ask Members to bear that point in mind and to make their remarks pertinent. Will they bear it in mind particularly when they take interventions? Lengthy interventions will not be favoured.

9.54 am

Mr. Tony Lloyd (Manchester, Central): I congratulate my right hon. Friend the Member for Manchester, Withington (Mr. Bradley) on his remarks. He has laid out in a detailed manner the grievances of the people of Manchester. I am sorry that my hon. Friend the Financial Secretary and not another Minister is present for this debate; she is probably sorry to be here too. One thing that the ONS could do is calculate the distances from a Minister of the Members complaining. That would provide a serious measure of the ice or warmth in any relationship.

We come to this debate with a strong sense of grievance. I am aware of many occasions when Members have been responsible for special pleading, but I do not think that this is a case of special pleading. There is a genuine sense that the ONS has made a complete mess of the census results at least in Manchester and—quite possibly, given the anecdotal evidence—elsewhere.

On my point that it might have been better had a different Minister attended the debate, our essential grievance is that an erroneous census matters because the Government will deprive Manchester of many resources as a result. The people of the city—it is still one of our poorer communities, despite the enormous progress that has been made in recent years—simply cannot afford the Government to make decisions about resourcing on the basis of an erroneous census.

My right hon. Friend the Member for Manchester, Withington has given a detailed description of the doubt and uncertainty. In fact, the problem is more than one

15 Jan 2003 : Column 255WH

of doubt and certainty; there are clear signs of a lack of credibility in the census process. I shall illustrate that point by making a few anecdotal observations. A friend of mine was a manager in the census process and he told me of the way in which it was mishandled. For example, the people in one inner-city area of Manchester are not particularly used, for all manner of reasons, to regular form filling. My right hon. and hon. Friends from Manchester know that the ethnic mix in that area is enormous and that a great many languages are regularly spoken. The consequence is that English is not always the language that is first used.

My friend told me that he had encountered about a 100 nationalities in his work. Somalians are now a significant group in my constituency and in those of some of my hon. Friends. However, because of the employment rules, my friend was not able to employ anyone from a Somalian background. That restricted the number of census enumerators able to work in the Somalian language and to determine whether the form returned from a particular household was accurate to any degree. Such details might seem trivial were it not for the fact that they are part of a systematic pattern of doubtful competence.

My friend made a further point that is relevant to the position of students. He told me that people employed by the census physically delivered the forms, but that they were returned by the Post Office. He made the obvious point that the system would have been far more credible if the Post Office—which generally tends to know where people are because it regularly delivers the post to them—had delivered the census forms and the enumerators had collected them. Although it was not the Post Office's fault, it took about two weeks for the forms to begin to be returned and it was only at that point that the enumerators could get out on to the streets to deal with the problems caused by missing forms.

My friend told me that he knew of at least one street in which not a single door was opened when the census enumerators returned. He did not know whether anyone lived in those houses. We have to face up to the fact that it is possible that no one lived in them and that a nil return was accurate. However, it is also possible that many of those houses were students' houses, because the process took place when the students were already returning to their parental homes or were involved in examinations. They may have been reluctant to open the doors. Again, that might seem a trivial point, but the number of students in Manchester is enough to make a whole constituency. An enormous number of them live in the city; we have the biggest concentration of students in western Europe in a small area. Therefore, if the way in which the student population has been accounted for contains a systematic error, it will make a significant difference to the overall census.

I know that there are other problems in determining a student's real residence. However, not all students should be reallocated to the home from which they came. Lifestyles are different these days, and many students are permanently resident in the town where they study. My points may seem trivial and anecdotal, but they illustrate the sloppiness and incompetence that ran through the whole census process.

15 Jan 2003 : Column 256WH

Fairly senior people working on the census were seriously unhappy about the process, and felt that they did not receive an adequate response from line managers in the ONS. The real charge against the ONS is that it was not fit to carry out the national census. That is an extremely serious charge, but it must be made. I know that you are a traditionalist on such matters, Mr. Cook, but if the census in Bethlehem had been conducted in such an illegitimate fashion, the whole of Christendom would have been remarkably different in the past 2000 years. That illustrates the scale of the problem.

I also want to comment briefly on some of the important points that my right hon. Friend the Member for Manchester, Withington has made. If the quality control system and the existence of a number of significant variables—child benefit, pensions and the school register—suggest that there are more people in certain categories than the census allows for and if every alternative indicator shows that the census has got things materially wrong and has underestimated the figures, it is bizarre for the ONS to decide to reject the quality control process. Instead, the ONS has insisted that it got it right in the first place. Common sense says that we should be prudent; that is a good Treasury term. Although we might accept that the census has not made a complete idiot of itself, there is a strong suggestion that it may have done that. We should seriously examine the ONS data.

My right hon. Friend has pointed out that Manchester feels that it has been blocked when it has tried to carry forward the debate. The ONS has not made available the information that the city needs. I despair of a retreat into an argument that is based on data protection. I agree that it would be outrageous for the ONS to give details about individual children and pensioners—of course, that would be illegitimate—but the provision of aggregate figures falls well within Parliament's intentions on the right to public confidentiality. We are talking about the protection of the ONS's confidentiality so that it can protect itself from the charge of gross incompetence. However, we are making that charge anyway, so it might has well come clean, own up and say, "Yes, we have probably got it wrong. Let us begin to talk seriously about the figures and the database." That would enable it to convince the public and the people of Manchester that it has got things right.

I understand the position of my hon. Friend the Financial Secretary. She will have received a briefing from her civil servants, but I am sure that she has already thought about the issues. It is not her first exposure to the problems that we have outlined, and I appeal to her sense of justice. The census matters because it is a once-in-a-while account of the population, and we need to know the national trends on the structure of the population and the data on matters such as ethnicity and gender. However, it also matters enormously because of its redistributive effect. It would be a bizarre consequence if a Labour Government, of all Governments, took part in a process that allowed an incompetent census to redistribute resources from one of the poorest communities in the country to those who are generally better off. We have no business being involved that.

My hon. Friend the Financial Secretary must assure us that there is a way of resolving our doubts about the ONS's competence and that a process will examine that

15 Jan 2003 : Column 257WH

point. We must also know that if the ONS is found to have been negligent, there will be proper recompense: my right hon. Friend has already made that point. It is no good saying—as happened in the past when Manchester lost money from previous Governments—that the loss will be made up in the future. That never happens and you know that as well as anyone Mr. Cook. Money is never made up. If the census is found to be inadequate, we want a guarantee that there will be proper and adequate compensation.

Perhaps the real point is that we should wean ourselves off the big-bang view that the census will determine all local government spending. If things go wrong, the process becomes incompetent. Even if it can be demonstrated that the census is absolutely right—I do not believe that that can be done—it is impossible to say to a community such as Manchester that it can afford to lose at the levels that my right hon. Friend has described. We need assurances that our points will be examined in a way that provides proper justice to the people of Manchester.

10.6 am

Mr. Graham Stringer (Manchester, Blackley): My right hon. Friend the Member for Manchester, Withington (Mr. Bradley) and my hon. Friend the Member for Manchester, Central (Mr. Lloyd) have enabled me to make my remarks relatively brief. They have comprehensively covered the inadequacies and inaccuracies of the 2001 census.However, I make it clear to you, Mr. Cook, that when I refer to a Mr. Cook of the ONS—some of my criticisms will be sharp because I am not one of his admirers—I am not undermining your position. I do not want there to be such a misunderstanding.

Since I have become aware of the problems in the 2001 census, I have tried to find out as much as I possibly can about the operation of the ONS and the census. On 28 October 2002, I went to a meeting of the Public Accounts Committee to watch it question Mr. Cook. It was a painful and revealing experience that shed light on why the census does not contain figures on which we can rely. At first, I thought that Mr. Cook just could not count up to 60 million. It is a large figure. However, he also has problems with other figures, statistical projections, administering the organisation and morale. I shall quickly go through the points that the Public Accounts Committee brought to light.

There is a limit of £20,000 on the letting of contracts in the ONS. However, the consultant firm, Vogue, spent nearly £1 million without going through the proper procedures. I remember from the arithmetic that I studied that there is a big difference between £20,000 and £1 million. Even after Mr. Cook initially checked the figures, he failed to detect that one of his officers had let a contract of £500,000, which led to £1 million of extra expenditure, to one of that officer's mates. The official lost a bonus of 4 per cent. but then received a bonus of 8 per cent. Therefore, he did not lose out, and he left the organisation without any penalties imposed on him. That did not reassure me that things were being carried out correctly.

One would expect a national statistician to come up with a correct estimate of the number of census forms that would be returned. He estimated that 70 per cent.

15 Jan 2003 : Column 258WH

of the forms would be returned. In fact, 88 per cent. were returned. Having got that wrong, one might suspect that it would make things easier, but it did not. Instead, the people who should have been checking up on those who had not returned their forms were kept in the office to sort out the forms that had been sent back. He also estimated that there would be 600,000 phone calls. Wrong again, Mr. Len Cook; there were 2.6 million calls, more than half of which went unanswered.

In addition, 30 per cent. of the 70,000 or so staff employed were not paid for a considerable time. Chessington, which was responsible for making the payments, ended up overpaying them by £500,000. Mr. Cook could not get the money back because he could not provide the statistics and information to Chessington to allow him to do that. In the light of that background of comprehensive incompetence in the ONS, it should come as no surprise that the figures do not match our expectation and knowledge of what is happening on the ground.

Mr. Cook estimates that 1 million people did not return the forms, so he is 2 million short. How many people were prosecuted for that? A mere 39. We must consider how effective the incentive was to persuade people to return the forms, although as my hon. Friend the Member for Manchester, Central said, they were not delivered to many people because the normal postal service was not used. The 39 who were prosecuted were honest people. They did not agree with the census and told Mr. Cook that they were not filling in his form. For that, they were prosecuted. The 1 million people who did not fill them in because they wanted to dodge something in the system or were just too lazy to do so were not prosecuted. That is a strange way of ensuring an accurate census.

As I have become more and more interested in the ONS, I have started to collect paper cuttings. Almost daily I see it retract information and I see people complaining about it. The latest complaint is that Mr. Cook has double-counted pension funds. He has counted the transfer of pension funds as new contributions to pensions, which gives a double estimate of people's savings in pensions. That is not a man who is doing a job properly. Professor Coleman is professor of demography at Oxford university and a professional in the sector. He simply cannot credit that 1,000 young men a week—1 million in all—have left the country to go to Ibiza to enjoy discos or whatever.

Mr. Lloyd : All of them from Manchester.

Mr. Stringer : Quite. It just is not credible and no other source of information supports that evidence.

Right hon. and hon. Friends have asked for a correlation between other sources of information. More people are paying council tax than are accounted for in the census for Manchester. Manchester people are very honest, but they are not paying council tax on behalf of imaginary people. There are more pupils, more pensioners and more voters than Mr. Cook accounts for in the census. It is more likely that he has got it wrong because the people who went to collect the forms were frightened of going into certain parts of Manchester, probably unreasonably. They did not know the area. Students who had temporarily left the city and people who were dodging the council tax by not registering on the electoral register did not appear on the census.

15 Jan 2003 : Column 259WH

All the sources from city council underestimate the figures. It is possible that people saw the forms but did not want to fill them in. Recent asylum seekers who have the right to stay here were probably worried about them. The common factors in all areas where the population has been underestimated are large ethnic minority populations, large student populations and, in some cases, poor populations, which has an effect because the second person in the household does not register.

My hon. Friend the Financial Secretary was decent enough to meet Manchester MPs at short notice. I ask her to look thoroughly at the census. I understand that she cannot put herself in the place of the officer of national statistics. I accept that the Government must have an independent source of information with which they do not interfere, but that does not mean that they should not take seriously our concerns that Mr. Cook has got it wrong and has done an extraordinarily bad job. The Government cannot rely on the information in the census when they plan future services. Cities such as Manchester do not like some of the services that they provide to be based on population because they are not related to the number of people in the population. The census would distort that further.

Mr. Cook has managed to make more people disappear from Manchester than General Galtieri ever managed to make disappear from Argentina. Although he has not had the same tragic consequences, there will be other tragic consequences for services if the census is not reviewed. It is the worst census for 200 years.

10.17 am

Paul Goggins (Wythenshawe and Sale, East): I congratulate my right hon. Friend the Member for Manchester, Withington (Mr. Bradley) on obtaining this timely and important debate, and other colleagues on their forceful arguments. No one could argue that Manchester lacks leadership and determination. We are certainly not a city that whinges about what other people do to us, or what happened in the past. The city is increasingly successful. We continue to combine, as we have over the centuries, the quest for economic progress with a commitment to fairness.

Last Saturday, I went to a concert in the Bridgewater hall, which is a world- renowned facility. Next season, I will support my football team in the new 50,000-seater City of Manchester stadium, the centrepiece of the Commonwealth games that grabbed the nation's attention last summer. It also showed that the regions can lead successful national and international events. As my hon. Friend the Member for Stafford (Mr. Kidney) said, Manchester's universities are among the finest in Europe. The metrolink system is doing well and is about to be extended to bring world-class public transport to our constituents. While the rest of the country was consulting—or not consulting, as it turns out—on the future of aviation, Manchester airport in my constituency, with two runways and three terminals, continued to deal with its 19 million passengers a year, which will double over the next 30 years.

All those successes and many others that I do not have time to mention have been achieved by a city that punches well above its weight in terms of its population.

15 Jan 2003 : Column 260WH

I will not use the census figures because I concur with my right hon. and hon. Friends on their reliability, but in the mid-year estimates for 2000, Manchester had 440,000 people, Leeds had 726,000 and Birmingham around 1 million. So there are far fewer people in our city than there are in others. As my hon. Friend the Member for Manchester, Central (Mr. Lloyd) said, many of our citizens are relatively poor. Indeed, about half rely on some form of means-tested benefit.

My right hon. and hon. Friends have argued forcefully that the census is a travesty, with 46,000 fewer people than the 1991 census and 47,000 fewer than the 2000 mid-year estimates. As I go around the city, I see new businesses rising up and new homes being built. So common sense tells us that the census cannot be true. My constituency experience also tells me the same thing.

I hope hon. Members forgive me for taking a little time to share facts about my constituency. When the boundary changed in 1993, about 3,200 people who were previously in the city of Manchester were moved out of it. They did not move away. Someone just drew a line on a different place on the map. Since 1997, however, because of new investment in housing and the growth of owner-occupation, 3,000 people have moved into Wythenshawe. At the same time, some buildings have been demolished and probably about 1,500 people have moved out. That process over the past few years has meant two things; an extra 1,500 people are living in Wythenshawe and about 1,700 fewer people are living in Manchester. There has been a small reduction in the population, which perhaps even the city council could accept. However, the reduction is considerably less than the 11 per cent. suggested by the census.

If my hon. Friend the Financial Secretary is not persuaded by the common-sense observations or the details of my constituency, I hope that she will be swayed by the information provided by my right hon. Friend the Member for Manchester, Withington and other colleagues on the statistic provided by the city council on the missing 10,000 properties on which council tax is being paid. The ONS evaluation report observes:

Let me translate the jargon. That means that properties and people that do not exist in the virtual world of the census do exist in the real world of the city of Manchester. The city's own evidence shows that there are more pensioners, more children and 5,000 more adults are on the electoral register than the census shows. It is not good enough for Mr. Cook to say that all previous surveys have been wrong or that 1,000 people a week are leaving for sunnier climes. The evidence simply is not there.

As we heard, Manchester is not alone. According to the census, Westminster has lost a quarter of its population. Kensington and Chelsea, Cambridge, and Oxford have also lost considerable numbers. Again according to the census, there has been a fall in the population of every one of the 10 districts in Greater Manchester. A third of my constituency covers the borough of Trafford, which is supposed to have lost 10,000 people from its population; about 5 per cent. of the total. Altogether, Greater Manchester is supposed

15 Jan 2003 : Column 261WH

to have lost more than 100,000 people. Again, that cannot possibly be true according to common sense or the data.

All of that would be interesting were it simply an academic debate, but population totals mean hard cash. Manchester is set to lose millions of pounds over the next decade if we continue to rely on the figures in the census. For the current year, that would have meant a loss of more than £18 million. We are told that there is some protection for next year because of the capping that would limit the loss to £8.2 million, but that is considerable in terms of the amount of resources that would be lost to the city. Who knows what might happen? The protection could disappear and we would have to rely on the statistics for grant distribution for many years to come. It would be a travesty if that were allowed to happen.

What can be done? The Government need to take a long-term view on the census. We need a longer gap between the publication of the census results and the implementation of the statistics for the distribution of grant. That would allow disputes, such as this one, to be resolved. It would also allow other factors such as ethnicity to be included. We need greater scrutiny of the process of the census; the methodology, the data sets and the conclusions. There has to be fuller public examination.

We also need a formal disputes procedure. That would deal with all the rowing between the city and Mr. Cook about what information the city can and cannot have. We should establish an independent system of assessing alternative sets of data when they are so at variance with what the census has uncovered. In the medium term, we need greater scrutiny of the ONS. The integrity of statistics is vital in the development of public policy, whether that is on unemployment, crime, poverty, homelessness or population trends as a whole. It is essential that the statistics are reliable. Government and Parliament need to do everything that they can to scrutinise the statistics.

We need more clarity on the rules on data sharing. We heard about Manchester's efforts to obtain data from the ONS. Either the supply has been slow or the ONS has refused outright to provide information on the grounds that it would breach confidentiality. Let me give one example of that. The city asked for data on general practitioner lists, but it has not been given. I do not know whether the membership of the national health service security and confidentiality advisory group is an official secret, but its members should be told that we are not after information on individual patients. We think that we have a right to know the total figures and expect to receive them. I fully accept that the registrar general cannot disclose personal information because it is an offence to do so, but the code of practice on access to Government makes it clear than the ONS and others should "respond openly" to requests for information.

On immediate action, I suggest that my hon. Friend the Financial Secretary does three things. First, she should inform the Deputy Prime Minister of the serious concerns about the accuracy of the 2001 census. She should advise him to make interim arrangements for next year's revenue support grant settlement. It is unacceptable that those with a variation of more than

15 Jan 2003 : Column 262WH

10 per cent. in their populations should be expected to have to live with the new figures without a thorough evaluation of what has happened.

Secondly, a thorough review of the 2001 census should be undertaken and all alternative data carefully scrutinised. Thirdly, my hon. Friend should do everything she can to ensure that Manchester has access to all the information that it has requested. If she is prepared to do those things, she will begin to restore faith in the integrity of statistics and ensure that the city of Manchester gets the resources it needs to run the services required in what is still a great city.

10.28 am

Mr. Andrew Stunell (Hazel Grove): May I add my condolences to the family of Stephen Oake, the constable who died tragically in the course of duty yesterday? I am sure other hon. Members will want to associate themselves with that.

The Financial Secretary may not know that my constituency is called Hazel Grove and is part of the metropolitan borough of Stockport, which is adjacent to Manchester. I am sure Manchester Members would agree that people from Stockport do not automatically have the same view of the world as people in their city. There tends to be a rivalry between us which I hope is always friendly. On this particular occasion, however, I want to make it clear that I strongly support the points made by Manchester Members.

I am sure that the Financial Secretary will agree that the right hon. Member for Manchester, Withington (Mr. Bradley) gave a measured, thoughtful and thorough presentation of an almost incomprehensible set of figures. I am sure that the hon. Lady will respond in an equally measured and thoughtful way; she is noted for that. Furthermore, I hope that she will not be content with speaking to a bland ministerial brief but that she will come through with positive suggestions and an acknowledgement of the need for changes in the current approach.

The right hon. Member for Manchester, Withington made clear the statistical differences between previous estimates and the numbers in the census. He drew attention to one or two obvious gaps. I want to build on one point that he touched on. He noted that the initial explanation for the discrepancy was that a large number of young men had gone overseas. It is true that large numbers of young men and women go away on gap years, perhaps remaining overseas for some time. However, in Stockport, which is adjacent to Manchester, a high number of young people stay on at school and go on to university and many of them take a gap year. It would thus be common sense to expect that a higher proportion of young men and women from an area such as mine in Stockport, with its middle-class and higher education background, its mobility and resources, would go overseas and stay there than from inner-city Manchester. Given the differences between Manchester and Stockport, if it is valid that 40,000 young people from Manchester opted to go overseas, where are all the young people from Stockport who should have done so? Some of the figures that have been chucked into the debate by people who are trying to justify the census figures need serious academic criticism.

15 Jan 2003 : Column 263WH

The right hon. Gentleman said that there was a problem with the release of data. Will the Financial Secretary confirm that she will expedite the release of information about the total number of patients on GP lists? The right hon. Gentleman mentioned that postcode samples are not being carried out. The Financial Secretary now has the opportunity to tell us that she will expedite that matter.

The right hon. Gentleman said that there were 10,000 more council tax payers than the number of enumerated households. The hon. Member for Manchester, Blackley (Mr. Stringer) pointed out that although his constituents were honest, even they were hardly likely to pay their council tax twice, or as phantom residents. In fact, most people who read the Manchester Evening News might say that the problem is underpayment rather than overpayment of council tax.

When we consider issues such as data checking and quality control, we have to acknowledge that the argument is on the side of the Manchester Members who have spoken in this debate, rather than on that of the ONS. The number of child benefit cases exceeds the total shown on the census, as does the number of people receiving pensions. It might be argued that as those benefits must be claimed, there may be fictional claimants who would not show up on the census. However, the figures on council tax payers must give the knockout blow to that argument.

The number of people on the electoral roll exceeds the number in the census, yet ever since the days of the poll tax we have heard the story that, especially in inner-city areas, there has been serious under-recording on electoral rolls. For the past 10 years and more, many of us have bemoaned that fact; yet Manchester, which has many inner-city characteristics, turns out to have 4,000 to 5,000 more people on its electoral roll than were recorded in the census. That does not make sense.

There is undoubtedly an enormous student population in the Greater Manchester area, especially concentrated in the city itself. Obtaining accurate records from students is bad enough for their tutors; I do not suppose it is any better for census enumerators.

All those things have dire financial consequences. The problems are not exclusive to Manchester, or to the university and other towns that were mentioned earlier. May I remind the Financial Secretary of the purpose of the census? It is not simply a neat academic exercise that continues a series of historical studies stretching back 200 years. It was not designed to enable genealogists to work out who was descended from whom and who moved from where. It was originally designed so that Governments could measure and design public services that were appropriate for communities and arrange for their delivery. The census was not designed for some other purpose, which has now thrown up a bizarre side effect. The core point of the census is to enable the Government to allocate resources and decide on appropriate services. On this occasion, it has clearly failed to do that job.

I have some questions for the Financial Secretary. If she cannot answer them now, I hope that she will be able to deal with them later. First, is she satisfied that the process of quality control—the check-off against the

15 Jan 2003 : Column 264WH

raw census figures—is adequate? Plenty of evidence has been produced today to show that even where there is quality control, it is either ignored or fails to deliver the necessary checks and balances.

Secondly, does the Financial Secretary accept that the rush to produce results has been at the expense of accuracy and checking? Does she agree with the hon. Member for Wythenshawe and Sale, East (Paul Goggins) that it might have been better to hold back from using the data until there had been a proper opportunity to check the discrepancies? Can the Government answer for the discrepancies that were found on child benefit, pensioners and school rolls?

Incidentally, the sum of school rolls in Manchester does not include all children of school age in Manchester, as those of us in Stockport know well. Plenty of children from Manchester go across the border to Stockport or to Trafford. If the sum of the school rolls is greater than the sum of children, we can be sure that the total number of children in Manchester is even greater.

Finally, will the Financial Secretary order the release of the additional data on GP lists, commission work on postcode samples and cut through the smokescreen of data protection? Will she do all that she can to restore faith in the census as a sound basis for describing the demographics of our complex society? It is nowhere more complex than in the mosaic of different communities in Manchester and Greater Manchester.

10.39 am

Mr. Howard Flight (Arundel and South Downs): I, too, express my condolences to the family of Stephen Oake.

I congratulate the right hon. Member for Manchester, Withington (Mr. Bradley) on securing this important debate, which, as we all know, has a wider application, beyond Manchester. We are expecting the Financial Secretary to tell us how the Government propose to resolve the problems caused by self-evident errors in the 2001 census figures. The Government's current position—as I think The Times reported on 9 January—is to dismiss pleas from councils for extra cash to cover multi-million pound funding shortfalls triggered by the census figures unless the mistakes are admitted by Len Cook, the national statistician. The Government have also said that they will not override the census figures, despite all the evidence and a vigorous campaign by individuals such as the two chief executives of Manchester.

As we have heard, the population discrepancy for Manchester is more than 10 per cent.—47,000 people—which translates into what is likely to be a recurrent funding shortfall of about £18 million a year until the next census in 10 years. In six other authorities there were even greater proportional discrepancies between the census figures and the 2000 estimates. The largest discrepancy was in Westminster, at 25.9 per cent., but it was also more substantial than Manchester in Forest Heath, Kensington and Chelsea, Cambridge, Richmond upon Thames and the City of London. Unless the problem can be resolved, the annual loss for Westminster is £63 million per annum. In Birmingham, the census discrepancies were fewer, but it faces losses in Government grants of £40 million per annum and could suffer a reduction in its European Union regeneration funding.

15 Jan 2003 : Column 265WH

Prima facie, the inaccurate census results seem to have occurred where returns were poor. As has been well aired, those results matter to citizens because of their importance not only in the recurring allocation of local authority grants but in the funding of primary care trusts. That point has not been adequately explored. I do not know how much it matters, but it would be logical to assume that in Manchester 46,000 patients will not be funded for their GP or hospital needs unless a correction is made.

The whole matter is in the same territory as the Government's control mantras and their somewhat discredited targeting process. What sounds fine in principle—the requirement for evidence for central Government policy in order to direct funding—is in practice proving counter-productive for deprived inner-city areas when based on census figures that are, prima facie, incorrect. As Members have already argued, transient populations are reluctant, for a variety of reasons, to complete complex official forms. People are daunted by a census form that has grown from five questions in 1801 to 40 questions at present; not all of which are necessary, especially as regards the important funding decisions for which the census is used.

The central criticism of the ONS and of Len Cook, its head, is that they failed to adjust the figures to reach a sensible, overall estimate of population where census participation seems to have been adequate and where the census results contradict reliable and established property-based data. In Westminster, the response was only 74 per cent; in Manchester, as we have heard, there are more council tax payers, more voters, more children, more pensioners and more children at school than the census figures show. I cannot understand why the ONS has not sorted out the problems in Manchester where the prima facie evidence shows that the figures were incorrect.

Attempts have been made to restructure the ONS and to improve its independence and reliability. However, as we all know, there have been disturbing spats with the Statistics Commission, which was set up to monitor the work of the ONS. Decisions have been questioned on the classification of receipts from the third-generation mobile telephony licences, on the status of Network Rail and, especially, on data relating to pension saving.

The ONS could have increased the accuracy of the status. There should have been more door knocking and the ONS should have been willing to examine substantiated evidence of mistakes when it was presented with unexpectedly large changes in population. The ONS could have made greater use of new survey techniques and could have adopted administrative records that have proved accurate in the past, especially for pinpointing data in inner-city areas.

As we have heard, the ONS figures for Manchester do not match property-based data. One would have thought that that problem was capable of speedy resolution and reconciliation but the ONS appears to have dragged its feet. In that context, I challenge the Financial Secretary's recent response to my hon. Friend the Member for Chipping Barnet (Sir S. Chapman) that the 2001 census had undergone vigorous testing and quality checking in both fieldwork and processing. The evidence that we have heard today suggests that that was not the case. However, I am glad that the Financial Secretary has advised us that the Treasury is working

15 Jan 2003 : Column 266WH

with local authorities that experienced unexpectedly large and, prima facie, unlikely population shifts to find explanations. I look forward to hearing the results of that exercise.

After some unsatisfactory correspondence with the ONS, Westminster has started judicial review proceedings against Len Cook for alleged non-discharge of his duties under the Census Act 1920. The council is also considering an application to the Chancellor, under section 6 of the Act, to rerun the census in Westminster. If Westminster is successful but the issue remains unresolved for Manchester, I am sure that Manchester would follow suit; so rather than wasting more time and effort, the pressure is on the Government to sort out the problem.

Manchester faces a further problem because the council pledged to keep its council tax increase modest—about 2.5 per cent.—yet, on its own figures, its budget needs to increase by more than 10 per cent. to offer the same level of services as last year. It is not satisfactory or sufficient for the Government to stall by arguing that the local government grants for 2003–04 will provide a one-year floor of a 3.5 per cent. increase. The Government have acknowledged that, mostly due to their own actions, there will be significant cost increases throughout the country of about 7 per cent. a year, so the floor may not even cover those increases.

The Government should accept the limitations of the census data when allocating funds for a wide range of policies. Manchester will lose at least £18 million this year and possibly even more each year for the next decade unless the process is corrected. Furthermore, as I stressed, its primary care trusts will be substantially short-changed.

I hope that we shall now hear from the Financial Secretary how these unfortunate and unnecessary problems are to be resolved.

10.48 am

The Financial Secretary to the Treasury (Ruth Kelly) : May I begin by saying how pleased I am to be serving yet again under your chairmanship, Mr. Cook?

I, too, add my condolences to the family of Stephen Oake.

I congratulate my right hon. Friend the Member for Manchester, Withington (Mr. Bradley) on securing this important debate. I am, of course, aware of the economic significance of Manchester; indeed, my constituency, Bolton, West, is within the boundaries of Greater Manchester. I also understand the significance of the census results for the allocation of resources, although the Treasury is not directly responsible for that allocation and I would expect hon. Members not to hold me answerable today for the way in which the census results are used by the Office of the Deputy Prime Minister.

I am also aware of the interest that hon. Members have taken in these issues. I met a delegation of Manchester MPs, including those present in this Chamber, and MPs from other local authorities that have been affected.

Before turning to the detail of the Manchester census, I affirm that we are confident that the census figures for England and Wales, released on 30 September last year,

15 Jan 2003 : Column 267WH

provide the best ever estimate of the population nationally and in each local authority. I wholly reject the charge of gross incompetence that has been made against the Office for National Statistics.

Of course, serious concerns have been expressed by local authorities whose census results differ in particular from the previous year's mid-year estimates. The vast majority of local authorities, however, are satisfied with the census results. Any Government must have authoritative figures that establish the size, make-up and geographic distribution of their people to provide a benchmark population derived on a coherent basis across the country. That benchmark has been established according to agreed statistical methodologies and has been subject to rigorous scrutiny, not just by the Government but by outside professionals and academics. In the United Kingdom, the accepted definitive benchmark for the population is the 10-yearly census.

Carrying out a census of the population is a significant task—it is the largest civilian exercise undertaken by any Government and involves a huge amount of planning, not simply in data collection but in the statistical methodologies used to determine outputs. The difficulty facing statisticians is that no census in the world ever achieves a 100 per cent. response rate. If such a result could be achieved, conducting a census would be relatively easy, but the world does not work like that. The 2001 census was designed to achieve the most accurate result using the best methods available. It was designed from the outset to take full account of what users of statistics really want—figures representing everyone. As a result, the 2001 census is the first in this country to represent the entire population.

That representation was achieved through a new and rigorous statistical strategy known as the one-number census. Census response in England and Wales—the proportion of people returning a valid form—was 94 per cent. The 6 per cent. of people estimated to have been missed by the census have been added to the counts using that statistical process which, prior to its adoption, had been the subject of extensive consultation.

One of the key elements of the one-number census is an independent follow-up survey. The census coverage survey involved an intensive face-to-face survey of a sample population of 300,000—by far the largest independent check on a census ever carried out. By combining the results of the census and the census coverage survey it was possible in 2001 to estimate the total resident population to a high level of precision, plus or minus 0.2 per cent.

Mr. Stringer : Is my hon. Friend prepared to provide the information on which those assertions are based? My understanding is that in Manchester the checks were done in areas such as the constituency of my right hon. Friend the Member for Manchester, Withington where there were very few problems, but not in areas where there were difficulties, such as my constituency or that of my hon. Friend the Member for Manchester, Central. Will the Minister check that?

Ruth Kelly : Of course I will. That is one of the issues that has come up in the attempt to reconcile

15 Jan 2003 : Column 268WH

administrative data with the census figures while respecting confidentiality—an issue that I shall return to if I have time.

I shall not dwell on the survey's statistical basis—I know that hon. Members want to hear more about the way in which the census was conducted in Manchester—save to say that in England and Wales there was an extensive quality assurance process for comparative data sets. I shall address some of the points made by my right hon. Friend the Member for Manchester, Withington, such as the preparedness and speed with which the ONS responded to requests for data.

The first letter from Manchester city council was received by Len Cook, the head of the ONS, on 7 October. He responded on 11 October, and meetings between him and council representatives took place on 4 November and 6 December, and there were further working meetings and correspondence. Available information was supplied at each stage of the process. The only outstanding information that has not yet been supplied, but which has been requested, is on NHS patient data, as genuine issues of confidentiality are involved. My office has tried to discover the position on that data set, and has been assured that the data will be released soon. I am hopeful that they will be released today, but I assure hon. Members that there have been serious confidentiality considerations that could not be treated lightly and had to be taken into account before the data were released. However, the ONS has assured me that it will be released as soon as it has received it.

Turning to the administrative data, it appears odd that many of the figures used by Manchester city council for population estimates do not at first sight tally with the census results, but there are serious problems with much of the administrative data. Many of the data sets are prone to data inflation—when people move house, for instance, they do not automatically remove their names from a register. That is particularly true of GP registration lists, the numbers on which are commonly much higher than the underlying population estimate in many local authorities.

Electoral registers were not used by the ONS as comparative data sets in the one-number census because of well-documented data deficiencies. For example, students may register both at their term-time and vacation addresses, but the published census figures count them only at their term-time address. People with two homes may register at both addresses—the census only counts people at the address where they spend the majority of their time. There are problems then, and Manchester is certainly not the only local authority whose electoral register exceeds the estimate of the underlying population base.

Manchester city council and hon. Members have raised with me and the ONS the issue of whether the property data sets held by the city council and the ONS could be reconciled. The ONS is considering that, and is prepared to undertake such an exercise. However, that would not be simple. A significant proportion of the difference between the two figures can be accounted for. For example, much of the discrepancy between the 177,000 properties on the Manchester list and the 168,000 on the census list can be explained by communal residences being counted as individual households by Manchester but as communal residences by the census.

15 Jan 2003 : Column 269WH

I am advised that if the communal residences that Manchester, according to ONS, may have counted as individual households housed an average of 38 people, that would account for far more than the 6,000 difference between the two sets of figures. Reconciling the lists is not straightforward and would involve an extensive process of checks taking several months that would have to be agreed by both parties. The ONS, however, has agreed to co-operate as much as possible in that process.

Another issue that has been raised is the release of postcode data. Again, the ONS is prepared to look at that, but I am advised that some postcodes include only one address, so the issue of data confidentiality arises. In such cases, postcode data could not be released on a straightforward basis.

Turning to the follow-up to the process, it is clear that the ONS mid-year population estimates have in many areas overestimated both the size and growth of the population for the past few areas. The national statistician considers that the reason for that is not so much the addition and subtraction of figures for births and deaths but migration—both external migration between countries and internal migration between local authority areas. He has already set up a fundamental national statistics quality review into the international migration figures, which is expected to report in the second quarter of this year. In addition, we need a more far-reaching review of cross-Government requirements of population figures to support Government objectives. The national statistician has discussed with the permanent secretary and his colleagues the need for strong action to find solutions.

11.00 am

Sitting suspended till Two o'clock.

Next Section

IndexHome Page