Skip to main content

Census (Local Government Funding)

Volume 549: debated on Tuesday 4 September 2012

I assumed there would be a large crowd interested in every word on this subject, but, unfortunately, I often have this effect when speaking in the House: the Chamber empties.

The census particularly affects one of the two cities I represent: the city of Westminster. Unfortunately, the hon. Member for Westminster North (Ms Buck) cannot be here today, but she associates herself with most of my speech. We have worked together throughout the 11 years that I have been in Parliament to try to ensure that the problems of previous censuses are ironed out. We have fears for the future, given the funding arrangements that have been put in place.

The city of Westminster is one of the most complex and diverse areas of the United Kingdom. As the cultural and political hub of our capital city, Westminster attracts a vast number of people each and every day, on top of the residential population, who come either to work or to visit Westminster’s wealth of attractions. Unsurprisingly, Westminster is a destination of choice for people arriving in the United Kingdom for the first time. Many plan to work or study for a short period before returning home; others hope to make a permanent new life here. The true extent of that population is unknown. Also hidden from official statistics are asylum seekers awaiting a decision from the Home Office, countless migrant workers from the European Union who are often willing to sleep in crowded rooms, illegal immigrants working in the black economy and those whose application for leave to remain has been rejected but who are yet to be removed. That huge tide of humanity must be catered for, no matter the unique difficulty of measuring its extent. For Westminster city council, that means funding services for a population well beyond that catered for by central Government money.

I made that point in Westminster Hall some four years ago. In 2008, Westminster city council, one of the two local authorities in my constituency, spent an estimated £6 million looking after that unaccounted-for population. The council had repeatedly warned the previous Government that methods of counting migration numbers were not keeping pace with modern patterns of population movement. That became especially problematic for the council after the 2004 accession of 10 new member states to the European Union, bringing a wave of immigrants to the capital that included Poles, Latvians and Estonians. Lessons were not learned and services came under renewed pressure when, only a few years later, a fresh influx of people came to London when Bulgaria and Romania were brought into the European Union fold.

There was enormous optimism that the 2011 census would at last provide an accurate indication of the numbers living in central London and ensure that the council’s funding settlement from central Government finally provided sufficient moneys to cover the cost of services for all those using them. Undoubtedly, the 2011 census was handled differently from the 2001 census. The Office for National Statistics made significant changes to address some of the obvious shortcomings of the previous population count, to which the hon. Member for Westminster North and I referred in the debate four years ago. We were pleased that Westminster was designated in the hardest-to-count category across all of its wards and received considerable resources to conduct the census.

Nevertheless, although the 2011 census estimate of 219,400 represents an increase in population since 2001, when the figure stood at 181,300, it comprises a large reduction of 21,800, or 9%, from the previous revised 2010 mid-year estimate of 241,100. The situation looks worse if one considers the figures used in the last local government finance settlement, which was based on 2008 mid-year estimates projected forward. Against those numbers, the 2011 estimate represents a drop of 43,500. Despite the resource given to counting Westminster’s population, the council understandably believes that the ONS remains wedded to a one-size-fits-all methodology that does not properly recognise the specific problems of areas such as central London.

Westminster city council believes that the 2001 experience set a precedent that should not be ignored by the Department for Communities and Local Government when determining which data to use when allocating local government resources. Westminster, with its thriving economy and world-class, highly regarded universities—Imperial college London, King’s college London and the London School of Economics, to name but three—is an especially attractive destination to live and work in. Westminster draws people from across London, the UK and the world and consequently provides services to the largest volume of non-residents in the country.

Historically, through a commitment to efficiency and innovation, Westminster has managed to soak up many of the cost pressures that go hand in hand with providing a high-quality cityscape. With the tightening of public finances, however, the council can no longer meet those costs. Westminster city council is underfunded by the revenue support grant and is unable to increase council taxes owing to the Government’s commitment to freezing the levy. I do not disagree with that commitment, but local authorities find themselves wearing that straitjacket.

The subsequent impact on services for people living in central London has been significant. The costs, however, are much wider and include deteriorating environmental services that affect almost 50,000 businesses, and the risk that the 22 million foreigners who visit the borough each year will form less favourable impressions not only of London but of the UK as a whole. That may seem a relatively trivial risk, but think of the impact of the magnificent showcasing of the capital and the UK during the Olympics that will fashion the world’s view of Britain for years to come.

I accept that there will always be a multitude of difficulties in collecting accurate population data in places such as Westminster compared with local authorities such as your local authority in Kettering, Northamptonshire, Mr Hollobone, or the relatively leafy suburb that the Minister represents in the south-east London borough of Bromley. Those problems include getting people to open their door—an estimated 89% of properties in Westminster operate multiple door entry systems. Westminster had the second-lowest response rate to the national place survey, and the profile of those who responded was overwhelmingly white, with almost every other ethnic group and those in the 20 to 34 age bracket under-represented. Similarly, in 2010, Westminster city council conducted a mini census coverage survey in four key areas. Some 54% of Edgware road respondents were found to be white, with only 35% identifying themselves as Asian despite the surveyed area being in the heart of London’s Arab community. In Soho, where Chinatown is based, one enumerator reported that she encountered many doorbells with Chinese names but few people answered the door. Yet, according to the 2009 pupil level annual school census, English was found to be the main home language for only three in 10 children attending Westminster’s schools. One consequence of an over-representation of white respondents is likely to be a reduction in average household size because migrant groups tend to live in larger households.

Although the estimated response to the 2011 census was better, at some 85%, than a decade earlier, when the response was only 74%, response rates still look very low for some demographic profiles, with a particular under-representation of young males. That is especially critical in Westminster, where those with the lowest response rates—males aged 25 to 44—are the most prevalent in the population and the most unlikely to register with comparator data sets such as GP lists.

Although I am sure that the Minister will update us, I understand that the ONS has not yet published the detail necessary for Westminster or any other local authority to evaluate and initiate the adjustment process, which strives to deal with the difficulties to which I have referred. However, Westminster city council is concerned about the methodology that is used. For example, part of the bias adjustment is the within-household bias, which refers to census returns that report a lower number of residents than are present, resulting in an under-count. To correct that particular bias, the ONS matches social survey data to the census data and makes an imputation based on the characteristics of respondents. However, it is not clear whether the data will sufficiently correct the bias for an area such as Westminster, where the number of visitors, migrants and large households is unprecedentedly high and unlikely to be accurately represented in voluntary social surveys.

The council also has concerns about administrative data. Although comparisons of administrative data to census outputs should be treated with a certain amount of caution until we are able fully to understand the ONS’s assumption, there are some anomalies that Westminster believes require further investigation. For example, 2011 census data claim that there are 4,800 fewer occupied households in Westminster than were identified on Department for Communities and Local Government council tax lists, meaning that almost 5,000 fewer households completed census forms than pay council tax. Why is the shortfall so large even after vacant and second homes have been formally accounted for? It is plausible that those properties are only partly occupied. If that is the case, it takes us back to concerns that the city is providing services for part-time people who are never captured in population counts and therefore never properly funded.

Other comparative data require close examination. The register of patients for Westminster GPs, for instance, listed 19,400 more people than the census outputs. GP lists in some parts of the country are deemed to be inflated, as people fail to deregister when they move out of an area, but Westminster is not like other parts of the country. It is likely instead that a substantial part of the population never registers with a GP due to population churn, migration and the prevalence of walk-in and accident and emergency services.

Department for Work and Pensions data on the number of people aged 80 to 84 in Westminster are 50% higher than the census output. The 2011 census estimates that only 3,100 over-85s live in Westminster, despite the fact that 6,800 claim a pension there. The census included questions on length of stay in the country for the first time, thereby providing some estimate of short-term migration, yet only 6,900 in that category were suggested as living in Westminster. Again, that is likely to be a massive underestimate, as it implies that a little over 1% of the number of people employed by Westminster businesses are short-term migrants. A cursory look at any restaurant or shop in the centre of London suggests that that is a underestimate.

Furthermore, the annual population survey 2008-09 estimated that 15,500 people aged 18 to 24 were full-time students in Westminster and might be here for only part of the year, adding a further burden to the issue of short-term migrants. Of the estimated 442,000 illegal migrants living in London, we believe—on a pro rata basis, although there is likely to be a bias towards central London—that 20,000 to 25,000 live in Westminster. All those uncounted and under-counted people are not represented in the funding formula.

For all those reasons, the council requested continually that the 2011 census be tested in Westminster, yet the ONS refused. If data based on the census are used in central Government’s calculation of Westminster’s funding grant, the local authority will be affected to its detriment. It is hard to grasp the exact financial implications, as the details of the revenue support grant model are not yet published. However, I believe that it is likely that Westminster’s RSG will drop substantially.

I calculate that Westminster’s new, substantially reduced census population, when compared with the population estimates used in the previous financial settlement for local authorities, could result in an estimated annual loss of £10 million to £15 million in funding, which will disproportionately affect the most vulnerable in the community. Although I am sure that the Government’s damping mechanism would mitigate some of that loss— the vagaries of the funding mechanism may mean that the estimates could change considerably—those sums are substantial, particularly in a time of tightening financial settlements, which I support, understanding the reasoning behind them.

As I see it, three options now face the Minister and DCLG. The Government can plough ahead with the 2011 census data and lock Westminster into perhaps seven years of underfunding via the business rates retention model. Alternatively, it can recognise that for a very few local authorities—a small number of special cases—the one-size-fits-all 2011 census model may not have worked well enough, and there should be an opportunity to discuss a population top-up. The third option is to continue to use the 2010 mid-year estimates for all authorities until the 2011 census can be adequately quality assured.

DCLG ought to give more recognition to the fact that modern population movement—many UK cities are hyper-mobile and hyper-diverse communities—means a constant turnover of people in inner urban authorities. Those people should be counted as part of the resident population. The Department should, after six years of ONS deliberation, include short-term migration in its funding formula. I must confess that I had hoped wistfully that the 2001 experience would have set a precedent that could not be ignored. I respectfully suggest that the coalition Government now seize the chance to rectify the situation. Otherwise, they face fundamentally undermining one of their flagship local authorities, which has for years served as a beacon of best practice in spite of its gargantuan task in managing this fine capital city of ours.

It is a pleasure, as always, to serve under your chairmanship, Mr Hollobone. I thank my hon. Friend the Member for Cities of London and Westminster (Mark Field) for raising this issue, which is important for his constituents and generally. I recognise, as do the Government, that it is important to get the most accurate census data possible, because they are a key element that feeds into the distribution of public funding. As I will explain, they are not the only element, but they are important, so I understand his concerns.

I agree with and accept his analysis that it is more difficult to conduct a census in an inner-city area with population churn such as Westminster than in some parts of the country. I hope that I can demonstrate that efforts have been made to take that on board. It is worth starting with the situation that confronted us after the 2001 census, when there were concerns and steps were taken by the Office for National Statistics, which is a body independent of Ministers, to improve the methodology and quality assurance underpinning the figures.

Since 2001, as my hon. Friend recognised, considerable effort has been made to improve census returns, as has been demonstrated. He is right that particular resource was put into Westminster to reflect the difficulties, resulting in a significant increase in the response rate. Although it is less than in many authorities, the gap has diminished considerably. The response rate in Westminster in 2011 was 85%, compared with 66% in 2001. Both those responsible for the census and those on the city council are to be congratulated on the hard work that they did collaboratively to achieve that. I hope that gives a better starting point.

Generally, the 2001 census figures have been welcomed by local authorities. There have been a small number of areas—Westminster is one of them—where issues of concern have been raised.

I appreciate that these issues are sensitive and that the Minister will therefore not necessarily want to name each and every local authority, but would it be fair to say that the local authorities that have expressed concern tend to be urban and therefore have the particular characteristics I referred to in my speech?

Some, but not exclusively—some are inner city authorities, but some are district authorities. There may be other causes of churn, which we will continue to look at. Other urban authorities—for example, Newham—have welcomed what they regard as an improvement in methodology. I recognise that particular circumstances apply to each case, but it is worth putting the Westminster situation into that broader context. Where there are issues, ONS is working collaboratively with those authorities who have sought clarification. Westminster is one of those authorities, and a meeting recently took place between the officials of Westminster city council and ONS. Westminster—I hope I can put this fairly—has said that it would like to consider its position further, in light of the discussions that have taken place. Of course, ONS stands ready to continue those discussions, as does my Department.

It has always been our intention—given the nature of a census count, there are certain caveats—to use the most up-to-date and nationally consistent data available. We are consulting on proposals for the basis of calculating the next local government finance settlement. The consultation will close on 24 September, and I know that Westminster council will respond to it in detail. Population is one figure. As you will know from your local government experience, Mr Hollobone, a number of other factors are churned into the regression analysis—the bane of all of us who deal with local government finance—and the product then emerges. As my hon. Friend says, the Government intend to continue the protections of floor damping—we have made it clear that that will be the case.

We have consulted on the proposal to use the interim 2011 census base population projections. I confirm that it is the intention for them to be released by ONS on 28 September—they will be available in good time for use in the settlement. Of course, other factors will have to be taken into consideration: assessments of relative need and resources, as you will know, Mr Hollobone; the operation of the floor damping; and, because we are setting up a baseline for the new system of retained business rates, the calculation of the tariffs and the top-ups that will apply for the reset period. It is our intention not to reset those baselines until 2020 at the earliest, subject to wholly exceptional circumstances. The new system will be an incentive for go-ahead authorities such as Westminster—I am the first to recognise that Westminster is a flagship authority in many respects—and an incentive for those local authorities to drive growth in their area over the period of the seven-year reset period. It is estimated that this new means of financing local government, with 50% of the growth in the business rate being retained, will increase gross domestic product by approximately £10 billion. There will also be a safety net, which will protect local authorities from unexpected volatility in their rate base.

On quality assurance—a particular issue that has been discussed with Westminster—we have seen considerable improvement in the census take-up. The issues raised by my hon. Friend were discussed with officials of the city council and ONS—the officials in my Department will obviously want to continue to discuss them with ONS, too—to ensure that they were taken on board. As my hon. Friend said, what we have is not in fact a decline, necessarily, in the population, but an estimate based on the census that is less than a projection anticipated it to be. Therefore, the projection must be treated with some caveats, too.

After the 2001 census, there was an independent review of the means of quality assessment used to double-check the reliability of the census figures. The independent body suggested 23 actions to improve quality assurance, all of which were taken on board by ONS. Significant steps have been made to improve the quality, and other measures can be looked at. For example, my hon. Friend referred to patient registers. He is right to say that they must be treated with caution. I agree that the level and risk of inflation may vary from place to place, which is something that we can discuss. Across London, it is thought that GP registers can be inflated by approximately 8%. I am not saying that that is necessarily the figure in Westminster, but that is a reason why one must approach them with caution. Although that is a factor, there is not one single figure that can be used as an alternative benchmark.

It is worth saying that we have done checks, with that caveat, against the council tax data. The census estimates are in line with those sources, if allowance is made for some known differences, including an allowance made for the very high proportion of short-term residents, of which, as my hon. Friend says, Westminster has particular numbers. There is also a question concerning second residences.

The Department believes that this approach represents a considerable improvement on the methodology of the previous census. In the Westminster scenario, I agree that it is always difficult to get as high a return as one would wish—I think my own local authority’s return is approximately 95%; it is approximately 85% in Westminster. None the less, in sheer numerical terms the number of questionnaires returned increased from 134,200 in 2001 to 186,800 in 2011—a 39% increase thanks to the work both of ONS and Westminster city council. I hope that that demonstrates that we are going in the right direction—it has generally been well received.

The constructive way forward is this: I will take away the specific points raised by my hon. Friend and I will liaise with him. Westminster city council is going to come to ONS directly after it has reflected on their conversation. That liaison will continue, because we want to see what can be done further to explain and clarify apparent or potential differences between the census and other data sets. ONS remains confident that improved methodology more accurately captures the figure on the ground. Of course, the census is in effect a snapshot taken on one day. I understand my hon. Friend’s concerns. As a former leader of the Conservative group on the London assembly, I raised the issue when I was wearing that hat. The issue ongoing in London.

On the liaison to which the Minister referred, will he confirm that we can expect something in writing? I accept, given the representations I have made today, that he and his officials will want to consider this. I also accept that, in allowing a particular change to be made in relation to Westminster, there is a risk of setting a precedent. Can I expect a formal letter in relation to the liaison to which he referred?

I am very happy to do that. My hon. Friend makes an important point. We must have methodology that can be applied consistently across the country. I can say both to him and to the hon. Member for Westminster North (Ms Buck) that if I remain in a position to do so, I am more than happy to continue to discuss this ongoing issue with them.