1.8 million Britons 'disappear'

Almost 1.8 million people have disappeared from official records in England and Wales, many of them to avoid paying poll tax, according to the preliminary findings of last year's census.

Efforts have been made by the Office of Population Censuses and Surveys (OPCS) to account for the missing people - 3.5 per cent of the population - but a very large number are certain to remain in official limbo.

The implications are serious. These 'lost' people, many of them thought to be needy, are unable to claim benefits. Local authority spending levels, which are based partly on population figures, are likely to be plunged into deeper confusion. And most controversially, the argument is strengthened that the poll tax drove poorer voters off the register in marginal constituencies and helped the Conservative Party to win the last election.

The figure of 1.8 million is the difference between the raw returns from the national headcount carried out on 21 April last year and the best previous official estimate of the total population of England and Wales. The estimate, dated mid-1990, was based on routine updates of the last census, in 1981.

Local authorities are certain that many of the missing people deliberately avoided filling in the 1991 census form because they did not want to be located and obliged to pay the poll tax. They are thought to include a high proportion of poor young people, single mothers and immigrants. Many are likely to be Labour voters.

Before last April's general election David Blunkett, then Labour's local government spokesman, claimed that, as a result of the poll tax, a million people had 'disappeared' from electoral rolls. Mr Blunkett, now the party's health spokesman, said last night that the new figures cast a shadow over the election result.

'When we were originally talking about a million we were concerned that a very substantial proportion of those would have voted and were therefore ineligible,' he said. 'The figure of 1.8m makes the case much worse and makes the result of the general election, particularly in very marginal seats, less legitimate than it otherwise would have been.

'In 17 seats where less than 1,000 votes determined the outcome, the Government's overall majority was at stake and you only have to predicate 1.8m to the 636 mainland constituencies to see that could well have had a substantial effect on the outcome of the general election.'

The population shortfall was identified by the Conservative- controlled London Boroughs Association, which will publish an analysis of the enormous problems involved later this month. These arise because the Government uses census data to help estimate what councils need to spend - the Standard Spending Assessment - and thus the level of Government grant. Mike Dubock, the London Boroughs Association finance officer, estimates the population shortfall will cost London alone pounds 135m and warned it would be a 'minefield'.

By region, the recorded 'falls' in population were: 744,000 in the South-east (415,000 in London); 265,000 in the North-west; 174,000 in Yorkshire and Humberside; 152,000 in the West Midlands; 120,000 in the East Midlands; 92,000 in the South- west; 70,000 in the North; 52,000 in East Anglia and 130,000 in Wales.

There is concern now about the manner in which the anomaly is being corrected. This is being done by the OPCS on the basis of a follow-up survey to the census, kown as the Census Validation Survey, which was conducted in July last year. The Labour-controlled Association of Metropolitan Authorities has suggested that the figures should also take into account statistics such as information from GPs' records and school rolls. It has also asked the Department of the Environment to change the way it calculates the spending assessments and to delay fixing them until more accurate population estimates can be made.

In a confidential report to the DoE the AMA warns: 'Unless these adjustments can be made using all the available evidence, the quality of the resident population data will be lower than we, or OPCS, would like - with the consequent catastrophic effects on the finances of local authorities who will be called upon to provide services for people whom the census does not believe exist.'

The department replied that it would rely on the OPCS's 'rebased' estimates and had no plans to cushion authorities from any sustained drop in population since 1981.

Council tax uproar, page 8