Reliable migration statistics, relating to both international migration and internal population movements within the UK, are an essential requirement for estimation of the national and local population and their characteristics. Good quality and reliable population and migration data are fundamental to the fair and effective resourcing of public services and to many other aspects of national policy.
The currently available statistical information on migration is not satisfactory. Data on international migration is mostly drawn from the International Passenger Survey, a relatively small sample survey that cannot offer sufficiently reliable estimates. The sample for this survey is currently being expanded in size which will improve estimates at national level, but is unlikely to solve problems at local level. Data on population movements between areas is drawn from administrative records that are not currently sufficiently accurate.
The Office for National Statistics cannot resolve this on its own. It will need the active support of departments and Ministers across Whitehall and the devolved administrations. ONS recognised this in December 2006 when it published an interdepartmental report on ways to improve migration statistics using administrative records from Whitehall departments.
The Royal Statistical Society welcomes the efforts of the ONS to improve national and local migration statistics. However we are concerned about:
-
the long time scales envisaged for the development work;
-
aspects of the report that may need further development; and
-
the lack of any clear government commitment to implement the report’s key recommendations or provide funding for this.
Without a concerted effort involving several government departments, the problems with estimating migration will not be resolved. A satisfactory solution is likely to involve both better administrative records on people entering, leaving and moving between areas of the country supplemented by better targeted statistical surveys.
Adopted by the Council of the Royal Statistical Society, 31 January 2008
|
