§ Mr. Timms
[holding answer 4 May 1999]: A considerable amount of work has been done to estimate the number of duplicate and fraudulent National Insurance Numbers in the Departmental Central Index.
Since 1995, the Departmental Central Index Data Cleaning Project has checked the database by selecting cases which fit certain risk criteria. The project has so far resolved over 750,000 duplicate records held in the database which were mainly created when details of customers were accidentally recorded incorrectly. Since the project started, it has detected only about 10 cases of the fraudulent use of National Insurance Numbers.
National Insurance Contributions Office has worked with employers to check their records and looked at over half a million cases: less than 0.1 per cent. of the items were referred for investigation of potential National Insurance Number irregularities.
The best available measure of the scale of fraud is provided by DSS benefit reviews. Since 1995, two Housing Benefit reviews have been carried out and in no instance was a case of identity fraud confirmed. Two Income Support benefit reviews have been carried out since 1995. The second review reported in 1997, and in 99.1 per cent. of cases randomly selected, identity was established to our satisfaction. The remainder included cases in which the customer could not be found.
The studies carried out indicate strongly that identity fraud is one of the least common types of fraud, and the fraudulent use of National Insurance Numbers will form only a proportion of identity fraud. Nevertheless, we are determined that the Welfare System should be secure against fraud of this kind. Our best defence against such fraud is through our current policy of rigorously checking all applications for new National Insurance Numbers, and continuing our programme of data-cleaning departmental databases, and ensuring records are reliable.