- 13 Nov 07, 06:36 PM
The home secretary's story today was that she didn't reveal the problem of illegal immigrants working in the security industry as she thought it best to act first to measure the scale of the problem and then to sort it out.
The figures she produced indicate that the figure of 5,000 illegal immigrants licensed to work in the security industry may turn out to be an underestimate. Based on the outcome of checks made so far the worst case scenario could be over 8,000. Ministers insist that it is impossible to be more precise than they have been so far since they will only have accurate figures once checks are complete in December and say that 5,000 is still the best estimate at the moment.
This is how the worst case scenario (not, I should stress either a Home Office or a BBC estimate) is calculated:
Officials in the Borders and Immigration Agency are still checking the immigration status of 40,000 people from outside the European Economic Area who have been licensed to work in the security industry.
10.5% of a sample of 6,000 cases have been shown not to have a right to work.
Checks are ongoing on 12.5% who simply do not appear on the Border and Immigration Agency records. A Home Office official has conceded that they are "almost certainly not entitled to work".
If this proves to be correct then the total figure of those not entitled to work will be 23% (10.5% + 12.5%).
If the sample of 6,000 cases is representative of the 40,000 backlog and around 23% turn out not to be entitled to work the figure would be around 8,800.
There are a couple of reasons for treating this worst case scenario with caution:
The figure is reached by extrapolating from a sample. Later checks may reveal that a lower proportion are illegals. The Home Office have indicated that the Borders and Immigration Agency started examining those cases they regarded as the highest risk - ie looking at certain nationalities and postcodes which are associated with illegal immigration.
The figure is based on an assumption that all 12.5% of "unknowns" prove not to be entitled to work. This may prove to be inaccurate.