My Lords, Amendment 10 is in my name and that of my noble friend Lord Tyler. The Government have been trailing data sharing with the DWP database since orders were passed through this House last year. We very much welcome their aspiration to data match some two-thirds of eligible voters from the old household register on to the new individual register using this process. However, we worry that this process will not prove to be as robust or successful as everyone hopes it will be. Other databases are, in our view, needed to make a success of this project. We have talked many times at the various stages of this Bill about the need for the electoral register to be complete. I believe this amendment about the use of other databases will show whether it is really the intention of the Government to walk the walk on this issue, as opposed just to talk the talk on it.
It will be a matter of judgment as to which databases may be appropriate for automatic registration, as the DWP’s will be, and which should only provoke invitations to register from electoral registration officers. What is clear is that to restrict ourselves to the DWP’s database, in either endeavour, is missing a real opportunity to improve the completeness of the registers, even from their present positions. For all the talk there will be about the dangers of the new system, we have to recognise that the old system has proved quite unsatisfactory. We now know that the electoral register is complete up to only 82% of eligible voters, as opposed to the 92% quoted by Ministers very frequently a year ago. Whether we have the old or the new system, we need better and more comprehensive data matching and data mining in order to help overcome the difficulties of registering voters.
We believe in particular that the information held by the DVLA—a comprehensive database of drivers—could provide a rich source of information better and more diverse than that of the DWP. Its database of national insurance numbers is of course notoriously unreliable: there are 80 million national insurance numbers in a population of only 51 million. We know there are many people on the DWP database who will have real trouble voting, since they died a long time ago. It would be particularly worrying if we restricted data matching to the DWP database only, as the Government could give the impression that they were keen only to see one demographic group of voters registered and not so keen on seeing other demographic groups registered.
Pensioners are not generally underrepresented on the voting registers or in the votes on election day. It is other groups where there is a more significant problem. There is a danger of unintended consequences in proceeding only with the DWP records, because they deal disproportionately of course with retired people. It is known that they vote disproportionately, although not exclusively, more in favour of the Conservative Party than perhaps other social groups. I know that our coalition partners would not want to give the impression that they are particularly keen on assisting
with the registration of voters that may aid their cause and not with the registration of voters in general, in accordance with healthy democratic principles.
It therefore seems very important that the Department for Transport allows use of the DVLA’s database in the same way and with all the appropriate safeguards about personal data that the DWP applies. We are told by the Electoral Commission that the Department for Transport does not wish the DVLA database to be used in this way. However, the DWP has given permission for its database to be used in this way. My proposition is simple: that there should be consistency across government databases, using all of them to maximum effect, with the proper safeguards about personal data, in order to ensure that as many people as possible are registered.