Protection of Freedoms Bill

My Lords, the amendment addresses a range of issues on which we have had long discussions at Second Reading, in Committee and on Report, on the clauses dealing with proposed extensions of the application of the Freedom of Information Act 2000 to research data sets. I thank the Minister especially for his agreement at Report to delay implementation of this part of the Bill until the completion of the post-legislative review of the workings of the Freedom of Information Act. Delay alone is not, however, enough, but it may be fruitful if used actively to deal with issues that have been raised in the course of the passage of this Bill. Amendment 19 sets out some conditions for a process for using this delay constructively. I hope that it builds upon the Minister's indication at Report that the delay would be used. So far amendments proposed to this part of the Bill have mainly met a uniform response that the exemptions in Freedom of Information Act already cover the case. In some instances perhaps they do, but it has not been easy to see that they do. There have been very few detailed explanations of how they do so, and no arguments that they can continue to do so in the world in which we find ourselves. That is a world in which new players, often in the Far East, combine high-quality IT with ambitions to be scientific and technological innovators, yet do not respect others' IP and are protected by jurisdictions that provide no remedies for IP violations. These circumstances mean that arguing that there have been few requests for research data in the past is no adequate guide to the future. I ask the Minister for assurances that the code of practice that will be revised will address the range of problems discussed, and take account of the results of post-legislative scrutiny, the Government's response to that scrutiny and other relevant evidence. In particular, the process needs to take account of some questions that have been raised repeatedly. The first is that of timing, a point raised successively in Committee and at Report by the noble Baroness, Lady Brinton. Will the code clarify the point at which data sets that are either incomplete or have not been checked must be released? This matters for individual researchers and research teams. Secondly, on costs, I am grateful to the Minister for his clarity at Report about the costs of locating data. He said: "““The Freedom of Information Act, passed by the previous Government, provides that where it would cost more than a certain amount to locate information that has been requested, there is no obligation to provide it””.—[Official Report, 15/2/12; col. 837.]" Will the code of practice to be introduced before implementing this extension of the 2000 Act provide comparable clarity about dealing with certain other costs? How will the IT costs of rendering data reusable—for example, if non-standard software has been used or if data sets are massively large—affect obligations to provide data, and how will the redaction costs of removing personal identifiers from data sets work? In each case, costs can run to very large sums. Will researchers be able to refuse to release data if these costs exceed a set threshold? I believe the matter has remained unclear, both in Committee and at Report. Thirdly, there is the question of licences for reuse. The Minister said at Report that, "““the Bill provides that public authorities, when making their data sets available for reuse, must use one of the specified licences which will be set out in the code of practice in due course … Published licence terms will provide clarity, certainty and consistency as to how information and data can be reused””.—[Official Report, 15/2/12; cols. 835-36.]" It is hardly surprising that a requirement to use licences yet to be specified has raised fears among UK researchers and among the institutions in which they work, not to mention in those companies that might otherwise wish to fund research in UK universities. What steps will the Minister take to ensure that these licensing requirements do not damage UK research? Is the office of the Information Commissioner equipped to judge specifications in licences that are likely not to be respected in certain jurisdictions? Can the office judge the reliability with which the conditions in the licences are likely to be respected? I would not want these arguments to be misinterpreted as opposition to data sharing, but there are many models for that in addition to the freedom of information model. Some are already in regular use in a range of scientific communities. Examples include: the conditional, regulated access to data used by the genomics community; the use of so-called data safe havens, for example by the Scottish Health Informatics Programme; and the use of a trustee model for data access, as for example by the UK Biobank. Freedom of information is a model of information sharing that is at one extreme end of a spectrum, since it requires publication to the entire world. That is why it threatens the capacity of institutions to commercialise their research. In the long march from work that first assembles data to the point at which it becomes feasible to file for a patent, others will be entitled to obtain researchers' data and may file prior patents—or patents that prevent researchers from filing for those patents they would otherwise seek. If we do not see a convincing code before implementation, we shall be committed to policies that will undermine two government initiatives for which I have great respect. The first and more general of these initiatives—which was introduced by the previous Government—is that universities and other publicly funded research institutes are now required and incentivised to commercialise their research. Yet here we see provisions to regulate them in ways from which their commercial competitors will be exempt. The second is the Government's Strategy for UK Life Sciences, published on 5 December last year, which is predicated on quite a different approach to data sharing. The right honourable David Willetts, writing in the spring 2012 issue of Science in Parliament spoke of, "““launching a new secure service to link primary and secondary care data at an unidentifiable … level””—" in effect, a safe-haven approach. It may be thought that this is quite another matter because we are talking about patient data rather than scientific data, but the life sciences strategy is about using patient data for research, and biomedical research draws not only on patient data but on a range of wider socioeconomic and biological and demographic data. If we want the life sciences strategy—an admirable strategy—to succeed and to benefit the UK biotech industries, we shall need a code that incorporates robust responses to the results of post-legislative scrutiny of the Freedom of Information Act, which may require further legislation. We shall also need a code that achieves a high degree of clarity about the conditions under which research data must be disclosed to the entire world, and the conditions under which they may not be disclosed. Amendment 19 seeks to set out steps by which these objectives might be achieved. I beg to move.