The Home Office is refusing to publish a report it commissioned on the retention policy of the DNA database.

The report analyses the justification for retaining DNA profiles, for various periods of time, from people arrested but not subsequently convicted of any offence. There are more than a million such profiles on the database and under a ruling from the European Court of Human Rights they cannot be retained indefinitely.

Remarkably, even the Ethics Committee responsible for monitoring the conduct of the database has been denied access to the report, forcing its chairman, Chris Hughes, to submit a Freedom of Information request in order to see it. See the committee minutes here.

The request was submitted at the end of March, and was refused. During the formulation of government policy information may be exempt under clause S35 (2) of the Freedom of Information Act, but once a decision on government policy has been taken, statistical information used to provide a background to the decision is no longer exempt. So Mr Hughes believes this decision is wrong in law.

It is known that such a report was prepared, because in a Parliamentary answer on behalf of the Home Office by Damian Green MP, its existence was disclosed. He told David Ruffley, Conservative MP for Bury St Edmunds that so far in 2011 just one report written by the Home Office’s Research, Development and Statistics Directorate remained unpublished. It is entitled: “DNA retention policy: results of analysis relating to the protections of the ‘Scottish model’”.

While the report has languished, Parliament has been considering the very issues it was commissioned to address. The Protection of Freedoms Bill, which has now passed its Second Reading and Committee Stage in the House of Commons, provides for a more liberal regime for the retention of “innocent”  profiles than the previous government proposed.

For those arrested but not convicted, samples may be compared to existing crime-scene profiles to see if there is a match, but not retained. For those arrested for a serious crime but not convicted, the profiles may be retained for three years, with a possible extension to five years with the approval of a court. The same applies to those arrested for a terrorist offence, but there the two-year extension is made by the Commissioner for the Retention and Use of Biometric Material (a new office created by the bill).

The key issue over the retention of “innocent” DNA profiles, and the period for which they are retained, is how many crimes they make it possible for the police to detect. Many sweeping claims have been made, most recently by Yvette Cooper MP, that the changes proposed by the Government in the bill will allow “numerous” rapists to escape detection. But the last Government failed to provide persuasive evidence that this was the case, as Straight Statistics has reported on several occasions.

Previous statistical analysis undertaken under the previous Labour administration was criticised because it considered only the risk that an individual who had previously been arrested but not found guilty of an offence might subsequently be arrested for a different crime. This was said to be because the follow-up period for the arrest dataset used was considered not long enough to permit the outcomes of those subsequent arrests to be resolved. Unfortunately, this meant that the report only considered individuals who were only ever defined in law as innocent.

Presumably, therefore, this new report analyses data relating to the risk of subsequent conviction of those previously arrested but not found guilty, rather than just their risk of subsequent arrest. This would make sense given the additional amount of time that has passed since the initial analysis, during which the outcomes of those subsequent arrests might be resolved. Theresa May, the Home Secretary, did indicate she had requested an assessment of the feasibility of undertaking further analysis, in a response to a parliamentary question tabled by Alan Johnson in June last year.  And James Brokenshire said he continued to review the Home Office hazard-rate analysis, during a debate on the Protection of Freedoms Bill in March this year.

So if this new analysis addresses the criticisms previously raised, and considers directly the coalition’s proposals to adopt ‘the protections of the Scottish model of DNA retention, what reason could the government have for continuing to withhold the report, and risk breaking the Freedom of Information laws in the process? One can only presume that it is because the analysis came up with results which do not support the government’s proposals.

What might be the source of the government’s difficulties? The set of measures relating to DNA retention in the Freedoms Bill could aptly be termed a "raft". Aside from the headline policy relating to the retention of  ”innocent” DNA profiles, there are differential treatments relating to whether retention is on charge (as under the Scottish model) or arrest (as was proposed by Labour); whether retention should be for all offence types or for a subset of serious crimes; and what should happen to those individuals who have previously received fixed penalty notices, or those who are juveniles and so on.

So there is significant potential for the government’s proposals to be confounded by “unhelpful” evidence. In some respects, their hands are tied – the European Court of Human Rights judgement (S and Marper) effectively requires more lenient treatment of juveniles, for instance, even though most criminological evidence indicates that the younger you get involved with the police, the more likely you are to become a persistent offender in later life.

But there are other potential areas of disagreement. For instance, is it justified to retain DNA for those charged but not found guilty of an offence, but not those who are arrested but released without charge – both groups are innocent in law, and where is the evidence that the former group is higher risk than the latter (and that the latter group poses no additional risk)? Is it justified to retain DNA of those charged with a serious offence, but not those charged with other offences – the range of severities within a given offence classification is significant, and previous criminological evidence indicates persistent offenders tend to be “generalists”, not “specialists” (a point made in the original Home Office analysis and reiterated by Labour MPs in the Freedoms Bill debates)? The Scottish Model retains DNA for those charged with common assault, even though this is a summary offence.

Finally, there is the question of whether indefinite retention is in fact justified even for those who have been convicted of an offence. This is the question which every commentator on the debate has apparently seen fit to ignore. A policy of indefinite retention is tantamount to saying: “Once a criminal, always a criminal.” This might well be a fair summary, but there has been no presentation of any evidence supporting it. Who is to say the suppressed Home Office report did not have something to say on the matter?

Whatever its conclusions, the report would have been valuable to MPs considering the bill. As Mr Hughes put it in his letter requesting sight of the report: “The publication of the Protection of Freedoms Bill would therefore seem to be an appropriate moment (if indeed it had not been appropriate before) for the statistical information in this area to be made available”.

Now, he says, the refusal of his FoI request is in the process of internal review. It rather looks as if the Home Office hopes to keep the report under wraps until the Bill has completed its Parliamentary journey.