Dear Ed,

Thank you for your letter of 21 November.

I welcome this engagement and discussion of the proper use of statistics. In a large operational department like our own there is a great deal of information which is essential to the running of our business, and it is right and proper that we pause and summarise the key areas of the operation of the immigration system, as we do each quarter in our departmental statistics (the most recent, published on the 24th November). I believe that the UK already publishes more data on the operation of the immigration system than any other country in the world (although I would be happy to be contradicted).

I also strongly agree that our data should be accessible, with the context and sources clearly explained, as is the case with the excellent statistical products produced by my statisticians. There is an important onus on us to ensure that the content of those statistical publications remain relevant and address key areas of public concern, and in general I believe that they do. If that is the case, then our strong preference, like yours, is that public comments by Ministers or senior officials be based on those statistics already in the public domain, so that they can be properly verified.

However, given the importance and immediacy of some of the issues this department is dealing with there will be times when in response to questions on pressing matters, information is provided which is drawn from other information shared within the department. When that is necessary, then we can consider whether routine publication of this data is useful and add it to our regular statistics; or on occasion publish a one-off or ad hoc release of that data, as we did for Albanians in small boats on 2nd November. Ad hoc releases are however an additional cost in terms of our limited analytical resources, so my preference would be to regularise this as far as possible. An alternative might be to release the
internal data on its own, but doing that will often lead to a compromise in our aims of providing the public with clear statistics placed in their proper context. Nonetheless, we can continue to consider both of these options.

In the three examples you give, I will ask my statisticians to look at whether we might incorporate this data more routinely into our statistical releases and once they have done so I will write to you again with their conclusions.

Given the large amount of data flowing in my department, some of the briefing for Ministers often comes direct from operational managers who are perhaps less expert in statistics and their uses. Hence, I also propose to invite my chief statisticians to set out again some of the principles of intelligent transparency for the department so we can ensure that we avoid any uses of the data which risk undermining public confidence in our statistics.

I would of course be very happy to discuss any of these issues further with you.

I am copying this response to Simon Palmer, Jon Simmons and Amy Baxter, as was your original letter.

Matthew Rycroft CBE
Permanent Secretary, Home Office