House of Commons - Constitutional Affairs

Select Committee on Constitutional Affairs Written Evidence

Evidence submitted by the XIX Article 19

1. STATEMENT OF INTEREST

1. ARTICLE 19 is an international, non-governmental human rights organisation which works around the world to protect and promote the right to freedom of expression and information. We are well known for our expertise in the area of access to information legislation, and have played an important role in the adoption of a great number of domestic access laws in recent years. We are also a leading member of the Global Transparency Initiative, which has successfully pressured international financial institutions—entities such as the World Bank and regional development banks—into adopting or improving their disclosure policies.

2. ARTICLE 19 is concerned that the proposed new regulations would undo the newly-won benefits of a vibrant freedom of information regime, and on dubious grounds. In our view the overall cost of the FOI Act is modest, the consultation document overstates the savings that would result from the regulations, their enactment would have a disproportionately negative impact on public interest requests and alternative cost-saving measures exist which are less detrimental to the right to know.

2. SUMMARY OF RELEVANT FACTS

3. The Freedom of Information Act was passed in 2000 and entered fully into force on 1 January 2005. The Act was adopted, in the words of the Home Secretary, in order to "transform the culture of Government from one of secrecy to one of openness."[13] The Department of Constitutional Affairs (DCA) has praised the "constructive and positive way" the Act has been used and described its operation in its first year as a "significant success".[14]

4. Under current rules, requests for information under the Act can be refused if the cost of processing them exceeds the "appropriate limit", which has been set at £600 for central Government and Parliament, and £450 for the wider public sector. Costs are assessed at £25 an hour. In calculating the overall cost of a request, regard may only be had to time spent on determining whether the requested information is held, and then locating, retrieving and extracting it. To prevent requesters from circumventing the appropriate limit by breaking up their request, the cost of multiple requests may be aggregated. Aggregation is possible when two or more requests relate to similar information, are received within a period of 60 working days, and are made by the same person, or persons who appear to be acting in concert or in pursuance of a campaign.

5. In December 2006, the Government published a consultation paper proposing to amend the way the costs of processing requests are estimated. Broadly speaking, the existing rules will remain in force, but public authorities will be permitted to take additional factors into consideration when calculating whether a request exceeds the appropriate limit.

6. First, the time spent examining documents, consulting with others and considering whether the information is covered by an exemption will be included in the calculation of the total, up to a maximum of £400 for central Government and Parliament and £300 for the wider public sector. Reading, consulting and consideration costs will not be taken into consideration if they fall below a floor, set at £100 for central Government and Parliament and £75 for the wider public sector.

7. Second, public authorities will be permitted to aggregate unrelated requests if it is "reasonable in all the circumstances" to do so. The Draft Regulations do not elaborate on what constitutes reasonableness, although the DCA proposes a number of possible considerations.[15]

8. The proposed changes are based on recommendations from an independent economic review of the operation of the FOI Act commissioned by the DCA.[16] Salient conclusions from this review are that the overall cost of administering the FOI Act is approximately £35.5 million annually; that 5% of requests account for 45% of this amount; and that the proposed measures would lead to annual savings of £11.8 million.

9. The stated purpose of the amendments is to address the top few percent of requests which "are imposing a disproportionately large burden on public authorities."[17] Under the heading "Types of Requesters", the independent review of the FOI Act lists a number of examples of frivolous requests, such as one for the total amount spent on Ferrero Rocher chocolates in UK embassies and another for the contact details of eligible bachelors in the Hampshire Constabulary. This has led to a widespread public belief that the amendments are simply designed to put a stop to abuse of the Act. In fact, the four categories of requesters identified by the government itself as making up the lion's share of "disproportionate" requesters are journalists, MPs, campaign groups and researchers.[18] It concedes these groups would be hardest hit by the proposed changes.[19]

3. THE PROPOSAL IS BASED ON A FALLACIOUS COST-BENEFIT ANALYSIS

10. The Government has rejected the option of leaving the current cost rules intact because "[public authorities] would continue to be obliged to comply with requests that impose disproportionate burdens on them, which would in turn affect their ability to deliver other core public services effectively and efficiently."[20] This argument turns on three claims: that the overall cost of the Act is too high; that the 5% of requests which cost more than £1,000 impose a "disproportionate", in other words unjustified burden; and that complying with these requests compromises the delivery of public services.

11. As noted, the cost of implementing the FOI Act is £35.5 million annually or about 67p for every person in England and Wales. Total government expenditure stood at £555 billion in 2006,[21] of which the FOI Act accounted for just 0.0064%. This seems to us a modest amount and good overall value for a law which has exposed serious inefficiencies in the way larger budgets are used, such as the £96 billion spent on health services.

12. The economic review commissioned by the DCA proceeds from the implicit assumption that all requests are of equal value, meaning that a loss of 5% of requests leaves 95% of the Act's utility intact. But, in practice, it would appear that the most useful requests are often precisely those that are most expensive. The fact that it is mainly journalists, MPs, campaigners and researchers—the groups which hold government to account—who make up the top 5% of most costly requesters is strong evidence of this claim. While the amendments might have the incidental consequence of ruling out some mischievous and genuinely "disproportionate" requests, their principal effect would be to prevent difficult questions from being posed on behalf of the public. Furthermore, there is no evidence to suggest that disproportionate or frivolous requests are concentrated among the more expensive requests and these measures do nothing to address such requests where they fall below the cut-off level.

13. Furthermore, the estimated figure of £11.8 million saved ignores the potential of those public interest requests which will be ruled out to save costs elsewhere in the long run. A robust freedom of information regime fosters efficiency in the public sector, generally by putting civil servants on notice that waste or corruption may not go undetected and specifically by exposing and thus bringing to an end inefficient and corrupt practices.

4. ANY COST-CUTTING SHOULD BE MINIMALLY DELETERIOUS TO THE RIGHT TO KNOW

14. Even supposing that the proposed amendments were based on a sound cost-benefit analysis, their simple economical rationale fails to take due account of the importance of the right to access information, which has been recognised as a fundamental human right under international law.[22] The Inter-American Court of Human Rights, which fulfils a role similar to its European namesake, recently held that:

It may be doubted that any court applying this rule would find the nominal savings of £11.8 million (0.0021% of government expenditure) a sufficiently "compelling public interest" to justify a significant curtailment of the right to know.

15. It is a fundamental principle of international law that restrictions on freedom of expression should be carefully designed to cause the smallest possible degree of harm to the right. In the case noted above, the Court stated:

The European Court of Human Rights has frequently reiterated that restrictions must be proportionate.[25]

16. The setting of an appropriate limit above which requests may be refused is highly unusual and may well be unique to the UK. ARTICLE 19 is not aware of any other law where a similar ceiling operates, with the exception of Scotland, which is governed by a law from the same pedigree. In Australia, the FOI Act specifically states that no financial limit of this kind will be set by the government.[26] It is notable that many of the EU's poorer members, including Bosnia and Herzegovina, Bulgaria, Estonia and Romania, will process requests up to any cost, while charging the excessive part to the requester.

17. We believe the appropriate limit is not in keeping with the proportionality principle. It operates as a complete bar to expensive requests, regardless of the public interest therein and regardless of whether the requester is willing to carry the excess part of the cost him/herself. Should the Government insist on reducing the already modest cost of the FOI Act, a less harmful approach from the perspective of the right to know would be to use the power given to it under Section 13 of the FOI Act, which authorises the imposition of fees when the cost of complying with a request exceeds the appropriate limit. Moreover, to ensure that legitimate complex requests are not discouraged, the American model could be followed. Under the US FOI Act, requests from educational, non-commercial, scientific and news media representatives enjoy a discounted rate, while requests deemed in the public interests are processed for free.[27]

5. COUNTING CONSIDERATION TIME WILL WORK AGAINST PUBLIC INTEREST REQUESTS

18. It is reasonable to assume that, in general, the amount of time required to read, consult and consider increases with the sensitivity of a request. This means that the new rules, far from ruling them out, would actually favour light-hearted requests when compared to serious political issues.

19. The draft regulations permit only time spent determining the applicability of a Part II exemption to be counted as consultation or consideration. While this helps bring some thorny requests back within the appropriate limit, it creates a bias for bland questions since these require proportionately less consideration under Part II. In a similar vein, the cost thresholds (below which consideration and consultation time may not be taken into consideration) and cost ceilings (over and above which any additional costs cannot be counted) would act in favour of straightforward requests while doing little to salvage contentious ones.

20. At a recent adjournment debate in the House of Commons, the Parliamentary Under-Secretary of State at the Department for Constitutional Affairs, Vera Baird, held the media culpable of wasting taxpayer money through "open-ended trawling and unspecific and unfocused inquiry."[28] The proposed regulations will do little to reward more targeted requests. Time spent locating, retrieving and extracting information is already counted under the existing rules; consultation time depends on the sensitivity of a request, much less on how well defined it is.

21. The proposed regulations also reward public authorities which engage in excessive consultation or consideration with a view to breaching the appropriate limit so as to deny access. The DCA consultation paper is confident that abuse can be prevented by providing proper guidance to public bodies. We do not share this optimism, which is not supported by the experience of other established democracies with long-standing FOI regimes, such as Canada and Australia. There will always be strong secretive tendencies within government, particularly where corruption, mismanagement or even simple waste is involved, and no amount of guidance will prevent this. We note that for similar reasons the US Freedom of Information Act specifically excludes such costs[29] and this was presumably the reasoning behind the current exclusion of such costs from the appropriate limit calculation.

22. It has been suggested that any threat of abuse of the new cost rules is effectively countervailed by the possibility of appeal to the Information Commissioner. This is at best only a very partial solution. The time and cost, including in terms of human resources, of such an appeal act as a significant disincentive to many requests. For many information requesters, especially the journalists who will be amongst those most affected by the new rules, time is of the essence and information loses its value if it cannot be obtained quickly. Furthermore, it will often be very difficult for the Commissioner to identify clearly abuse of this nature. These problems are exacerbated by the fact that the Commissioner is already burdened with a serious backlog of cases.

6. AGGREGATION OF NON-SIMILAR REQUESTS MAY BE APPLIED SELECTIVELY

23. The DCA has set out four criteria which public authorities may employ to decide whether aggregation of non-similar requests is "reasonable in all the circumstances": (1) the level of disruption caused by dealing with the requests; (2) whether the applicant is acting in an individual capacity or for a business or professional reason; (3) the number of requests the applicant has made in the past; and (4) whether the applicant has previously been "uncooperative or disruptive".

24. Lord Falconer has said that:

This statement reflects the position in a great majority of jurisdictions with access to information legislation.

25. Points 2-4 above contradict the principle of applicant-blindness, since in varying degree they entail value judgements on the person and motive of the requester. In effect, point 2 allows an assessment of whether the information is of sufficient importance to the requester; point 3 permits a judgement of whether he/she has already benefited from the Act enough; and point 4 whether he/she is a good partner and warrants further assistance.

26. The combined import of points 2-4 is that public authorities may place institutional requestors, such as journalists and NGOs, on an "FOI diet" of as little as £600 per 60 working days, although "cooperative" partners may be rewarded with a higher allowance. It is not hard to see how the discretionary element in disclosing information might be used to favour "loyal" reporters.

27. To conclude, the right to information is now widely recognised not only as a fundamental human right, but also as a key underpinning of democracy, a central tool in the fight against corruption and incompetence and an invaluable means of promoting public accountability. To save what is ultimately a very minor sum of money, the government is proposing changes which will undermine the country's fledgling FOI regime just two years after it was put into place. We strongly urge the withdrawal of the proposed regulations.

Dr Agnès Callamard
Executive Director

February 2007

13 Jack Straw, then Home Secretary, on the introduction of the Freedom of Information Bill: Hansard, House of Commons Debates, 7 December 1999, col 714. Back

14 Constitutional Affairs Committee, Freedom of Information-One Year On, June 2006, HC 991, p 9. Back

15 Consultation paper 28/06, pp 13-14. Back

16 Available online at http://www.foi.gov.uk/reference/foi-independent-review.pdf. Back

17 Partial Regulatory Impact Assessment, para 20. Back

18 Ibid, para 40. Back

19 Ibid, para 41. Back

20 Ibid. Back

21 Pre-Budget Report 2006, Summary, at http://prebudget2006.treasury.gov.uk/page08.html. Back

22 See Toby Mendel, Freedom of Information as an Internationally Protected Right, available at http://www.article19.org/pdfs/publications/foi-as-an-international-right.pdf. See also Claude Reyes et al v Chile, judgement of 19 September 2003, Case 12.108, Inter-American Court of Human Rights (not yet published). Back