Holland v The Information Commissioner & Anor

The origins of the EIR are to be found in the Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters 1998 (referred to in this decision as “the Aarhus Convention” or “the Convention”). The objectives of the Aarhus Convention include rights of access to information (article 1) and article 4 requires Parties to ensure that, in response to a request for environmental information, public authorities make such information available. There is an express exemption at article 4.3(a) where the public authority “does not hold the environmental information requested”, subject to an obligation (article 4.5) to inform the applicant of the public authority to which it believes it is possible to apply or transfer the request to that authority.

The UK and the European Union are signatories to the Aarhus Convention and it has been implemented by the EU by means of Directive 2003/4/EC on public access to environmental information (“the Directive”).

Article 1 of the Directive provides that its objective includes

“to guarantee the right of access to environmental information held by or for public authorities…”

Article 2.1 of the Directive defines “environmental information” as follows:

“…any information in written, visual, aural, electronic or any other material form on–

(a)

the state of the elements of the environment, such as air and atmosphere, water, oil, land, landscape and natural sites including wetlands, coastal and marine areas, biological diversity and its components, including genetically modified organisms, and the interaction among these elements;

(b)

factors, such as substances, energy, noise, radiation or waste, including radioactive waste, emissions, discharges and other releases into the environment, affecting or likely to affect the elements of the environment referred to in (a);

(c)

measures (including administrative measures), such as policies, legislation, plans, programmes, environmental agreements, and activities affecting or likely to affect the elements and factors referred to in (a) and (b) as well as measures or activities designed to protect those elements;

(d)

reports on the implementation of environmental legislation;

(e)

cost-benefit and other economic analyses and assumptions used within the framework of the measures and activities referred to in (c); and

(f)

the state of human health and safety, including the contamination of the food chain, where relevant, conditions of human life, cultural sites and built structures inasmuch as they are or may be affected by the state of the elements of the environment referred to in (a) or, through those elements, by any of the matters referred to in (b) and (c);”

Article 2.3 defines “information held by a public authority” as

“environmental information in its possession which has been produced or received by that authority”.

I also set out article 2.4 because, although Mr Holland has not applies in this case, he has referred to its terminology as shedding light on the meaning of Article 2.3. Article 2.4 defines “information held for a public authority” as

“environmental information which is physically held by a natural or legal person on behalf of a public authority.”

Article 3 requires Member States to ensure that public authorities are required to make available environmental information held by or for them to any applicant on request. Article 4 provides for exceptions, including if the information is not held by or for the public authority.

The EIR implement the UK’s obligations under the Directive. The relevant provisions are as follows

Regulation 2(1) defines “environmental information” in the same terms as Article 2.1 of the Directive.

Regulation 3 defines the scope of the EIR’s application:

“(1)

Subject to paragraphs (3) and (4), these Regulations apply to public authorities.

(2)

For the purposes of these Regulations, environmental information is held by a public authority if the information –

(a)

is in the authority's possession and has been produced or received by the authority; or

(b)

is held by another person on behalf of the authority.”

Regulation 5(1) requires that

“a public authority that holds environmental information shall make it available on request.”

This is subject to exceptions including the following in Regulation 12:

“(1)

Subject to paragraphs (2), (3) and (9), a public authority may refuse to disclose environmental information requested if –

(a)

an exception to disclosure applies under paragraphs (4) or (5); and

(b)

in all the circumstances of the case, the public interest in maintaining the exception outweighs the public interest in disclosing the information.

(2)

A public authority shall apply a presumption in favour of disclosure.

…

(4)

For the purposes of paragraph (1)(a), a public authority may refuse to disclose information to the extent that –

(a)

it does not hold that information when an applicant's request is received;

…”

It is also relevant to note that the Convention, Directive and EIR each define the term “public authority”.

As the EIR implement similar provisions in the Directive, which itself gives effect to the Aarhus Convention, it is relevant to have regard to the UN Economic Commission for Europe’s document, “The Aarhus Convention: an Implementation Guide” (“the Implementation Guide”), although it has no binding force: Solvay v Region Wallonne C-182/10, [2012] 2 C.M.L.R. 19, at paragraph 27. At page 83 the Guide addresses Article 4.3(a) of the Convention as follows:

“A public authority is required to give access only to the information that it “holds”. This means that if a Party decides to provide for this exception, it will need to have defined what is meant by “holding” information. However, information that is held is certainly not limited to information that was generated by or falls within the competency of the public authority. The Convention provides some guidance in article 5, paragraph 1 (a), which requires Parties to ensure that public authorities possess and maintain environmental information relevant to their functions. In practice, for their own convenience, public authorities do not always keep physical possession of information that they are entitled to have under their national law. For example, records that the authority has the right to hold may be left on the premises of a regulated facility. This information can be said to be “effectively” held by the public authority. Domestic law may already define conditions for physical and/or effective possession of information by public authorities. Nothing in the Convention precludes public authorities from considering that they hold such information, as well as the information actually within their physical possession.

If the public authority does not hold the information requested, it is under no obligation to secure it under this provision, although that would be a good practice in conformity with the preamble and articles 1 and 3. However, failure to possess environmental information relevant to a public authority’s responsibilities might be a violation of article 5, paragraph 1 (a). Moreover, where another public authority may hold the information, the public authority does have a duty under article 4, paragraph 5, to inform the applicant which public authority may have the information. Alternatively, it can transfer the request directly to the correct public authority and notify the applicant that it has done so. In either case, the public authority must take these measures as promptly as possible.”

Article 5.1(a) of the Aarhus Convention, to which the above refers, provides that parties to the Convention shall ensure that “Public authorities possess and update environmental information which is relevant to their functions.”

Requests for information which is not environmental information within the EIR are dealt with under FOIA. Section 3(2) provides:

“… information is held by a public authority if –

(a)

it is held by the authority, other than on behalf of another person, or

(b)

it is held by another person on behalf of the authority.”

The meaning of “held” in FOIA has been considered by the Upper Tribunal in University of Newcastle upon Tyne v Information Commissioner and British Union for the Abolition of Vivisection [2011] UKUT 185 (AAC) (“BUAV”) and Department of Health v Information Commissioner and Lewis [2015] UKUT 159 (AAC). In BUAV at [27] the Upper Tribunal upheld the decision of the First-tier Tribunal which had reasoned:

“‘Hold’ is an ordinary English word. In our judgment it is not used in some technical sense in the Act. We do not consider that it is appropriate to define its meaning by reference to concepts such as legal possession or bailment, or by using phrases taken from court rules concerning the obligation to give disclosure of documents in litigation. Sophisticated legal analysis of its meaning is not required or appropriate. However, it is necessary to observe that ‘holding’ is not a purely physical concept, and it has to be understood with the purpose of the Act in mind. Section 3(2)(b) illustrates this: an authority cannot evade the requirements of the Act by having its information held on its behalf by some other person who is not a public authority. Conversely, we consider that s.1 would not apply merely because information is contained in a document that happens to be physically on the authority’s premises: there must be an appropriate connection between the information and the authority, so that it can be properly said that the information is held by the authority….For example, an employee of the authority may have his own personal information on a document in his pocket while at work, or in the drawer of his office desk; that does not mean that the information is held by the authority.”

(cited in BUAV at [23])

Upper Tribunal Judge Wikeley continued:

“28.

…The test is not whether the public authority “controls” or “possesses” or “owns” the information in question; simply whether it “holds” it … “Hold”… is an ordinary English word and is not used in some technical sense in the Act. That construction is also supported by one of the leading texts, Information Rights: Law and Practice by Philip Coppel QC (3^rd edn, Hart Publishing, 2010), which observes that FOIA “has avoided the technicalities associated with the law of disclosure, which has conventionally drawn a distinction between a document in the power, custody or possession of a person” (p.339, para. 9-009)….

29.

… I do not regard the tribunal’s reference to the need for “an appropriate connection between the information and the authority” as a misguided attempt to replace the statutory language with its own “rather nebulous” test (as Mr Pitt-Payne put it). On the contrary, the tribunal was simply pointing to the need for the word “hold” to be understood as conveying something more than the simple underlying physical concept, given the intent behind section 3(2).”

In that case Judge Wikeley confirmed the First-tier Tribunal’s decision that information contained in project licences issues in respect of experiments carried out at the University, was “held” by two Professors who held the licences on behalf of the Registrar and that they were acting under their contracts of employment.

In Lewis Charles J said at [96] that the application of the “ordinary English word” approach in BUAV indicates that:

“a)

before an authority can be said to “hold” information as a matter of ordinary usage of language it will have been given it, or have obtained it, or have created it, and

b)

the reasons why it was given it, or obtained it or created it inform on whose behalf it holds the information and thus whether it holds the information solely for another person, or solely or partly for itself.”

Charles J said that this means that possession is not enough; there must be “a sufficient connection between the information and the authority to found the conclusion that it holds that information”, but it is not necessary to find that the information is held solely or predominantly for the authority’s purposes.

I have drawn attention to these decisions because of the apparent similarity in the scope of FOIA and the EIR (despite some differences in the wording), in that they both apply to information which is held by a public authority. However neither of the Respondents relied on these decisions in support of their position in respect of the EIR. The EIR must be interpreted in accordance with the EU Directive which they implement. FOIA is an entirely separate domestic regime and cannot affect the meaning of the EIR and Directive which is a matter of EU law. Nonetheless, the Respondents have derived some reassurance from the fact that the result for which they contend as a matter of EU law chimes with the approach under FOIA, albeit by a different legal route. In particular Ms John noted that the observations at paragraph 96 of Lewis outline an approach which is very similar to the “produced or received by” test in Regulation 3(2) (as to which, see below).

Before turning to the grounds of appeal Ms John addressed me on the question whether this Tribunal should be determining the appeal at all, at least in relation to request 1. After the date on which the request was refused by the University, the information in request 1 was published on the IPCC website. Ms John submitted that this does not render the appeal academic. I agree. The relevant date for determining the appeal is the date of the refusal: R (Evans) v Attorney General [2015] AC 1787 at [72]. There is a live issue of law between the parties which merits the Upper Tribunal’s attention.

This ground relates to requests 2 and 3. Requests 4 and 5 were answered in the University’s review decision.

In the First-tier Tribunal Mr Holland said that he disputed the Commissioner’s conclusion that FOIA applied to all of the requests other than the first part, but he did not advance any basis for doing so. He said that he could not classify the documents without seeing them. He repeated that submission in the Upper Tribunal. But to require disclosure of documents to the requestor in order to allow him to form a view as to or make submissions on which regime is applicable would defeat the purpose of the legislation. In any event, in this case at least, it is possible to determine the applicable regime by reference to the requests.

In the absence of any substantive challenge to the Commissioner’s classification, it seems to me to be difficult to criticise the First-tier Tribunal for accepting that classification. In the Upper Tribunal appeal Mr Holland has not explained why the First-tier Tribunal was in error in that respect. I have no difficulty in concluding that there was no such error. The regulation should be liberally construed. But the information requested related to administrative matters which were incidental to the production and handling of the reports and were not within any part of the definition of environmental information.

I have not dwelt on this issue because it can make no difference to the outcome of the appeal. The decision of the First-tier Tribunal that the information was not held by the University would have applied to all parts of the request even if all addressed under the EIR. Indeed, at paragraph 11 the First-tier Tribunal appears to have addressed the whole of the request under both statutory regimes.

Mr Holland submitted that information is held if it is in the physical possession of the authority. He said that “possession” should be given its ordinary meaning: that a person “has” the information or it is “in their control”. He said that the University of Cambridge held the information because it was physically in its possession in that it had been sent to Professor Wadhams at his University email address which operated through the University’s server and so it was held on a computer owned by the University. Any information which was on its computer system was either received or produced by it. Moreover, the postal address used by Professor Wadhams for his IPCC work was that of the University and so documents would have been received by the University when posted to him. In addition, Mr Holland said that it is significant that FOIA specifically excludes information held on behalf of another person, but that the EIR do not.

Central to his case are his submissions that the EIR should be construed consistently with the provisions and the policy objectives of the EU Directive 2003/4/EC and the Aarhus Convention as further explained or elucidated by the Implementation Guide. Mr Holland made extensive reference to the preambles to the Directive and Convention to support his submission that exceptions should be interpreted restrictively. In particular, he referred to the 17^th preambular paragraph to the Convention (“acknowledging that public authorities hold environmental information in the public interest”) as explained at page 34 of the Implementation Guide, which states that “underlying the principle of openness is the notion that information held by public authorities is held on behalf of the public” and that it is “improper to talk of ownership of such information”. He relied on Exports Credits Guarantee Department v Friends of the Earth [2008] EWHC 638 (Admin), in which Mitting J explained that the right to information under the Directive meant that disclosure should be the general rule and emphasises the public interest in disclosure. Mr Holland said that there is a heavy burden on an authority seeking to rely on exceptions. He sought support from the travaux preparatoires for the Directive, and in particular that the draft Directive did not originally define “held” but in later drafts identified two ways in which information can be held – either by or for a public authority – as in what are are now articles 2.3 and 2.4.

Mr Holland also relied on the Almaty Guidelines on Promoting the Application of the Principles of the Aarhus Convention in International Forums”, produced by the Economic and Social Council of the UN in 2005. Paragraph 23 provides that “any member of the public should have access to environmental information developed and held in any international forum upon request”. In addition he said that the First-tier Tribunal had implicitly adopted the Commissioner’s reliance on Article 5.1(a) of the Aarhus Convention, but that this was wrong because Article 5 is directed to a different issue and cannot modify the meaning of “held”.

Mr Holland submitted that the University’s claim that it did not physically have the information was not tested by the Commissioner or the First-tier Tribunal because of their erroneous approach to the meaning of “held”.

Ms John, on behalf of the Commissioner, submitted that Mr Holland’s approach was inconsistent with the language of the Directive and the Regulations. She conducted a detailed review of the Travaux Preparatoires for the Directive and said these show that a clear decision was made by the EU institutions that mere possession of information was insufficient. She submitted that Mr Holland’s position would deprive the words “produced or received by” of any meaning or function and would mean that personal communications received by employees at their place or work would be “held” by the authority as would communications about wholly independent activities. Information is not held by a public authority merely because it finds its way on to the authority’s premises or computer systems in relation to activities of individuals which are not connected with the authority. In deciding whether information is “held”, the capacity in which a public authority possesses the information must be considered. It will be necessary to determine whether a particular piece of information was produced or received by an individual in their personal capacity, in their professional capacity as an employee of the authority, or in some other capacity such as MP or school governor. This is a question of fact for the Commissioner or, on appeal, the First-tier Tribunal to determine. In the present case, the First-tier Tribunal made determinations of fact and correctly concluded that the information was not held by the University.

On behalf of the University, Mr Cross supported Ms John’s submissions. In particular, he said that the EIR do not mean that any information produced or received by an employee of a public authority is produced or received by the authority. So, even if (which the University denied) the information was on its computer system at the time of the refusal, it was not produced or received by the University: it was produced or received by Professor Wadhams. The words “by the authority” in regulation 3(2)(c) and article 2.3 of the Directive are significant and exclude information produced and received by an employee in an independent capacity. The text of the Directive and the background documentation show that the Commission gave careful consideration to the definition of “public authority” which is clearly set out in the Directive, but the approach in relation to “held” is quite different because it is difficult to lay down hard edged rules as to the degree of association required. It is a matter left to be determined on the basis of factual inquiry. Finally, he said that Mr Holland’s position was not supported by a purposive construction of the Convention and the Directive. The focus of those instruments is not on those who are not public authorities.

I start with the language of regulation 3(2)(a), which mirrors article 2.3 of the Directive. It requires that the information is in the authority’s possession and produced or received by it. The latter phrase must have been intended to add something to the fact of possession. This is reinforced by the traveaux preparatoires which show that the words “and produced or received by” were included after much deliberation. The original Commission proposals for the Directive did not include the words. The European Parliament introduced the concept of “received or produced by the public authority” at First Reading. During the subsequent legislative process involving the Commission, Council and Parliament the words were removed and then reinstated, and finally all institutions agreed to the final version of article 2.3 In the light of the deliberate inclusion of the words, they should not be ignored nor given an interpretation which renders them superfluous. Mr Holland did not say what the force of “produced or received by” is but, if he was right that the phrase would be satisfied where information is received by means of an electronic communication on an authority’s computer system, this would render the phrase superfluous.

In addition, Mr Holland’s approach ignores the specific and important word “by” in the phrase “produced or received by”. This shows that the authority must itself be the producer or recipient of the information. Ms John pointed out, and I agree with her, that this is particularly clear in respect of the words “received by”. They must mean more than “came into the possession of”, or they would be superfluous and regulation 3(2) would apply to information which is “in the possession of the public authority and which has come into its possession”.

Ms John gave a number of examples illustrating the correct approach. Where a student places an essay in the pigeon hole in the Porters’ Lodge for the attention of a tutor, the essay is received by the tutor on behalf of the University and the University can be said to have “received” the document. It possesses it and it is the recipient of it. However, when a personal bank statement is sent to an academic at the University, and delivered to that person’s pigeon hole, the bank statement is not received by the University. It is sent to the academic in his or her personal capacity and is received by that person in that capacity. Although it is physically on the University’s premises, the University is not the recipient. The same approach applies if these documents are received by way of emails on the University’s server. The same approach also applies to documents produced by the academic. A set of lecture notes handed to students, sent to them by email or posted on the University’s website or a shared drive, is produced by a lecturer in the capacity of University employee and produced by the University. An email sent by a lecturer to his or her spouse or partner about a social engagement after work is not produced by the University even though a copy may be on its server. These examples do not involve environmental information, but the position would be the same where they do. Thus if a lecturer was involved in a purely personal capacity in a local campaign against a proposed road development near where his or her home, and used the University email address to send a letter from the campaign group to the lecturer’s local MP, the mere fact that a copy of the letter was on the University’s server would not mean that it was produced by the University.

Of course, there will be cases which are less straightforward than the examples above, but the approach is the same. A factual determination is required as to how the information has come to be in the possession of the authority. The question is whether the information was produced or received by means which were unconnected with the authority, for example by an individual in their personal or other independent capacity; or whether it was produced or received by means which were connected with the authority, for example by someone acting in their professional capacity in relation to the authority (such as an employee of the authority). The connection must be such that it can be said that the production or receipt of the information is attributable to (“by”) the authority.

This approach is consistent with the Convention and the Directive. Those instruments clearly limit the scope of the right of access to information to that which is held by public authorities as defined. They do not provide for a general right to all environmental information regardless of the circumstances in which it arose or exists.

Mr Holland had submitted that, because article 2.4 is expressly limited to information which is “physically held” but article 2.3 is not, it can be inferred that article 2.3 applies to information which is physically held (including on its computer systems) or if it is held elsewhere under its control or the control of its officers and employees such as Professor Wadhams. This submission reads too much into the use of the word “physically” in article 2.4. The European Commission’s proposals for the Directive explain that the purpose of this provision is to ensure that environmental information is accessible where a public authority is entitled to hold it on its own account but it is kept physically elsewhere. It does not assist with the meaning of “held” in article 2.3. Further, if that submission were correct it would introduce the concept of holding information for the authority into article 2.3 and essentially elide the meanings of “held by” and “held for”, contrary to the structure of Article 2.

It was not entirely clear at the hearing whether Mr Holland was seeking to make a separate submission that the information was held by Professor Wadhams on behalf of the University in accordance with article 2.4. If so, it was not a submission properly open to him. That was not his case before the First-tier Tribunal and, at the permission hearing before me, Mr Holland was clear that he did not rely on that provision.

Finally, on this point, even if Mr Holland’s approach to the construction of article 2.3 and 2.4 was correct, he could not succeed in this appeal unless Professor Wadhams performed his role on the IPCC in some capacity as officer or employee of the University (see his Reply at paragraph 27), but the tribunal found to the contrary as a matter of fact.

Mr Holland’s reference to the judgment of Mitting J in ECGD does not take matters any further. That case was concerned with the public interest in maintaining an exception to disclosure. It was not concerned with the prior question as to whether information is held by a public authority.

My conclusion is not affected by the Almaty Guidelines. The guidelines were adopted by the parties to the Aarhus Convention in order to “provide general guidance to the Parties on promoting the application of the principles of the Convention in international forums” and are not applicable in the present context. In any event, there is no indication that the Guidelines were intended to expand the scope of the Convention. Indeed, the adoption of Convention definitions (see paragraph 8) indicates the contrary. Paragraph 23 does not qualify the exclusions in article 3 of the Convention.

I also reject Mr Holland’s submission that the First-tier Tribunal erroneously adopted the Commissioner’s reliance on article 5.1(a) of the Aarhus Convention. In paragraph 6 of its decision the First-tier Tribunal recorded the Commissioner’s submission that his interpretation of regulation 3(2)(a) EIR was consistent with article 5.1(a) of the Convention, but that aspect of the submission does not appear to have formed any part of the tribunal’s reasons. In any event, the Commissioner was correct to submit that his approach to the EIR was consistent with the Aarhus Convention. Article 4.3 of the Convention does not define the term “held”. Page 83 of the Implementation Guide, cited above, recognises that it is for the parties to the Convention to define what the term means. The EU has done this in Article 2.3 of the Directive. The Guide also explains that article 5.1(a) may shed light on whether information is “held”, so that it includes information which is relevant to an authority’s functions. In order to comply with article 5.1(a) States must ensure that public authorities possess and update environmental information which is relevant to their functions. That information will be produced or received by the authorities and so the public will have a right of access to it. This is entirely consistent with the Respondents’ approach and my analysis.

I conclude my analysis of the meaning of “held” by noting that, if Mr Holland’s approach was correct, it would mean that, although Professor Wadhams was not a public authority (as Mr Holland rightly conceded) and so environmental information held by him in connection with his role on the IPCC was not subject to the EIR regime, that same information would be accessible because it was held by the University. This would have the effect of undermining the purpose of the carefully crafted definition of “public authority” in the Convention, Directive and the EIR. Regulation 2(2)(d) means that the EIR applies to those who would not otherwise be defined as a public authority, in respect of their public role, but only where they are under the control of a public authority. To construe regulation 3(2)(a) as applying the EIR to information produced by such a person who is not (in relation to that information) under the control of a public authority would defeat the careful crafting of the public authority definition.

The First-tier Tribunal’s decision was consistent with the above approach. It found as a matter of fact that Professor Wadhams acted in a private capacity when assisting the IPCC. It is not open to the Appellant to go behind those findings of fact, there being no suggestion of any error of law in making those findings. In the light of the proper approach to the meaning of “held” there was simply no need for the First-tier Tribunal to interrogate the University any further as to whether any of the information could be found on the University’s computer system. As the information was not produced or received by the University, it did not need to determine the question of possession.

It follows that there was no error of law by the First-tier Tribunal and this appeal is dismissed.