Recently, David Holland reported the unedifying spectacle of John Mitchell, Chief Scientist of the Hadley Center, an institution which proclaimed itself to be the “most significant” contributor to WG1, attempting to circumvent mandated requirements that IPCC be “open and transparent” and that all “written comments” be archived, by claiming that his email correspondence with IPCC Authors using his Hadley Center email account were somehow his “personal” property and therefore not subject to UK Freedom of Information legislation.
Elsewhere, we’ve seen the equally unedifying spectacle of Caspar Ammann, a
U.S. federal employee, an employee of a federally funded research and development center (NCAR), seeking to circumvent the IPCC requirement that comments be “open and transparent” by sending comments to IPCC author Briffa at his CRU email, rather than through the IPCC registry (as even Susan Solomon did). Even though IPCC is required to archive all written comments, Ammann and Briffa regard themselves as above IPCC policy and entitled to exchange confidential comments, sort of like little school children whispering at the back of the classroom.
While the spectacle is obviously unedifying in itself, I thought that it would be interesting to examine government policies on e-mail correspondence to see whether there is any policy basis for either theory.
Canadian Government Policy
I’m going to start by discussing Canadian government policy on e-mail involving government employees at their place of business, not because it is directly applicable to any of the parties, but because, in this case, the Canadian government appears to have spelled out their policies in more detail. Given the close similarity of applicable legislation in the U.K. to Canadian legislation, it becomes a useful reference point. (My take in this respect is that, by issuing the manuals, the Canadian government is not promulgating new policy, but merely providing information to employees on policies that are already required by the legislation. That’s why one can reasonably expect the principles in these manuals to carry over to the U.K. situation,)
A first principle is that most email messages are “records” (the first test in FOI legislation):
1. Most email messages are records: Email messages, including any electronic attachments, created, collected, received or transmitted in the normal course of government business which reflect the functions, business activities, and decisions of government are records.
Second, e-mail received at the Hadley Center comes under the “control” of the institution:
A record is under the control of a government institution when that institution is authorized to grant or deny access to the record, to govern its use and, subject to the approval of the Librarian and Archivist of Canada, to dispose of it. Regarding the question of physical possession, a record held by an institution, whether at headquarters, regional, satellite or other office, either within or outside Canada, is presumed to be under its control unless there is evidence to the contrary. A record held elsewhere on behalf of an institution is also under its control, for example at an employee’s home or on business travel.
The distinction between “official” and “personal” records is discussed as follows:
Email messages created, collected, received or transmitted during the normal course of government business are records of the Government of Canada. Such messages and their attachments reflect the functions, business activities, and decisions of the government. They must be kept to ensure the integrity of the corporate memory of government. Records, including Transitory records , regardless of format, must only be disposed of according to Records Disposition Authorities approved by the Librarian and Archivist of Canada.
Email messages whose content is of a personal nature are not records of the Government of Canada and therefore are not covered by the Library and Archives of Canada Act. Examples include email messages regarding arrangements for lunch, an employee’s personal information such as email relating to hobbies, extracurricular activities, announcements, unsolicited advertising, etc. Such messages should be deleted once their usefulness is completed. However, a user must not delete records where the institution has received a formal request, under the Access to Information or Privacy Acts, relating to these records
Assuming a similar U.K. policy, Mitchell’s claim seems absurd on its face. His correspondence with IPCC authors presumably rose above being “arrangements for lunch” and his correspondence with IPCC authors in his capacity as Review Editor hardly relates to a hobby or extracurricular activity (by which one would think of things like squash games). I see no conceivable basis under which Mitchell’s claim that his IPCC correspondence were his “personal” property could possibly stand up. Accordingly, by asserting title to the correspondence that is inconsistent with the title of the true owner (his employer), Mitchell has, in effect, converted the property. The conversion of company property by its executives is a touchy issue. My advice to Mitchell, for whatever it’s worth, would be not to go down this road. The risks and rewards are worth considering; maybe there’s an even a precautionary principle.
But aside from anything else, the spectacle of Mitchell arguing that the correspondence is his “personal” property is going to appear very distasteful to any member of the public who encounters the argument. Considerable interest has arisen in the UK about FOI legislation in connection with MP salaries and benefits. Would the tabloid newspapers get interested in the evasions of the U.K. Met Office? Who can tell. I was surprised at the traction of the Y2K issue last summer.
U.S. Government Policy
Here is a statement summarizing model U.S. policy.
It states explicitly that employees do not have any right or expectation of privacy when using government e-mail, which are waived. So Ammann, at his end, has little basis for claiming an expectation of privacy vis-a-vis U.S. FOI. Given a comparable Canadian policy, it seems unlikely that U.K. policy would deviate much from these principles.
Executive Branch employees do not have a right, nor should they have an expectation, of privacy while using any Government office equipment at any time, including accessing the Internet, using E-mail. To the extent that employees wish that their private activities remain private, they should avoid using an Agency or department’s office equipment such as their computer, the Internet, or Email. By using Government office equipment, executive branch employees imply their consent to disclosing the contents of any files or information maintained or pass-through Government office equipment.
By using this office equipment, consent to monitoring and recording is implied with or without cause, including (but not limited to) accessing the Internet, using E-mail. Any use of government communications resources is made with the understanding that such use is generally not secure, is not private, and is not anonymous.
This language is adopted by a number of institutions. Here’s similar language at Stanford, for example.
Regulations on public companies in the U.S. are very severe under Sarbanes-Oxley.
Financial regulations, such as Sarbanes Oxley and Basel II, stipulate that companies must store all emails for at least seven years. Although none have been convicted yet, chief executive officers found to have failed in such auditing regulations can face heavy fines and jail sentences.
Sarbanes-Oxley obviously does not apply to executives of the U.K. Met Office, but readers will undoubtedly understand why the destruction of e-mail correspondence by a Hadley Center executive is something that appears very unsavory to someone familiar with securities regulations.