16 September 2010
By Dan Herbert
The government’s data-sharing revolution has begun and armchair auditors have an array of website weapons to choose from. This is not a threat to the public sector but an opportunity for greater accountability, argues Dan Herbert
One of David Cameron’s first actions as prime minister was to write to all government departments outlining plans to open up government data. He believes greater transparency leads to improved accountability and better value for money. His vision is that individuals, groups, companies and not-for-profits will build websites and applications to analyse the newly available data.
One of the more significant moves was the requirement for public bodies to disclose far more detail about their spending. For central government, this started with the decision to publish the Treasury’s Combined On-line Information System. Coins contains, among other data, details of spending plans and outturns across central and local government: an estimated 3.3 million data items just for the 2009/10 financial year.
Almost immediately, groups, newspapers and companies started to mine the data. The Guardian produced a tool that allows individuals to explore the complex figures, and spending analysts Rosslyn Analytics used their software to study the data in detail. A more novel approach is the wheredoesmymoneygo.org website, which produced a ‘dashboard’ of the spending data broken down by function.
Local authorities are being asked to go even further than central government. By January next year, they will have to publish details of all purchases from suppliers worth more that £500. Many authorities have released this information already.
When Local Government Secretary Eric Pickles announced the abolition of the Audit Commission, he said his vision was for an ‘army of armchair auditors’ to do the work of the watchdog using this newly released spending data. And there is indeed at least one armchair auditor. Adrian Short’s website, armchairauditor.co.uk has taken the spending data for the Royal Borough of Windsor & Maidenhead and presented it in a format that allows users to explore it right down to individual payments to suppliers. More importantly, he is distributing his application under an open licence, allowing others to use it for any other authority.
Another website, openlylocal.com, provides comprehensive and accessible information on local authorities. It has used the data to allow users to see spending at transaction level, alongside other detailed information about each authority. Then there’s spotlightonspend.org.uk, developed by data analysts Spikes Cavell. This publishes data for authorities and lets users explore spending by type, using Standard Industrial Classifications.
There have been teething problems and not all data has been published in user-friendly formats. For example, with PDF files, analysts have to extract the data manually before they can use them. Much more useful is a machine-readable form such as .csv, which can be used for automated analysis applications.
Sometimes the categorisation of data itself can lead to misunderstandings. On Spotlight On Spend, Woking Borough Council appears to have spent £18,254 on lingerie. The payment is actually a refund of business rates but the use of Standard Industrial Classifications as the basis for analysis does not make this clear.
These issues aside, it is clear that, in a relatively short space of time, what Adrian Short refers to as a little platoon, rather than an army, of armchair auditors has appeared. In a slightly haphazard and uncoordinated way, they are producing applications that provide a whole new approach to public accountability. Bodies publish their data and the armchair auditors do the rest.
As Lisa Evans, the lead researcher for Where Does My Money Go?, says: ‘If you publish your data in machine readable form, you pretty much don’t have to worry about presenting this data — there is already a community who will do it all for you for free. Everyone wins.’
Once the data is available and presented, the real power of web-based disclosures kicks in. Public sector accountants have long struggled to involve the public in discussion about financial matters. But the perceived complexity of accounting reports, the professional jargon and the degree of aggregation is offputting to non-experts.
The annual reporting cycle also limits the extent to which the public can be involved in discussion of accounting reports. Evidence of this is the poor use of the public right to inspect local authority accounts. The discursive possibilities of web-based disclosures overcome this to some extent. The Armchair Auditor allows comments on individual transactions. Openly Local contains not just spending data but details of councillors, with their e-mail addresses and committee memberships. Asking direct questions is easy and quick, using information from a single source.
It would be easy to dismiss the work being carried out by these new analysts but it has struck a chord not just with the coalition government but with the public. In many ways, it is the antithesis of the approach taken to reporting by the accounting profession. It is unregulated, without standards and based on granular, rather than aggregated, information. The data released is unaudited. The analysis undertaken is personal and subject to possible error. The open data community would not see these as particular weaknesses. If data is made open, then the community can comment on it, improve analyses and correct errors. The power of web communities to develop complex applications should not be underestimated as Wikipedia and others have shown.
But is Pickles’ claim that the armchair auditors can replace the Audit Commission valid? If transactions are being scrutinised by individual members of the public, then there are checks on the probity of spending. As Jeremy Bentham said in 1791: ‘The more closely we are watched, the better we behave.’ More problematic is the judgement of processes and controls, which cannot be assessed merely from looking at transactions. Nor can the outcomes of spending. The information being released is limited; there is no salary data, nothing on small transactions, nothing on internal transfers and nothing that links budgets to outturns. There is no information on asset management, borrowing or reserves. So, yes, there is a limited armchair auditing role but not a full replacement for audit yet. However, there is huge potential to extend and improve the open data made available.
The focus on auditing is perhaps missing the most important implication of all this – that a new channel of financial accountability is being opened up. For example, Freedom of Information requests about specific transactions can be made more easily. Combining spending analysis websites with sites such as whatdotheyknow.com, which provides a channel for automated FoI requests, has built a new accountability mechanism.
These developments give public bodies an opportunity to involve citizens in financial matters and, at a time of spending cuts, to explain the decisions that have to be made. Bodies that treat the release of spending data as a tick-box exercise will miss out on the possibilities. The people working on this are highly skilled at analysing and communicating data but they are not finance professionals. And they are certainly not just local busybodies with an axe to grind.
Chris Taggart, who runs Openly Local, emphasises that there is an opportunity here for forward-looking finance professionals to liaise with analysts and the wider community to improve accountability and public participation. Perhaps the failure of the public to embrace accounting information, as noted in David Walker’s article, is due to the way that the profession has presented the information. Armchair auditing could be a powerful tool to build involvement through improved communication; an opportunity rather than a threat.
Dan Herbert is MSc Accounting programme director at the School of Business, Oxford Brookes University