What to Do When Open Data Is Either Wrong, Scandalous, or Both?
BY David Eaves | Thursday, August 23 2012
So here's an interesting fact: The Guardian quotes a now-defunct UK charity watchdog website as saying, "We would normally have concerns about a charity spending greater than 15% of its income on administration.
"Such charities," reads charityfacts.org, "we would ask to justify their level of expense." Other newspapers, like The Independent, state that somewhere between 5 and 20 percent is considered acceptable.
That sounds reasonable.
So imagine the surprise of open data enthusiast and Listpoint champion David Mitton when, while sifting through local UK government budgets, he discovered that 40 percent of spending on homelessness by local authorities is allocated to administrative costs. The 33 local authorities he sampled from Openly Local, a fantastic UK website that aggregates open data from local governments across the UK, had cumulatively spent £67,591,000 ($107,239,880) of which £27,129,000 ($43,042,871) went to administrative costs.
It's a pretty stunning conclusion. It suggests that if these government programs were run as charitable organizations, leading experts would be concerned about their management.
Of course, as David himself pointed out to me, there is a good chance that some of these "excess" administrative costs are a result of poor labeling. In some jurisdictions, like Sunderland City Council and Knowsley Metropolitan Borough, administrative costs were 100 percent. In others, like Solihull Metropolitan Borough Council, administrative costs hit 93 percent. If this is at all accurate it should be scandalous. Hopefully it is just an inaccurate or lazy description of the budget.
But herein lies a prefect example of a second-order open data problem. It is fantastic that there are large swaths of budget data available in the UK thanks to legislation and the great work of websites like Openly Local. But having the data means little if it is poorly coded, and if it is hard to standardize across jurisdictions.
It could be that David Mitton has stumbled upon a fantastic scandal in local politics — that hundreds of thousands, if not millions, of pounds of taxpayer money has been wasted on administrative costs rather than helping those it sets out to assist. Then again, it may be that the money is being well spent.
One promise of open data is that it is supposed to enable citizens to better understand how their governments spend money.
In practice, this problem remains fiendishly difficult, even under the best of circumstances. At the very least it can help flag interesting questions.
What David Mitton really wants to know now is who in government is supposed to be holding someone to account for either terrible performance or for terrible data. The answer to that remains very unclear.
It is worth remembering that it takes more than open data to shift a government. Just because we think we have some knowledge or intelligence doesn't mean that taking action around it becomes radically easier. Someone, somewhere, will still likely have to raise awareness or hound some official. But at least this knowledge can focus that action in a way that was previously not possible, and that is where I think the benefit of sites like Openly Local will, over time, be empowering.
If the spending question David Mitton discovered has a tame answer, hopefully that can be coded in the data, so at least taxpayers can get a better sense of how their money is being spent. If it has a more scandalous answer, well, that will be interesting as well, and a good example of how open data really is letting citizens ask better questions of their government.
Personal Democracy Media is grateful to the Omidyar Network for its generous support of techPresident's WeGov section.