OK, I know I’m beating a dead horse. Within the last month I’ve already posted several times (see here, here, and here) about bogus anticorruption statistics, as has Rick. And I promise that after this post, I’ll move on to other topics. But I can’t help commenting on this latest release from Transparency International, criticizing the recent World Economic Forum (WEF) meeting for not explicitly addressing corruption. As its lead example, TI faults the WEF for not addressing issues like open data (and openness more generally). I’m sympathetic to TI’s policy position, but in making the case, TI asserts, “One study suggests that open data could reduce the costs of corruption by about 10 percent.”
I was curious (and, admittedly, skeptical) about yet another seemingly precise estimate of something that’s inherently hard to measure. So I clicked on the link to the “one study” that “suggests” that open data technologies would reduce the costs of corruption by 10%. This “study” is actually a report (really, an advocacy document) from an Australian consulting firm (Lateral Economics), commissioned by a philanthropic fund (the Omidyar Network) that invests in open data initiatives. How does this “study” reach its conclusion that open data could reduce the costs of corruption by 10%? I will now quote in full the entirety of the evidence and analysis supporting that conclusion:
There is now a growing body of evidence that open data plays a role in reducing corruption. Open data can reduce the extent of corruption by both reducing its private returns and making it easier to detect. Based on the evidence, we think it reasonable to suggest that the costs of corruption would be reduced by of [sic] the order of 10%. [p. xiv.]
…
[And again, later in the document:] To estimate the benefits of open data in reducing corruption, we begin with estimates of the costs of corruption in developed countries…. We think a reasonable contribution of open data in reducing corruption is on the order of 10%. [p. 58]
That’s it. What’s “the evidence” on which this 10% estimate is “based”? Who knows. There’s essentially no discussion, save for one anecdote (repeated several times throughout the document) about how open data helped prevent a $3.2 billion tax fraud in Canada in 2007–which, it should go without saying, is not actually a measure of the costs of corruption. As far as I can tell, the 10% figure is completely made up.
I’m not sure I can really come down too hard on TI for including this bogus figure in what is, after all, an advocacy-oriented press release… But really, I mean, come on. This casual invention and repetition of fabricated figures is ridiculous, embarrassing, and–as Rick warns–could come back to haunt the anticorruption movement, ultimately undermining its credibility. That TI press release could have worked just as well, and been just as persuasive in presenting the case for the importance of open data, without all the nonsense statistics thrown in just to sound impressive.
Seriously, guys, cut it out.
GAB | The Global Anticorruption Blog 
No, you are not beating a dead horse. The inclusion of bogus, made-up, phony, crazy, irresponsible (choose your adjective) numbers in discussions of how to combat corruption is, as your example shows, unfortunately alive, well, and even spreading to heretofore responsible voices. The ten percent claim should be WAGed.
I honestly doubt that Open Data has the impact Transparency International and other organisations (read Open Government Partnership) claim. Although there is not much (if any) available data to go around, it is difficult to see how the introduction of a tool could make any difference where there are no operators; this is, aside from the developed world, countries struggling with pervasive low and high-level corruption do not have a strong-enough civil society to take advantage of governmental data. Information overload allows corruption as much as a culture of secrecy does.
On the other hand, when we look at specific cases, it is easy to see how political and social institutions in the developing world reflect a distressing level of corruption tolerance. Therefore, to suggest that the presence of open data will automatically reduce corruption in any significant way is a clear overstatement. At the very least, open data needs to be discussed within pertinent conditions, and not as a standard fix.
Mathew, I am afraid I am with you on this one. There is no point in making bold statements of ‘fact’ that undermine the credibility of your program like TI appears to have done. A little bit of due diligence and thought would have avoided this overstatement of the case. Keep flogging this one as I don’t think it is dead yet.
In a recent large scale lab-in-the-field experiment in Tanzania they showed that transparency of information did not decrease the level of embezzlement – so I also doubt that this link between open data and corruption really exists.
(Di Falco, S., Magdalou, B., Masclet, D., Villeval, M.C. and Willinger, M., 2015. “Embezzlement: Does transparency of information matter? An experiment in Tanzania” URL: http://asfee2015.sciencesconf.org/61557/).
Pingback: What’s The Harm? | Anti Corruption Digest