Many prominent scholars, Nobel laureates Joseph Stiglitz or Amartya Sen among them, emphasised the role of the access to government information for democracy. The emergence of new digital technologies has made seeking and disseminating information as easy as just a few clicks, and therefore, also created new opportunities for governments to become more transparent and for citizens to engage with the published data. Open government data has become central to the global anti-corruption agenda (see strategic granting priorities of the World Bank, international foundations etc.) and political campaigns (see Barack Obama’s presidential campaign in 2008 and the UK Conservative Party’s general election campaign in 2010). David Cameron, the British Conservatives’ leader at the time, argued that the citizens would act as ‘an army of armchairs auditors’ and monitor how public money is being spent.
However, due to the novelty of open government data and a lack of measurements, these assertions have only seen a preliminary investigation. This research has two primary aims. First, benefiting from the emergence of new cross-national data on the availability of specific government datasets produced by the World Wide Web Foundation, this study examines if and under what conditions this new digital transparency impacts the levels of perceived and actual corruption. As from a principal-agent perspective, greater access to information has the potential to address an information asymmetry between the government (agent) and voters (principals), and as a result, discourage corrupt behaviour. The study also examines the effects of the publication of individual government datasets that may be crucial for the public in monitoring the use of public resources, for instance, data on spending, contracts, company registration data and land data. Second, drawing from semi-structured qualitative interviews with civil society representatives and investigative journalists, the study identifies barriers that prevent the public from benefiting fully from this new digital transparency.
The results from quantitative cross-country comparison suggest that the availability of open data poorly explains the actual levels of corruption but matters significantly for the perception of thereof. Therefore, the government’s strategists are probably wise in supporting open data agenda if they seek a public image of being accountable. Another prominent finding is the conditionality of the effects of open data on the levels of perceived and actual corruption and their strength upon the presence of free media. These findings are consistent with previous research on more traditional transparency measures, such as freedom of information legislation. They have important policy implications. In particular, digital transparency reforms will not yield greater government accountability and diminish corruption unless robust provisions safeguarding media freedom complement them. The results from qualitative interviews also confirm that the role of intermediaries in interpreting open government data is crucial, as cumulative disadvantages (social, economic, digital) make open government data inaccessible to the broader public.
