Intervention(s)
Corruption is – like most illegal transactions – hard to observe and quantify systematically. Given that the ‘true’ level of corruption is unobservable, corruption perceptions are widely used as a substitute in research and policy discussions. Perceptions have been shown to be biased measures of corruption (Olken 2009, Gutmann et al. 2020) but they matter in their own right. Domestic corruption perceptions may sway the outcome of elections while international corruption perceptions may, in the case of Ukraine, affect willingness to supply aid, foreign direct investment, reconstruction funding and the prospects of EU accession.
Over the past decade, Ukraine has implemented a range of reforms intended to combat corruption, including the creation of new, independent agencies tasked with anti-corruption enforcement. One consequence of their activities has been a steady stream of news about the discovery and investigation of corruption cases. The publication of news about prosecutions and convictions is an important component of law enforcement because it is a prerequisite for deterrence. News about corruption cases may, however, also affect corruption perceptions. Individuals who learn about a new corruption case may focus on the details of the crime and their perception of corruption in the country may worsen. Others may focus on the fact that it has been detected and is being prosecuted, causing their perceptions to improve. Our study aims to evaluate these responses empirically and in particular, to measure the net effect of news about enforcement on corruption perceptions.
Information about corruption cases can be delivered in different forms. Individuals may be most frequently exposed to news stories about a single corruption case. Such news reports may have a different effect from aggregate statistical information about the total number of prosecutions or convictions. For example, news stories about individual cases may encourage a focus on the crime while statistical information may be viewed through the lens of enforcement. Our study includes treatment arms that provide both kinds of information in order to evaluate whether the nature of the news affects the response.