The most exciting time for a researcher is when a project comes together and there are actual outputs. So it is with our ‘Islands of Integrity‘ research project…
The research – which has been done with Dr Caryn Peiffer, Dr Rosita Armytage and (in South Africa) Prof Trevor Budhram – looks at anti-bribery positive outliers, or cases where bribery has (unexpectedly) reduced relative to other sectors in a country where the governance environment is challenging: the Uganda health sector and the South African police. The research uses a novel three-stage methodology that allowed us to uncover ‘hidden’ outliers, and the 2 papers are for the country case study papers where we tested the methodology out to see if it does indeed throw up genuine positive outliers (spoiler alert: it does…).
This case study is part of a research project that used a novel mixed-methods approach to identify potentially unrecognised instances of development progress – specifically, of bribery reduction. Analysis of Transparency International’s Global Corruption Barometer flagged Uganda’s health sector as a potential ‘positive outlier’ because its bribery rate had halved between 2010 and 2015, which did not fit bribery patterns in other sectors of the country. Qualitative fieldwork then examined whether and how bribery had been reduced in Uganda’s health sector. It highlights the work of a relatively newly established Health Monitoring Unit (HMU). The HMU’s corruption control strategy exemplifies a ‘principal-agent theory-inspired approach’. Principal-agent theory suggests that it is possible to control corruption when potentially corrupt actors (agents) are made to be fearful of their respective principals holding them to account for engaging in corruption. Principal-agent theory-inspired anticorruption efforts have attracted criticism recently for being largely ineffective. Nevertheless, this study finds that requests for bribes for health services had decreased as a likely consequence of the HMU’s efforts, especially its high-profile publicised raids, which made health workers more fearful of being caught. However, apparent unintended consequences mean that this may not be a simple success story. Principal-agent theory anticorruption approaches that target frontline service workers may bring unintended consequences that outweigh the benefits of bribery reduction. There is evidence that the HMU’s activities may be harming morale among health workers and undermining citizens’ trust in the sector. Moreover, the HMU’s approach has not sought to address bribery’s function as a mechanism through which health workers supplement extraordinarily low wages. The HMU’s successes in reducing bribery may therefore be difficult to sustain.
This case study examines police-related bribery in Limpopo province, South Africa. It asks why such bribery reduced by almost 15% between 2011 and 2015, while the rate in the rest of the country reduced by on average less than 4%. This bribery reduction in Limpopo’s police took place during the time that the national government led an unprecedented high-level anticorruption intervention in several provinces, but a specific intervention in Limpopo, unrelated to the police, impacted bribe levels dramatically in a relatively short period of time. The research suggests the police in Limpopo may have been especially reluctant to engage in bribery then because of uncertainty as to whether they were also under investigation and because of the heightened anticorruption action. The bribery reduction in the police was likely an unanticipated ‘benign side effect’ of a separate intervention. This case shows that certain types of disruption may reduce bribery patterns, but only for a relatively short time. For longer-term impact, disruption strategies likely need to be continuously inventive and re-inventive, and/or to be driven by strong leadership that can and will continually disrupt corruption patterns. The research also highlights how new insights can emerge from anticorruption research that focuses on specific forms of corruption in a specific place, sector and time. Characteristics specific to the police sector—such as a high degree of discretion, peer solidarity and regular contact with criminals—can make fighting entrenched corruption particularly difficult. While the long-term sustainability of the impact of this unusual intervention is questionable, the case teaches a wider lesson: a disruptive event can counteract sector-specific factors that likely enable entrenched patterns of corruption, and this can happen more quickly than expected.