Rotten Apples or Systemic Failures?Assessing Risk Factors for Public Corruption

Abstract

The main purpose of this research is to develop an understanding of how some elements of the demographic, political, and economic context, as well as the institutional design of public entities, can explain variation in public corruption. In other words, the study intends to give new evidence on some factors that can affect the levels of corruption besides anti-corruption mechanisms or cultural aspects. The dissertation is focused on the study of Spain, a country that does not have a high level of corruption from a global perspective but that stands out in this matter in relation to other European countries, which makes it a relevant case to explore why relatively high levels of corruption can persist even in democratic and developed countries. The thesis is centred on the use of an objective measure of corruption based on the analysis of final judgements for corruption felonies. The court cases are thoroughly analysed to build a dataset with detailed information on the corrupt activities, the convicted person, the administration where the actions happened, and the judicial sentence. Thus, the dissertation is one of the first research works on corruption in Spain using a large-N design based on the study of judicial condemnations. The thesis is organized into three main chapters that focus on two kinds of public administrations that appear as salient considering their levels of corruption: local administrations, and public companies and foundations. Each of the chapters combines the use of different statistical techniques to get general conclusions on the relevance of the studied factors to explain corruption. The results are not conclusive in relation to the potential impact on local corruption of the population size and the concentration of political power in one party but, considering the measure of corruption used in the research, they suggest a more nuanced relationship between these variables. Interestingly, the results point to the fact that smaller municipalities and those where the same party governs for long periods have more difficulties to make corruption be prosecuted and condemned. In consequence, we cannot definitively state if there is more corruption in those kinds of cities and towns but the empirical data seems to show that public integrity controls fail in those contexts. The study also provides some evidence suggesting that the dependence of city councils on the revenues coming from the construction industry and real estate sector, likewise the relevance of tourism in a city, are factors that could increase corruption risk. Regarding public companies and public foundations, the results indicate that, indeed, corruption is more prevalent in these kinds of public entities than in the rest of the public sector. The empirical data shows that this fact could be, at least in part, related to a lower level of effectiveness of the public activity control mechanisms in place in these entities in relation to the ones in other public bodies. The findings of this dissertation can make several contributions to the research on corruption. From a methodological approach, they show the advantages of objective measures based on court cases that make us aware of some tendencies of corruption that do not appear in survey or experience-based methodologies. In the substantive aspect, this work extends our knowledge of some contextual and institutional factors that can explain variation in public corruption. This evidence can have relevant implications for public integrity policymaking since, on the one hand, it shows the need for some administrative reforms to overcome or mitigate the factors facilitating corruption and, on the other hand, it points to some kinds of public entities that should receive special attention in anti-corruption controls with a risk assessment strategy.