Red tape, bribery and government favouritism: evidence from Europe

Red tape has long been identified as a major cause of corruption, hence deregulation was advocated as an effective anticorruption tool, an advice which many country followed. However, we lack robust systematic evidence on whether deregulation actually lowers corruption. This is partially due to the difficulty of defining what is good regulation, but also to the lack of theoretical clarity about which type of corruption regulations impact on and to the deficient measurement of different types of corruption. In order to address the latter two gaps, we differentiate petty corruption from government favouritism and propose novel measurement of the latter by developing two objective proxy measures of favouritism in public procurement: single bidding in competitive markets and a composite score of tendering 'red flags'. Using publicly available official electronic records of over 2.5 million government contracts in 27 EU member states and two European Economic Area countries in 2009-2014, we directly operationalize a common definition of favouritism: unjustified restriction of access to public contracts to favour a certain bidder. Petty corruption is measured using business surveys while the extent of business regulation is measured by Doing Business expert assessment of precise regulatory costs. Using country-level panel regression analysis, we find that deregulation has a heterogeneous impact on both low and high level corruption. It is largely ineffective in tackling government favouritism, with business start-up deregulation even facilitating such corruption. Whereas deregulating the various channels through which governments and businesses interact (e.g. obtaining construction permits) often decreases the perception of bribery and petty corruption. Policy consequences are profound and point at a more targeted and context-dependent promotion of the deregulation agenda. Full public procurement database is available at http://digiwhist.eu/resources/data/.