In 2004 and 2005, Bulgaria was unexpectedly hit by several waves of torrential rain that caused extensive flooding. To deal with destruction sustained during the floods, the central government awarded nearly 67 million euros to 257 flood-stricken municipalities (see the figure below). For many of Bulgaria’s impoverished municipalities, the awarded funds represented a substantial financial injection; mayors and local councils were the final authority on how this windfall would be spent. Not surprisingly for the country ranked second for corruption in the EU, the media uncovered multiple instances when local politicians pocketed the money instead of using it for disaster relief. According to estimates by opposition parties, 59 million euros disappeared into the pockets of firms related to the ruling political coalition.
In a new working paper, we use this episode to examine how intra-governmental transfers affect local corruption, using a unique data set on the amount and use of disaster aid distributed to local governments. What was not known at the time of disbursement was that an independent national auditing watchdog, the Bulgarian National Audit Agency, would be summoned in 2006 to issue detailed reports on how the money was used for 227 out of 257 municipalities receiving assistance (comprising 96.8 percent of disbursed funds). The resulting public reports chronicle a variety of infringements including: (1) contracts not awarded to the highest bidder or no bidding, (2) money channeled for the repair of buildings experiencing no damage, and (3) money given for no work. The figure below shows how our corruption measure varies within Bulgaria.
According to our estimates, a 10 percent rise in per capita flood-related relief funds increased corruption by 12.2 percent in the average municipality, which is a sizable effect. The interesting thing is that we observe this result in municipalities controlled by the ruling coalition as well as in those affiliated with the opposition.
What are the mechanisms through which this effect operates? Suggestive evidence implies that corruption had little effect on the reelection chances of the incumbent mayor, partly because mayors in high-corruption municipalities were less likely to run for reelection. We also find that this effect is stronger in municipalities in which the incumbent had previously won by a narrow margin. More tentatively, we also show that the link between funds and corruption varies with municipal characteristics. The negative impact of flood assistance on governance is more pronounced in municipalities that are more developed economically and that have a weaker media. All in all, our suggestive evidence on mechanisms indicates that accountability informs the decision to steal: mayors engage in corrupt behavior when the prize of stealing more outweighs the benefit of running for office again, or when they feel they can get away with it.
Our results imply that potentially beneficial transfers such as disaster relief or aid from international organizations may unintentionally cause governance in weak democracies to deteriorate. Although our analysis is focused on a single country and covers a relatively narrow time period, our findings are relevant for a much broader set of developing and transition countries in which financial windfalls, whether derived from natural resources or not, are particularly high. Many countries, Bulgaria included, re-distribute public funds, such as foreign aid or budgetary transfers, to the local level. Continuing such transfers without strengthening mechanisms of control, such as the judiciary or creating independent budgetary watch-dogs, can create and deepen tendencies to engage in fraud. More broadly, our results also relate to the policy debate among international financial institutions as to whether weather-related insurance should be provided publicly or by the market. While the Eastern European experience demonstrates that privatization is far from a panacea, this work warns that local governments may be tempted to pocket at least some of the disaster relief, thus stalling vital reconstruction efforts.