It is widely recognized that politicians deliberately allocate goods and services just prior to the election, and road investments are arguably among the most visible infrastructure to influence voters. Using a comprehensive dataset on Italian municipalities over the period 2010-2015, we test whether investments in roads and transport services are affected by political manipulations close to elections using as independent variables the year-in-term dummies. We exploit the staggered time of local election to show, indeed, that investment spending on road and transport in the year before election is 30% higher than in the electoral year. Further analyses suggest that our results are more marked (i) in cities guided by a mayor who can run for re-election and (ii) in municipalities with a lower share of educated voters. We isolated the portion of the (exogenous) correlation between the probability of observing an accident and the amount of expenditure on road services that is induced by the political cycle by using the year-in-the-term dummies as instruments. We did not detect any relationship between the increase of investments in road services induced by the political cycle and the local need for road safety, as the probability of having an accident in local roads remained unchanged. Taken together, these findings suggest that politicians manipulate the budget only for re-electoral purposes. Therefore, it is needed a rule, binding visible expenditures, such as those on road services, of the year before the election, or allowing visible expenditures not to exceed those of the previous year within the mandate of the mayor. Such rules would let avoid or at least reduce the estimated inefficient spending by properly programming investment according to real needs and not to electoral convenience.
|