Suchergebnisse

We investigate the effects of the Indonesian decentralization and democratization process on budget allocation at the sub-national level. Based on panel data for 271 Indonesian districts for the years of 1994 to 2009, we address the determinants of local investment expenditures in public infrastructure in the sectors of education, health, and physical infrastructure. We find that after the dramatic expenditure decentralization of 2001, districts with relatively lower levels of public infrastructure started to invest more in these sectors. In contrast to the marked budgeting changes following the fiscal and administrative decentralization, we find no consistent effects of the democratization process on local public investments. Our results reflect initial improvements in local targeting but show no evidence of increasing electoral accountability.

Analyzing law enforcement data on corruption incidents for a panel of 79 Russian regions for the period 2004-2007, we find that the relative salaries of bureaucrats determine corruption levels: Corruption declines as relative salaries rise up to a turning point, beyond which corruption rises again. Other important determinants are the strength of law enforcement, available rents through government budgets and natural resources, education levels, unemployment rates, and income inequality.

We analyze the determinants of the excessive administrative spending of local governments in Indonesia. In an unbalanced panel data set of 399 districts for 2001-2009, we show that the proliferation of districts has not led to increased administrative spending; instead a lack of political accountability is responsible for this excess. The degree of political competition influences the level of administrative spending significantly; newly introduced direct elections of district heads, however, did not curtail the waste.

We study political business cycles (PBC) in newly democratized and decentralized Indonesia at the local level. As the timing of indirect and later on direct elections of district heads differs exogenously across jurisdictions, the political business cycle effect is clearly identified . Our data allow distinguishing PBCs for indirect and direct elections. We find significant budget cycles for discretionary budget categories at the disposal of the district head (not for the overall budget) only for direct elections, not for indirect elections. These cycles are much larger if the incumbent runs for reelection. Our results underline the conditional nature of PBC.