Horizon problem and capital expenditures: evidence from the public sector in Indonesia

Purpose - This study investigates whether horizon problems affect the allocation of capital budgets and their implementation in a government setting. Design/methodology/approach - We use data from 2005 to 2020 for local governments in Indonesia, which apply a limit of two five-year terms for mayors. We use regression analyses for panel data with total observations of 4,541 local government years from 448 unique local governments. We also use graphical analyses andt-tests to provide robustness to our results. Findings - Mayors allocated lower capital expenditures in the second term than in the first. Capital budget allocation is lower for local governments whose mayors are older than 60. Our additional analysis shows that incumbents seeking re-election allocate more capital expenditure than those not seeking re-election. Research limitations/implications - This study contributes to the literature on the behavioral effect of term limits on local government's allocation and implementation of capital budgets. Limiting elected government officials to a certain number of terms will prevent the monopoly of power. However, it may negatively affect budget allocation on capital programs in their last term. Our findings should interest public policymakers in discerning the costs and benefits of term limits for elected offices. Originality/value - Most studies on horizon problems have focused on the corporate setting. This study provides evidence of the effects of horizon problems in the government setting, especially in Asia.