Working Paper: NBER ID: w26150
Authors: M. Chatib Basri; Mayara Felix Rema Hanna; Benjamin A. Olken
Abstract: Developing countries collect a far lower share of GDP in taxes than richer countries. This paper asks whether changes in tax administration and tax rates can nevertheless raise substantial additional revenue – and if so, which approach is most effective. We study corporate taxation in Indonesia, where the government implemented two reforms that differentially affected firms. First, we show that increasing tax administration intensity by moving the top firms in each region into “Medium-Sized Taxpayer Offices,” with much higher staff-to-taxpayer ratios, more than doubled tax revenue from affected firms over six years, with increasing impacts over time. Second, using non-linear changes to the corporate income tax schedule, we estimate an elasticity of taxable income of 0.59, which implies that the revenue-maximizing rate is almost double the current rate. The increased revenue from improvements in tax administration is equivalent to raising the marginal corporate tax rate on affected firms by about 23 percentage points. We suggest one reason improved tax administration was so effective was that it flattened the relationship between firm size and enforcement, removing the additional “enforcement tax” on large firms. On net, our results suggest that improving tax administration can have significant returns for developing country governments.
Keywords: Corporate taxation; Tax administration; Tax rates; Indonesia
JEL Codes: H25; H26; O23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
Cause | Effect |
---|---|
tax rate change (H29) | reported taxable income change (H26) |
MTO introduction (Y20) | tax revenue increase (H29) |
MTO introduction (Y20) | higher reported gross incomes (D31) |
MTO introduction (Y20) | higher taxable incomes (H29) |
improved tax administration (H26) | more substantial revenue increases (H29) |
long-term effects of MTO (R48) | increasing tax revenue (H29) |