Working Paper: NBER ID: w19995
Authors: Sebastian Galiani; Iván Torre; Gustavo Torrens
Abstract: We exploit three natural experiments in Argentina in order to study the role of legislative malapportionment on the biased federal tax sharing scheme prevalent in the country. We do not find support to attribute it to legislative malapportionment during periods when democratic governments were in place; nor did we find any evidence that the tax sharing distribution pattern became less biased under centralized military governments. We argue that these results are attributable to two of Argentina's institutional characteristics: first, the predominance of the executive branch over the legislature; and, second, the lack of any significant difference in the pattern of geographic representation in the executive branch under democratic and autocratic governments. Thus, the observed biases in the distribution of tax revenues among the Argentine provinces are not caused by legislative malapportionment, but are instead the result of a more structural political equilibrium that transcends the geographic distribution of legislative representation and even the nature of the political regime. Our findings also illustrate the importance of informal institutions and the interactions between formal and informal institutions. Legislative malapportionment could have an innocuous influence in countries with a strong executive branch but could become a key factor in countries with an empowered Congress.
Keywords: Fiscal Federalism; Legislative Malapportionment; Natural Experiments; Argentina
JEL Codes: D72; D78; H3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Legislative malapportionment (D72) | Distribution of federal tax revenues (H29) |
| Centralized military governments (H56) | Tax sharing distribution pattern (H22) |
| Legislative malapportionment (D72) | Biases in tax revenue distribution (H20) |
| Structural institutional characteristics (D02) | Biases in tax revenue distribution (H20) |