copy the linklink copied!2.7. Revenues and expenditures structure by level of government
In general, sub-central governments are responsible for delivering services to the population. Their ability to collect revenues depends on the degree of fiscal federalism. In countries where sub-central levels have limited ability to collect taxes, their main source of revenues are central government transfers, which are mostly earmarked in the central budget.
Recent research has found that decentralising spending and revenue collection to the same extent contributes to economic growth and spending efficiency, especially in economies that are highly integrated to global markets (Kim and Dougherty, 2018). LAC countries, however, have traditionally suffered from an imbalance between resource-generation and spending at sub-central levels of government, rendering local and state levels dependent on transfers from the central government (Cibils and Ter-Minassian, 2015). As a result, these have limited autonomy in their public finances.
Among LAC countries with available data in 2017, the central level generated 69.7% of the revenues, the state level contributed 19.1% and the local government 10.4%. This shows a higher decentralisation than in 2007, when the central level collected 70.7% of the revenues, the state 21.2% and the local level 7.5%. Similarly, in OECD countries, the state level collected 19.2% of the revenues and the local level contributed 10.2% in 2017.
There are differences in decentralisation of revenue collection among LAC countries. In Chile, in 2017, the central government collected 91.9%, and the local government only 8.1%. Regional governments in Chile are “mixed entities” (both decentralised and deconcentrated) and the only self-governed entities are municipalities (which manage the communes). However, these are highly dependent on transfers. Hence, they serve mainly as public service providers (OECD, 2017). On the contrary, in Brazil, a federal country, the central government contributed 65.5%, the state level 24.0% and the local 10.4%.
In LAC, spending is more decentralised than revenue collection. In 2017, local levels spent, on average, 18% of total expenditures. By contrast, the central level spent 62.7% and the state level 18.5%. The imbalance is more apparent in Mexico, where the central government spent 38.6% and the state level 35.0%, although the former collected 80.5% of the revenues and the latter only 7.2%.
Similarly, in Peru, the central government collected 82.6% of the revenues, and spent 53.3% of the total expenditures. The state level collected 0.7% of the revenues and spent 19.1%, and the local level raised 4.5% and spent 15.9%. Such an imbalance is due to the dependence on extractive industries (whose revenues are shared), the lack of sources of revenue at the regional level (e.g. specific taxes) and the lack of autonomy in defining municipal taxes (Cibils and Ter-Minassian, 2015).
Data are from the IMF Government Finance Statistics (IMF GFS) database, which applies the concepts set out in the Government Finance Statistics Manual (GFSM). The GFSM provides a comprehensive conceptual and accounting framework suitable for analysing and evaluating fiscal policy. It is harmonised with the other macroeconomic statistical frameworks, such as the System of National Accounts (SNA). Some differences exist between the GFS and the SNA frameworks in several instances, which led to the establishment, to a large extent, of correspondence criteria between the two statistical systems. The GFS and SNA frameworks have been recently revised and several statistical standards were implemented by the countries.
General government consists of central, state and local governments and social security funds. State government is applicable to the federal states of Brazil and Mexico and the highly decentralised countries of Colombia, Paraguay and Peru. For detailed information on the components of revenues and expenditures, see methodology and definitions’ paragraphs of Sections 2.4 and 2.6 respectively. Data across levels of government exclude transfers between levels of government in order to see the contribution of each sub-sector in general government total revenues/expenditures, which are at this level consolidated. For the OECD average, data are derived from the OECD National Accounts Statistics database, which is based on the SNA framework.
Further reading
Cibils, F. and Ter-Minassian, T. (eds.) (2015) Decentralizing Revenue in Latin America: Why and How, Inter-American Development Bank, Washington, DC.
Kim, J. and S. Dougherty (eds.) (2018), Fiscal Decentralisation and Inclusive Growth, OECD Fiscal Federalism Studies, OECD publishing, Paris/KIPF, Seoul. https://doi.org/10.1787/9789264302488-en.
OECD (2017), Making Decentralisation Work in Chile: Towards Stronger Municipalities, OECD Multi-level Governance Studies, OECD Publishing, Paris. https://doi.org/10.1787/9789264279049-en
Figure notes
Data for Mexico, Peru and Paraguay are recorded on a cash basis. Transfers between levels of government are excluded. Data for Mexico are not included in the LAC average. LAC and OECD averages are weighted.
Data for Honduras refer to 2015 rather than 2017 and data for Paraguay refer to 2016 rather than 2017; and for Colombia refer to 2008 rather than 2007.
Social security funds are included in central government for Brazil, Chile and Colombia.
Source: Data for the LAC countries: IMF Government Finance Statistics (IMF GFS) database. Data for the OECD average: OECD National Accounts Statistics (database).
Source: Data for the LAC countries: IMF Government Finance Statistics (IMF GFS) database. Data for the OECD average: OECD National Accounts Statistics (database).
Metadata, Legal and Rights
https://doi.org/10.1787/13130fbb-en
© OECD 2020
The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.
