Posted by Francois Michel
In a recent paper published in Public Choice, Konstantinos Angelopoulos, Apostolis Philippopoulos and Efthymios Tsionas provide a welcome addition to the empirical growth literature by enlarging the traditional focus on fiscal size to introduce measures of public sector efficiency. Two such measures of public sector efficiency are used.
The first one follows the “output-to-input” (or, as it might be more appropriately denominated, outcome-to-input) approach used by Afonso, Schuknecht, and Tanzi in two famous papers ("Public sector efficiency: an international comparison," Public Choice Volume 123, Number 3-4, June 2005; and "Public Sector Efficiency: evidence from new EU members states and emerging markets," ECB Working Papers number 581, January 2006) governements’ socio-economic outcomes are related to resources used, proxied by public sector spending, for major functions—limited by data availability, in the article’s case, to four basic roles of administration, education, infrastructures, and economic stabilization. The sample consists of 64 developed and non developed countries during four five-year periods between 1980 and 2000.
The second measure of government efficiency to which the model is applied is obtained by applying a stochastic production frontier approach as discussed in Lovell and Kumbhakar’s book (2000). The sample consists here of yearly data between 1995 and 2000 for 52 countries.
Despite important data limitations, most of which stem from the “coarseness” of efficiency scores, the authors find evidence that the size-efficiency mix—namely its country- and time-period product in the growth regression model—is critical for understanding the medium-term implications of fiscal policy. As the authors conclude, what matters for growth is not only how much resources should be spend, but where and how those resources should be spent.
This represents another evidence of the major progress made in recent years on understanding the efficiency of the public sector and calculating efficiency indexes. To a certain extent, similar advances have been achieved in the domain of corruption to calculate transparency and governance indexes, as I discussed in a recent post on the subject.
More advances should also be expected in the near future from the development of standardized assessment tools—such as the PEFA framework for public financial management—that will allow investigating relationships between efficiency and structural reforms, but also from the development of performance budgeting, which is expected to clarify and strengthen the link between public sector spending and related outcomes. The already large body of literature available on performance budgeting and performance indicators will certainly help bridging these conceptual gaps.
