Luiz on “What is governance?”
John Luiz of the University of Cape Town responds to Francis Fukuyama’s “What is governance?”, published by Governance this week:
I welcome this important contribution to the broader discussion of governance and development, and indeed the use of institutional measures in general within the social sciences. In his commentary Pollitt quite rightly reflects that “Fukuyama does not ask the fundamental question of why we would wish to measure governance in the first place?” This is not a matter of semantics as any critique will depend upon the reason why we want these measures. Potential problems of endogeneity loom large in some of these measures depending upon how they are utilized.
An industry has developed within economics and political science using these mass generated institutional measures for cross-sectional statistical analysis often without much regard for what is actually being measured or reflecting on the methodology of how these measures have been developed. In a forthcoming paper in Governance, Constructing Institutional Measures: Indicators of Political and Property Rights in Mozambique, 1900–2005, we construct a series of indices for a single country over a protracted period of time using historical records. We argue that it provides us with a richness of institutional data previously not available and allows us to explore the dynamics of development over time: “This will allow us to examine directions of association and feedback loops. Sifting through the political history of individual countries is a valuable exercise in itself.”
What I particularly appreciate in the Fukuyama piece is the focus on state capacity, a much under-looked institutional measure. State capacity is important irrespective of one’s political ideology and where one stands on the state versus market debate. A strong, capable state is needed, be it to direct economic processes directly, or to enforce the rules of the game which make markets work. An immediate concern which arises from within this discussion is the apparent (implied) expectation that state capacity is merely an economic issue which we need to guard against as it is also a sociological and political issue (this coming from an economist). The early literature on the developmental state highlighted the importance of socio-political and administrative factors in determining state capacity. In particular, it emphasized the role of the elite in the development process, the necessity of a competent and insulated economic bureaucracy, and the significance of ‘embedded autonomy’ for the state.
Turning to the measures themselves there are obvious problems. To highlight a few:
The use of tax extractive capacity is problematic for a host of reasons. Firstly, it captures a very limited dimension of state capacity. Secondly, there is the danger of the more is better phenomenon which is clearly an issue with this measure. Thirdly, the actual evidence points to its limitations. Two countries which stand out in the literature as possessing significant levels of state capacity are South Korea and Sweden with polar opposite results on the tax extractive measure. Likewise, South African tax revenue as a proportion of GDP has been rising of late and few would see this as a measure of enhanced state capacity (other than measuring the improved capacity of a particular department of revenue collection).
I would argue that a better measure is the efficiency with which state resources are employed and this has the added advantage of possessing both an input and output dimension. For example, using production functions we can measure the efficiency with which inputs are translated into outputs. So more revenue is not necessarily better and indeed it is possible that tax revenue declines whilst capacity, as measured by efficiencies, improves.
One last point is that the discussion applies mostly to developed countries. In developing countries one would need to account for more obvious measures of state capacity including corruption, service level protests, ‘kangaroo’ courts, rising crime levels and drug cartels, smuggling, and the use of private security firms for protection.
Despite the issues that I raise, the Fukuyama piece provides a useful basis with which to continue an important debate.
John Luiz is a professor of business administration at the University of Cape Town. He is co-author, with Luis Brites Pereira and Guilherme Oliveira, of “Constructing Institutional Measures: Indicators of Political and Property Rights in Mozambique, 1900-2005,” Governance Early View.