What is the evidence for a link between institutions and economic growth across countries and over time? This paper, prepared for the 1997 World Development Report, reviews the economic literature that tries to link quantitative measures of institutions with growth of gross domestic product. It concludes that the evidence does suggest a link between the quality of institutions and investment and growth, but that the evidence is by no means robust.
North provides an institutional framework that can be understood with the economic literature to provide a rich interpretation of the effects of institutions on growth. The institutional framework affects growth because it is integral to the amount spent on both the costs of transactions and the costs of transformation in the production process. Institutions may be weaker in some countries because rules are simply absent, suboptimal or poorly enforced.
The way that empirical institutional measures are categorised is important for interpreting their effects. It is important to distinguish between measures that describe the attributes of institutions and those that evaluate their performance. Descriptions of features of institutions reveal nothing about how they perform. Another important distinction is between the performance of institutions and measures of social characteristics and political instability. Informal constraints are difficult to measure but some measures of social capital capture the extent of civic activity and associations, which in turn imply the existence of social constraints that may influence growth.
Correlations in growth models can suggest a relationship, but not the direction of causality. Also, adjustments are required to growth models to control two problems in particular:
- Endogeneity: Institutions are considered ‘endogenous’ variables in growth models, as their value is partly determined within the structural system of the model. Few researchers take endogeneity seriously. The problem can be reduced by measuring institutional quality at the beginning of the period being studied.
- Ordinality: Some indexes rank countries on a given criterion without specifying the degree of difference between countries, simply giving each country a ranking number. This can give misleading results.
- Earlier data for institutional quality are positively correlated with subsequent growth. There are interesting positive correlations between institutional quality and social capital, and negative correlations with various social characteristics and political instability.
- There are weak positive correlations between social capital and growth. Increased ethnic diversity is negatively and significantly correlated with institutional quality, and has a negative but insignificant impact on growth. Real growth in 1980-89 is negatively correlated with earlier measures of political instability.
The evidence for causation in the literature on growth may be compromised by deficiencies in institutional measures, as well as by poor quality data for other variables. Notable findings include:
- Weak institutions may affect growth by affecting investment efficiency, and there could also be an indirect effect on growth from a decline in investment. The literature has mostly used reduced form models, where the direct and indirect effects of institutions are difficult to disentangle. Solow-type models are more useful.
- Some studies find that political characteristics and stability are important determinants of the quality of economic institutions.
- Evidence suggests that quality of institutions has a robust and significant relationship to growth via its effect on the volume of investment. There is also weak evidence for a direct relationship to growth.
- Recent work is paying more attention to methodological difficulties, with more thorough specification, more attention to endogeneity and more guarded policy conclusions.
- Institutionalists emphasise that institutional change is a long, slow process that requires time to evolve.
