Where do Institutions Come From?

By Sergei Guriev (Sciences Po, Paris)

This essay is written for sioe.org. It is not intended to provide a survey of the relevant academic literature; rather, it is a point of view on one of the key pillars of the Society’s research agenda.

The debate on the role of institutions has changed dramatically since ISNIE was established in 1996. Research on institutions has become part of economics mainstream. Qualitative and descriptive studies have been complemented by increasingly rigorous quantitative work applying modern and sophisticated econometric methods. Academics in general and ISNIE members in particular, have learnt a lot, not only about correlations but also about causal relationships. There are still many open questions: we do not have a consensus on measuring institutions, on quantitative importance of different kinds of institutions for growth and development and on the factors that may drive the heterogeneity of the impact of institutions. Yet, there is no longer any doubt that institutions are important for economic performance and social development. This leads to a key challenge for our Society – as we are certain that institutions matter we should also try to understand where the institutions come from. Given the importance of this question for policymaking, it must be addressed using research methods of highest academic quality, whether they are qualitative or quantitative studies.

The question of where institutions come from is certainly not new. We know that formal institutions are shaped by political choices. We also know that there are informal institutions, cultural and social norms that evolve through social interactions and changes in attitudes. Therefore the analysis of the formal institutions and that of the informal ones are both important. These two research agendas are not mutually exclusive, in fact, they are complementary. Informal institutions influence the effectiveness of formal ones – and therefore their determinants. Simultaneously, formal institutions provide threat points for informal institutions and thus affect their evolution as well.

This implies that the analysis of origins of institutions has to be multi-disciplinary. Institutional economics has to learn from political science, legal studies, and from sociology, psychology and anthropology. In order to study the emergence and dissemination of attitudes and beliefs we also need to learn from the recent developments in computer science that looks at the information flows within networks and formation of various kinds of networks. Finally, as institutions are persistent and change slowly, we necessarily need to learn from history.

Institutional economics' eternal problem has been that of measurement. By now, we have a number of approaches to measure formal institutions. These are imperfect and yet, recent studies show that they are meaningful – although we do have to recognize that their bite depends on the extent and effectiveness of the informal institutions. There is much less agreement on how to measure informal institutions. Should we rely on experiments? What is the external validity of experiments in one country? In one city? Shall we rely on surveys of self-reported attitudes and intentions? What is the relationship between self-reported behavior and actual behavior?

Recent technological developments allow major breakthroughs in both running much larger-scale experiments (e.g. those online) and in observing very large datasets on revealed preferences (such as users’ behavior in online commerce, social networks and search engines). This may create an impression that Big Data will necessarily turn institutional economics into a highly quantitative field. The move to becoming more quantitative and rigorous will indeed happen – and is certainly welcome. However, this does not mean that theoretical and qualitative studies will become irrelevant. It is exactly the unmanageable scale of “Big Data” that makes it crucial to have a conceptual framework in place before crunching the numbers.

Along with high quality of research and multidisciplinarity, the third defining feature of ISNIE/SIOE is its international nature. This is also going to stay. Emergence of some institutions can be better tracked and studied at the subnational level (e.g. at the level of US states, cities or counties). These studies are very useful, in particular, because they allow comparing institutions within a single polity and within the same culture. But the most important formal and informal institutions are at the national level – in particular, the formal institutions of the nation states and informal institutions intrinsic to national culture. Therefore institutional economics will have to study different countries. And, in order to understand institutional and cultural contexts, the scholars will always rely on local knowledge. This will be especially important for carrying out qualitative studies and linking their results with those of the quantitative ones.

While institutional economics has made great progress in the last two decades, its research agenda still includes many exciting open questions. Our Society, continuing to stick to its three key principles of high academic quality, interdisciplinarity and international approach, is well positioned to contribute to addressing them.