ECONOMIC GROWTH CENTERYALE UNIVERSITYP.O. Box 208629New Haven, CT 06520-8269http://www.econ.yale.edu/~egcenter/CENTER DISCUSSION PAPER NO. 928Institutions and Development:A View from BelowRohini PandeYale UniversityandChristopher UdryYale UniversityNovember 2005Notes: Center Discussion Papers are preliminary materials circulated to stimulate discussionsand critical comments. We have greatly benefitted from discussions with Daron Acemoglu, Tim Besley andJames Robinson. We are grateful to our discussant Orazio Attanasio for comments. Finally, we thank James Fenske and Pinar Keskin for fantastic research assistance.This paper can be downloaded without charge from the Social Science Research Networkelectronic library at: http://ssrn.com/abstract=864044An index to papers in the Economic Growth Center Discussion Paper Series is located at:http://www.econ.yale.edu/~egcenter/research.htmInstitutions and Development: A View from BelowRohini Pande and Christopher UdryAbstractIn this paper we argue the case for greater exploitation of synergies between research on specificinstitutions based on micro-data and the big questions posed by the institutions and growth literature.To date, the macroeconomic literature on institutions and growth has largely relied on cross-countryregression evidence. This has provided compelling evidence for a causal link between a cluster of‘good’ institutions and more rapid long run growth. However, an inability to disentangle the effectsof specific institutional channels on growth or to understand the impact of institutional change ongrowth will limit further progress using a cross-country empirical strategy. We suggest two researchprograms based on micro-data that have significant potential. The first uses policy-induced variationin specific institutions within countries to understand how these institutions influence economicactivity. The second exploits the fact that the incentives provided by a given institutional contextoften vary with individuals’ economic and political status. This can help us better understand howinstitutional change arises in response to changing economic and demographic pressures.JEL codes: O11, O12, O17, P51Keywords: Institutions, Growth, Cross-Country RegressionsInstitutions and Development: A View from BelowRohini Pande and Christopher UdryYale University∗November 18, 2005AbstractIn this paper we argue the case for greater exploitation of synergies between research onspecific institutions based on micro-data and the big questions posed by the institutions andgrowth literature. To date, the macroeconomic literature on institutions and growth haslargely relied on cross-country regression evidence. This has provided compelling evidencefor a causal link between a cluster of ‘good’ institutions and more rapid long run growth.However, an inability to disentangle the effects of specific institutional channels on growth orto understand the impact of institutional change on growth will limit further progress usinga cross-country empirical strategy. We suggest two research programs based on micro-datathat have significant potential. The first uses policy-induced variation in specific institutionswithin countries to understand how these institutions influence economic activity. The secondexploits the fact that the incentives provided by a given institutional context often varywith individuals’ economic and political status. This can help us better understand howinstitutional change arises in response to changing economic and demographic pressures.1 IntroductionRecent years have seen a remarkable and exciting revival of interest in the empirical analysis ofhow a broad set of institutions affects growth. The focus of the recent outpouring of research∗We have greatly benefitted from discussions with Daron Acemoglu, Tim Besley and James Robinson. We aregrateful to our discussant Orazio Attanasio for comments. Finally, we thank James Fenske and Pinar Keskin forfantastic research assistance.1is on exploiting cross-country variation in ‘institutional quality’ to identify whether a causaleffect runs from institutions to growth. These papers conclude that institutional quality is asignificant determinant of a country’s growth performance.These findings are of fundamental importance for development economists and policy prac-titioners in that they suggest that institutional quality may cause poor countries and peopleto stay poor. However, the economic interpretation and policy implications of these findingsdepends on understanding the specific channels through which institutions affect growth, andthe reasons for institutional change or the lack thereof. However, for reasons discussed b elow, weargue that the coarseness of cross-country data limits its usefulness for such research. Instead,we suggest that a more fruitful research agenda is to exploit the synergies b etween researchbased on micro-data and the questions posed by the institutions and growth literature.North (1981) defines an economic institution as “a set of rules, compliance procedures andmoral and ethical behavioral norms designed to constrain the behavior of individuals in theinterests of maximizing the wealth or utility of principals.” (p.201-202). We adopt his definitionof institutions as sets of rules, procedures or norms that constrain behavior but disagree withthe notion of agency embodied in this definition. Institutions need not be ‘designed’, and evenif they are, their actual operation may be quite different than intended. For this reason, weemphasize a research agenda on institutions which pays attention to de facto rather than dejure institutions and one that pays attention to how changes in resource endowments can causeindividuals to change their economic behavior within a given institutional context and potentiallycause the institution itself to change in the longer run. Such a focus is particularly relevant whenthinking about institutions in low income countries – since development, by definition, is aboutchange.In section 2 we summarize the main insights from the cross-country literature on institutionsand growth. This literature has successfully focussed attention on the complex interactions b e-tween economic growth and institutional development. It has uncovered important correlationsacross countries between growth and the nature and quality of a core set of economic, politi-cal and social institutions. This literature has also been