Disentangling the Economic Effects of Political Institutions

This project, Disentangling the Economic Effects of Political Institutions (DEEPI), has studied whether and how particular political institutional characteristics -- for example concerning the ruling party, election systems, and the protection of civil liberties -- affect economic policies and development outcomes. More specifically, the project analyzed how different institutions, both in democracies and dictatorships, affect economic growth and redistributive policies, thereby addressing issues of immense importance for the welfare of citizens across the world. To arrive at empirically well-founded answers to these questions, data with extensive coverage across space and time are needed. The project therefore engaged in the collection of historical data for polities across the world, notably including the Historical Varieties of Democracy dataset on hundredsof detailed institutional characteristics going back to 1789. The project also contributed to the Social Policies around the World dataset, also going back to the 19th century, and covering features of different social policy programs. The project further engaged in the development of theoretical arguments on why and how different institutions may matter for growth and redistributive policies, and followed up with empirical analysis and systematic testing of these arguments by drawing on the above-mentioned and other data sources. Examples of research questions that have been addressed are whether particular leader-selection institutions in autocracies affect growth, and whether electoral system characteristics in democracies or ruling party characteristics in autocracies affect what social groups are covered by redistributive policy programs. Yet, growth and redistributive policies are, likely, interrelated, and they also have "feedback effects" on how political institutions evolve in the longer run. The project therefore put an emphasis on studying the dynamics of these relationships, and how such phenomena reciprocally affect each other. Key examples of findings from the project include (but are not restricted to) the following seven, whereof the first four pertain to economic growth, and the final three to redistributive policies: First, focusing on the provision of particular civil liberties such as freedom of expression, travel and association, we displayed how these may improve long-run economic development through improving the diffusion of ideas and technologies into and throughout the economy. Second, we showed that it is less likely that democracy increases economic growth through improving so-called "human capital", as democracy does not systematically relate to education quality. Third, we showed that strong. Institutionalized parties have a clear, positive effect on economic growth, both in autocracies and in democracies. Fourth, we documented how the sequence of historical-institutional development related to obtaining key democratic institutions versus building the state bureaucracy is not systematically related to subsequent growth Fifth, regarding redistribution and redistributive policies, we found no robust link between democracy, on the one hand, and redistribution or overall income inequality, on the other. Yet, there were indications in the data that certain features of democracy reduce income inequalities measured between groups of capital owners and wage workers. Sixth, we found that autocracies are no less likely than democracies to provide different social policy programs, notably including nation-wide old-age pension programs. However, pension programs in democracies tend to cover more social groups than programs in autocracies. Seventh, within democracies, however, the type of electoral system in place is very important for the introduction of major social policy programs, according to our results. Majoritarian electoral systems have historically seemed to hinder the development of encompassing welfare states, though only in societies where rural interest groups are relatively strong. In sum, this project, DEEPI, has contributed with both data collection and studies of political regimes and their effects, which have shed new light on what specific institutions affect economic growth and distributive policies, and which ones that have no effects.

Hva angår virkninger har DEEPI bidratt til å høyne synligheten og fremme karrieren til prosjektleder, samt fremme karrierene til en rekke yngre forskere. Hva angår effekter av DEEPI, kan nevnes at datasettene samlet under DEEPI har fått svært mye oppmerksomhet i fagmiljøet. Disse bidrar dermed indirekte til videre kunnskapsutvikling, på samme måte som studiene publisert i DEEPI siteres og benyttes videre av andre forskere. Hva angår videre effekter, har prosjektet -- via flere formidlingstiltak -- hevet offentlig bevissthet rundt hvordan demokratiske institusjoner fungerer, og dette kan ha ringvirkninger på det politiske klimaet. Videre har prosjektet formidlet direkte mot ulike beslutningstakere i Norge og utenlands. Dersom disse formidlingstiltakene har påvirket politikkutformingen til bistandsaktører, regjeringer, eller andre aktører, kan det potensielt ha influert demokratisk styring og økonomisk utvikling. Hvorvidt dette er tilfelle, er imidlertid vanskelig å si.

DEEPI will study the impact of regime components and particular institutions on economic growth and redistributive policies. Political institutions are widely thought to matter; yet, it remains unclear which exact institutional structures matter, what outcomes they matter for, and which background factors condition these relationships. This knowledge gap stems, in part, from the lack of data tracking detailed institutional features over time. DEEPI will contribute to the Historical Varieties of Democracy dataset with app. 280 indicators coded back to 1800 for all states and some semi-autonomous polities, and expand the Social Policies around the World dataset on eligibility and redistributive potential of different social policy programs (from 19th century). These datasets, with long time series, provide unprecedented opportunities to identify effects of particular institutions, which are often slow-moving and highly correlated. DEEPI will make important contributions by developing precise arguments on how specific institutions incentivize politicians and others to pursue actions with consequences for growth and redistribution, and properly test implications from these arguments. DEEPI divides into three work packages (WPs). The first disentangles effects on growth; it will e.g. analyze whether the participation-, contestation- or civil liberties component of democracy matters more for growth, and whether particular leader-selection institutions in autocracies affect growth. The second analyzes how regime components and specific institutions, e.g. authoritarian party characteristics, affect features of redistributive policies, including what social groups are covered. Yet, growth and redistributive policies are, likely, interrelated, and have feedback effects on institutional design. The third WP will thus link the areas and study dynamics and long-run effects of institutional changes through simulation models incorporating the complexities of these relationships