Research Briefs in Economic Policy No. 46

The Evolution of Culture and Institutions: Evidence from the Kuba Kingdom

By Sara Lowes, Nathan Nunn, James A. Robinson, & Jonathan Weigel
March 2, 2016

Increasing evidence suggests that both culture and institutions are important for economic development. Institutions are typically defined as the external “rules of the game” that shape individuals’ expected material payoffs from different actions. Culture, by contrast, is often defined as the collection of internal values and beliefs of individuals in a society. While culture and institutions are often treated as separate, they plausibly interact. Our work considers the effects that institutions might have on culture: institutions that incentivize people to take particular actions may also generate cultural norms that lead individuals to want to choose these same actions. For example, institutions incentivize certain behaviors that, over time, come to be viewed as “right” such that individuals may experience disutility when they deviate from these behaviors.

We consider one possible relation between institutions and culture by testing the impact of state formation on individuals’ propensity to follow rules and obey laws. We do this by exploiting a unique historical episode in Central Africa: the creation of the Kuba Kingdom in the 17th century.

During the medieval period, about two centuries prior to the formation of the kingdom, a large migration of related Mongo peoples settled near the confluence of the Kasai and Sankuru rivers. These groups, according to oral histories, are descendants of a mythical ancestor named Woot. Thus, prior to the formation of the kingdom, societies in the region were a culturally similar, homogeneous group of peoples who had recently migrated from the northwest. The kingdom was formed when Shyaam, an institutional entrepreneur and an outsider, united a group of villages and small chieftaincies under the rule of the Bushong clan.

Geographical features of the area had a decisive impact on the boundaries of the Kuba Kingdom: the region where the “children of Woot” settled is naturally divided by the Kasai and Sankuru rivers. Shyaam established the kingdom to the south of the Sankuru and to the east of the Kasai. During his reign and those of subsequent kings, the kingdom expanded to the natural limits defined by these rivers. The boundaries of the Kuba Kingdom remained stable and clearly defined by the Kasai river (to the west), Sankuru river (to the north), and Lulua river (to the south). Although the specific location of these natural boundaries was otherwise unimportant, the rivers determined which villages were inside and outside of the Kuba Kingdom, implying that establishment of the kingdom to the east of the Kasai, rather than to the west, occurred for idiosyncratic reasons.

The Kuba Kingdom had many characteristics that are associated with modern states, making the estimated effects of the Kuba state on cultural norms of general interest. The kingdom had a capital city where the king and members of numerous executive councils lived. Additionally, it had a professional bureaucracy, an unwritten constitution, a sophisticated legal system that featured trial by jury and courts of appeal, a professional police force, a military, a system of taxation, and extensive public goods provision.

Thus the idiosyncratic origins of the Kuba Kingdom, amidst the culturally homogeneous Mongo peoples, provides a compelling natural experiment that we use to estimate the impact of state formation on rule-following. While the villages on the Kuba side of the rivers that formed the kingdom’s boundaries were unified into a centralized state, the villages on the other side of these rivers—including a group called the Lele who are also descended from Woot—continued to exist in small chieftaincies. The fact that the Kuba state approximated a legitimate bureaucratized state with the rule of law makes it an attractive context in which to study our hypotheses of interest.

Our analysis estimates the effect of originating from the Kuba Kingdom on the propensity to obey rules even when there is a monetary incentive not to do so. We do this by conducting two behavioral experiments with current residents of Kananga, the largest city in the region. The first is the resource allocation game. In the experiment, there is a cost to following the rules and a benefit to cheating. Although it is impossible for us or anyone else to know if any specific experimental participant cheats, we are able to measure whether groups of participants cheat by observing the frequency of outcomes in a larger sample. The second experiment is a version of the standard ultimatum game, in which participants physically allocate money in a private setting (rather than play on a computer). This provides an opportunity for participants to steal money during the course of the experiment.

We compare the experimental behavior of individuals whose ancestors lived within the Kuba Kingdom with that of individuals whose ancestors lived just outside the kingdom. The sample is taken from the local provincial capital, Kananga. Thus, individuals in our sample no longer live in their ancestral villages; they or their ancestors migrated to the provincial capital. Examining this population has several benefits. First, it is logistically easier to work in one city, rather than a number of villages in a region in which the transportation and communication infrastructure is extremely poor. Second, considering a population of immigrants holds current institutions constant and allows us to better estimate the impact of historical institutions on cultural norms, which migrants take with them when they move to the city.

We consider three samples of interest. The first is the largest and compares all individuals whose ancestors lived inside versus just outside the Kuba Kingdom. The second compares the central Kuba (the Children of Woot who were inside the Kuba Kingdom) to the Lele, who lived just on the other side of the kingdom’s borders. And the third focuses on the core peoples of the Kuba Kingdom, the Bushong, and compares them to the Lele.

According to our experimental measures of rule-following, a consistent and robust set of empirical results emerges. We find that Kuba ancestry—that is, growing up under a state—is associated with more rule breaking, more theft, and more cheating. This is true for both experiments and for all three samples of interest, most notably when we compare the central Kuba to the Lele and the Bushong to the Lele. These findings demonstrate that culture can be shaped by state institutions. They also show that culture and state institutions are not necessarily complements. Our findings are consistent with the Kuba state crowding out internal norms of rule-following.

These findings are also consistent with some existing evidence showing that socially desirable laws and regulations (as well as monetary incentives) can crowd out beneficial norms. Although this remains an open field of inquiry, a number of studies find that the presence of laws and institutions—relative to their absence—can at times undermine intrinsic motivation. In other words, in certain settings, institutions can crowd out cultural norms and intrinsic motivations.


This research brief is based on Sara Lowes, Nathan Nunn, James A. Robinson, and Jonathan Weigel, “The Evolution of Culture and Institutions: Evidence from the Kuba Kingdom,” National Bureau of Economic Research Working Paper no. 21798, December 2015.

Read the Full Research Brief in Economic Policy

Sara Lowes, Harvard University; Nathan Nunn, Harvard University and National Bureau of Economic Research; James A. Robinson, University of Chicago and National Bureau of Economic Research; and Jonathan Weigel, Harvard University.