Chiefs Again
Thursday, October 3, 2013
Daron Acemoglu and James Robinson

In our early posts in February 2012, we discussed the economic institutions that keep Sierra Leone poor, and pointed out that many of these are not just national but also local. These village-level institutions are dominated by chiefs, who raise taxes, hire the local police, deal with conflicts and control the land.

We have been working on these local institutions and chiefs in Sierra Leone with Tristan Reed. The final version of our paper on this topic is here and the online appendix can be downloaded here.

Our strategy in this paper is to use the colonial organization of the chieftaincy in Sierra Leone to investigate whether unconstrained power of chiefs holds back economic development.

In 1896 the British empowered paramount chiefs as the main, perhaps only, authority in rural Sierra Leone. Chiefs, and the sub-chiefs and headmen, continued to retain their power to this day. British also developed a system in which paramount chiefs are elected for life by a ‘Tribal Authority’ made up of local notables. But importantly, only those from ‘ruling families’ of a chieftaincy designated by the British can compete to become paramount chiefs.

At one fell swoop, the British thus created the rural elite of Sierra Leone that has persisted to this day. But also useful for our identification strategy, there is considerable, and mostly historically idiosyncratic, variation across chieftains in the number of ruling families. And the number of ruling families matters greatly. In chiefdoms with few ruling families, chiefs and their ruling family can act more despotically, with much less constrained power. In chiefdoms with many ruling families, there is fierce competition and much more coalition-building when the time comes for the election of a new chief, and we document that chiefs or ruling families that ask despotically tend to be kicked out of power (but of course this is slow since chiefs are in it for life and good alternatives are not always abundant).

This institutional structure suggests that political competition will be greater in chiefdoms with more ruling families and economic institutions less extractive.

This is exactly what we find. In chiefdoms with fewer ruling families, villagers do indeed have less secure property rights. Moreover, they are less educated, have lower literacy, are more likely to work in agriculture, have less wealth and lower quality housing, and their children are less healthy. The effects are quantitatively sizable. For example, moving from the bottom quartile to the top (from 1.8 ruling families to 7.7) is associated with a seven percentage point increase in literacy.

All this seems to confirm our expectations that, regardless of its exact form, political competition tends to put constraints on how the elites can use their power and how extractive economic institutions can become.

But we also did find some very puzzling patterns.

We have access to information on civic participation, social capital and information on villagers’ respect for the authority of chiefs and the chiefdom institution. Inspired by Putnam and his collaborators’ work in Making Democracy Work, one might have expected that where chiefs have less constraints on their power, there should be less civic participation, less social capital and less respect for their authority.

We find exactly the opposite pattern. In chiefdoms with fewer ruling families, there is more social capital of both the “bridging” kind (linking elites to non-elites such as community meetings) and of the “bonding” type (where non-elites participate more intensively in civic activities and collective action). Perhaps even more surprisingly, when they have less constrained power, chiefs appear to be more highly respected.

So what’s going on?

We suggest a simple explanation: a bridge can be crossed in either direction — that is, bridging social capital can be used as a vehicle to assert social control by the elites rather than as an instrument for constraining elites as Putnam suggested.

If this view is correct, powerful chiefs may not just distort the allocation of resources to education or discourage t the development of the non-agricultural sector. In order to enhance their control over society, they may also need to monitor it and bring people together so as to tell them what to do. And the more they do this, the more deep-rooted their control over every aspect of village life becomes.

Though we have no direct evidence proving that this is the right explanation for our at first puzzling findings, our interpretation is consistent with the anthropological literature. For example, William Murphy, in his important work on politics in rural Sierra Leone, emphasizes that committee meetings in Sierra Leone are often used as a form of social control, and are used by elites to construct the appearance of governance based on community consensus. He states:

public forms [of discourse] are often recognized as an illusion masking alternative commitments arranged in secret. … [A] key attribute of the mature person or a successful group is the ability to strategically construct … public appearances.

We believe that what we have uncovered in rural Sierra Leone may be a much more general pattern. It’s similar to what Lungisile Ntsebeza describes in Democracy Compromised: Chiefs and the Politics of the Land in South Africa as:

traditional authorities derive their authority from their control of the land allocation process, rather than their popularity amongst their subjects … the need for land … compelled rural residents willy-nilly to cooperate with the traditional authorities.

Or as succinctly put in a related study by Jesse Ribot:

legitimacy follows power.

It may not be confined to sub-Saharan Africa either. For example, work by Siwan Anderson, Patrick François and Ashok Kotwal finds that in parts of western India where landownership is dominated by elites, development outcomes are worse, but measured social capital is higher. They also suggest that this might be related to the capture of social society by elites.

It’s too early to say, but this pattern may be the rule and Putnam’s the European exception.

If it is, some recent trends in foreign aid delivery might need rethinking. For example, The World Bank is investing heavily in Community Driven Development schemes based on the idea that these will tap into the independent power of civil society in less developed countries, especially in rural areas. But if civil society is captured by chiefs or elites, then efforts to strengthen it without freeing it from the control of traditional elites might just strengthen the power of chiefs.

Alas, there may be no silver bullets in kickstarting institutional change in societies where extractive institutions have deep roots.

Article originally appeared on Why Nations Fail by Daron Acemoglu and James Robinson (http://whynationsfail.com/).
See website for complete article licensing information.