Social networks near and far: The role of bonding and bridging social capital for assets of the rural poor

• Author: Tewodaj Mogues
• Published date: 01 February 2019
• DOI: http://doi.org/10.1111/rode.12529
• Date: 01 February 2019

REGULAR ARTICLE

Social networks near and far: The role of bonding

and bridging social capital for assets of the rural

poor

Tewodaj Mogues

International Food Policy Research

Institute (IFPRI), Washington, DC

Correspondence

Tewodaj Mogues, International Food

Policy Research Institute (IFPRI), 2033 K

St NW, Washington DC 20006, United

States.

Email: t.mogues@cgiar.org

Abstract

With formal insurance and credit markets either absent or

inaccessible to rural agents in most poor rural economies,

social networks play a highly important role in mitigating

the risks that agricultural households face. These kinds of

informal insurance schemes are presumed to be most

effective in the face of idiosyncratic risk. However, social

mechanisms also exist in developing countries that may

reduce locally correlated risk such as the adverse eco-

nomic effects of climatic conditions that affect multiple

residents in a village. This paper analyzes the role of

localized (bonding) and of spatially dispersed (bridging)

social capital in mitigating the impact of idiosyncratic and

of locally correlated shocks on farm households’livestock

endowments. Using dynamic panel generalized method of

moments (GMM) system estimation with seven‐period

panel dataset of over 400 households, we find that bond-

ing social capital is able to protect households’livestock

assets against idiosyncratic shocks, but bridging social

capital does not play a role in mitigating the impact of

correlated shocks. The results hold up to multiple robust-

ness checks. A test of different hypotheses about the nat-

ure of these assets’trajectories rejects the asset poverty

trap hypothesis, and instead finds that livestock asset

dynamics are characterized by a single stable equilibrium.

DOI: 10.1111/rode.12529

Rev Dev Econ. 2019;23:189–210. wileyonlinelibrary.com/journal/rode © 2018 John Wiley & Sons Ltd

|

189

1

|

INTRODUCTION

In most poor rural economies, social networks play a highly important role in mitigating the risks

that agricultural households face. With formal insurance and credit markets either absent or inac-

cessible to rural agents, the ties forged through common social and cultural institutions among

members of a kinship group, ethnic group, or village enable households to transcend some of the

information problems barring the development of impersonal markets. An often complex system of

social exchange is an integral part of rural households’ex ante risk reduction and ex post coping

strategies. Social networks thus have an important place in resource allocation and risk manage-

ment, and in that sense can be treated as an economically valuable asset.

Informal insurance schemes are presumed to be most effective in the face of idiosyncra tic risk,

such as illness or death of a household member, theft of livestock or other assets, etc.

1

However,

social mechanisms also exist in developing countries that may reduce locally correlated risk such

as the adverse economic effects of climatic conditions or social conflict that affect multiple resi-

dents in a village. This issue has been explored primarily in the migration literature, for example

by addressing the extent to which remittances from household members who migrated can partially

insure origin households from the effects of correlated shocks. Besides this body of wor k, the bal-

ance of the social capital literature presumes that social networks are necessarily localized; they

may well frequently be, given the obvious problems of fostering and maintaining ties over large

distances, but the existence of social institutions or relationships that traverse space, and their use-

fulness in controlling the effects of covariate shocks for their members, is an empirical question

that continues to merit attention.

This constitutes the primary agenda of this paper. Given the potential importance of individual

and locally correlated shocks for asset dynamics of the poor, it analyzes the role of different types

of social capital in mitigating the impact of each of these types of shocks on asset endowments

and trajectories. In order to determine how social networks bear not only on assets, but also on

asset dynamics, we first develop a model of what these dynamics may look like—specifically,

whether or not asset dynamics are characterized by the existence of multiple equilibria and asset

poverty traps—and empirically test different hypotheses about the nature of asset dynamics. Based

on this, the results on the effects of social capital and shocks are used to simulate their effect on

asset dynamics. The use of a seven‐period panel dataset of over 400 households allows a fine‐

grained examination of the evolution of household assets over time, and especially enables the

application of a dynamic estimation model using the generalized method of moments (GMM)

approach to the empirical questions at hand.

This paper's contribution lies in conducting analysis that integrates important strands of the

social‐networks’literature in development, and of the literature on poverty traps, to address the

research questions laid out above. In so doing, we distinguish both between local bonding social

capital (measured as households’engagement in and contribution to informal local social institu-

tions) and bridging social capital (measured as households’interactions with social ties who are

geographically dispersed, and their consideration to migrate away), as well as between idiosyn-

cratic shocks (crop loss owing to drought suffered by the household) and locally correlated shocks

(the share of households in the respondent's village that suffered crop loss owing to drought). We

find that bonding social capital is able to protect households’livestock assets against idiosyncratic

shocks, but bridging social capital does not play a role in mitigating the impact of correlated

shocks.

The next section briefly discusses existing empirical and theoretical work relating to the mecha -

nisms of social networks in the presence of shocks, and the implications of theories of asset

190

|

MOGUES