Land rental markets and rural poverty dynamics in Northern Ethiopia: Panel data evidence using survival models

• Author: Stein Holden,Hosaena Ghebru
• Published date: 01 February 2019
• Date: 01 February 2019
• DOI: http://doi.org/10.1111/rode.12548

REGULAR ARTICLE

Land rental markets and rural poverty dynamics in

Northern Ethiopia: Panel data evidence using

survival models

Hosaena Ghebru

1

|

Stein Holden

2

1

International Food Policy Research

Institute, Washington, DC, USA

2

Norwegian University of Life Sciences,

Ås, Norway

Correspondence

Hosaena Ghebru, International Food

Policy Research Institute, Development

Strategy and Governance Division, 1201

Eye Street, NW, Washington, DC 20005.

Email: hosaena.ghebru@cgiar.org

Funding information

The United States Agency for

International Development (USAID); The

CGIAR Research Program on Policies,

Institutions, and Markets, led by IFPRI.;

The Centre for Land Tenure Studies and

the School of Economics and Business at

Norwegian University of Life Sciences

Abstract

There is growing interest in understanding the links

between land reforms, land markets, and poverty reduc-

tion in Africa. The study uses four‐wave panel data from

the northern highlands of Ethiopia to assess the dynamics

of rural poverty taking into account the status of partici-

pation of rural households in the land rental market.

Applying both nonparametric (Kaplan–Meier estimator)

and semi‐parametric survival models that control for dura-

tion dependence of poverty transition, results show partic-

ipation and degree of participation on the supply side of

the tenancy market (landlords) have highly significant and

positive effect on the chances of escaping poverty while

the same cannot be said about the demand side of the

tenancy market (tenants). The empirical evidence also

confirms that households headed by older and literate

people have relatively larger exit rates from poverty as

compared with households headed by younger and illiter-

ate ones. Though transacting farmers may engage them-

selves in win–win rental arrangements by the time they

join the tenancy market, results indicate that gains are

unequal as those tenants who enter the markets from low

economic leverage (were poor) are liable to face a lower

margin of net gains, which may limit their ability to move

out of poverty.

DOI: 10.1111/rode.12548

Rev Dev Econ. 2019;23:131–154. wileyonlinelibrary.com/journal/rode © 2018 John Wiley & Sons Ltd

|

131

1

|

INTRODUCTION

Throughout the Sub‐Saharan Africa, access to rural land and its potential in reducing rural poverty

has long been a subject of high policy agenda and research interest (Haggblade & Hazell, 1988;

Holden, Otsuka, & Place, 2008a,b; Warriner, 1969). The increasing trend in the number of Afri-

cans living in poverty has made the focus of governments, development partners, and the

researcher communities towards piloting “pro‐poor”land policies (Cotula, Toulmin, & Hesse,

2004; Holden et al., 2008). The slow progress in redistributive and other land tenure reforms (land

rights formalization/regularization) has encouraged the consideration of land market transactions as

viable options (Adams, 2004) and caused major multilateral agencies like the World Bank to

rethink the role even temporary access via land (rental) markets can play in agricultural growth

and poverty reduction (Deininger & Binswanger, 1999).

The fact that land rental markets have relatively lower transaction costs than land sales markets

(Chamberlin & Ricker‐Gilbert, 2016; Deininger & Binswanger, 1999; Jin & Jayne, 2013;Sadoulet,

Murgai, & De Janvry, 2001; Zimmerman & Carter, 1999) has given way to recognition of the crit-

ical role these markets can play as a means for providing the poor with access to land. Rightly so,

since the turn of the millennium, land policies and legal reforms to liberalize and formalize land

rental markets have been top policy agenda in many countries of Africa South of the Sahara

(Cotula et al., 2004; Holden et al., 2008).

There is a large body of empirical literature on the welfare implications of land access and land

distribution in Africa (Carter & May, 1999;Haggblade & Hazell, 1988; Jayne, Zulu, Kajoba, & Weber,

2008; Jayne et al. , 2003; Karugia, Oluoch‐Kosura, Nyikal, Odumbe, & Marenya, 2006; Rigg, 2006), but

relativelyfew studies on the poverty‐reducing impacts of land rental markets. Despite the e arlier skeptical

view on the performance of land markets (Binswanger, Deininger, & Feder, 1995; Carter, 1988; Lipton,

1993; Sjaastad, 2003), empirical studies from Rwanda (Andre & Platteau, 1998; Blarel, 2004), Ghana

(Migot‐Adholla, Benneh, Place, & Atsu, 1994; Otsuka, Quisumbing, Payongayong, & Aidoo, 2003),

Kenya (Jin & Jayne, 2013; Muraoka, Jin, & Jayne, 2018; Yamano, Place, Nyangena, Wanjiku, &

Otsuka, 2008), Ethiopia (Deininger, Ali, & Alemu, 2008; Ghebru & Holden, 2008; Holden & Bezabih,

2008; Kassie & Holden, 2008), and Malawi (Holden, Kaarhus, & Lunduka, 2006; Lunduka, Holden, &

Øygard, 2008) show that land rental markets contribute to more equitable operational holding between

the poor and the rich. In contrast with these, studies from Ethiopia (Bezabih & Holden, 2006; Holden &

Ghebru, 2006; Kassie & Holden, 2006), Malawi and Zambia (Chamberlin & Ricker‐Gilbert, 2016),

Madagascar (Bellmare, 2006), Tunisia (Laffont & Matoussi, 1995) and Eritrea (Tikabo & Holden, 2003)

reveal land rental markets role in transferring land from relatively poorer landlords to wealthier tenants.

What is empirically common with these studies (with the exception of Chamberlin & Ricker‐Gilbert,

2016) is that they make deductions about the poverty‐reducing effects of land rental markets by

analyzing the allocative efficiency and equity implic ations of such markets.

While it is true that allocative efficiency and equity implications of land rental markets can be

implicative to suggest the poverty‐reducing potential, the story remains incomplete unless nonland

resource endowments are also considered. Hence, it is not always true to conceive the landless

(land‐constrained) as the poorest (Chamberlin & Ricker‐Gilbert, 2016). The rural landless in Africa

(mostly young and inexperienced in farming) may find alternative sources of income, either from

agricultural labor and/or employment in the rural nonfarm economy. Particularly, in countries with

an egalitarian land ownership distribution (like, Ethiopia), the poorest members of society can be

those with few capital or nonland assets and constrained access to micro‐finance credit. In other

words, the conventional approach of using own land holding for characterization of poverty may

not only be less informative but also misleading. See (Chamberlin & Ricker‐Gilbert, 2016).

132

|

GHEBRU AND HOLDEN