Land Tenure Regimes and Land Conservation in the African Drylands: the Case of northern Ghana

8/12/2019 Land Tenure Regimes and Land Conservation in the African Drylands: the Case of northern Ghana

1/22

This article was downloaded by: [197.251.171.52]On: 25 January 2014, At: 10:53Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Land Use SciencePublication details, including instructions for authors and

subscription information:

http://www.tandfonline.com/loi/tlus20

Land tenure regimes and land

conservation in the African drylands:

the case of northern GhanaEmmanuel Joseph Mensah

ab

aPolicy Fellow, Alliance for a Green Revolution in Africa (AGRA),Accra, GhanabExecutive Member, lOasis Development Group, Accra, Ghana

Accepted author version posted online: 06 Jan 2014.Published

online: 21 Jan 2014.

To cite this article:Emmanuel Joseph Mensah , Journal of Land Use Science (2014): Land tenure

regimes and land conservation in the African drylands: the case of northern Ghana, Journal of Land

Use Science, DOI: 10.1080/1747423X.2013.878765

To link to this article: http://dx.doi.org/10.1080/1747423X.2013.878765

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (theContent) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinions

and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,

systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
http://www.tandfonline.com/page/terms-and-conditionshttp://dx.doi.org/10.1080/1747423X.2013.878765http://www.tandfonline.com/action/showCitFormats?doi=10.1080/1747423X.2013.878765http://www.tandfonline.com/loi/tlus20http://www.tandfonline.com/page/terms-and-conditionshttp://www.tandfonline.com/page/terms-and-conditionshttp://dx.doi.org/10.1080/1747423X.2013.878765http://www.tandfonline.com/action/showCitFormats?doi=10.1080/1747423X.2013.878765http://www.tandfonline.com/loi/tlus20


2/22

Land tenure regimes and land conservation in the African drylands:the case of northern Ghana

Emmanuel Joseph Mensaha,b*

aPolicy Fellow, Alliance for a Green Revolution in Africa (AGRA), Accra, Ghana; bExecutiveMember, lOasis Development Group, Accra, Ghana

(Received 23 January 2013; final version received 17 December 2013)

In Africas drylands, threats to economic livelihood from climate change are exacer-bated by a low rate of investments in land conservation. The existing literatureidentifies land tenure insecurity as a major contributory factor. To help better under-stand this challenge, this paper investigated, empirically, the impact of land tenureregimes on investment in land conservation, focusing on northern Ghana. The study sfindings identified three dominant traditional institutions of land administration, viz.,tendana, chieftaincy and family. These institutions in turn define three broad categoriesof tenure regimes, namely, freehold, leasehold and lineage. It further finds that theseregimes exhibit a continuum of tenure security that impact differentially on investmentin land conservation. This is buttressed by the quantitative analysis, which reveals thatwhereas freehold (relative to lineage) significantly engenders investment in conserva-tion, leasehold shows no important impact. Among others, the paper recommends theinstitution of reform processes that promote durable and individualized land tenuresecurity.

Keywords:land tenure; land conservation; dryland; farm household; northern Ghana;Africa

1. Introduction

Contemporary adverse changes in global climate and land productivity continue to

threaten farm-based production and livelihood worldwide (UN WCED, 1987). As noted

by IFAD (2009), climate change and variability is unequivocal, accelerating and consti-

tutes a major threat to sustainable livelihood. In the drylands of Africa, this trend is

exacerbated by a relatively low rate of investment in land conservation in these regions.The existing literature hypothesizes land tenure insecurity as a major contributory factor.

In this paper, a quantitative assessment of the extent of impact of land tenure regimes on

household decision to make major investment in land conservation is provided. The aim is

to help enhance understanding of this challenge and distil the appropriate policy inter-

ventions necessary for stimulating investment in land conservation, thereby contributing

to existing knowledge and discourse on how derivatives of institutional arrangements such

as land tenure regimes influence household decisions on land productivity and resource

management.

Whereas the vulnerability of agro-ecological systems in arid and semi-arid regions of

Africa is attributable to the state of the environment and the increasing extremity in

weather patterns, investment in land conservation is endogenous and driven by factors

*Email:[email protected];[email protected]

Journal of Land Use Science, 2014

http://dx.doi.org/10.1080/1747423X.2013.878765

2014 Taylor & Francis
mailto:[email protected]:[email protected]:[email protected]:[email protected]


3/22

inherent in the behaviour and decisions of farm households. As discussed in Adams

(1974) and recounted by Goldstein and Udry (2006), key derivatives of institutional

arrangements that underlie access to and use of economic resources such as lands are

important in engendering their sustainable management.

In particular, land as an economic resource is important for agricultural production.

Investment in land conservation to help sustain their productive capacity is therefore a

non-trivial economic decision for all farm households. Like all economic actors, these

households require sufficient incentives to justify major investment in landholdings

(AGRA, 2012). Among such incentives is the extent and duration of control over the

given land resource. Indeed, such control, as defined by the prevailing land tenure regime,

provides the basis for capturing and internalizing the benefits arising from investment in a

given conservation practice (Deininger & Binswanger, 1999).

Hence, land tenure is hypothesized in the literature as critical for determining the rate

of investment in land conservation. However, empirical understanding of the structure of

such impact remains limited. Specifically, plural systems of land tenure administration in

most parts of Africa (including Ghana) tend to exhibit a continuum of user rights, whichranges from very strong to relatively weaker tenure security. To what extent then do such

different tenure regimes influence decisions on major investment in land conservation and

related productivity-enhancing practices? How could development policy properly opti-

mize such leverage to engender greater investments in land conservation, especially in one

of the worlds increasingly vulnerable agro-ecological zones Africas drylands (Deressa,

Hassan, & Ringler, 2009; Downing,1992; UNCCD, UNEP, & UNDP,2009)? These are

the core research questions underlining this paper.

The primary objective of the paper is therefore to establish the differential impact of

land tenure regimes on major investment in land conservation, focusing on the semi-arid

agro-ecological zone of Ghana. The specific objectives are to

(1) identify the dominant institutional arrangements for land administration in the

study area and the tenure regimes arising as key derivatives of these institu-

tions; and

(2) quantify the differential impact of land tenure regimes on the incidence of major

investment in land conservation by farm households in the study area.

Justification for this study is founded on several grounds. First, land rights and their

influence on agricultural production and conservation practices will remain an important

global development issue for the twenty-first century (Asabere, 1994; De Soto, 2000;

Maxwell & Wiebe, 1999). By focusing on investigating the nature of such nexus and

within the context of a typical semi-arid agro-ecological zone in Africa, this paper

responds to an important policy issue.

Second, for Africa generally and Ghana in particular, the strategic importance of

agriculture to sustainable development makes a better and thorough understanding of

the link between land rights and the adoption of land conservation and productivity-

enhancing practices extremely necessary. Also, the dominance of agriculture and the

espousal of the private sector as the engine of growth imply that policies that specifically

address bottlenecks to private investment in the sector needs to be urgently identified and

deployed. Obviously, this will involve policy initiatives that are sufficiently informed by

the results of studies of this nature.Third, some districts in the focal area already benefiting from major development

investments such as the Millennium Village Project1 and various national agricultural

2 E.J. Mensah


4/22

development programmes. With the strategic objective of increasing agricultural produc-

tion, productivity and competitiveness, the results of the study should prove useful and

timely.

Lastly, economic livelihoods in northern Ghana, like most of Africas drylands, are

identified to be systematically vulnerable to endemic poverty. GoG (2006) indicates that

overall poverty incidence in Ghana is highest in these regions, with relatively more fragile

environment for agro-based production. Thus, by exploring the extent of influence of land

tenure regimes, among other factors, on the adoption of major conservation measures by

landholders, the study contributes to knowledge on the subject while informing policy

strategies appropriate for promoting efficient land use and resource management in this

zone. This paper is therefore very relevant and timely.

The paper is organized into five sections. Section 2 provides a review of the focal

country, the literature on the nexus between land tenure and investment in land conserva-

tion and the study area. Section 3 discusses the analytical framework, methodology, data

source and definition of variables. Section 4 then discusses the results of the analyses,

with Section 5 concluding.

2. Literature review

This section provides a brief overview of the economic and policy background of the

focal country (Ghana), review of the existing literature on land tenure and land conserva-

tion with special reference to Ghana and an introductory note on the study area (northern

Ghana).

2.1. Ghana: a country backgroundGhana is a West African economy, located on the southern coast of the sub-region,

between latitude 444 N and 1111 N and longitudes 311 W and 111 E (FAO,

2006). A recent population and housing census in 2010 estimated the population of the

country at over 24.6 million, with an intercensal growth rate of 2.4% (GSS, 2012).

Economic development policy in Ghana since the post-independence era identifies pov-

erty as a major development problem requiring direct and methodical response. Towards

this end, economic policy in Ghana over that period has focused on accelerating growth in

income towards improved livelihood and significant reduction in poverty. Aside the

adoption of the global agenda of drastically reducing poverty and hunger by 2015 (as

specified in the United Nations Millennium Development Goals), major policy pro-

grammes have been implemented in more recent decades to specifically address poverty

and its associated challenges to the Ghanaian society.

Some of these programmes are marked out in the Human Development Strategy Plan

(NDPC, 1991) and the National Development Policy Framework (NDPC, 1994) of the

country. The latter formed the basis for the development of the various phases of the

Vision 2020 policy programme, which was implemented from 1995 to 2000 (NDPC,

2003). Succeeding these programmes has been the Interim Poverty Reduction Strategy

Paper (I-PRSP) and the Poverty Reduction Strategy Papers (PRSPs), the Millennium

Challenge Programme of the United States, the Ghana Shared Growth and Development

Agenda (GSGDA) and the Savanna Agricultural Development Programme.

Significantly, a key strategy cutting across all of these programmes has been thetargeting of agriculture as the primary sector to drive up income growth and thereby

improve living standards in the country. This is based on the view that an agro-based

Journal of Land Use Science 3


5/22

industrial development strategy, in the context of the countrys economy, has greater

potential to impact immensely on employment and wealth creation.

Ghanas agricultural sector has accounted for over 36% of national income and

employed more than two-thirds of the countrys labour force in the periods preceding

the export of crude oil in 2011. In terms of growth, the sector has grown at an average rate

of 6.1% since 2002, representing over 1 and 1.3 percentage points more than that reported

for industry and service sectors, respectively (BOG, 2008; CEPA,2007; ISSER, 2006).

Even more crucial is the fact that traditionally growth in the sector tends to have stronger

leverage on other sectors of the economy, especially on the basis of its contribution to

food security and raw material supply, foreign exchange earnings and savings, food prices

and improved household incomes. Furthermore, improved farm incomes create significant

demand for domestic manufactures and services, more so given the sheer size and

structure of demand of the rural sector.

However, agriculture in Ghana confronts major debilitating challenges that consider-

ably constrain the capacity of the sector to respond to the growing market opportunities

and policy initiatives. Dominant among these factors is the issue of land access and tenuresecurity.

Quite consistent with a sector characterized by the dominance of traditional production

practices, land constitutes a principal factor of production for the agricultural sector in

Ghana. In many instances, access to land alone determines to a large extent the poverty

status and degree of vulnerability of farm households to hunger. Beyond access however,

the right to use land and benefit from investments made in it constitute a defining factor

for production decisions and the very survival of most farm households. Underlining this

state is especially the trend in population growth and the ever-growing demand for land,

land productivity losses through degradation, overuse and extremities in precipitation and

other weather patterns.

2.2. Land rights and land conservation in Ghana

As noted by Goldstein and Udry (2006) in a study on land rights and agricultural

productivity in Akwapim (a farming community in southern Ghana), political influence

and social network significantly influence a households ability to exercise right to a piece

of land. Such rights then provide the incentive to invest in improving farm productivity

and income. The livelihood implication of this phenomenon is what the paper aptly

describes as the profits of power. This is very much consistent with the global view

expressed by Lappe (1998) thataround the world, the poorest of the poor are the landless

in rural areas, followed closely by the land-poor; those whose poor quality plots are too

small to support a family (FAO, 2005, p. 2). And, it is such global concerns that have

generated the intense policy debate and varied programmes on land tenure reforms in

most parts of the developing world, Africa not excluded.

For Ghana specifically, the debate on land has centred mainly on the plural2 land

tenure systems prevailing in the country and how they could be reformed and adminis-

tered to foster economic growth and development. For instance, theNational Land Policy

(GLAPS,2003) clearly articulates a general focus on stimulating economic development

through improving land markets, security of tenure and effective administration.

Nevertheless, an obvious trend in the policy debate seems to favour market-based

access and private entitlement to land. This is based on the notion that private ownershipof land, as expressed in well-documented and enforceable title, tends to engender greater

security of ownership and, in some cases, serve as useful certificate for accessing

4 E.J. Mensah


6/22

investable resources such as credit. Consequently, individuals with secured land title are

found to be better incentivized to invest in and optimally use existing land resources than

those without it (Cleaver & Schreiber, 1994; De Soto, 2000; Hitchcock,1981; Picardi &

Seifert, 1976).

A contrary perspective to the foregoing argument is provided by another strand of

literature led by Besley (1995). This literature posits that privatization of land cannot

constitute the primary solution to the complexities associated with land tenure problems in

Africa, and for that matter Ghana. Indeed, it argues that typical of most African societies,

goodwill and social bonds tend to provide sufficient protection for rights to land and

therefore the motivation to invest, not only for the individual concerned but for the

community and posterity. As illustrated by Besley (1995), since community-based risk

sharing forms an important feature of most African societies, it is quite outside expecta-

tion that the incentive to invest in land necessarily increase with formal rights to land as

against a communal (or the indigenous) tenure system. The paper further points out that

such formal right in land do not necessarily correspond with the rights that a farmer may

desire in planning to invest in that land.Additionally, beyond the matter of ensuring efficient land use administration, some

literature also express greater confidence in the ability of traditional institutions in readily

responding to local needs and the changing exogenous conditions that preserve efficient

use of land resources. This argument is best articulated in Dasgupta and Maler (1990).

Another dimension to the argument is the view that even where communal tenure

systems are alleged to have failed in providing sufficient protection to rights in land,

individuals have been able to enhance such rights by making long-term investments in the

land, such as tree planting. Bruce (1988), in a review of the African land tenure systems,

summarizes this view as follows:

Some observers have suggested that indigenous tenure systems pose a degree of insecuritythat destroys incentives to plant tree crops. This may be true in some cases, but the literatureis also replete with examples in which holders with temporary or fragile titles, havingsucceeded in planting trees, enhanced their tenure. The establishment of tree farms can bea critical step in the transition from shifting to stabilized cultivation, with trees definingpermanent holdings. Tree planting may initially be discouraged by insecurity of tenure, buttree planting can actually produce greater security of tenure and act as a way of claimingland. (Bruce, 1988, p. 41, in Besley, 1995)

In Uganda, Place and Otsuka (2002) also find that in enhancing tenure security farmers in

their study also tended to engage in coffee cultivation on the land.

Obviously, these varying perspectives provide important reflection of the conflicting

views associated with the nexus between land tenure systems, land use forms and

decisions regarding investments in land conservation by landholders. Empirical insights

to reinforce the fitting perspective, deepen evidence-based discourse and better inform

policy actions for efficient administration of land resources are sorely needed. For most

parts of Africa, the significance of this debate is already manifested by the increasing

vulnerability of food production systems to the rising trends in land degradation, soil

productivity losses and declining food security.

In fact, recent projections by Scherr and Yadav (1997) indicate that by the year 2020

land degradation may pose a serious threat to food production and rural livelihoods,

particularly in poor and densely populated areas like the arid and semi-arid regions of thedeveloping world. The paper further predicts that if accelerated erosion continues una-

bated, yield reductions by 2020 may reach 16.5%. Already, the AU Commission (2010)



7/22

observes that decades of nutrient mining has rendered farmlands in Africa the poorest in

the world. The Commission further estimates US$4 billion as the equivalent value of

nutrient lost on the continent annually.

In Ghana, the situation is even more pertinent in the semi-arid regions where pro-

ductivity losses of farmlands are accelerated by the increasing population pressure, limited

investment in external inputs like fertilizer, limited land cover and extensive exposure of

the top soils to rain and wind erosion, extreme rainfall patterns and high-temperature

conditions, obsolete land management know-how and poor land use practices. Some

notable land use practices that directly compromise the productivity and sustainable

management of these land resources include short-fallow shifting cultivation practices,

intensive farming on limited pieces of land, indiscriminate felling of trees for fuel wood

and charcoal production, bush burning and overgrazing. Certainly, these developments

pose significant challenge to the national agenda of agricultural transformation for

improved incomes and poverty reduction. At the household level, it also deepens the

vulnerability of households to hunger and chronic poverty.

2.3. Background to the study area: northern Ghana

The focal study area is the semi-arid agro-ecological zone of Ghana, which is made up

entirely of the landmasses of the Upper East, Upper West and Northern Regions of the

country. Together, these regions also constitute the northern corridor of Ghana.

On account of the distribution of the agricultural and climatic conditions in the

country, Ghana is organized into six agro-ecological zones. These zones exhibit different

attributes of vegetative cover, geological structure, natural resources and climatic condi-

tions. They are, from the north to south:

(1) Sudan Savannah

(2) Guinea Savannah

(3) Forest Savannah Transition

(4) Semi-Deciduous Rainforest

(5) Rainforest

(6) Coastal Savannah

The Sudan and Guinea Savannah agro-ecological zones of the country therefore constitute

the semi-arid (or the dryland) region of Ghana. Figure 1 is the map of the six agro-

ecological zones, showing the key administrative towns and cities as well as the range of

annual rainfall amounts for the respective zones. The zone covers a total area of about

97,702 km2 and an estimated population of 4.23 million, representing about 41% and

17.1% of the national landmass and population, respectively (GSS, 2012). This population

is dominated by two major ethnic groups. These are the Mole Dagbani and the Grunsi

ethnic groups.

The vegetative cover of this zone is characterized by the savannah grassland, thinly

interspersed with some thickets and drought-resistant trees such as the shea, dawadawa,

baobabs or acacias. The zone experiences unimodal rainfall distribution. This occurs

during the months of May to September. This period constitutes the wet or the principal

production season in this zone. The mean annual rainfall over the entire zone is

1000 mm, with a minimum of about 700 to a maximum of 1200 mm. The temperaturelevels vary significantly across the seasons, peaking at about 41C during the season

especially.

6 E.J. Mensah


8/22

3. Analytical framework

The evaluation of the extent of impact of land tenure regimes on farmer decision to

implement major land conservation measures has been the subject of both theoretical andempirical research. According to de Janvry and Sadoulet (2001), contentions over issues

of land access and user rights are now even more acute, driven especially by two principal

Figure 1. Map of the six agro-ecological zones of Ghana.

Source: www.ehs.cdu.edu.au/chromolaena/proceedings/third/3tim1b.html (Accessed 14 December2013).

http://www.ehs.cdu.edu.au/chromolaena/proceedings/third/3tim1b.htmlhttp://www.ehs.cdu.edu.au/chromolaena/proceedings/third/3tim1b.html


9/22

factors: the growth in human population and the rate of integration of markets for

agro-products.

In respect of population growth, this argument is bolstered by Boserup ( 1965). In

analysing the relationship between population growth and changes in agricultural produc-

tion systems, that paper contends that within traditional agrarian production systems the

scarcity of farmlands and the consequent reduction in fallow periods arising from popula-

tion explosion necessitates a shift from extensive to more intensive production systems.

Farmers thus compensate for the increased pressure on the available lands with the

deployment of more labour resources and investment in land conservation practices. By

this argument, Boserup contradicts Malthus (1998) that growth in human population

largely determines growth in agricultural production.

Considered within the context of land tenure and its relation to land use and produc-

tivity, Kabubo-Mariaria (2006) also notes that some studies provide further impetus to the

Boserup thesis by asserting that the increase in the relative price of food, arising especially

from increased demand, necessitates reforms in land tenure arrangements towards

improved incentives for increased investment in land conservation. Platteau (1996,2000) are cited as typical examples. Accordingly, arising from Boserups argument is

the consideration that rights in land could be endogenous; that is, derived on the basis of

previous investments in the land.3 From this conceptual premise, this paper proceeds with

the analysis of the factors influencing farmers decision to invest in land conservation by

evaluating the relationship between the expected right in land and previous investments in

conservation on the given land.

However, since the expected right in land is not directly observable but still dependent

on the prevailing tenure regime and the investment in conservation made on that land, a

relationship could be specified to measure the expected land right, following the metho-

dology of Otsuka, Quisumbing, Payongayong, and Aidoo (2003) and adapted by Kabubo-Mariara (2006) such that

Re 0 1X 2C (1)

whereRe refers to the expected land right;Xis a dummy that refers to the prevailing land

tenure regime;Crefers to investment in land conservation;0is the value of the land right

without any investment in conservation; 1is the marginal effect of the land tenure regime

on the expected land right; and 2 is the marginal effect of the incidence of investment in

land conservation on land right.

Following the conceptual arguments preceding Equation (1), 2 is hypothesized to bepositive,a priori; that is, as investment in land conservation increases, the expected right

increases,ceteris paribus. The relative effect ofX (i.e.1) however depends on the tenure

regime.

As noted earlier, a thesis spearheaded by the neoclassical school provides another

important basis for conceptualizing the relationship between land tenure and conservation.

From that perspective, the incentive to invest in land increases as the right to appropriate

and internalize the benefits of those investments rise (Adam, 1974). In the expression of

Uma (1992, p. 37), assured land rights ensure incentives to invest in land . Thus,

following the argument of Otsuka et al. (2003), a simplified linear expression of this

relationship could be specified such that

C 0 1Re (2)

8 E.J. Mensah


10/22

where 0 is the intercept term in the model and 1 is the marginal effect of the expected

right in land on the adoption of a land-conserving practice.

The dependent variable, C, is measured as a binary variable assuming the value 1

when a major land conservation practice is adopted, and 0 otherwise, while 1 is

hypothesized to be positive.

To make the two models tractable for empirical estimation, the unobserved variable Re

is eliminated by substitution such that

C 0 1 0 1X 2C (3)

Solving forC in Equation (3) yields a reduced form function, given as

C 0 1X (4)

where 0 010111

and

11=111

Equation (4) therefore constitutes the basic model of the study.

3.1. Methodology

Following standard practice in econometric analysis, the functional form of the model

above is estimated taking into account other factors that the existing literature identifies tohave potential effect on the dependent variable. Thus, at the level of a farm household i,

the model to be estimated is specified in matrix formulation as

Ci 0 1Xi 2Zi ei (5)

where Ci defines the adoption (incidence of investment) in land-conserving measure; Xidefines a vector of the prevailing tenure regimes, as a measure of right in the land; Zi is a

vector of factors other than the land tenure regime influencing the decision to adopt a

conservation practice by farm household i; S are the parameters (or coefficients) of the

model; and ei is the random error term of the model.

Factors represented in the vector Zi include the socioeconomic characteristics of the

household. These include the educational attainment of the head of the household,

household size and assets, the sex of household head, age and marital status of the

household head, proximity to major market centre, population density, land use form

and the extent of presence of public/private institutions in land conservation-related

interventions. As the availability and definition of these variables depend entirely on the

empirical data available, Section 3.3 provides extensive discussion of the definition and

construction of these variables.

In the empirical estimation of Equation (5), the binary adoption approach is employed.

Soule, Tegene, and Wiebe (2000), for example, have used this approach in modelling the

relationship between land tenure and adoption of conservation practices among a sampleof U.S. corn producers. The dependent variable in the model is assumed to be generated

by a latent variable,C, such that the incidence of a conservation measure on a farmers



11/22

field implies C is positive, generating the observation, C 1, in Equation (5).Otherwise, the conservation practice is not observed and thus C 0.

In Long (1997), the structural model is specified such that

C

i f0W

i 2

i i 1; :::; N (6)

where Wi is a vector of land tenure regime and other characteristics specific to the farm

householdi, f is a vector of coefficients while 2i is a random disturbance term. In effect,the relationship between Ci and Ci is stated as follows:

Ci 1 if Ci > 0;

0 if Ci 0:

(7)

and 2i ,NID 0; 2 .

As a result, for a farmer i, the latent variable Ci > 0 if the farmer adopts the

conservation practice on the field, which also implies Ci 1. The probability thatCi 1 is therefore given in Equation (8) as

PrCi 1 PrCi > 0

Prf0Wi 2i > 0

1 Ff0Wi

Ff0

Wi (8)

where Pr and F is the probability distribution and cumulative distribution functionsof the model, respectively. To ensure that 0 Ci 1, the cumulative distributionfunction F of the error term in the model is assumed to follow the logistic distributionin this study. For comparison purposes, the model will also be estimated under the

assumption of the cumulative standard normal distribution.

3.2. Source of data

Data for the analysis were drawn from a household sample survey carried out under the

ISSER/USAID4 project on Land Tenure and Land Policy Research in Ghana. The survey

was designed and conducted by ISSER. The overall survey covered 2690 households

nationwide. The data for the present analysis were therefore extracted from the national

database using the regional identification code for Upper West, Upper East and Northern

regions. As explained earlier, these regions constitute the northern corridor of Ghana and

makes up the semi-arid agro-ecological zone of the country. This sub-sample therefore

contained a total of 695 households, from 10 enumeration areas in 10 districts of the three

regions.

Key sections of the structured questionnaire included questions on household compo-

sition and characteristics, landholding, land use and land tenure regimes, evidence ofmajor investment in landholdings, land prices, land markets and institutional issues,

among others. Survey enumerators were trained with the aim of enabling them retain

10 E.J. Mensah


12/22

the same understanding of the concepts and questions as the researchers, communicate the

intent of each question to the respondents and elicit the appropriate responses.

As part of the overall project, key informant and focus group discussions were also

designed to generate additional information on land rights and land use practices, land

administration and other institutional issues. These data were organized and reported as

two other independent levels of analysis, namely, reports on statutory land institutions (or

institutional analysis) and key informant interviews (or community-level analysis).

However, as explained above, the data at the core of the present paper were based

primarily on the sub-sample data obtained exclusively from the quantitative survey. The

Statistical Package for the Social Sciences(SPSS) was used for processing the data, while

STATAwas used for the estimation of the models.

3.3. Definition of variables and hypotheses

Land conservation in the current analysis is measured as the incidence of major invest-

ment in conservation on the principal farmland of the household since the past 5 years.These measures are defined to mean those on-farm investments undertaken to help protect

or even enhance the productive capacity of farmlands, especially in relation to soil and

water conservation. Among such practices identified by the respondents are tree planting

and contour bands that are created to prevent soil erosion. Other practices such as ridges,

mounds, stone lines, basins rectangular mounds, contour stripping and the use of cover

crops are also reported to be in use in the area although these are noted to be short-term

conservation measures. Giving the long-term nature of investments in tree planting and

contour bands, widespread adoption of these measures is particularly constrained by land

tenure insecurity and limited resources.

Indeed, across the broad spectrum of respondents in the survey, the notion of majorinvestment in land conservation is well understood to be relative. That is, given the

differences in resource endowment among farmers, what constitutes a major outlay for

one farmer might be minor for the other. This is more so since the opportunity cost of

investing a specified amount of resources in land conservation is expected to be lower for

a resource-rich respondent, relative to her resource-poor counterpart, ceteris paribus.

Thus, for the purposes of the study, what constitutes a major investment in land

conservation is permitted to be self-evaluated by the farmholder during the interview.

Here, it is hypothesized that the stronger the security of tenure of a farmholder, the more

likely that such a holder will undertake what she evaluates (relative to her resource

endowment) to be a major investment towards the conservation of that land. The

variable is assigned the value 1 when respondents indicate they made such investment,

0 otherwise. Based on this methodology, only 15.5% of the respondents identified to have

made major investment in land conservation in the past 5 years.

Whereas the initial assessment of the study had considered the state of land market

(and prices) as important for consideration in the model, preliminary evaluation of the

empirical data suggested an absence of such dynamics within the focal area. For instance,

the difference between the price of migrant lands for farming in 1995 and 2005 was

considered a useful proxy for estimating the extent of vibrancy in the local land market.

Unfortunately, this variable was found to be highly collinear, as it showed little change.

The response rate on land price (values) was also very poor. This factor was therefore

excluded from the empirical model.For the tenure regime, three broad dummy variables are defined in the study, based on

the regime under which the land is operated. These are the freehold, lineage and leasehold



13/22

tenure regimes. By definition, the freehold regime involves cases where the landholder

expresses the right to own and use the land for unlimited period of time. In the sample

data, this regime was found to arise mainly from outright purchase of land by the

respondent or when the land is passed on to the holder as gift. Under such regimes, the

holder exercised full right to the use of the land and had no obligation to an institution or

person(s) regarding the form of use the land is applied, the payment of rent and duration

of access to the land.

The leasehold land tenure arises where the respondent exercised right to the use of the

land only for some period (whether such period is explicitly specified or not). In such

instances, the respondent acknowledged the fact that right to the use of the land is subject

to the discretion of the landlord or upon the fulfilment of some known obligations. Such

obligations were noted to include cash payments (occasional) assistance offered to the

owner and other forms of payment in-kind. In other instances where no specific obligation

is reported, respondents clearly acknowledged that the land will be returned to the land-

lord over time or on demand.

The last tenure regime is the lineage. This form of tenure arrangement involved rightsto use the land on the basis of family ties (lineage) or inheritance. Underlying this

arrangement is the fact that though no known restriction apply in the form and duration

for the use of the land (which could be for even unlimited time period), the land is still

regarded as a resource belonging to the family or the lineage. Any transfer or even

continued use is dependent on the state of association with the lineage. Thus, unlike the

freehold, the right to bequeath the land resource is limited but the form and duration of

use by the occupant is usually unrestricted.

Arising from these three different tenure regimes, the leasehold tenure regime is

hypothesized in this study to be inferior (in terms of tenure security and user rights) to

freehold and lineage regimes, in that order. This is more so as the choice of land use ishighly limited and may even involve restrictions on the type of crops to cultivate or even

the degree of transformation that is allowed on the land. Indeed, unlike the latter two, any

major investment on the land (and outside the original terms) may require prior and

explicit approval of the principal landowner. The order of tenure security and user right

for the three regimes is thus illustrated as follows:

In the estimation of the model, the relative strengths of these three tenure regimes in

influencing the decision to undertake major investment in land conservation are empiri-

cally investigated using the lineage regime as reference.

Among other independent variables, the marital status, age, gender and educational

attainment of the head of the household, location and size of the household were

investigated. The variable age is defined as the number of complete years of the head

of the household. This variable is introduced in the model to measure the cumulative

experience and hands-on skills in farm-based production of the household. Size is used

to measure the amount of labour resources available to the household, whereas genderis

defined as the sex of the head of household, assuming the value 1 if the head of the

household is male, 0 otherwise. For a typically patriarchal society like the communities inthe focal areas, the gender of the head of household could make considerable difference in

determining the quality and size of farmlands that the household could hold and operate. It

Land Tenure Security Very Strong Very Weak

Land Tenure Regime Freehold >> Lineage >> Leasehold

12 E.J. Mensah


14/22

could also influence the extent to which the household could assert its right in the use of

the land and the range of choices to which the land could be applied (Goldstein & Udry,

2006). All these variables are postulated to have positive impact on household decision to

invest in land conservation, ceteris paribus.

The level of educational attainment of the household head, as a measure of the

households knowledge frontier, capacity to properly analyse trends in the productivity

of their land, absorb new knowledge and implement major conservation measures, was

constructed as follows:

None, if household head had no formal education.

Basic, if the head had had at least primary, junior secondary, middle school or

Koranic education.

Secondary, if the head had received at least senior secondary, technical, vocational

training, ordinary level or advanced level schooling.

Post-secondary, if the head has attained at least teacher training, technical, profes-

sional or other forms of tertiary education.

These variables are constructed as mutually exclusive dummy variables (i.e. assuming

value 1 if applicable, and 0 otherwise), with none serving as the base or reference

variable. The variable on location is defined as the rural/urban location of the place of

residence of the household, which is assigned a unit value if household reside in rural

location and zero otherwise.Table 1summarizes the definition and descriptive statistics of

these variables.

According to Kabubo-Mariara (2006), though most household characteristics do not

significantly impact on the adoption of conservation practices in Kenya, the educational

attainment of the head of the household has significant positive effect. This is contrary tothe observation made by Mwakubo (2002) that educational level of the household head

reduces the likelihood of adoption of conservation practices by reducing the intensity of

terracing among farmers in Machakos and Kitui Districts. The paper explains that the

Table 1. Definition, measurement and summary statistics of the key variables.

Variable Basic definition and measurement Mean SD

Conservation Major investment in land conservation since the past5 years (1 = yes, 0 otherwise)

0.155 0.362

Lineage Using principal farmland from lineage/inheritance(1 = yes, 0 otherwise)

0.043 0.204

Freehold Using principal farmland under freehold terms (1 = yes,0 otherwise)

0.262 0.440

Leasehold Using principal farmland under leasehold terms (1 = yes,0 otherwise)

0.255 0.436

Married Marital status of head (1 = yes, 0 otherwise) 0.872 0.334Age Age of respondent (range 20100) 48.249 15.383Household Size Size of the household (range: 115) 6.436 2.843Male Sex of respondent (1 = yes, 0 otherwise) 0.901 0.299Rural Location of household (1 = rural, 0 otherwise) 0.458 0.499None No formal education (1 = yes, 0 otherwise) 0.550 0.498

Basic Educated up to basic level (1 = yes, 0 otherwise) 0.308 0.462Secondary Educated up to secondary level (1 = yes, 0 otherwise) 0.085 0.279

Post-secondary Educated beyond secondary level (1 = yes, 0 otherwise) 0.033 0.179

Source: Authors computation from the survey data.



15/22

existence of off-farm income generation opportunities for the relatively better educated

household heads accounts for that result. In this paper, it is postulated that younger and

more educated household heads adopt land-conserving measures than the older and less

literate heads.

In relation to education, it is further argued that the level of awareness of a farm

household about issues of environmental degradation and fertility loss could also influ-

ence a landholders decision to employ on-farm conservation practices. For instance,

relative to a farmer who confronts limited knowledge on sustainable land use and

conservation methods, the one with good exposure to such practices is likely to adopt

the most appropriate and cost-effective conservation method in response to actual or

perceived reduction in crop harvest. As a proxy of landholders exposure and knowledge

frontier, formal education is used here to provide an empirical test of this hypothesis.

4. Discussion of results

Preliminary analysis of the survey data shows that ownership and administration of landin the focal area is governed by three dominant institutions. These are tendana, chief-

taincy and family institutions. Thus, although private and public entities may declare

ownership of land in these areas, these three institutions constitute the principal source of

such access or ownership.

Unlike the other two, tendana is a traditional land administration authority which is

peculiar to northern Ghana. This authority is constituted by religious figures identified by

the people as representing the earth god. They are suggested to be the immediate

descendants of the early settlers on the land and thus exercise jurisdiction over a generally

agreed landmass, made up of a village or villages and its surrounding land area. Among

others, the tendana performs specific traditional and religious functions in their area ofjurisdiction, which reinforces their control and influence within the respective societies.

Some of these include the responsibility to allocate land and resolve land disputes, impose

sanctions in respect of infringements of land rights and abuses and perform religious rites

towards maintaining the sanctity and overall productivity of the land. As religious figures,

thetendanaalso acts as the intermediary between the people and the gods (or the supreme

spiritual authorities of the land).

Among the respondents, about 36% suggested the tendana as the principal institution

of land administration. Another 33% of the respondents have their lands owned and

administered by the family, whereas 24% reported administration by the relevant chief-

taincy institution. Ownership by individuals was marginal, forming only about 7%.

The focus group discussion revealed that cases of individual ownerships arise from

transfers from heads or representatives of these three institutions to individuals as gift

rather than on the basis of any commercial transaction. Indeed, as custodians of these

lands, these representatives are enjoined to administer the lands for the larger interest

of the people. Profiteering or commercial transactions in land is therefore rare in these

areas. Access to land by both indigenes and migrants therefore follows a basic procedure

of placing a request with the appropriate institution and the payment of a token of

appreciation. For indigenes or members of the same family, such requests are granted

mostly for free.

Inherently, these three systems give rise to the three distinct tenure regimes. As

discussed earlier, these regimes are hypothesized to exhibit a continuum of tenure security,with potentially significant implications for the incentive to invest in land conservation.

Among others, Table 2reports the result of the empirical test of these hypotheses.

14 E.J. Mensah


16/22

The probit model was estimated alongside the logit model to achieve some compar-ison of the appropriateness of the assumptions on the distribution of the error term. That

is, whereas the probit model assumes that the dependent variable of the model follows the

standard normal cumulative distribution function, the logit model assumes the logistic

function. The Wald 2 test statistic indicates that the coefficients of the independent

variables in the model are jointly significant in explaining the incidence of major invest-

ment in land conservation by landholders in northern Ghana. The result of the RESET test

also rejects the null hypothesis that the model is functionally misspecified. The results are

also shown to be homoscedastic in the standard error. There is therefore sufficient

statistical basis to assess the estimated models as well fitted, correctly specified and

good for the present analysis.Indeed, the probit model predicts the probability of a household making a major

investment in land to be 14.05%, which is comparable with the observed adoption rate of

15.52%. These test statistics also suggests the probit model has relative superiority over

the logit model. Results of the probit model are therefore selected for discussion, along-

side estimates of its partial effects.

Among the tenure regimes investigated, the freehold (relative to lineage) was found to

have a positive and significant impact on the probability of a household making a major

investment in the land. The result shows that farm households using land under freehold

tenure are 6.7% more likely to make major investment in the land than those under

lineage. In the case of leasehold tenure regime, the variable shows a positive but

statistically insignificant impact, even at the 10% significance level. The results suggest

that as far as major investment in land is concerned, whereas tenure security under the

Table 2. Results of the estimated models (logit, probit and the marginal effects).

Logit model Probit modelProbit model

(marginal effects)

Number of observations =683 =683Wald2(10) =30.98*** =32.56***

Pseudo R2 =0.051 =0.052

Conservation Coefficient Robust SE Coefficient Robust SE dy/dx Robust SE

Freehold 0.503* 0.263 0.281* 0.147 0.067 0.037Leasehold 0.314 0.272 0.174 0.148 0.041 0.036Married 0.438 0.344 0.263 0.190 0.065 0.051Age 0.008 0.008 0.005 0.004 0.001 0.001Size 0.117*** 0.042 0.066*** 0.023 0.015 0.005Gender 1.472** 0.598 0.784*** 0.277 0.122 0.027

Rural 0.425* 0.228 0.237* 0.126 0.053 0.029

Basic 0.019 0.264 0.004 0.143 0.001 0.032Secondary 0.879** 0.392 0.514** 0.221 0.141 0.071Post-Secondary 0.655 0.774 0.335 0.390 0.063 0.059Constant 3.595 0.766 2.014 0.391

Ramsey RESET:2 (1) =2.16 =2.62Observed probability =0.1552 =0.1552

Predicted probability =0.13850 =0.14047

Notes: *, ** and *** indicate significance at 10%, 5% and 1% levels, respectively.dy/dx is the discrete change in the dummy variable (Conservation) from 0 to 1.



17/22

leasehold regime (relative to the lineage) does not induce statistically significant differ-

ential impact on the decision by households to investment in land conservation, the

freehold regime does.

These observations reveal a very interesting aspect of the incentive structure of land

tenure security in northern Ghana. Given the perspective that freehold land rights offer

occupants much longer and stronger right to the land and the benefits accruing from any

investment made in it, this finding is consistent with the longstanding argument that

assured rights ensure greater incentive to invest in improving the productive capacity of

the land by farmers (Uma, 1992). Furthermore, in view of the fact that families or

communities tend to maintain ownership of (or at least, some interest in) lands used

under lineage tenure regimes, the incentive to make major investment in such lands by the

occupant is limited and not too different from the situation of landholdings under

leasehold regimes.

Thus, though lineage arrangements grant landholders considerable right to the land,

the fact that ownership is not individualized but remains largely a common property of the

larger family (or community) curtails the incentive to undertake major investments in landconservation. To the extent that the long-term use of that land is guaranteed only for as

long as the circumstances of the family or lineage permit, tenure security is in reality not

durable. As shown by the results, the incentive to undertake major investments in

conservation under lineage regimes is therefore still limited and akin to that found

under leasehold, statistically.

A related explanation provided by Bekele and Alemu (2010) in their study of the

Ethiopian Highlands observes that whereas a farmers decision to invest in land conserva-

tion depends generally on the immediate or shorter-term risk he faces, the intensity of

such investment rather depends on the longer-term risk. The paper further establishes that

intense or major investment in land conservation occurs when tenure security is assuredfor at least the next 5 years. Similarly, in the context of engendering greater investment in

land conservation by households in northern Ghana, the present findings suggest that

assured right to access and use farmlands even under lineage tenure regimes is not

enough. Durability of such rights in terms of stable access and use over predictably

much longer time is fundamental for engendering significant and non-random major

investment in land conservation. Following the expression of Uma, one could state that

only durable and assured individualized5 right to land ensures major investment in land

conservation in the study area.

Again, this finding is similar to the results of other studies. In most of these studies,

the general conclusion drawn has been that where the tenure security is strong, farm

owners are more disposed to undertake investment in land conservation. For instance, a

study by Kabubo-Mariara (2006) in Kenya reports that bequest right (which is synon-

ymous to rights under freehold) is important in determining whether a farmer invests in

land conservation practices or not, as against cases where the ownership of the land is

retained by other person(s), community or the state (like those lands obtained from public

schemes for enhancing land access). A meta-analysis of 46 cases studies provided in

Raquez and Lambin (2006) further supports this finding.

In the case of the household characteristics, the size and gender of the head of

household are found to be significant in explaining investment decisions in land.

Notably, both variables are positive and significant at the 1% level. In explaining the

result for the household size variable, it is relevant to note that smallholder farming in thefocal zone also tends to depend less on external inputs and more on family labour and

local resources. Thus, the larger the household, the more hands it is able to mobilize in

16 E.J. Mensah


18/22

working on the farm and implement investments in land conservation. It is therefore

consistent with general expectation that the larger the size of the household, the more

likely the incidence of major investment in land. This observation also provides empirical

support to the Boserups thesis, that within traditional agrarian production systems (as is

the case of the study area), population growth and declining access to land for extensive

farming engender intensification through the increased use of farm labour to compensate

for the declining sizes and productivity of farmlands.

This observation is also consistent with many others reported in the literature. In

Kabubo-Mariaria (2006), except for investments in the construction of soil bunds (as a

specific example of conservation investment), the variable was found significant in

explaining the adoption of other forms of investment in land, which are terracing, the

planting of drought-resistant vegetation and a combination of these methods.

In relation to the gender of the household head, the estimated model unravels the

gender dimension of the argument. In particular, the result indicates that relative to

female-headed households, male-headed households are 12.2% more likely to make

major investment in land conservation. Though this result is not consistent with theobserved non-significance of gender in farm-level investment in land conservation in

Kenya (as reported by Kabubo-Mariaria,2006), it supports assertions in various literature

on Ghana that female-headed households are more resource-poor and tend to lack capacity

to make major investment in livelihood support strategies than male-headed households.

It also supports the earlier argument that given the labour-intensive nature of farming

in the focal area, the capacity to invest in land conservation is enhanced when households

possess significant labour resources. This is found more with male landholders than their

female counterparts. For example, in discussing the gender dimension of household

poverty and capacity to invest in land conservation across African societies, Barbier

(1998) makes similar assertion and argues that labour market discrimination againstwomen as well as large labour demand on women in catering for the needs of their

households plays a very important role in constraining females from investing in land

conservation. To a large extent, such circumstances contribute to the well-noted gender

disparities in especially rural livelihood settings in Ghana (GSS,2000; NDPC,2005).

On the contrary, the marital status (i.e. married respondents as against never married,

divorced, widowed or other state of non-marriage) and the age of the household head were

found to be statistically insignificant in explaining farmer investment in land, not even at

10% significance level. In the study, it had been hypothesized that relatively younger

landholders would show higher probability of investing in farmland as compared with the

older ones. This result, as well as that for the marital status, is found to be similar to the

observation made by Kabubo-Mariaria (2006) in Kenya that both marital status and age of

the household head are not significant, statistically, in explaining investment in land

conservation.

In the case of the location dummy, households in rural locations are found more likely

to invest in improving the productivity of their farmlands than households in urban

locations. Given the similarity in the topography and climate of the communities in the

three northern regions, a possible explanation for this result is that the livelihood sources

of farm households in rural locations are less diverse and tend to depend more on farming

on these lands compared with those in urban areas who tend to have better access to non-

farm livelihood sources. Farmers in rural locations therefore have better motivation to

make major investment in conserving these lands than those in urban areas.The result on education was mixed, as some of the variables were significant and

showed the expected sign while others did not. Specifically, it is noted that whereas



19/22

landholders with basic education, relative to those with no education, showed no statis-

tically significant probability to make major investment in land conservation, those with

up to secondary level education were 14% more likely to invest in land conservation.

Similar effect was expected for even a higher level of educational training (i.e. post-

secondary education) but the variable was found to be insignificant, statistically. A

plausible explanation for this result is that landholders with at least post-secondary

education are more likely to be active in off-farm incoming-generation activities and

even more dependent on external inputs than those with limited education. Thus, major

investments in conservation may not occur as critical for sustaining production and

livelihood generally, unlike the less literate counterparts.

5. Conclusion and policy recommendations

The importance of agriculture is well noted in the pursuit of sustainable and pro-poor

economic development. For Ghana in particular, the growth of the sector is identified to

have significant leverage on enhancing household income and living standards. Thesenotwithstanding, major challenges confront the sector. A dominant one is the issue of land

tenure insecurity.

In arid and semi-arid agro-ecological zones such as northern Ghana, contemporary

trends in population growth, land degradation, productivity losses and extremities in

weather patterns make the subject of land tenure regimes and their implications on

investment in land conservation all the more important. In this study, this nexus was

investigated using data from a structured household survey obtained from the ISSER/

USAID Land Tenure Policy project.

Among others, the study reveals that land in the focal area is administered under three

dominant traditional institutions of land administration in Ghana. These are the tendana,chieftaincy and family institutions. Arising from these, three tenure regimes are identified

in the area, namely, freehold, leasehold and lineage. It is further found that only 15.5% of

landholders have made major investment in land conservation since the past 5 years. In

explaining the trend, the study finds a continuum of land tenure security exhibited by

these three tenure regimes, which then impact differentially on the decision to invest in

land conservation by the holders. In particular, the results of the quantitative analysis

reveal that whereas freehold tenure regime, relative to lineage, significantly increased the

likelihood of major investment in land conservation by households, leasehold shows no

important differential impact.

Thus, while the existing literature theorizes assured rights to land as important in

engendering increased investment in land conservation, these findings reveal that such

rights must be durable and individualized to help realize such investments in the focal

area. Development policy must therefore move beyond improving access and promote

durable and private (individualized) land tenure security to help sustain the incentive for

significant investments in land conservation. In other jurisdictions such as Ethiopia,

Deninger, Ali, and Alemu (2009) and Tsegaye, Adgo, and Selassie (2012) provide

evidence of the effectiveness of low-cost land certification systems in attaining similar

purpose. In the latter paper, land certification is identified to account for over 15%

increment in the adoption of land management practices among households in the dry-

lands of the Amhara region of that country.

The findings also reveal the gender dimension of this subject; which is, male-headedhouseholds are systematically more likely to make major investment in land conservation

than their female-headed counterparts. It is recommended for policy to institute support

18 E.J. Mensah


20/22

mechanisms targeted at female-headed households to enable them implement land con-

servation as a livelihood support intervention.

Consistent with the Boserup thesis, it is also found that the incidence of investment in

land conservation is significantly associated with households with more labour resources.

It is further found that rural-based farm units show greater tendency to make major

investment in land conservation than their urban-based counterparts. However, contrary

to expectation, age and marital status are not important determinants of the incidence of

major investment in land conservation by households.

In relation to education, a minimum threshold of formal training is required to

significantly engender a positive and significant impact on the probability of major

investment in land conservation by households. In promoting sustainable agriculture

and investment in land conservation, development policy must regard educational attain-

ment of up to such threshold as an important driver. Below such threshold, households

might require more specialized training and education to create awareness and incentives

for greater investments in land conservation.

5.1. Recommendations for future research

As the opportunity did not exist for the current study to evaluate the role of extension and

related training interventions in enhancing the capacity of landholders in investing in land

conservation, future studies could explicitly measure the impact of such training activities.

Also, notwithstanding the rigour of the test statistics of the estimated models and the

general trend of extensive dependence on own labour for such major investments in land

conservation by farm households in the study area, it is still conceivable that income or

wealth could influence the propensity to invest in land conservation. This is because even

in the face of weak tenure security, resource-rich smallholder households are likely to facedisproportionately lower opportunity cost of capital to undertake such major investments

in land conservation than their resource-poor counterparts. It is recommended for future

research to investigate this factor.

Acknowledgements

My sincere appreciation to the lOasis Development Group of the Global Youth Alliance forthe earnest support leading to this paper. Funding from the ISSER/USAID Land TenureReform Project for the original study underlying this paper is also appreciated as well as thecontribution of Mrs. Freda Asem to the same. Nonetheless, all errors and omissions herein are

entirely mine.

Notes

1. The MDP is a flagship project of the Earth Institute at the Columbia University, USA. Theproject adopts a community-led, integrated rural development model to fast-track the realizationof the UNs millennium development goals in selected rural communities in Sub-Sahara Africa.Recent models have included Haiti, Cambodia and Jordan.

2. Pluralis used in this context to refer to the coexistence of different systems derived especiallyfrom both formal and informal institutions and rules for land use and administration in thecountry.

3. Besley (1995) provides an extensive treatment of this argument with an application to Ghana.4. ISSER is the Institute for Statistical, Social and Economic Research, a research institute basedat the University of Ghana, Legon, Accra. USAID is the United States Agency for InternationalDevelopment, which also maintains a country office in Accra, Ghana.



21/22

5. Individualized in this context is used to emphasize exclusivity of ownership to an individual,which then allows the land to be applied and the benefits captured more as a private rather thana common resource.

References

AGRA. (2012). The policy and partnership programs of the alliance for a green revolution inAfrica. Nairobi: Author.

Adam, S. (1974).An inquiry into the nature and causes of wealth of nations (1). Book III, Chapter 2.London: Penguin Books.

Asabere, P. (1994). Public policy and the emergent African land tenure system: The case of Ghana.Journal of Black Studies, 24(3), 281289.

AU Commission. (2010).The Abuja declaration on fertilizer for African green revolution: Status ofimplementation at regional and national levels. Policy Briefs. Abuja: Author.

Barbier, E. B. (1998). The economics of land degradation and rural poverty linkages in Africa.UNU/INRA 1998 Annual Lecture. Japan: United Nations Institute for Natural Resources in

Africa, United Nations University.Bekele, G., & Alemu, M. (2010). Investment in land conservation in the Ethiopian Highlands: A

household plot-level analysis of the roles of poverty, tenure security, and market incentives.Discussion Paper Series EfD DP 10-09. Addis Ababa: Environment for Development.

Besley, T. (1995). Property rights and investment incentives: Theory and evidence from Ghana. TheJournal of Political Economy, 103(5), 903937.

BOG. (2008). Statistical bulletin 2008. Accra: Author.Boserup, E. (1965). The conditions of agricultural growth: The economics of agrarian change

under population pressure. New York, NY: Aldine Press.CEPA. (2007). Review and analysis of government economic policy for 2007. Accra: Author.Cleaver, K. M., & Schreiber, G. A. (1994). Reversing the spiral: The population agriculture, and

environment nexus in Sub-Saharan Africa. Washington, DC: The World Bank.

Dasgupta, P., & Maler, K. (1990). The environment and emerging development issues. Annual bankconference on development economics. Washington, DC: The World Bank.de Janvry, A., & Sadoulet, E. (2001). Access to land and land policy reforms A policy brief. UNU

World Institute for Development Economics Research. Helsinki: UNU/WIDER.De Soto, H. (2000).The mystery of capital: Why capitalism triumph in the west and fails everywhere

else. New York, NY: Basic Books.Deininger, K., & Binswanger, H. (1999). The evolution of the World Banks land policy:

Principles, experience and future challenges. The World Bank Research Observer, 14(2),247276.

Deninger, K., Ali, D. A., & Alemu, T. (2009). Impact of Land Certification on Tenure Security,Investment and Land Markets: Evidence from Ethiopia. Discussion Paper Series EfD DP 0911.Addis Ababa: Environment for Development.

Deressa, T. T., Hassan, R. M., & Ringler, C. (2009). Assessing household vulnerability to climatechange: The case of farmers in the Nile Basin of Ethiopia. IFPRI Discussion Paper 00935.Addis Ababa: International Food Policy Research Institute.

Downing, T. E. (1992). Climate change and vulnerable places: Global food security and countrystudies in Zimbabwe, Kenya, Senegal, and Chile. Oxford: Environmental Change Unit,University of Oxford.

FAO. (2005).Concept note International conference on agrarian reforms and rural development:New challenges and options for revitalizing rural communities. 128 Session, CL 128/9Supplement 1. Rome: Author.

FAO. (2006). Fertilizer use by crop in Ghana. Land and Water Division. Rome: Author.GLAPS. (2003).Review of studies conducted on gender and land in Ghana for LAP and support for

developing gender policy on land matters Draft final report. Accra: Author.GSS. (2000). Poverty trends in Ghana in the 1990s. Accra: Author.

GSS. (2012). Ghana 2010 population and housing census Final results. Accra: Author.GoG. (2006).Reducing poverty through agricultural transformation Millennium challenge Ghana

programme. Accra: Author.

20 E.J. Mensah


22/22

Goldstein, M., & Udry, C. (2006).The profits of power: Land rights and agricultural investment inGhana. Discussion Paper Number 32. Accra: Institute of Statistical, Social and EconomicResearch, University of Ghana.

Hitchcock, R. K. (1981). Traditional systems of land tenure and agrarian reform.Journal of AfricanLaw, 24, 2433.

IFAD. (2009). IFADs response to climate change through support to adaptation and relatedactions. Rome: Author.

ISSER. (2006). The state of the Ghanaian economy in 2005. Accra: Author.Kabubo-Mariara, J. (2006). Land Conservation in Kenya: The Role of Property Rights. AERC

Research Paper 153. Nairobi: African Economic Research Consortium.Lapp, F. M. (1998). World hunger: Twelve myths (2nd ed.). New York, NY: Grove Press.Long, J. S. (1997). Regression models for categorical and limited dependent variables. Advanced

quantitative techniques in the social sciences, number 7. Thousands Oak, CA: SagePublications.

Malthus, T. R. (1998). An essay on the principle of population. London: Electronic ScholarlyPublishing Project (Original work published 1798).

Maxwell, D., & Wiebe, K. (1999). Land tenure and food security: Exploring dynamic linkages.Development and Change, 30, 825849.

Mwakubo, S. M. (2002).Land tenure and natural resource management in Semi-Arid Areas, Kenya.Kenya: Agricultural Economics Department, Egerton University.

NDPC. (1991).Making people matter: A human development strategy for Ghana. Accra: Author.NDPC. (1994).National development policy framework. Accra: Author.NDPC. (2003). Ghana poverty reduction strategy (20032005): An agenda for growth and prosper-

ity. Accra: Author.NDPC. (2005).Growth and poverty reduction strategy (GPRS II), 20062009. Accra: Author.Otuska, K., Quisumbing, A. R., Payongayong, E., & Aidoo, J. B. (2003). Land tenure and the

management of land and trees: The case of customary land tenure Areas of Ghana. Environmentand Development, 8, 77104.

Picardi, A. C., & Seifert, W. W. (1976). A tragedy of the commons in the Sahel. Technology Review,78, 4251.

Place, F., & Otsuka, K. (2002). Land tenure systems and their impacts on agricultural investmentsand productivity in Uganda. The Journal of Development Studies, 38(6), 105128.Platteau, J. P. (1996). The evolutionary theory of land rights as applied to SUB-Saharan Africa: A

critical assessment. Development and Change, 27(1), 2986.Platteau, J. P. (2000). Institutions, social norms, and economic development. Amsterdam: Harwood

Academic.Raquez, P., & Lambin, E. F. (2006). Conditions for sustainable land use: Case study evidence.

Journal of Land Use, 1(24), 109125.Scherr, S., & Yadav, S. (1997). Land degradation in the developing world. Issues and policy options

for 2020. 2020 Brief 44. Washington, DC: International Food Policy Research Institute.Soule, M. J., Tegene, A., & Wiebe, K. D. (2000). Land tenure and the adoption of conservation

practices. American Journal of Agricultural Economics, 82(4), 9931005.Tsegaye, A., Adgo, E., & Selassie, Y. G. (2012). Impact of land certification on sustainable land

resource management in Dryland Areas of Eastern Amhara Region, Ethiopia. Journal of Agricultural Science, 4(12). doi:10.5539/jas.v4n12p261

UN WCED. (1987). Our common future. A Report of the UN World Commission for Environmentand Development. New York, NY: United Nations.

UNCCD, UNEP & UNDP. (2009). Climate change in the African drylands: Options and opportu-nities for adaptation and mitigation. A joint publication of the United Nations Convention toCombat Desertification (UNCCD), United Nations Environment Programme (UNEP) and theUnited Nations Development Programme (UNDP). New York, NY: UNCCD.

Uma, L. (1992). Structural adjustment and agriculture: A comparative perspective of performancein Africa, Asia and Latin America. IW92-30r, International Working Paper Series. Gainesville:Food and Resource Economics Department, University of Florida.


Land Tenure Regimes and Land Conservation in the African Drylands: the Case of northern Ghana

Documents