See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/326489415

Implications of Agricultural Land Subdivision on Productivity:Case Study of

Kajiado County

Article · June 2018

CITATION READS

1 1,334

2 authors:

Erastus Museleku Paul M. Syagga

University of Nairobi University of Nairobi
2 PUBLICATIONS   1 CITATION    63 PUBLICATIONS   125 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Property Taxation View project

Housing View project

All content following this page was uploaded by Paul M. Syagga on 23 July 2018.

The user has requested enhancement of the downloaded file.

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

INTERNATIONAL JOURNAL OF
INNOVATIVE RESEARCH AND KNOWLEDGE
ISSN-2213-1356 www.ijirk.com

DRIVERS OF AGRICULTURAL LAND SUBDIVISION

IN DRYLANDS OF KENYA:
A Case of Kajiado County, Kenya

ERASTUS KIITA MUSELEKU*, DR. MARY W. KIMANI,

DR. WINNIE N. MWANGI & PROF. PAUL M. SYAGGA
Department of Real Estate and Construction Management,
University of Nairobi, Nairobi, Kenya
* Corresponding Author

ABSTRACT
Private dry agricultural land sizes are reducing over the years mainly due to subdivisions, contrary to the
theoretical and general expectation that they are held and used in large contiguous sizes. This paper
investigates the significant drivers of agricultural land subdivisions in Kajiado County, Kenya. The paper
used a mixture of primary and secondary research methods to investigate this phenomenon. Primary data was
mainly obtained from 203 agricultural landowners in the study area. The data reveals that socio-cultural and
economic factors are the most significant drivers of agricultural land subdivision in the study area:
agricultural land inheritance practices, individualization of tenure, price of agricultural land and demand for
urban housing. Firstly, the paper recommends that the national and county governments should put in place
appropriate and clear policy, legal and institutional frameworks to prescribe allowable minimum
economical/optimal agricultural land sizes in various agro climatic zones. Secondly, individual titles should
have restrictions on the minimum allowable sizes depending on the location and use of the land. Thirdly,
government and private sector should support agricultural enterprises, including livestock production to make
agricultural activities economically viable and reduce the influence of attractive agricultural land prices.
Fourthly, urban revitalization strategies and urban land banking should be encouraged as means of providing
urban housing in the urban areas.

www.ijirk.com Page 195

International Journal of Innovative Research and Knowledge ISSN-2213-1356

Key Words: Subdivision of Agricultural Land, Drivers, Drylands

INTRODUCTION
Globally, agriculture remains the single largest employer and provides livelihoods for more than 40 per cent
of the globe’s population. Besides, agriculture is the world’s largest source of income and jobs for rural
communities. Agricultural land, therefore, remains a key resource for the rural communities as well as urban
dwellers. Thus, the size of agricultural land is as important as the distribution and access to this resource.
Over 80 per cent of Kenya’s population reside in rural areas and obtain their livelihoods from agricultural
land. Indeed, agricultural sector in Kenya directly contributes about 26 per cent to the Gross Domestic
Product (GDP) annually, accounts for approximately 65 per cent of national’s total exports and creates over
70 per cent of informal employment opportunities in the rural areas thus making it a backbone of Kenya’s
economy. Agricultural land in Kenya plays a key role in poverty reduction in the lives of vulnerable groups
such as the pastoralists and subsistence farmers who derive their livelihoods mainly from agricultural
activities (Government of Kenya [GoK], 2016; 2010).
Kenya’s landmass is approximately 582,646 square kilometres of which about 98 per cent is land and 2 per
cent is water surface. Only about 20 per cent of Kenya’s land is arable while the bulk of the land (over 80 per
cent) is arid or semi-arid (ASALs). Besides, about 75 per cent of the Kenya’s populace resides in the arable
lands, thus contributing to high population densities in those lands (GoK, 2009). Essentially, the medium to
high potential agricultural land in Kenya is already subdivided into small units which may be uneconomical.
Agricultural land in Kenya, including ASALs, however, is being subdivided into small sizes (sometimes
below 1ha) despite the enormous importance it has on economic development. At the national level, the
average farm size is approximately 2.5ha, with 98 per cent of agricultural land sizes being about 1.2ha (The
Centre for Land, Economy and Rights of Women, 2006; Syagga & Kimuyu, 2016). This phenomenon has
raised concerns among land administrators and managers, policy makers and general public that such
transformations may impact negatively on the productivity of the agricultural land (GoK, 2009; Kelleher, et
al., 1998). There is need therefore to understand significant drivers behind this phenomenon especially in the
dry agricultural lands which support extensive livestock production systems.

THEORY
Drivers of Agricultural Land Subdivisions
Identifying the drivers of agricultural land subdivision (ALS) requires an examination and understanding of
how people make land use decisions and how various factors interact in specific localities to influence their
decision making process (Lambin et al., 2003). The drivers of ALS vary with localities, for instance, the
Scottish government (2009) categorised the drivers in rural Scotland under environmental, demographic,
economic, technological, policy, institutional, cultural and social factors while Chazan & Cotter (2001)
categorised drivers of ALS in United States as population and household size, personal housing style
preferences (demographic), government policies and economic stimulus. Jiang et al. (2013), Lo & Yang
(2002) and Liu et al. (2004) identified the same drivers of land subdivisions in China.
Lambin et al. (2003) generalised the drivers and broadly categorised them into natural/environmental changes,
economic and technological, demographic, institutional, cultural and social factors. They went on to clarify
that these drivers are either direct or indirect and agricultural land subdivision into small sizes and subsequent
conversions in a locality is usually occasioned by a combination of several factors. Olson et al. (2004)
concluded that agricultural land subdivisions and subsequent conversions of use in East African countries,
Kenya included, are generally occasioned by drivers such as government policies and laws, economic factors,

www.ijirk.com Page 196

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

population growth and migration, land tenure arrangements, market access and environmental conditions.
From these broad categories of drivers of agricultural land subdivisions, it seems that the following are the
likely drivers in most localities.

Natural Environmental Factors

Physical or natural environmental changes have been observed to interact with the human decision making
processes that cause ALS. Highly variable environment conditions, for instance, usually occasioned by
changes in climate are likely to magnify the pressures of agricultural land subdivisions (Olson et al., 2004).
Although other drivers, such as socio-economic factors may operate independently, natural environmental
changes may interact with other drivers to influence ALS (Lambin et al., 2003). Environmental factors are
important due to their influence on agriculture, the main rural land use activity.
Specific and important environmental factors that may influence agricultural land subdivisions have been
observed to include quality/fertility of the land, terrain/topography, location (for example near road networks)
and climatic conditions (rainfall and temperature) (Chazan & Cotter, 2001; GoK, 2016; 2016a; Olson et al.,
2004). When the quality/fertility and topography of land and climatic conditions (rainfall and temperature) are
favourable, landowners may not be willing to subdivide their agricultural land since agricultural activities
may be viable.
Similarly, when farms are located near transport networks farmers can transport their farm produce to the
market with ease hence encouraging them to preserve the agricultural land. Concomitantly, these variables are
also favourable to real estate development and may encourage real estate developers to buy land from the
local landowners for real estate development hence fuelling agricultural land subdivisions.

Economic Factors
Broadly speaking, economic factors influence agricultural land subdivisions through market forces; supply
and demand for agricultural land (Lambin et. al. 2003). Since supply of land is usually static demand for it
becomes important in influencing the phenomenon of ALS. Economic drivers may also interact with
institutional factors and policies (McDonagh, 1997; Thuo, 2013; Olson et al., 2004).
In particular, economic factors that are likely to influence ALS have been noted to include per capita
income/poverty, demand for urban housing, agricultural productivity or farm-income and none/off-farm
income (Ayonga, 2008; Chazan & Cotter, 2001; GoK, 2016; 2016a; Henry et al., 2012; Lee, 1999;
McDonagh, 1997, Nkedianye et al., 2009). In East Africa, privatisation of former communal or group ranches
has been stated as an important economic factor (Olson et al., 2004). Per capita income/poverty may
influence agricultural landowners to subdivide their land and sell to the property developers, especially if the
return from the agricultural activities and off farm income is not adequate to support their livelihoods.
Similarly, per capita income growth of the urban dwellers may encourage them to look for housing in the
suburbs and nearby rural areas (away from pollution, congestion and general poor quality of life associated
with urban areas), thus influencing demand for housing and increasing the rate of ALS. Technology is also
likely to influence agriculture by intensifying land use and irrigation practices thus making it possible to
maintain agricultural production and productivity levels after subdivision of agricultural land.

Demographic Factors
Changes, either positively or negatively, in local populations are likely to influence subdivision of agricultural
lands. Important demographic factors in agricultural land subdivisions have been observed to include urban
and rural populations’ growth rates and may interact with government policies and economic drivers (GoK,
2016; 2016a; Henry et al. 2012; Jayne & Muyanga, 2012; Lambin et al. 2003; Lee, 199; Olson et al., 2004).

www.ijirk.com Page 197

International Journal of Innovative Research and Knowledge ISSN-2213-1356

Lambin et al. (2003) noted that growth of urban aspirations and urban-rural population distribution are
important factors in regional ALS within major urban centres, in peri-urban areas and even in remote rural
areas. Increase in urban population puts land in the surrounding rural areas under a lot of pressure to
transform to urban use such as residential use (Thuo, 2013).

Institutional Factors
Institutional factors mainly influence ALS indirectly but are influenced directly by political, legal and
economic drivers and their interactions with individual landowner decision making. The use of resources such
as agricultural land is facilitated by local and national policies thus institutions play a great role in the
designation of property rights (Ayonga, 2008; GoK, 2016; Lambin et al., 2003; Olson et al., 2004; Thuo,
2013).
Olson et al. (2004) postulated that key institutional factors that may influence agricultural land subdivisions
and transformations are technical capacity and involvement of public in land development decision making
processes. Besides, institutional arrangements create exclusive individual property rights in land which, if not
managed well, may result to too many owners holding small pieces of agricultural land that cannot guarantee
efficient and optimal agricultural production leading to a tragedy of spatial anticommons (Heller, 1998).

Cultural and Social Factors

Numerous cultural factors may influence decision making process of an agricultural landowner. These
variables are often related to political and economic conditions (GoK, 2016; Lambin et al. 2003; Nkedianye et
al., 2009; Thuo, 2013; Scottish government, 2009).
Important sociocultural drivers of ALS in East Africa may include land inheritance practices and land tenure
systems (for example customary rights), individualization of titles and acceptability to sell agricultural land
(commodification of land) (Mburu, 2009). Insignificant factors have been identified to include changing
distribution of land, wealth and power and commercialisation of labour and water resources. Competition or
cooperation between groups and changing gender roles and responsibilities may be relevant as well (Olson et
al., 2004).

Government Policy, Laws and Regulations (Political and Legal Factors)

Government policy, laws and regulations are important drivers of ALS. Policy is a course of action that is
formulated and adopted to promote a desired future outcome hence inevitably an indirect driver of ALS
(GoK, 2016; Olson et al., 2004). The influence of policy on ALS can be more meaningful when observed in
an historical perspective to show the outcomes of their interactions with other drivers.
ALS in East Africa, for instance, reflects the influence of both colonial policy and laws extending back to the
19th century. Such regimes have shaped the land tenure systems with wide implications on the access and
utilization of agricultural land. The colonial policies and legal frameworks used to enforce them led to
alienation of land for Europeans settlements and large scale farming and ranching in the East African region
(Partners News, Ogendo, as cited in Society for International Development, 2006). This affected the
distribution of agricultural land between the native people and the colonial settlers in Kenya, Tanganyika and
Uganda (Olson et al., 2004). Such regimes included, for example, the treaties between the Maasai and British
in 1911 and 1912 in Kenya (Koissaba, 2015).
Other land laws have granted natives power to subdivide and transfer agricultural land. Essentially, spatial
anticommons property rights are created by government policies, laws and regulations under political and
economic constraints (Heller, 1998).

www.ijirk.com Page 198

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

RESEARCH METHODOLOGY
Research Design
Cross-sectional survey and case study designs were utilized in this study. Babbie (1994) postulates that survey
design is probably the best method available for studying social phenomena because it allows researchers to
collect original data for describing a population too large to observe directly. Subdivision of agricultural land
into small sizes is a social phenomenon and a survey approach is appropriate. The choice of the study designs
is influenced by the nature of the data and the essence of meeting the study objectives in a cost efficient
manner.

Target population, sample size and sampling techniques

The total target population in this study included all agricultural land parcels and their respective owners in
the study area who were accessible at the period of the study. Other respondents included Kajiado County
land administration and management officials (County physical planners and surveyors; District physical
planners and surveyors; County Land Management Board and Land Control Board).
To estimate the population of the agricultural land parcels and their owners in the study area, the study used
the number of households in the study area according to the latest Kenya’s national housing and population
census statistics of 2009. The total number of agricultural land parcels and their owners was estimated to be
approximately 5,000, which is less than 10,000 cases. According to Mugenda & Mugenda (1999), when the
population is more than 10,000 a sample size of 384 is adequate. When the population is less than 10,000
cases, however, the following formula should be used to estimate sample size.

Box 1: Sample size for agricultural landowners and land parcels

nf = n/1+n/N Where:
nf = desired sample size when the population is less than 10,000
n = desired sample size when the population is more than 10,000
N = estimate of the population size.

Using the above formula sample size was calculated as follows:

Nf = 384/1+384/5,000 = 356.6 = 357 Agricultural land parcels & owners

Source: Adapted from Mugenda & Mugenda, 1999

The study used simple random sampling technique to access the targeted agricultural land parcels and
landowners living in their land in the study area at the time of the field survey. This sampling technique was
used to survey a total of 39 villages in the study area. Out of the 357 targeted agricultural land parcels and
owners, 203 were accessible, resulting to a response rate of approximately 57 per cent which was adequate for
analysis purposes (Mugenda & Mugenda, 1999).

Data collection tools

In this study, collection of data was done using both qualitative and quantitative methods to collect
information on agricultural land subdivision. The data sought was mainly primary and secondary data.
Primary data was sourced from the survey respondents while secondary data was sourced from libraries,
internet and public/government offices, mainly from the Kajiado County government and local land control
board. The tools that were used for data collection are as follows.

www.ijirk.com Page 199

International Journal of Innovative Research and Knowledge ISSN-2213-1356

Structured observation method

Non-participant direct observation of documents and the land parcels in the study area was done in a
structured manner and data recorded in the process of observations using a note book. The unit of observation
was the existing agricultural land sizes and their proximity to services, among other physical factors that may
influence demand for agricultural land or the rate of subdivisions. Structured observation method is advisable
since it eliminates bias and relates to current information which is not complicated by past events or future
aspirations. Besides, it is not dependent with respondent’s willingness to participate in a study, unlike in
questionnaire method (Kothari, 2004). Consequently, structured observation was used to gather data relevant
to the study objectives.

Semi-structured personal interviews

Interview method involves presentation of oral questions and responses given in the same way by the
respondents. Structured interviews are more economical, easier to analyse for generalization purposes and
ensures high response rate. In addition, they allow the interviewee to clarify questions hence the researcher is
able to gather more data than is possible using observation method (Babbie, 1994; Kumar 2005; Kothari,
2004; Nachmias & Nachmias, 2000). Thus structured interviews were conducted with the key study
informants who include the county and district land officials (planners and surveyors), chairman of land
control board and county land management board.

Semi-structured self-administered questionnaires/Schedules

Questionnaire is a proforma with a set of well sequenced questions relevant to the study objectives.
Schedules/self-administered questionnaires are more appropriate where the respondents are not well educated
than use of questionnaires (Kothari, 2004). Schedules are faster and ensure that data collected is complete
without omissions/unanswered questions. They also enable high response rates and enables combination of
different methods and personal contact possible (Babbie, 1994; Kumar, 2005). This study used schedules due
to the above reasons to collect data from the agricultural landowners.

Data Analysis and Presentation

The analysis of the drivers of ALS was carried out by presenting the typical drivers identified through
literature review to the study respondents to select the ones that are applicable in the area of study. The
respondents were then asked to use a numerical horizontal scale of 1 to 4 to rank the drivers in their order of
significance whereby 1=Not Important; 2=Less Important; 3=Important and 4=Very Important. Several
researchers have used and recommended a horizontal scale of 1 to 4 to represent two extremes; ‘not
important’ and ‘very important’, respectively. This scale is appropriate to avoid confusion on the respondents
(Alreck & Seattle, 1995; Masu, Murigu, Talukhaba, as cited in Kieti, 2015). In so doing, the significant
drivers of agricultural land subdivisions were identified and ranked in their order of importance.
Identification of significant drivers of agricultural land subdivision was done using the population mean score
and the critical Z-value. Firstly, to further analyze and determine significant drivers of the ALS, the study

used the population mean score ( X == 2.5),

52 kthat
g. is the middle point of the 1- 4 numerical score, to act as a
decision point, whereby any driver whose mean score was found to be below this point was considered to be
not significant and any driver with a mean score equal to or above this point was considered to be major or
significant driver in influencing ALS. Masu, Murigu, Talukhaba, as cited in Kieti, 2015 have used similar
approach. Thus, all the 25 drivers had two assumptions; that all the drivers are not significant in influencing
subdivisions of agricultural land and all the drivers are significant in determining the subdivisions of
agricultural land in the study area. A driver with a mean score of about 2.5 is considered to be average and

www.ijirk.com Page 200

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

important. At this point, the insignificant drivers were dropped for purposes of subsequent further data
analysis. Insignificant drivers are those with a mean score below 2.5.

Secondly, identification of significant drivers of agricultural land subdivision was further carried out using z
scores. Z-test is a statistical test used to determine whether two means are different. This test is best used
when the sample size is large (greater than 30 cases) because under the central limit theorem, as the number
of samples gets larger, the samples are considered to be almost normally distributed a requirement for z-test.
Besides, for each significance level, the z-test has a single critical value which makes it more convenient to
use than the t-test which has separate critical value for each sample size and is best suited for small sample
sizes (Kingoriah, 2004). Z-test was therefore used after setting the confidence level at 95%. According to
Masu, (as cited in Kieti, 2015), confidence levels help in reducing chances of identifying a particular
driver/factor to be significant when actually it is insignificant (Alpha error or type 1 error) or concluding that
a particular driver is insignificant while it is actually significant (Beta error or type II error) (Harper, as cited
in Kieti, 2015; Kingoriah, 2004).
Since the analysis of the drivers influencing subdivisions of agricultural land were only meant to provide
policy direction to the land managers, confidence level was set at 95% to identify significant drivers. Thus z-
test analysis provided a decisive way of identifying significant drivers of agricultural land subdivision in the
study area. Z-test was performed only on the average/moderate and the major significant drivers as analyzed
using the population mean score. The formula for computing z-value calculated for each average/moderate
and the major significant driver is shown below, as suggested by Kingoriah, 2004.

z = ( X -=µ)52
/ (δk /√n);
g.
Where z = Calculated z-value
X == Mean
52 k gscore
. for each driver
µ = Population mean score (x for this study is 2.5)
δ = Standard deviation
n = Sample size (n for this study is 357 cases)

Sirkin (as cited in Kieti, 2015) indicates that critical z-value at 95% probability/confidence level is 1.65. This
was used as a decision point, whereby the z-value calculated for each driver was then compared with the
critical z-value at 95% confidence level in one-tailed z-test. Where the z-value calculated for each driver was
greater than the critical z-value at 95% confidence level, the study was confident that the particular driver was
significant in influencing subdivision of agricultural land. Therefore, the critical z value (one-tailed test) at
95% confidence level is 1.65 hence any driver whose computed z value was found to be less than 1.65 was
decisively considered to be less significant in influencing agricultural land subdivisions and vice versa.

RESULTS
Significant Drivers of Agricultural Land Subdivisions in Kajiado County
Through literature review, 25 typical drivers were identified to be possible drivers behind the phenomenon of
agricultural land subdivisions in the study area. The drivers of agricultural land subdivisions in Kajiado
County were hypothesized to be a function of demand and supply of agricultural land, which in turn is
influenced by various physical, economic, demographic, sociocultural, institutional and political/legal factors.
In a nutshell, these are the drivers that either motivate or block farmers from subdividing their land into small
pieces. Significant drivers of agricultural land subdivision identified by use of population mean score are
presented in table 1 below.

www.ijirk.com Page 201

International Journal of Innovative Research and Knowledge ISSN-2213-1356

Table 1: Identification of Significant Drivers using Population Mean Score

Mean ScoreType of Driver
Driver Rating Percentage (%) ( X )= 52 k g.

Land inheritance practices Not important 0.0% Socio-cultural factor

Less important 0.5%

Important 10.7%

Very important 86.8%

Total = 3.8)
( X = 52 k g.

Individualization of titles Not important 0.5% Socio-cultural factor

Less important 0.0%

Important 7.6%

Very important 87.8%

Total = 3.7)
( X = 52 k g.

Price/Value of agricultural land Not important 6.6% Economic factor

Less important 4.6%

Important 34.7%

Very important 50.0%

( X ==3.2)
Total 52 k g.

Demand for urban housing Not important 6.1% Economic factor

Less important 10.7%

Important 32.7%

Very important 48.0%

Total = 3.2)
( X = 52 k g.

Future expectations on value ofNot important 10.2% Economic factor

agricultural land
Less important 6.6%

Important 22.4%

Very important 56.1%

( X ==3.2)
Total 52 k g.

www.ijirk.com Page 202

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

Customary land tenure systems Not important 5.1% Socio-cultural factor

Less important 4.6%

Important 30.5%

Very important 54.3%

Total = 3.2)
( X = 52 k g.

Off-farm income Not important 6.1% Economic factor

Less important 8.7%

Important 42.3%

Very important 40.8%

Total = 3.1)
( X = 52 k g.

Price/Value of urban land Not important 6.1% Economic factor

Less important 8.7%

Important 42.9%

Very important 38.3%

( X ==3.1)
Total 52 k g.

Rural population growth rate Not important 8.7% Demographic factor

Less important 6.7%

Important 34.4%

Very important 45.1%

Total = 3.1)
( X = 52 k g.

Urban population growth rate Not important 6.2% Demographic factor

Less important 11.3%

Important 51.8%

Very important 25.6%

( X ==2.9)
Total 52 k g.

Supply of agricultural land Not important 10.7% Economic factor

Less important 8.7%

Important 43.4%

www.ijirk.com Page 203

International Journal of Innovative Research and Knowledge ISSN-2213-1356

Very important 31.6%

Total = 2.8)
( X = 52 k g.

Proximity of agricultural land to servicesNot important 12.8% Natural/Physical factor

Less important 21.4%

Important 48.5%

Very important 15.3%

Total = 2.6)
( X = 52 k g.

Agricultural land use policies and laws Not important 20.8% Political/legal factor

Less important 17.8%

Important 35.0%

Very important 21.8%

Total = 2.5)
( X = 52 k g.

Source: Field Survey, 2016/2017

The factors that had a value rating with a mean score of less than 2.5 (population mean score) were considered
to be statistically less significant in agricultural land subdivisions and included: temperature ( X == 52 k g.
1.1),
topography ( X ==1.2),
52 krainfall
g. ( X ==1.3), = 52 kdemand
g. of agricultural finance / interest rates ( X =1.5),
52 kCost g.

for agricultural products ( X ==1.6),

52 kavailability
g. of agricultural finance / credit / capital ( X ==1.6),
52 kpublic
g.

participation in agricultural land development decision making process ( X ==1.6), g. income ( X =

52 kFarm 52 k, g.
=1.7)
= 52 klocal
Off-farm income ( X =1.7), g. land institutional technical capacity ( X ==1.9),
52 kquality
g. / fertility of land
( X =
= 2.0) g. commodification of land ( X =
52 kand 52 k g.
= 2.3).

The analysis of 25 factors by the use of population mean score failed to conclusively isolate the significant
factors influencing agricultural land subdivision in the study area. The reason was due to the fact that
confidence levels had not been put into consideration. This would help in lowering the errors that come with
identification of the significant factors. The errors can be reduced by using Z test for the statistical
significance on these factors to further identify the significant ones.

The drivers whose mean score rating were equal to or more than the population mean score ( X == 2.5) 52 kare
g.
significant in influencing subdivisions of agricultural land in the study area. Z test was done on the factors
which were significant and also average in terms of their influence on the agricultural land subdivision and are
shown in table 1 above. Table 2 below indicates the results for the Z test for each of the factors with mean
score rating equal to or more than the population mean score (2.5).

www.ijirk.com Page 204

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

Table 2: Identification of Significant drivers using Critical and Calculated Z values

Variable Critical Z value at Calculated Remarks
95% confidence level Z value
(one-tailed test)
Agricultural land inheritance 1.65 42.32 Factor is significant
practices
Individualization of titles 1.65 32.06 Factor is significant
Price of agricultural land 1.65 30.89 Factor is significant
Demand for urban housing 1.65 26.54 Factor is significant
Future expectations on the value of 1.65 25.89 Factor is significant
agricultural land
Customary land tenure systems 1.65 22.56 Factor is significant
Off-farm income 1.65 22.09 Factor is significant
Price of urban land 1.65 21.67 Factor is significant
Rural population growth rate 1.65 20.78 Factor is significant
Urban population growth rate 1.65 16.52 Factor is significant
Supply of agricultural land 1.65 1.254 Factor is not significant
Proximity of agricultural land to 1.65 0.89 Factor is not significant
services
Agricultural land use policies and 1.65 -1.53 Factor is not significant
laws
Source: Field Survey, 2016/2017

From the above analysis, it is clear that 10 out of the13 factors have the calculated Z values greater than the
critical Z value at 95% confidence level. The ten factors were found to be significant in influencing
agricultural land subdivisions.
A summary of responses from the Kajiado land officials indicates that the key drivers/challenges facing
administration and management of agricultural land in Kajiado County is inadequate institutional capacity
mainly in form of inadequate technical staff and finances to carry out proper and effective control of land
development. The local land control board, for instance, has no physical planner and has got only one land
surveyor. Similarly, the Kajiado County Land Management Board is facing similar technical staff deficiency
since it has no physical planner and has got only one land surveyor. This may affect negatively development
control efforts in the area. The chairman of the local land control board also lamented lack of clarity on the
role of the National Land Commission [NLC] or the newly established county land management boards and
the local land control boards, citing unnecessary interference from the County government.
In addition, the chairman of the local land control board, the local land institution under which agricultural
land subdivision falls directly, cited government’s limited role in control of private property development.
This challenge could be based on the erroneous assumption that ownership right, as guaranteed in the Kenyan
Constitution 2010, is equal to user rights hence private landowners are expected to use their private property
as they deem fit, without state interference (neoliberalism tendencies). These frustrations could also imply
inadequate land administration and management policy and regulatory guidelines.

DISCUSSION
The important drivers of agricultural land subdivision in the study area, as ranked by the respondents and z
test, are discussed below.

Land inheritance practices

Agricultural land inheritance practices was rated as the most significant driver of agricultural land subdivisions

in the study area ( X == 3.8,

52 kZg=. 42.32). Indeed, a total of 120 landowners (59%) had acquired their land

www.ijirk.com Page 205

International Journal of Innovative Research and Knowledge ISSN-2213-1356

through inheritance while 19 landowners (9%) reported to have subdivided their agricultural land among their
sons without selling a portion. This finding concurs with findings of a previous study by Mburu (2009) who
found land inhertaice practices to be key driver of agricultural land subdivision in Gatundu district, Kenya.
Thuo (2013) also established land inheritance to be an important driver of agricultural land subdivision in
Kiambu County. Similarly, the Kenya draft National Land Use Policy (NLUP) and National Spatial Plan
(NSP) have all identified land inheritance to be important driver of subdivision of agricultural land (GoK,
2016; 2016a). Elsewhere (Lambin et al., 2003; Scottish Government, 2009) socio-cultural factors have been
found to be important drivers in agricultural land subdivision. Subdivision of agricultural land for inheritance
purposes without enclosures/fencing off may not be a bad thing as it may not hinder large scale agricultural
production and movement of livestock. This is made possible by the close relationship of the family members
thus enabling land to be used as a whole and avoiding tragedy of the spatial anticommons (Heller, 1998).
Families are sometimes however not close-knit. Besides, private property rights grants landowners exclusive
right of access and use. Moreover, some agricultural land may be idle or owned by non-agricultural owners.
All these scenarios are likely to reduce cooperation in the community. Essentially, in the long run agricultural
production may be curtailed due to increased transaction costs/cost of agricultural production (for example the
time spent in informal negotiations and informal land management efforts).

Individualization of titles
Individualization of titles was ranked as the second most significant driver of agricultural land subdivisions

( X == 3.7,
52 kZg=. 32.06). A total of 198 landowners (98%) had reported to have private titles to their land. A
similar finding was determined by Ayonga (2008) who found private title deeds without restrictions on the use
of land to be important driver of land subdivision and land use conflicts in the peri-urban areas of Nairobi and
Kajiado County. Thuo (2013) also found individualization of land ownership to be an important factor in
influencing land subdivisions in urban fringes of Kiambu County while Olson et al., (2004) established private
titles to be key driver of agricultural land transformation in east Africa. Previously the agricultural land in
Kenya was mainly communally owned and issues of land subdivisions were not very common.
The land tenure in Kenya, however, has changed for most of the communal agricultural land to private
ownership, making it easier to transact with agricultural land. This has largely been facilitated by the
Registered Land Act, cap. 300 (repealed) which granted agricultural landowners freehold interest, with powers
to subdivide and transfer land. The government intended to foster economic development by ensuring that
agricultural landowners could access finance using their land as collateral. The private individual titles,
however, do not have provisions on the allowable minimum land sizes, a situation that has partly led to the
phenomenon of agricultural land subdivisions into small/uneconomic sizes. Therefore, private land rights in
agricultural land appear to have promoted neo-liberalism tendencies by promoting superiority of private
property, market forces, commodification of agricultural land and entrepreneurial character of landowners.

Price/value of agricultural land

The third most significant driver of agricultural land subdivisions is price/value of the agricultural land ( X == 52 k g.
3.2, Z = 30.89). This is an economic factor which signifies presence of neo-liberalism through market forces of
demand for agricultural land. Some previous studies have found a positive correlation between price of
agricultural land and rate of subdivision (Chazan & Cotter, 2001; Lee, 1999; Henry et al. 2012; Olson et al.,
2004). Locally, Thuo (2013) has determined price of agricultural land to be important in influencing
subdivision of agricultural land, especially in the urban fringes of Nairobi City and Kiambu County.
Nkedianye et al. (2009) also found price of agricultural land to be an important driver in agricultural land
subdivision in some parts of Kitengela area near the Nairobi National Park. Data from the estate agents in the

www.ijirk.com Page 206

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

study area revealed that the average price of the agricultural land in the study area has been increasing over the
years from an average of about 1 million per ha in 2006 to an average of 7.5 million per ha in 2015, translating
to approximately 650% increase over a period of 10 years.
On one hand this positive trend in agricultural land value is likely to entice the landowners to subdivide and
sell off portions of their agricultural land. The increasing average agricultural land value coupled with the
perception that there is plenty supply of land in the study area could be motivating farmers to subdivide and
sell off part of their land. These trends may also create favourable environment for private individuals to
speculate on agricultural land as land investors expect to reap higher prices in the future. Lack of an effective
taxation instruments and a regulation on the minimum land holding acreages are likely to further fuel these
trends over time. On the other hand, however, high price/value of agricultural land is likely to restrain
expansion of agricultural production since land would be expensive to acquire in the future. Economies of
scale would thus be absent or expensive to attain for the farmers who would wish to expand their operations.

Demand for urban housing

Demand for urban housing was ranked as the fourth most significant driver of agricultural land subdivisions in

the study area ( X == 3.2,

52 kZg=. 26.54). Previous studies which have identified this factor to be important in
agricultural land subdivision elsewhere include Lee, (1999), McDonagh, (1997); and Thuo, (2013). Demand
for urban housing could signify expansion of urban areas, pointing to a possibility of application of the urban
growth theories such as concentric rings, sector and multiple nuclei or residential location theory.
This driver, however, could easily be confused with speculation on the agricultural land. In-depth interviews
with key informants and some landowners revealed that some developers were buying land from the farmers
then subdividing into small sub-plots to sell to private individuals for housing development yet the area may
be lacking basic services to support real estate development. From personal observations, some developers
have constructed earth access roads to hoodwink land buyers that the agricultural land is ripe for development.
From observations it was evident, however, that after the small sub-plots are sold no housing development
takes place, possibly due to lack of supporting services. This leaves the small sub-plots ‘undeveloped’ for a
long time thus taking agricultural land out of production prematurely.
Besides, all agricultural land should not be considered as future urban land. Agricultural land will continue to
be important to both rural and urban dwellers thus vital agricultural land should be protected from subdivision
into small sizes and subsequent conversion into urban use.

Future expectations on the value of agricultural land

Demand for urban housing and the increasing average price/value of agricultural land could explain why the
respondents ranked future expectations on the value of agricultural land as the fifth most significant driver of

agricultural land subdivisions ( X ==3.2,

52 Zk g=.25.89). This finding is similar to some previous studies (see
Ayonga, 2008; Chazan & Cotter, 2001; Henry et al., 2012; Thuo, 2013). This driver, an economic factor, again
shows application of neo-liberalism theory since change in market prices and values indicate market forces to
be important market regulator.
Land investors are likely to influence high demand for agricultural land in the hope of reaping higher prices in
the future. This seems to have created a ‘buy-and-wait’ trend in the agricultural land, making agricultural land
not to be used productively and sustainably as per the prescriptions of the Kenya Constitution and the National
Land Policy (GoK, 2010; 2009). This is because once agricultural land is subdivided into bare minimum sub-
plots it is likely to influence future land use patterns and may hinder agricultural production or proper future
urban development by creating incompatible land uses (Ayonga, 2008).

www.ijirk.com Page 207

International Journal of Innovative Research and Knowledge ISSN-2213-1356

The ‘buy-and-wait’ trends in agricultural land appear to assume that residential development is the only urban
development necessary both currently and in the future. Through personal observations, for instance, there
were no many ‘industrial or institutional land for sale’ adverts in the study area. This may create problems in
the future as urbanization engulfs the rural areas by making it difficult to plan for the various urban land uses.
This has been evident in the peri-urban areas in Kenya (Ayonga, 2008).

Customary land tenure systems

Customary land tenure systems (men owning land on behalf of family) was ranked as the sixth most

significant driver of agricultural land subdivisions in the study area ( X ==3.2,

52 Zk g=. 22.56). Olson et al., (2004)
& Nkedianye et al., 2009 have found customary land tenure systems to be important drivers of agricultural
land subdivision in east Africa. This was mainly so because most of the decisions regarding land, including
ownership and transfer of agricultural land, were found to be a prerogative of the male gender. Out of the 203
agricultural land parcels surveyed, 187 (92%) were registered in the name of the male head of the household
and only 6 parcels (3%) were registered under the female head of family, mostly the female headed
households.
Interviews with the land officials revealed that sometimes the men, under whom the agricultural land is
registered, make decision to subdivide and sell/transfer portions of their land without consulting their wives
and children. This allegation was further verified by existence of cases and caveats/prohibitions on land sales
at the Kajiado lands office. Customary land tenure system is a sociocultural factor which could be used by
landowners to achieve economic self-gratification, as per the dictates of the neo-liberalism theory. Besides, a
total of 19 landowners (9%) reported to have subdivided their agricultural land among their sons.

Income of the agricultural landowner

Poverty/per capita income was ranked as the seventh most significant driver of agricultural land subdivisions

in the study area ( X ==3.1,

52 Z
k g=. 22.09). The respondents reported that sometimes they sell agricultural land to
meet their immediate domestic needs like buying clothes and paying school fees for their children. This
finding concurs with some previous research (see Chazan & Cotter, 2001; Henry et al., 2012; Nkedianye et al.,
2009; Thuo, 2013). The government of Kenya has also taken income of the landowner to be an important
driver of agricultural land subdivision (GoK, 2016; 2016a). The total average annual income bracket (from
both farm and off-farm ventures) for majority of the landowners (51%) were reported to be between Kshs.
201,000 – 250,000, mainly from agricultural activities.
This finding suggests that majority of the households could be living below the international poverty line of
US$ 1.90 per day, considering the average size of the household of 5 members. According to the County
Government of Kajiado (2013), and the Kenya National Bureau of Statistics [KNBS] & Society for Internal
Development [SID], (2013), almost 50% of the Kajiado County’s population lives below poverty line. Thus,
agricultural landowners are likely to subdivide and sell portions of their land as a source of livelihood, yet
majority derive their livelihood from the agricultural land through agricultural activities. Given the tendencies
of neo-liberalists, market forces are bound to control the property market where the property owners are poor.

Price/value of urban land

Price of urban land was ranked as the eighth most significant driver of subdivisions of agricultural land in the

study area ( X ==3.1,

52 kZ g=. 21.67). Ayonga (2008) established that high price of urban land in Nairobi City has
forced urban dwellers to look for cheaper land in the urban fringes of Nairobi and Kajiado County. Similarly,
Thuo (2013) identified increase in price of urban land in Nairobi City to be an important driver of land

www.ijirk.com Page 208

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

subdivision in Kiambu County. Elsewhere (see Chazan & Cotter, 2001; GoK, 2016; 2016a; Lee, 1999;
McDonagh, 1997; Olson et al., 2004) previous studies have found value/price of urban land to be important
driver of subdivision of agricultural land.
This finding could be explained by the rising prices/value of urban land in Nairobi City (the largest and capital
city/urban area of Kenya) and the surrounding/satellite urban areas. According to a property price index
published by the HassConsult Limited (2016), the average price per acre of urban land in Nairobi increased
from Kshs. 74.87 million in December, 2007 to Kshs. 441.32 million/ha in June, 2016 translating to an
increment of 7.12 fold over the stated period.
Similarly, the price of urban land in the satellite urban areas, including Kitengela Township, went up from an
average of Kshs. 5.93 million/ha to 39.04 million (6.47 fold) per ha over the same period. The increasing
prices of urban land is making land within the urban areas to be way beyond the reach of the low and middle
urban income earners, forcing them to look for affordable land in the surrounding rural areas hence fuelling
subdivisions of agricultural land. Market land price/value is a function of the market forces of demand and
supply. Therefore, price of urban land is an indicator of neo-liberalism tendencies.

Rural population growth rate

Rural population growth rate ( X ==3.1,
52 kZ g=. 20.78) was ranked the ninth most important factor in influencing
subdivision of agricultural land. This finding coincides with Jayne & Muyanga’s (2012) who found rural
population growth rate to be significant driver of agricultural land subdivision in the densely populated areas
of central parts of Kenya. Lambin et al., (2003) and Olson et al., (2004) have also determined rural population
growth rate to be significant driver of agricultural land transformation.
Positive growth of rural population coupled with agricultural land inheritance practices is likely to fuel
subdivisions of agricultural land in the study area. Indeed, this is likely to be the future trend, if policy
interventions are not put in place by the rural land managers. The population of Kajiado County, as well as the
Kenyan population, has been growing over the years. The current population growth rate in Kajiado County is
at 5.5% per annum which is almost twice higher than the national’s population growth rate of 2.9% per annum
(County Government of Kajiado, 2013). Rural population growth rate is a demographic factor which is likely
to be influenced by the economic and other factors.

Urban population growth rate

Urban population growth rate was ranked as the last/tenth most significant driver of agricultural land

subdivisions in the study area ( X == 2.9,

52 kZg=. 16.52). This driver signifies application of the urban growth
theories which postulate that urban areas will always expand into the surrounding rural areas. Kenyan urban
population has changed from 7% in 1960 to 26% in 2016 with an annual urban population growth rate of
4.2%. Kajiado County is, however, mainly rural with 76.2% of its population being rural while urban
population stands only at about 23.8% up from 19.8% in 2009. This is an indication that Kenya is urbanizing
over the years and urban land uses are likely to push away agricultural land use through agricultural land
subdivisions and subsequent conversions. Increasing urban land prices appears to have aggravated the
situation.
Urban population growth rate has been assumed to be an obvious and important driver of subdivision of
agricultural land in many parts of Kenya (GoK, 2016; 2016a; Lambin et al., 2003; Olson et al., 2004).
Previous studies in New Zealand have also found urban population growth rate to be significant driver of
subdivision of agricultural land (Henry et al. 2012; Lee, 1999). The finding of this study has however shown
that even though this is a significant driver in the study area it is not the most important. This could be due to

www.ijirk.com Page 209

International Journal of Innovative Research and Knowledge ISSN-2213-1356

the location of the study area which focused on the rural areas only. It appears thus urban population growth
rate may be most significant in rural/agricultural lands near urban areas but not in remote rural areas.

CONCLUSIONS AND RECOMMENDATIONS

The paper has revealed that socio-cultural and economic factors are the most significant drivers of agricultural
land subdivisions in the study area. Based on the first five most significant drivers of agricultural land
subdivisions, the paper proposes the following recommendations:
The national and county governments should thus put in place appropriate and clear policy, legal and
institutional frameworks to prescribe allowable minimum economical/optimal agricultural land sizes in
various agro climatic zones in Kenya.
Individualization of title was cited as the second most significant driver of agricultural land subdivisions in the
study area. Since it is difficult and unnecessary to reverse agricultural land privatization trends, in any case it
is assumed to be a cure of the tragedy of the commons, individual titles should have restrictions on the
minimum allowable sizes depending on the location and use of the land. Where agricultural land is used for
extensive livestock production system, for example, large tracts should be encouraged unless and until the
owner wants to change the user.
Government and private sector should support agricultural enterprises, including livestock production, to
make agricultural activities economically viable and competitive to reduce the influence of attractive
agricultural land prices, the third most significant driver of agricultural land subdivisions. This policy measure
would in turn raise the income of the landowners and help them invest in agricultural production.
Appropriate property taxation policies should also be used to discourage speculation on agricultural land. This
intervention would discourage idle agricultural land thus avoiding land fragmentation. The allowable
minimum economical agricultural land sizes in various agro climatic zones would also go a long way in this
endeavor.
To reduce the influence of demand for urban housing on the agricultural land subdivisions, urban
revitalization strategies and urban land banking should be encouraged as means of providing urban housing in
the urban areas. Agricultural land should only be developed once basic services are provided thus public
facilities requirement ordinances may go a long way in achieving this quest.

REFERENCES
Alreck, P. L. & Settle, R. B. (1995). The survey research handbook (2nd ed.). Chicago: Irwin.

Ayonga, N. J. (2008). Land use conflicts and in-optimal spatial patterns in peri-urban areas of the City of Nairobi, Kenya
(Unpublished PhD Thesis). University of Nairobi, Nairobi.

Babbie, E. (1994). The practice of social research. (7th ed.). California: Wadsworth Publishing Company.

Chazan, D. A. & Cotter, A. A. (2001). Evaluating the impacts of proposed land conversion: A tool for local decision-making.
Michigan: Center for Sustainable Systems.

County Government of Kajiado. (2013). Kajiado County Integrated Development Plan 2013 – 2017. Nairobi: Government Printer.

www.ijirk.com Page 210

International Journal of Innovative Research and Knowledge Volume-3 Issue-6, June-2018

Government of Kenya. (2009). The National Land Policy. Nairobi: Government Printers.

Government of Kenya. (2010). The Constitution of Kenya. Nairobi: Government Printers.

Government of Kenya. (2016). The Draft National Land Use Policy. Nairobi: Government Printers.

Government of Kenya. (2016a). National Spatial Plan (2015-2045). Nairobi: Government Printers.

HassConsult Limited. (2016). The Hass property index: Land price index quarter two report 2016. Nairobi: HassConsult Limited.

Heller, M. A. (1998). The tragedy of the anticommons: Property in the transition from Marx to markets. Harvard Law Review.
111(3), 621-688.

Henry, M., Rusell, M. & Wood, B. (2012). Drivers of Change: Farmer Experiences and Perceptions of Subdivision, Working
Paper No. 2. New Zealand: Living Lab.

Jayne, T.S. & Muyanga, M. (2012). Land constraints in Kenya’s densely populated areas: Implications for food policy and
institutional reforms. Food Security. 4(3), 399-421.

Jiang, L., Deng, X. & Seto, K. C. (2013). The impact of urban expansion on agricultural land use intensity in China. Land Use
Policy. 35, 33–39.

Kelleher, F.M., Chant, J. J. & Johnson, N.L. (1998). Impact of rural subdivision on agriculture: A report for the rural industries
research and development corporation. Sydney: University of Western Sydney.

Kenya National Bureau of Statistics & Society for International Development. (2013). Exploring Kenya’s inequality: Pulling apart
or pulling together? Nairobi: Ascent Limited.

Kenya National Bureau of Statistics. (2010). The 2009 Kenya Population and Housing Census. Nairobi: Government Printers.

Kieti, R. M. (2015). Urban housing affordability in Kenya. A case study of the mortgage housing sector in Nairobi (Unpublished
PhD Thesis). University of Nairobi, Nairobi.

King’oriah, G. K. (2004). Fundamentals of Applied Statistics. Nairobi: Jomo Kenyatta Foundation.

Koissaba, B. (2015). Elusive justice: The Maasai contestestation of land appropriation in Kenya; a historical and contemporary
perspective. Washington: Centre for World Indigenous Studies.

Kothari, C. R. (2004). Research methodology: Methods and techniques (2nd ed.). Jaipur: New Age International Publishers.

Kumar, R. (2005). Research methodology: A step-by-step guide for beginners (2nd ed.). London: SAGE Publications Limited.

Lambin, E., F., Helmut, J. G., & Erika, L. (2003). Dynamics of Land-Use and Land-Cover Change in Tropical Regions. The
Annual Review of Environment and Resources. 28, 205-241.

Lee, K. M. (1999). Rural subdivision: A case study of farmers’ accounts of rural subdivision in the Selwyn district (unpublished
Master’s Thesis). Lincoln University, Lincoln.

Liu, Y. L., Gan, H., Jay, G. & Deng, X. (2004). The causes and environmental effects of land use conversion during agricultural
restructuring in Northeast China. Journal of Geographical Sciences. 14(4), 488-494.

Lo, C.P. & Yang, X. (2002). Drivers of land-use/land-cover changes and dynamic modeling for the Atlanta, Georgia metropolitan
area. Photogrammetric Engineering & Remote Sensing. 68(10), 1073-1082.

www.ijirk.com Page 211

 International Journal of Innovative Research and Knowledge ISSN-2213-1356

MacDonagh, J. (1997). Theories of urban land use and their applications to the Christchurch property market. Lincoln: Lincoln
University.

Mburu, F. (2009). Land consolidation for sustainable rural development in Kenya: Examining the role of Kenyan partible
succession in land fragmentation. A case study of the larger Gatundu District (master’s Thesis). Institute of Geodesy, GIS and Land
Management, München.

Mugenda, O. M. & Mugenda, A.G. (1999). Research methods: Quantitative and qualitative approaches. Nairobi: Acts Press.

Nachmias, C. F. & Nachmias, D. (2000). Research methods in the social sciences (6th ed.). New York: Worth Publishers.

Nkedianye, D., Radeny, M., Kristjanson, P. & Herrero, M. (2009). Assessing returns to land and changing livelihoods strategies in
Kitengela. Nairobi: International Livestock Research Institute.

Olson, J. M., Misana, S., Campbell, D. J., Mbonile, M. & Mugisha, S. (2004). The spatial patterns and root causes of land use
change in East Africa, Land Use Change Impacts and Dynamics (LUCID) Project Working Paper #47. Nairobi: International
Livestock Research Institute.

Scottish Government. (2009). Changing land use in rural Scotland – Drivers and decision-making: Rural Land Use Study Project
1. Edinburgh: Government Printers.

Society for International Development. (2006). Readings on inequality in Kenya: sectoral dynamics and perspectives. Nairobi:
Regal Press Kenya Limited.

Syagga, P. M. & Kimuyu, J. S. (2016). Report of desk study on minimum and maximum land holding acreage in Kenya: A Guide to
Policy Makers (Unpublished study report). University of Nairobi, Nairobi.

The Centre for Land, Economy and Rights of Women. (2006). Land policies in Sub Saharan Africa. Nairobi: The Centre for Land,
Economy and Rights of Women.

Thuo, A. D. M. (2013). Impacts of urbanization on land use planning, livelihood and environment in the Nairobi rural-urban fringe,
Kenya. International Journal of Scientific & Technology Research. 2(7), 70-79.

www.ijirk.com Page 212

View publication stats