Can Mobile Communication Technologies EnhanceKnowledge and Technology Transfer for ImprovedAgricultural Productivity in Developing Nations? APreliminary Macro-Economic Assessment in Kenya

Item Type text; Electronic Thesis

Authors Vancel, James Hugh

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 18/05/2018 18:25:34

Link to Item http://hdl.handle.net/10150/244839

1

CAN MOBILE COMMUNICATION TECHNOLOGIES ENHANCE

KNOWLEDGE AND TECHNOLOGY TRANSFER FOR IMPROVED

AGRICULTURAL PRODUCTIVITY IN DEVELOPING NATIONS?

A PRELIMINARY MACRO-ECONOMIC ASSESSMENT IN KENYA

By

James Hugh Vancel

______________________

A Thesis Submitted to the Honors College

In Partial Fulfillment of the Bachelor’s degree

in

Interdisciplinary Studies/International Studies

THE UNIVERSITY OF ARIZONA

MAY 2012

Approved By:

_______________________________________

Barron J. Orr, Ph.D. Office of Arid Lands Studies School of Natural Resources and the Environment International Studies

2

Table of Contents Abstract ............................................................................................................................... 4

1. Introduction ............................................................................................................... 5

1.1 The Problem .......................................................................................................... 5 1.2 Agricultural History of Kenya............................................................................... 6 1.3 Agricultural Productivity in Kenya ....................................................................... 9 1.4 Agricultural Extension in Kenya ......................................................................... 10 1.5 The Relationship between Agricultural Technology and Productivity ............... 11 1.6 Mobile Communication Technology in Agriculture ........................................... 12 1.7 Mobile Communication Technologies and Agricultural Productivity? .............. 13

2. Methods................................................................................................................... 15

2.1 Study Area ........................................................................................................... 15 2.1.1 Relevant History ............................................................................................ 18

2.2 Public Investments in Technology and Agricultural GDP .................................. 19 2.2.1 Analytic Framework ...................................................................................... 19 2.2.2 National Macro-economic Data ..................................................................... 20 2.2.3 Model Design ................................................................................................. 21

2.3 Case Study Analysis of the Extension Services Dimension ............................... 24 2.4 Assessing Mobile Phone Market Penetration in Kenya ...................................... 27 2.5 Assessing the Capabilities of New Mobile Services ........................................... 27

2.5.1 Awaaz De ....................................................................................................... 27 2.5.2 FarmFox ......................................................................................................... 28 2.5.3 Community Knowledge Workers .................................................................. 28

3. Results ..................................................................................................................... 30

3.1 National Level Macro-econometric Model ......................................................... 30 3.1.1 Statistical Summary ....................................................................................... 30 3.1.2 Regression Results ......................................................................................... 31

3.2 Extension Services Case Study Review .............................................................. 32 3.3 Mobile Phone Sector Trends ............................................................................... 36 3.4 The Potential of New Mobile Phone Services .................................................... 38

4. Discussion ............................................................................................................... 39

5. Recommendations ................................................................................................... 43

6. Conclusions ............................................................................................................. 44

7. Appendix: Case Study Analysis.............................................................................. 45

8. References Cited ..................................................................................................... 47

3

Table of Figures

Figure 1. Agricultural GDP growth from 1963-1988 in Kenya...................................................... 7 Figure 2. Total debt outstanding for the period 1970-2007. ........................................................... 8 Figure 3. Agricultural spending as a percentage of total government spending following implementation of SAPs in 1988. ................................................................................................... 8 Figure 4. Real agricultural GDP from 1980 to 2005. ................................................................... 10 Figure 5. Agro-climatic zone map of Kenya. ............................................................................... 17 Figure 6. Total research investment stock for a given year t. ....................................................... 24 Figure 7. Gross mobile phone subscriptions in the developing and developed world in 2000 and 2011............................................................................................................................................... 36 Figure 8. Mobile phone subscriptions per 100 inhabitants in East Africa, 2000 to 2010. ............ 37 Figure 9. Mobile money users in Africa, percentage of total population, 2011. .......................... 38 Figure 10. Indexed seed input levels 1980-2005 (1982=100) ...................................................... 41

Table of Tables

Table 1. Availability of variables of interest on national and provincial levels for period of 1980-2005............................................................................................................................................... 20 Table 2. Cobb-Douglass production function variable descriptions. Data provided for period 1980-2005 unless otherwise stipulated. ........................................................................................ 22 Table 3. Case Studies Overview ................................................................................................... 26 Table 4. Agricultural GDP and input quantities in Kenya, 1980-2005 (n=26)............................. 30 Table 5. Correlation of agricultural GDP with variables of interest. ............................................ 32 Table 6. Econometric results obtained in the case studies evaluated describing the relationship between extension services and agricultural production. ............................................................. 32 Table 7. Challenges facing extension services. ............................................................................ 34 Table 8. Recommendations for extension services. ...................................................................... 35 Table 9. Capabilities of mobile services. ...................................................................................... 39

Table of Equations

Equation 1. Cobb Douglass production estimation for agricultural GDP in Kenya. .................... 21 Equation 2. Agricultural research stock for a given year t. .......................................................... 23

4

Abstract Agriculture is central to the economy of Kenya, but as the population dependent on the

same land grows, farmers have been compelled to develop more intensive practices to produce similar yields as in the past. Investments in agricultural research and technology facilitate this intensification and have historically resulted in substantial gains. However, efforts to promote effective transfer of new technologies and methodologies have been stifled by financial challenges to traditional agricultural extension delivery approaches. Extension services involve significant face-to-face contact in farmer fields, through farmer field schools, or at regional events. However with rising costs, the demand for more mobile and adaptable mediums is apparent. This study examines the potential for new methods of knowledge and technology transfer and diffusion of innovation within Kenya, most notably the use of mobile phone services. In order to frame the potential of such services, a national level macro-econometric analysis was undertaken to assess the effects of investments in agricultural research on agricultural gross domestic product. These results were then overlaid on an analysis of 14 case studies examining the effects and challenges of agricultural extension in East Africa. Three potential services, Awaaz De, FarmFox, and Community Knowledge Workers (CKW), were then assessed on a capabilities standard in the context of the rise of the mobile communication markets in Kenya for potential application to extension services. The results indicate that agricultural research is strongly correlated with growth in agricultural GDP and that effective extension services have strongly facilitated that in the past. However, limited flexibility, lack of farmer input, and high costs limit that potential contribution to growth. Mobile communication services may be part of the solution; capabilities of Awaaz De, FarmFox, and CKW are promising, but further research is needed to accurately recommend a service for Kenya’s extension services.

5

1. Introduction 1.1 The Problem

Agriculture has been central to the social and economic infrastructure of Kenya over

centuries past and continues to play a major role in the contemporary economy of Kenya.

Agriculture contributes 22% of GDP and 72% of the population is engaged to some extent in

agricultural production or marketing (World Bank 2012a). However, with population growth has

come greater and greater demand for the same plots of land farmed in the past, resulting in

decreasing marginal returns to productivity and diminishing incomes. In other parts of the world

the scenario of increasing population densities have led to what Boserup (1965; 1981) termed a

spontaneous movement towards agricultural intensification (increased average inputs of labor or

capital in order to increase the value of output) in response to shortened fallow periods and

declining yields. New technologies demand more farm labor and investment in land that leads to

higher crop yields and household income. In a more techno-centric approach, Simon suggests

this process is driven by invention, a function of population density where more people produce

more ideas, which in turn drives growth (Simon 1986). Technological advancement has the

potential to promote positive growth in productivity, yet a great deal of that relies on the

effective diffusion of information through agricultural extension services. At present, extension

services are costly and challenging to implement effectively, but can result in great returns to

investments (Evenson and Mwabu 1998). However due to their high costs , governments across

Sub-Saharan Africa have been cutting funding to extension services, resulting in significant

problems for rural farmers. Using Kenya as an example, this research explores the potential for

the new wave of mobile communication technologies to make up some of the difference,

resulting in a net positive affect on agricultural production.

6

1.2 Agricultural History of Kenya

The importance of agriculture in Kenyan society, particularly since Independence in

1963, is fundamental to understanding contemporary conditions. Kenya experienced strong

economic growth immediately following independence, however the legacy of the state-

controlled colonial economic model during the colonial period that disenfranchised the majority

of the population remained. Determining how to address this led initially to a struggle between

socialist and capitalist philosophies; the new government of Kenya ultimately settled upon a

market-based mixed economy espoused by Tom Mboya and other others which placed a large

emphasis on investment in the agriculture sector (Republic of Kenya 1965).

After a period of significant fluctuations immediately following independence, the

growth trend was strongly positive and was relatively stable by 1988 (Figure 1). These gains

were attributed to an overall expansion of land under cultivation, policy reforms enabling under-

utilized land to be harnessed once again, an improved supply of inputs, and heavily subsidized

extension services (Okech et al. 1996). The growth was largely in the small-scale farming sector,

which went from contributing just 22% of the marketed agricultural production in 1963 to 50%

in 1967 (Inukai 1974).

In comparison with other sub-Saharan African nations, the Kenyan government has

played a minimal role in agricultural interventions, primarily limited to facilitating rather than

directing the growth of the agricultural sector (Johnston 1989; Alila and Atieno 2006). The elites

of Kenya during this timeframe staked a large claim in the agricultural sector, channeling private

investment into agriculture, creating a policy environment friendly to large producers (Lofchie

1989).

7

Figure 1. Agricultural GDP growth from 1963-1988 in Kenya Source: World Bank 2012a

Following a long series of external debt accumulation due to drought, rising oil prices,

and a global recession during the 1970’s and early 1980’s (Figure 2), Kenya was subjected to a

series of Structural Adjustment Programs (SAP) recommended by the World Bank and

International Monetary Fund (IMF). These programs were characterized by the liberalization of

prices and marketing systems, reforms in international trade and financial regulations,

government budget rationalization, and civil service reforms (Kenya Central Bureau of Statistics

1997). Government expenditures subsequently shifted towards attempting to balance the budget.

In the period of 1964-1969 servicing the debt constituted 8.3% of the Kenyan National budget;

this number jumped to 31.8% in the 1990-1991 period (Rono 2002). This shift in realities left the

Ministry of Agriculture with significantly less funding, resulting in major cuts in services (Figure

3). This is a growing concern considering the clear link between research, extension, agricultural

intensification, productivity, and household food security in Kenya (Sutherland et al. 1999).

-15

-10

-5

0

5

10

15

20

25Ag

ricul

tura

l GDP

Gro

wth

(%)

8

0

2

4

6

8

10

12

14

0 5 10 15 20 25

% o

f Tot

al G

over

nmen

t Sp

endi

ng

Years Since Adoption of Structural Adjustment Programs

Figure 2. Total debt outstanding for the period 1970-2007. Source: World Bank 2012a

Figure 3. Agricultural spending as a percentage of total government spending following implementation of SAPs in 1988. Source: Republic of Kenya and KIPPRA 2009

0

1000

2000

3000

4000

5000

6000

7000

8000To

tal D

ebt O

usta

ndin

g U

SD 2

010

(Mill

ions

)

9

1.3 Agricultural Productivity in Kenya

Kenya has experienced annual population growth rates ranging from 2 to 4% since

independence (World Bank 2012b). Kenya has also experienced agricultural intensification over

the past thirty years, with agricultural labor climbing from 5.71 million people in 1981 to 12.6

million in 2007 (IFPRI 2012). Traditional economic theory suggests that increased labor will

typically result in increased production. However, where arable land remains relatively constant,

and there are few alternatives to agriculture for employment, the agricultural sector population

density will experience significant increases as population grows. Boserup (1965) posited in her

seminal work The Conditions of Agricultural Growth that as population increases, society will

rise to meet production demands through agricultural innovations which ultimately intensify the

use of land. Kenya has followed this model with significant progress in terms of agricultural

productivity, however the surplus in available agricultural labor has resulted in smaller and

smaller landholdings through family inheritance sub-divisions. Advances such as high yield seed

varieties, fertilizers, and pesticides have facilitated the intensification of agriculture, but there are

significant diminishing marginal returns and high costs associated with these inputs (Tilman et

al. 2002). Furthermore, effective implementation of new farming practices and associated

technologies requires extensive knowledge and resources which are not always available. The

relative stagnation in the growth of the agricultural sector in Kenya since 1980 suggests that high

input prices and poor implementation practices may have prevented effective long-term

agricultural productivity gains (Figure 4).

10

Figure 4. Real agricultural GDP from 1980 to 2005. Source: World Bank 2012a

Low or negative growth in the agricultural sector effects the livelihoods of the 72% of the

population engaged in farming, reinforcing poverty and negatively affecting other sectors of the

economy (Thirtle et al. 2001). The substantial role agriculture plays in Kenya’s economy makes

agricultural productivity an extremely important issue in the development of Kenya’s future.

1.4 Agricultural Extension in Kenya

Technological innovations in agriculture can promote gains in productivity, however for

benefit of these innovations to be fully realized, effective knowledge and technology transfer and

diffusion to the farming population is necessary. The primary mechanism which facilitates

agricultural knowledge and technology transfer in Kenya is agricultural extension. Following

independence, agricultural extension services were given a high priority. During this era, East

Africa as a regional bloc was spending more than double industrialized and semi-industrialized

countries on extension services, relative to the value of agricultural Gross Domestic Product

(GDP) (Schwartz and Kampen 1992). However, since the implementation of the SAPs of the

World Bank and the IMF in the late 1980’s, investment in agricultural extension services has

0

1000

2000

3000

4000

5000

6000U

SD 2

012

(Mill

ions

)

11

been reduced in an attempt to meet new budget realities (World Bank OED 1990). In a

household survey, the World Bank found that only 2% of all farmers regularly met with

extension agents at least once a month in their own or a neighbor’s field (World Bank OED

1990), further evidence that the role of extension services was diminishing in Kenya. Extension

services have since been perceived as top-down and inflexible and strongly linked with the

limited growth of the agricultural sector (Republic of Kenya 2005).

1.5 The Relationship between Agricultural Technology and Productivity

There has been considerable past research that suggests investment in agricultural

technology promotes agricultural productivity. Extensive literature concerning the Green

Revolution in Asia and elsewhere has pointed to substantial returns on investments in

agricultural technology (e.g., Foster and Rosenzweig 1996; Huffman and Evenson 2006; Alston

2010; de Janvry and Sadoulet 2002). This body of work explores the depth of direct and indirect

effects agricultural technology has had on global poverty in South Asia, Africa, and Latin

America with varying levels of success. For example, Asia stands out for high levels of success,

as documented by Jin et al. (2002) who assessed the effects of China’s national investments in

agricultural technology on total factor productivity, concluding that China’s growth in wheat,

maize, and rice production was mostly attributable to the investments in agricultural

technologies. Agricultural research in particular has played a major role; in the case of China, it

led to a significant share (20%) of increases in agricultural productivity in China since 1965 (Fan

and Pardey 1997).

While there is considerable literature connecting public investments in agricultural

research to productivity gains in Asia, the returns in Africa are not as clear. In the first decade

after the transition to independence in most African nations (which occurred for many countries

12

in the early 1960s) agricultural growth was strongly correlated with initial investments in

agricultural technologies, particularly high yield varieties of maize, helping agricultural GDP

grow significantly (Bigsten and Collier 1995). Since that time there has been very limited

documentation on the impact of modern public investments in agricultural research and

technology on the macro-level. Isolated breakthroughs have led to positive returns to agricultural

investment, but on aggregate there is no strong indication that unequivocally demonstrates

whether public investment in agricultural technologies have resulted in increased productivity in

Africa.

1.6 Mobile Communication Technology in Agriculture

Mobile communication is increasingly becoming viewed as an indispensable necessity in

many developing countries (Kriem 2009). Africa is experiencing exceptional growth in the

mobile phone sector, leading some to suggest a narrowing of the digital divide through new

digital opportunities (Kyem and LeMarie 2006). Mobile phones have begun to infiltrate markets

in and improve market efficiencies (Jensen 2007; Donner 2008; Aker and Mbiti 2010). The

associated increased access to information has changed the capacity for individuals and small

businesses to improve income and expand markets (Salia et al. 2011), though not necessarily

with a corresponding improvement in labor productivity (Chowdhury 2006). There are signs that

impacts are being felt across the entire rural agricultural value chain as well as the fundamental

livelihood construct of small holders, accompanied by the potential for increased opportunities

and reduced risks that come from more consistent and rapid access to information and the

capacity for information sharing (Furuholt and Matotay 2011).

In developing nations, communication between extension workers and farmers is

essential, but declining resources make this problematic. Mobile communication has been

13

demonstrated to be an effective means to help extension workers bridge this gap through more

reliable and rapid communication options with farmers ranging from product and input price

fluctuations to sharing advice on addressing immediate farming practice challenges (Dey et al.

2011). Mobile communication services have been developed to allow farmers to access up-to-

date market information to ensure that they are selling their product at a reasonable price subject

to the prices of other urban centers. Other services help microfinance borrowers to conveniently

receive and repay loans, such as the M*PESA (mobile money) system in Kenya. It has facilitated

growth of the agricultural sector by providing fast and effective means of transferring money

from any location in the nation, making it increasingly easy for farmers to access or send

remittances nationally.

1.7 Mobile Communication Technologies and Agricultural Productivity?

The potential benefits accruing from the mobile communication boom in Africa seem to

suggest that the agricultural sector stands to benefit. Access to new markets requires access to

relevant information and the knowledge associated with innovations that may be applicable to

local farmers – something that requires active and regular information and knowledge transfer

(Sutherland et al. 1999). In order to understand the potential for mobile communications to

enhance knowledge and technology transfer, more in depth study is required. Kenya is ideal for

such a study because of its historical agricultural investment trends, the initial high investment

and then diminishing support for extension services, and its extraordinary growth in mobile

phone access and innovation in mobile communications. Drawing from the most recent

framework provided by Aker (2011), this research attempts to explore these relationships in

order to shed light on the question “Can mobile communication technologies be used as a means

14

of enhancing agricultural extension services in Kenya, and in so doing, boost agricultural

productivity?”.

Due to the fast-growing pace of the mobile phone networks and the lack of quantitative

data concerning the level and types of usage, particularly concerning agricultural utility, it is not

possible to take a direct approach to the question. Therefore a more indirect approach comprised

of a simple national-level macro-econometric model that takes a more comprehensive look at the

state of agricultural productivity within Kenya is employed in order to infer whether recent

public investment in agricultural technologies have resulted in higher yields and income amongst

farmers. This is followed by an exploration of the extent to which the diffusion of agricultural

information has contributed to the growth (or lack of growth) in agricultural productivity as a

result of public investment in agricultural technology. Because accessible information at this

scale is limited, cases studies from other parts of the world concerning varying methods of

agricultural extension and diffusion were assessed in detail. The inferences drawn from these

case studies was then applied to the mobile phone market in Kenya in order to propose the

parameters for a comprehensive field study designed to interpret the relationship between mobile

phone based agricultural extension services, farm productivity, and farmer welfare. The methods

used in this study serve to draw substantive conclusions on national macro-level data, refine the

focus area, draw further conclusions from the case studies, analyze the capabilities of three

potential mobile-based extension providers, and then make recommendations on their

applicability in addressing the connection between the potential of mobile communications,

agriculture extension services and agricultural productivity in Kenya.

15

2. Methods 2.1 Study Area

The study area for this analysis was Kenya, located along the equator in eastern Africa.

Kenya’s land area is 56.914 million hectares, of which only 5.3 million are arable land (World

Bank 2012a). A major share of agricultural activity is conducted in central and western Kenya.

Kenya can be divided up physiographically into four zones, beginning with the Lake Victoria

basin in the west, the Rift Valley basin, the western and central highlands, and eastern and

coastal plains. Elevations range from 0 m at the Indian Ocean to 5,199 m at Mt. Kenya; the

variation is considerable where low plains rise to central highlands bisected by Great Rift Valley.

The soils of Kenya vary in terms of topography and in their agricultural potential (Figure 5).

FAO Aquastat (2006) statistics provides a useful summary of these soils:

The soils in western parts of the country are mainly acrisols, cambisols and their mixtures, highly weathered and leached with accumulations of iron and aluminum oxides. The soils in central Kenya and the highlands are mainly the nitosols and andosols, which are young and of volcanic origin. The soils in the arid and semi-arid lands (ASAL) include the vertisols, gleysols and phaeozems and are characterized with pockets of sodicity and salinity, low fertility and vulnerability to erosion. Coastal soils are coarse textured and low in organic matter and the common types are the arenosols, luvisols and acrisols. Widespread soil salinity, which has adversely influenced irrigation development, is found in isolated pockets around the Lake Baringo basin in the Rift Valley and in the Tavetta division in the coastal provinces. (FAO Aquastat 2006).

The climate of Kenya is as varied as its physiography and soils. Temperatures range from

9°C (48°F) to 29°C (84°F) in January and 7°C (45°F) to 26° (79°F) in July in the highlands,

while the low plateaus range from 19°C (66°F) to 37°C (99°F) in January and 19°C (66°F) to

34°C (93°F) in July. Mean annual rainfall ranges for 200mm in the arid and semi-arid regions

(about 80% of the country) primarily in the northern part of Kenya to over 1,800 mm on Mt.

Kenya. The rainfall pattern for most of the country is bimodal, with the heaviest rains occurring

16

between March and June, with more limited rains in October to November. About 80 percent of

the country is arid and semi-arid, while 17 percent is considered to be high potential agricultural

land. Agriculturally, the Kenyan Highlands and western plateau are among the most productive

areas in Africa, sustaining 75% of the population (FAO Aquastat 2006). Forest cover in Kenya is

relatively limited (~3% of the total land area), but the highly varied relief and climate results in a

highly diverse biota.

The population of Kenya was estimated to be 39,802,000 in 2009, with a mean density of

68.9 per sq, km. (UNSD 2009). About half of Kenya’s population is considered poor, with 79%

of the people living in rural areas (IFAD 2011). Subsistence agriculture is the primary source of

income for most of these people, and they combine to produce over 75% of Kenya’s total

agricultural output. Administratively, the country is subdivided into 8 provinces and 70 districts.

The selection of Kenya was paramount due to its vibrant mobile phone market, and

access to some of the essential comprehensive data sets necessary to support this kind of

research. Mobile phone market expansion is high across the developing world, however the

performance of this sector in Kenya has been exceptional having some of the highest coverage

rates as well as intensity of usage in the developing world (ITU 2012). Readily available macro-

level data also can prove problematic in conducting national studies, but the Kenya National

Bureau of Statistics has comprehensive statistical datasets on major development and economic

indicators dating from 1980, and even some as far back as 1963 (Republic of Kenya and

KIPPRA 2009). In addition, relevant data sets from the World Bank (2012a) and the

International Food Policy Research Institute(IFPRI) (2012) were available for public

consumption.

17

Figure 5. Agro-climatic zone map of Kenya. Source: (Braun 1980)

18

2.1.1 Relevant History

Kenya has enjoyed a relatively stable past, politically and economically. Kenya has often

been viewed as a bastion of political stability in East Africa due to the turbulent socialist

experiment in Tanzania, the discriminatory and extremist policies of Idi Amin in Uganda, the

civil wars and political crises in Rwanda, Burundi, and the Sudan, the collapse of the state in

Somalia, and the famine and subsequent unrest in Ethiopia. However, in 2008, political unrest

unfolded in Kenya following the December 2007 presidential elections. The incumbent, Mwai

Kibaki, was re-elected for a second term under what some viewed as suspicious polling

behaviors (BBC News 2008). News of Kibaki’s reelection ignited supporters of the prominent

opposition candidate, Raila Odinga, sparking riots in Kisumu and eventually many other urban

centers. The post-election violence that ensued would eventually claim over 1,500 lives and

force over 600,000 to flee their homes before a power sharing agreement was brokered by Kofi

Annan on February 28, 2008 (BBC News 2008). This devastation left the Kenyan political

system and economy severely damaged, upsetting agricultural productivity and investment for

both 2008 and 2009.

This anomalous period is therefore omitted from this analysis; however it is relevant to

the contextualization of Kenyan agriculture and it is important in understanding changes in the

way people communicate in Kenya. It has been argued that the post-election violence of 2008

demonstrated the level at which mobile phone penetration facilitated the expansion of many-to-

many communication both in the organization of ethnic-based hate crimes and some of the

positive responses, such as the civic behavior of professional and amateur journalists and in the

documentation of the crimes committed (Goldstein and Rotich 2008).

19

2.2 Public Investments in Technology and Agricultural GDP

The first approach taken in this study is to explore the overall effects of public

investments in agricultural research and their effects on agricultural gross domestic product.

Public investments in agricultural research are largely funneled through the Kenya Agricultural

Research Institute (KARI). KARI was founded in 1979 as a semi-autonomous government

institution to address the needs of agricultural research within Kenya. The majority of

government agricultural research funding was funneled through this agency.

2.2.1 Analytic Framework

In developing the baseline framework for analyzing whether, in general, public

investments in agricultural technology lead to increased agricultural productivity, several similar

studies conducted in the United States and Asia were evaluated to determine the most

appropriate model to approach this question. Notable studies have focused on household level

survey data in order to draw conclusive results about the population based on significant

sampling (e.g., Evenson and Gollin 2003). Access to data sets with the information necessary to

replicate the level of detail required in the Evenson and Gollin (2003) household survey

methodology was limited, so a national macro-level econometric model based on socioeconomic

indicators was adopted. The proposed national macro-level indicators model was initially to be

conducted using a collection of panel-series data based on the provincial breakdowns of the

variables of interest. This was intended to account for variations in region as well as time,

however the availability of provincial level data was limited to only a few variables, so this study

employed only national level data in this step (Table 1).

20

2.2.2 National Macro-economic Data

National macro-economic data were obtained from IFPRI (2012), World Bank (2012a),

Republic of Kenya and KIPPRA (2009) in order to compile one comprehensive dataset for the

purposes of analyzing the full effects of agricultural research on agricultural GDP. In the national

level macro-econometric analysis, variables identified as essential to explaining change in

agricultural GDP from pinnacle studies on agricultural production were agricultural labor, arable

land for production, total use of fertilizer and seed inputs, and a measure of total rainfall

(Griliches 1964). Annual data was collected for a period of 26 years over the time frame from

1980 to 2005. These were treated as control variables in order to isolate the variable of interest,

the public stock of investments in agricultural research, and examine its full effects on the

dependent variable, agricultural GDP, over the time frame 1980-2005.

Agricultural labor statistics and public investment in agricultural research were gathered

for the period 1980-2005 from the International Food Policy Research Institute’s (IFPRI)

subsidiary research database, Agricultural Science and Technology Indicators (ASTI) (IFPRI

2012).

Table 1. Availability of variables of interest on national and provincial levels for period of 1980-2005.

Variable Name National Data Available Provincial Data Available 𝐴𝑔𝐺𝐷𝑃𝑡 Yes No

𝑃𝑢𝑏𝐼𝑛𝑣𝑆𝑡𝑜𝑐𝑘𝑡 Yes No

𝐴𝑔𝐿𝑎𝑏𝑜𝑟𝑡 Yes Yes

𝑅𝑎𝑖𝑛𝐼𝑛𝑑𝑒𝑥𝑡 Yes Yes

𝐼𝑛𝑝𝑢𝑡𝑠𝐹𝑒𝑟𝑡𝑡 Yes No

𝐼𝑛𝑝𝑢𝑡𝑠𝑆𝑒𝑒𝑑𝑡 Yes No

𝐴𝑟𝑎𝑏𝑙𝑒𝐿𝑎𝑛𝑑𝑡 Yes Yes

21

Rainfall data was refined to create a measure of erratic weather patterns as an indicator

variable within the national-level econometric model (Bewket 2009).

Accurate rainfall data for 21 weather stations as well as total use of fertilizer and seed

inputs were gathered from the Kenya Ministry of Agriculture’s Agricultural Sector Data

Compendium for the period 1980-2005 (Republic of Kenya and KIPPRA 2009).

Agricultural gross domestic product and arable land measures were obtained from the

World Bank’s African Development Indicators dataset (World Bank 2012a). Agricultural GDP

provided by the World Bank was only given in percent change for each year from 1965-2009,

measures of agricultural GDP from the Ministry of Agriculture for 2009 were coupled with the

data from the World Banks and then regressed based on the percent change measures. This also

provided real values which ensured the measures reflect the value of agricultural GDP without

inflation.

2.2.3 Model Design

The national level macro-econometric model employed in this study explores the

relationship between agricultural investment, and more specifically public investment, in

agricultural technology and agricultural GDP at a national level in Kenya.

The most common approach to assessing agricultural productivity involves a Cobb-

Douglass Production Function (Cobb and Douglass 1928). The model employed here was a

multiple log-log regression based on the Cobb-Douglass Production Function (Equation 1).

Equation 1. Cobb Douglass production estimation for agricultural GDP in Kenya.

log (𝐴𝑔𝐺𝐷𝑃)𝑡 = 𝛽0 + 𝛽1log (𝑃𝑢𝑏𝐼𝑛𝑣𝑆𝑡𝑜𝑐𝑘𝑡) + 𝛽2𝑙𝑜𝑔(𝐴𝑟𝑎𝑏𝑙𝑒𝐿𝑎𝑛𝑑𝑡)+ β3log (𝐴𝑔𝐿𝑎𝑏𝑜𝑟𝑡)

+ 𝛽4 log(𝐼𝑛𝑝𝑢𝑡𝑠𝐹𝑒𝑟𝑡𝑡) + 𝛽5𝑅𝑎𝑖𝑛𝐼𝑛𝑑𝑒𝑥𝑡 + 𝜀

where the variables are described in Table 2.

22

Table 2. Cobb-Douglass production function variable descriptions. Data provided for period 1980-2005 unless otherwise stipulated.

Variable Name Description

𝐴𝑔𝐺𝐷𝑃𝑡 Total Agricultural GDP for time t in USD (2005)

𝑃𝑢𝑏𝐼𝑛𝑣𝑆𝑡𝑜𝑐𝑘𝑡 National Public Investment Stock in Agricultural R&D in Millions USD (2005) for year t based on Equation (1)

𝐴𝑔𝐿𝑎𝑏𝑜𝑟𝑡 Total Agricultural Labor Force (millions) for a given year t

𝑅𝑎𝑖𝑛𝐼𝑛𝑑𝑒𝑥𝑡 Indicator variable accounting for number of weather stations with unstable rainfall for a given year. Index indicates number of weather station s not within 50% of the average rainfall from 1963-2005 out of 21 total weather stations.

𝐼𝑛𝑝𝑢𝑡𝑠𝐹𝑒𝑟𝑡𝑡 Total Expenditures on Fertilizer Inputs based on an Index (1982=100) for year t

𝐼𝑛𝑝𝑢𝑡𝑠𝑆𝑒𝑒𝑑𝑡 Total Expenditures on High Yield Seed Inputs based on an Index (1982=100) for year t

𝐴𝑟𝑎𝑏𝑙𝑒𝐿𝑎𝑛𝑑𝑡 Total Arable Land (Hectares) for a given year

The variables selected for this model attempt to control for factors other than the stock of

public investment in agriculture research which might influence agricultural gross domestic

product each year. In order to run a multiple linear regression using the Cobb-Douglass

production function, the logarithm of the main contributing independent variables to production

were taken, namely the variable of interest: public investment stock in agriculture

(PubInvStockt); and the explanatory variables controlling for other essential factors: labor

(AgLabort), farm fertilizer (InputsFertt) and seed (InputsSeedt) inputs, an index of rainfall

patterns (RainIndext), and total land area cultivated (ArableLandt). Various other studies have

utilized the pure measure of rainfall (cms/year) for each year as a variable within their models,

however in looking at national data it is extremely difficult to normalize the inherent variability

in this variable across regions and through time (Foster and Rosenzweig 2004). As such, it ws

necessary to compile a measure of erratic rainfall (RainIndext) for each weather station based on

23

their mean rainfall over the period 1963-2005 (SEI 2009). This approach provides a measure that

linked rainfall to agricultural GDP impact by capturing rainfall patterns associated with flooding

and drought. If rainfall exceeded 150% of or fell below 50% of the mean rainfall, this implied

that there was a significant impact on farmers in the region, and the observation was assigned a

value of 1 in the model. This was then compiled for all weather stations and used as a measure of

weather stability, with values ranging from 3 to 18 over the time frame.

Public agricultural research spending (PubInvStockt) was provided for each year between

1980 and 2005 in 2005 Kenyan shillings; however it is clear that investments in research in the

year it was undertaken will generally have little effect on agricultural GDP for that same year. In

order to effectively capture the long term investment effects of agricultural research, a measure

was constructed for the agricultural research stock, based on a trapezoidal structure introduced

by Evenson (2001). Using the trapezoidal research stock approach, the total research for a given

year was calculated by the sum of weighted investments over the previous 14 years (Equation

2).

Equation 2. Agricultural research stock for a given year t.

𝐴𝑔𝑟𝑖𝑐𝑢𝑙𝑡𝑢𝑟𝑎𝑙 𝑅𝑒𝑠𝑒𝑎𝑟𝑐ℎ 𝑆𝑡𝑜𝑐𝑘(𝐴𝑅𝑆)𝑡

= 0.1𝐴𝑅𝑆𝑡−14 + 0.3𝐴𝑅𝑆𝑡−13 + 0.5𝐴𝑅𝑆𝑡−12 + 0.7𝐴𝑅𝑆𝑡−11 + 0.9𝐴𝑅𝑆𝑡−10

+ �𝐴𝑅𝑆𝑡−(10−𝑛)

5

𝑛=1

+0.9𝐴𝑅𝑆𝑡−4 + 0.7𝐴𝑅𝑆𝑡−3 + 0.5𝐴𝑅𝑆𝑡−2 + 0.3𝐴𝑅𝑆𝑡−1

+ 0.1𝐴𝑅𝑆𝑡

This equation followed the traditional trapezoidal format of net research effects (Figure 6). The

trapezoidal research stock method allowed measurement of long term effects of the change in

agricultural research funding on agricultural GDP. However, since the data was limited to the

24

period of 1980-2005, it was necessary to generate values for the period 1966-1979 based on a

regression of the linear trend line from 1980-2005 in order to calculate the stock starting in 1980.

Fertilizer and seed inputs were indexed on a scale with a 1982 base of 100 in order to

demonstrate change in total output relative to previous use. Adding in a time-trend variable was

also considered, but due to the general increase in all independent variables over the course of

1980-2005, there would be significant collinearity in the regression, causing errors in the results.

Figure 6. Total research investment stock for a given year t.

2.3 Case Study Analysis of the Extension Services Dimension

Following the econometric analysis of the role of public investments in agricultural

technology, it was necessary to assess the impact of attempts to transfer agricultural knowledge

and technology on agricultural productivity. Limited household level survey data prevented

carrying the econometric analysis using the methodology recommended by Evenson and Gollin

(2003). However, it was possible to gain insights from the work of others in other countries in

order to extrapolate what was learned to the Kenya situation. Specifically, 14 case studies were

assessed to understand the effects of extension services on agricultural productivity, with

0

0.2

0.4

0.6

0.8

1

1.2

t-14 t-13 t-12 t-11 t-10 t-9 t-8 t-7 t-6 t-5 t-4 t-3 t-2 t-1 t

25

emphasis on East Africa (Table 3). In each study dominant themes, strengths, and shortcomings

of extension efforts were tabulated. Two major forms of case studies were considered, with the

first taking a more strict analytic approach at quantifying the effects of extension interventions in

the target country, and the second focusing more on policy proposals concerning agricultural

extension based on the trends in agricultural development in the region. The case study analysis

included documentation of the relationship between extension services and productivity from

any econometric model provided, the challenges attributed to extension services in the region

and the recommendations for future action provided. The goal of this qualitative analysis is to

infer what sorts of policy actions are consistent across evaluations of extension services, and if

they provide evidence to suggest the potential for the use of extension services through mobile

phones.

26

Table 3. Case Studies Overview Title Author(s) (Year) Paper Type Study Area “The Social Structure of the Agricultural Extension Services in the Western Province of Kenya”

Leonard (1972) Policy Framework

Kenya

“The Efficiency of Women as Farm Managers: Kenya”

Moock (1976) Econometric Analysis

Kenya

“The Economic Impact of Agricultural Extension: A Review”

Birkhaeuser et al. (1991)

Econometric Analysis

Global

“The Impact of T&V Extension in Africa: The Experience of Kenya and Burkina Faso”

Bindlish and Evenson (1997)

Econometric Analysis

Kenya, Burkina Faso

“The Effects of Agricultural Extension on Farm Yields in Kenya”

Evenson and Mwabu (1998)

Econometric Analysis

Kenya

“Agricultural Extension: The Kenyan Experience”

World Bank OED (1999)

Policy Framework

Kenya

“Changing Incentives for Agricultural Extension: A Review of Privatized Extension in Practice”

Chapman and Tripp (2003)

Policy Framework

Global

“Agricultural Extension: Good Intentions and Hard Realities”

Anderson and Feder (2004)

Policy Framework

Global

“Challenges Facing Agricultural Extension Agents: A Case Study from South-western Ethiopia”

Belay and Abebaw (2004)

Policy Framework

Ethiopia

“Agricultural Extension in Kenya: Practice and Policy Lessons”

Muyanga and Jayne (2006)

Policy Framework

Kenya

“Farmer Field Schools: An Alternative to Existing Extension Systems? Experience from Easter and Southern Africa”

Anandajayasekeram et al. (2007)

Policy Framework

Eastern and Southern Africa

“Extension in Sub-Saharan Africa: Overview and Assessment of Past and Current Models, and Future Prospects”

Davis (2008) Policy Framework

Sub-Saharan Africa

“Impact of Farmer Field Schools on Agricultural Productivity and Poverty in East Africa”

Davis et al. (2010)

Econometric Analysis

Kenya

“The Association of Agricultural Information Services and Technical Efficiency Among Maize Producers in Kakamega, Western Kenya”

Nambiro et al. (2010) Econometric Analysis

Kenya

27

2.4 Assessing Mobile Phone Market Penetration in Kenya

The next step in the analysis was to assess the overall level of mobile phone market

penetration in Kenya in order to gage the potential for mobile phone services to be used as a

medium for knowledge and technology transfer that effectively reaches and informs farmers.

Key factors documented are level of adoption, intensity of use, and market trends. The data were

obtained from the International Telecommunications Union (2011), the Communications

Commission of Kenya (2012), and the World Bank (2012a). Preliminary results on the overall

effectiveness of various other mobile-based services, such as M*PESA, the money transfer

service provided through Safaricom, are then used to hypothetically explore potential application

to mobile extension services.

2.5 Assessing the Capabilities of New Mobile Services

After analyzing the level of mobile phone penetration and access in Kenya, three distinct

services were assessed for their relative capabilities as well as their overall applicability to the

challenges facing extension services reduced in the case study analysis. The services were

analyzed using a capabilities assessment approach. They included Awaaz De, a Stanford and

UC-Berkeley voice-based extension service provider, FarmFox, a text-based group messaging

system developed at the University of Arizona’s Bio-computing Facilities, and Community

Knowledge Workers, a program implemented through the Grameen Foundation in Uganda which

provides selected community workers in communities smart phones loaded with applications to

access agricultural data and advisory services to disseminate to local farmers.

2.5.1 Awaaz De

Awaaz De is an agricultural extension service developed by Neil Patel and Tapan Parikh

of Stanford and University of California, Berkeley respectively (Patel et al. 2010). This service

28

was designed as a way of incorporating ask-and-reply services through a voice-based provider in

order for farmers to be able to access agricultural extension officers at any point or time. Awaaz

De has already been implemented in India, however the effects of this service are still unknown,

though they are currently being measured by Professor Shawn Cole from the Kennedy School at

Harvard University in a randomized control trial (Gautam Bastián 2012, personal

communication).

2.5.2 FarmFox

FarmFox is an android cell phone application suite developed by the Bio-Computing

Facilities at the University of Arizona for the USDA funded project Stealth Health: Youth

Innovation, Mobile Technology, Online Social Networking and Informal Learning to Promote

Physical Activity (Going et al. 2008). FarmFox is designed as a group-based text message service

facilitator, permitting individuals to create and subscribe to various “tags”, similar to listservs in

email, to send and receive short message service (SMS) messages freely (Dimuth Kulasinghe

2012, personal communication). Development is still in the early stages, however preliminary

uses of the service have been promising for the intended functionality within the Stealth Health

application suite.

2.5.3 Community Knowledge Workers

Designed by the Grameen Foundation’s Technology Center, Community Knowledge

Workers (CKWs) is a program designed to build a nationwide web of information intermediaries

in Uganda to deliver modern information concerning agricultural technology and innovations to

and from farmers (Grameen Foundation 2009). Information is collected by agricultural research

organizations and reviewed before being submitted to online databases which are accessible to

29

CKWs. The project is still in the very preliminary phases of implementation, with the goal of

having a fully established country-wide network in place by 2014 (Grameen Foundation 2009).

30

3. Results 3.1 National Level Macro-econometric Model

In running the national level macro-econometric model, statistical significance,

magnitude, and the sign of the coefficient for Public Spending on Agricultural Technology were

assessed, while controlling for several other contributing variables to maize production. To

ensure the validity of the national level macro-econometric model, coefficients should have

values which reflect existing knowledge of agricultural GDP growth in Kenya.

3.1.1 Statistical Summary

Descriptive statistics resulting from an analysis of the national level macro-economic

data variables of interest over the 26 year time period between 1980 and 2005 are depicted in

Table 4. The mean annual arable land was 4,743,259 hectares, with a small range indicating

minimal change in cultivated acreage over the time period (Table 4). The Agricultural Labor

Force grew a significant amount during this period, ranging from a minimum of 5.71 million and

reaching a maximum in 2005 at 12.12 million. Fertilizer inputs were placed on an index scale,

with 1982 =100, so by 2005 Fertilizer Inputs had reached 271% of the 1982 levels, indicating a

significant rise over time.

Table 4. Agricultural GDP and input quantities in Kenya, 1980-2005 (n=26) Variable Min Max Mean Std. Dev.

Agricultural Gross Domestic Product (2009 USD millions) 3,884.7 4,997.9 4,409.2 319.1 Arable Land (Hectares) 3,725,000 5,452,000 4,743,259 573,279.4 Fertilizer Inputs Index (1982=100) 87 271.7 138.3423 41.36231 High Yield Seed Inputs Index (1982=100) 37.3 145.4 111.33 25.75 Agricultural Labor Force (Millions) 5.71 12.12 8.8192 2.010873 Stock of Public Spending on Agricultural Technology (2005 USD) 34711.2 55902 41150.8 7486.3

Unstable Weather Station Count 8 14 10.5 2 Source: Author Compilation from World Bank 2012, Republic of Kenya and KIPPRS 2009, and ASTI 2012

31

3.1.2 Regression Results

The results of the regressions are displayed in Table 5. The coefficient for public

spending on agricultural technology is strongly positive and statistically significant at a level of

1%, indicating that there is strong reason to reject the hypothesis that public spending on

agricultural technology has no effect on maize production.

Furthermore, the national level macro-econometric model fits well into accepted

understanding of general agricultural production. The coefficient for the logarithm of fertilizer

input index is strongly positive and statistically significant, in concordance with what is expected

from understanding the impact of fertilizer and the law of diminishing marginal returns. The

coefficient for agricultural labor force was strongly negative and statistically significant at the

1% level, displaying a strong negative relationship with agricultural GDP. Arable land, seed

inputs, and unstable rainfall all had a positive coefficients, however the coefficients were not

statistically significant, failing to reject the hypothesis that each of them has no impact on the

total agricultural GDP. Lastly, the R2 is relatively high at 0.782 indicating that this model

explains for a great deal of the variation between variables.

32

Table 5. Correlation of agricultural GDP with variables of interest.

Variables Maize Production

(Metric Tons) Log of Stock Public Spending on Agricultural Technology (2005 USD) 0.5933***

(.2048)

Log of Agricultural Labor Force (Millions) -0.7236***

(0.2052)

Logarithm of the Fertilizer Input Index (1982=100) 0.0572*

(0.0315)

Logarithm of the Arable Land (Hectares) 0.2614

(.1722)

Logarithm of Seed Input Index (1982 = 100) 0.0190

(0.0401)

Lagged Unstable Rainfall at Weather Station Count 0.0040

(.0050)

Observations 26 R-squared 0.728

Note: Standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1 3.2 Extension Services Case Study Review

Among the 14 case studies under consideration (Table 3), only 6 employed extensive

econometric models, the results for which are displayed in Table 6. In each of these six cases, the

evaluations attributed a positive production correlation to the independent variable for extension

services, indicating that extension services contribute positively to agricultural production in all

six evaluations. Furthermore, five of the six studies indicated that the coefficient was also

statistically significant, given added weight to the sign of these coefficients. The fact that the

magnitudes varied could be attributed to the variance of other factors controlled for in the

models.

Table 6. Econometric results obtained in the case studies evaluated describing the relationship between extension services and agricultural production.

Survey Production Correlation Magnitude Evenson (1998) Positive, Statistically Significant – 5%

Bindlish (1997) Positive, Statistically Significant – 10% 1.0 Nambiro (2010) Positive 0.2850 Hopcraft (1974) Positive, Statistically Significant – 10%

Moock (1976) Positive, Statistically Significant – 10% 0.03 IFPRI (2010) Positive, Statistically Significant – 1% 0.80

33

Both the econometric analyses and policy framework form cases were analyzed for

dominant challenges facing the agricultural extension industry in East Africa as well as their

individual recommendations for policy changes to improve extension services. These

components are documented in detail for each study in the Appendix, while the common themes

are tabulated in Tables 7 and 8.

The most frequently challenges facing extension services noted in the case studies were

limited contact with extension officers, inequitable distribution of services, and high costs with

diminishing marginal returns to the impact of services (Table 7). Additional concerns were the

establishment of poor incentive systems for extension officers and the lack of farmer to farmer

lead diffusion of information. The most commonly cited recommendations in the case studies

were promotion of privatization, improved utilization of information technologies, increasing the

role of the farmer in the process, and ensuring that extension services are more flexible in nature

(Table 8).

34

Table 7. Challenges facing extension services.

Study High

Costs

/Dim

inish

ing R

eturn

sIn

equi

tabl

e Dist

ribut

ion

of S

ervi

ces

Poor

Ince

ntiv

e Sys

tem

s for

Agricu

ltura

l Ext

ensio

n Offi

cers

Lack

of T

ime b

y Far

mer

sLi

mite

d Role

in C

onta

ct w

ith O

ffice

rs

Lim

ited

Farm

er to

Far

mer

Spr

ead

of

Info

rmat

ion

Lack

of C

redi

t to F

inan

ce F

arm

ers

Lack

of N

ation

al C

oord

inat

ion

Leonard 1972 x x xMoock1976 x x

Birkaheuser et al. 1991 x xBindlish and Evenson 1997 x x x xEvenson and Mwabu 1998 x x

World Bank OED 1999 x x x xChapman and Tripp 2003 x x x xAnderson and Feder 2004 x xBelay and Abebaw 2004 x x x x xMuyanga and Jayne 2006 x x x

Anandajayasekeram et al. 2007 x x x xDavis 2008 x x x x x

Nambiro et al. 2010 x x x x xDavis et al. 2010 x x x x x x

35

Table 8. Recommendations for extension services.

Study Exten

d Ro

le of

Cre

dit

Targ

et Rem

ote A

reas

Impr

oved

Info

rmat

ion D

iffus

ion

Mec

hani

sms

Give F

arm

ers a

n In

crea

sed

Voice

Mak

e Ext

ensio

n Se

rvice

s Mor

e

Flex

ible

Supp

ort F

arm

er O

rgan

izatio

nsIn

vest

in In

fras

truct

ure i

n Rur

al

Areas

Priva

tizat

ion of

Ser

vice

s

Leonard 1972 x x xMoock 1976 x x

Birkaheuser et al.1991 x xBindlish and Evenson 1997 xEvenson and Mwabu 1998 x xWorld Bank OED 1999 x x x x

Chapman and Tripp 2003 x x xAnderson and Feder 2004 x x xBelay and Abebaw 2004 x x xMuyanga and Jayne 2006 x x x

Anandajayasekeram et al. 2007 x x xDavis 2008 x x

Davis et al. 2010 x x xNambiro et al. 2010 x x

36

3.3 Mobile Phone Sector Trends

Mobile phone markets in the developing world have grown remarkably over the past ten

years (Figure 7). Ownership rates in the developing world are rapidly overtaking those of the

developed world, and have quickly become the majority of the global market.

Figure 7. Gross mobile phone subscriptions in the developing and developed world in 2000 and 2011. Source: International Telecommunications Union 2011

Kenya is both the largest and fastest growing user of cell phone services in East Africa.

Mobile phone subscriptions per 100 inhabitants jumped from 12.95 in 2005 to 61.63 in 2010, an

over 375% increase in five years (Figure 8), well above neighboring Tanzania, Uganda, and

Rwanda.

469

250

2000

Total: 719 million users

Developed

Developing

1,641

4,520

2011

Total: 6 billion users

Developing

Developed

37

Figure 8. Mobile phone subscriptions per 100 inhabitants in East Africa, 2000 to 2010. Source: International Telecommunication Union 2011

Analysis of the mobile phone sector reveals that Kenya also is advanced in terms of its

mobile applications. Mobile-based financial services have begun to play a dominant role in the

banking sector, and Kenya is far more advanced than any nation in the world. As of 2012, over

68% of Kenyans declared that they used mobile money services, the highest rate in the world

(Figure 9). Services such as Twitter have also become prominent in Kenya; Twitter recorded

over 2 million tweets from Kenya alone last year, the second highest in Africa behind South

Africa (The Economist 2012b).

0

10

20

30

40

50

60

70

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Subs

crip

tions

per

100

inha

bita

nts

Kenya

Tanzania

Uganda

Rwanda

38

Figure 9. Mobile money users in Africa, percentage of total population, 2011. © The Economist Newspaper Limited, London (The Economist 2012a).

3.4 The Potential of New Mobile Phone Services

From a baseline capabilities analysis of Awaaz De, FarmFox, and Community

Knowledge Workers, a list of strengths and weaknesses was generated which was particularly

focused towards the challenges outlined in the case study analysis of extension services in East

39

Africa (Table 9). Ratings of effectiveness vary across services, with some being more

appropriate for various challenges.

Table 9. Capabilities of mobile services.

Aspect/Specification Awaaz De FarmFox

Community Knowledge Workers

Responsive to Farmer Inquiries + + + + + Information Diffused Promptly - + + + Ability to address complicated problems + + - 0 Puts the Power in the Farmer’s Hand + + + - - Promotes Farmer to Farmer Spread 0 + + + Incentivizes Extension Workers + + - + + Actively (Not Reactively) Diffuse Information

- - + + -

Curb Bias in Extension Service Targeting + + - - - Reduce Costs + + + + + Reduce Officer Staffing Demands - - + + + Complement Existing Extension Services + + + + + + Coordinate Nationally + + + + Complement Existing Technologies + + + 0 Legend: - - extremely ineffective, - ineffective, 0 neutral, + effective, and + + extremely effective 4. Discussion

The results of the national level macro-econometric model regression analysis suggest

that public investment in agricultural technology (PubInvStockt) contributes positively to the

agricultural gross domestic product (AgGDPt) in Kenya. The high magnitude and statistical

significance of the coefficient for the stock of public spending on agricultural demonstrates that

investments do produce high returns (60%) to agricultural gross domestic product. Furthermore,

the results fall in line with common theory regarding the effects of agricultural fertilizer

(InputsFertt) and seed inputs (InputsSeedt) and agricultural production (Tilman et al. 2002). Only

the coefficient for fertilizer was statistically significant, but it demonstrates significant

diminishing marginal returns. This could be due to agricultural intensification and nutrient

leeching, or improper fertilizer use due to poor access to information (Zhu and Chen, 2002;

Duflo et al. 2008).

40

The statistically insignificance of the arable land (ArableLandt), seed input index

(InputsSeedt), and unstable rainfall index (RainIndext) can be explained by their lack of variation

over the time period. Arable land is extremely limited in Kenya, and the growth over this time

period was not substantial. Furthermore, a great deal of past expansion of arable land was due to

major agricultural investment such as irrigation schemes which would reduce the agricultural

GDP at first due to their investment expense. Sub-Saharan Africa has been shown to have

significantly higher irrigation costs per hectare than North Africa, South Asia, East Asia, and

Latin America (FAO 1986; Brown and Nooter 1992; Jones 1995). The statistical insignificance

of the seed input index can be explained by the sporadic level of seed adoption over this period.

While the range did vary extensively from 1980 to 1982, once the base index was calculated,

overall seed input use remained relatively constant, not producing significant variance to draw

from (Figure 10). The lack of significance in the coefficient of the rain index is attributable to the

high variability in rainfall during this time period. The mean value for rain index was 10.5 out of

a total of 21 rain stations, indicating that on average, half of the meteorological stations were

experiencing erratic rainfall for any given year. This suggests a high level of irregularity in

rainfall during this time period.

41

Figure 10. Indexed seed input levels 1980-2005 (1982=100) Source: Republic of Kenya and KIPPRA 2009

The statistical significance of the negative coefficient for the agricultural labor force

(AgLabort) demonstrates that in fact labor had negative returns to scale during this time period.

This can possibly be explained in that relative increase in surplus of available agricultural labor

during 1980-2005 resulted in smaller and smaller landholdings through family inheritance sub-

divisions. This then reduces the total product on existing landholdings, thus reducing agricultural

GDP.

The overall results of the national level macro-econometric model suggest with

confidence that public investment in agricultural technology contributes positively to the

agricultural gross domestic product in Kenya. However, due to the limited availability of data

which prohibited a household-level analysis, it is not possible to state conclusively how

knowledge and technology transfer efforts to affected agricultural GDP in Kenya during this

same period.

The case study analysis, however, sheds some light on this. The results of six studies

which were able to conduct a household level analysis found that extension services had a strong

0.00

20.00

40.00

60.00

80.00

100.00

120.00

140.00

160.00

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

42

connection with agricultural productivity, with all six econometric analyses indicating a positive

relationship between extension services and productivity (Table 6). This suggests that both

investments in research as well as extension may prove to be vital to productivity in Kenya, but

the stagnant growth level in the past decade indicates there are still major impediments to

growth. While the case study analysis did indicate positive returns to extension, it also explained

substantial challenges to the effective implementation of extension services, demonstrated by the

relatively low level of extension service coverage in Kenya to date (World Bank OED 1990).

Lack of incentive systems for extension officers, lack of inequitable distribution of

services, limited role of farmers, and high costs all represent serious systemic problems with

agricultural extension services. The frequency with which these issues were cited across the 14

case studies lends credence to the claim that extension services are lacking the logistical and

financial capacity to maintain the same level of interaction necessary to achieve the effective

knowledge and technology transfer documented in the past. The incorporation of services which

could provide the farmer with more flexibility, a greater voice, and greater responsiveness to the

needs of farmers could significantly strengthen the impact agricultural research funding has on

final agricultural gross domestic production levels in Kenya agrarian society, promoting rural

development, poverty alleviation, and hopefully social mobility.

Analysis of market penetration in the mobile phone sector in Kenya makes it clear that

mobile phones have reached the level of adoption necessary to address several of the challenges

highlighted in the case study analysis. Rapid growth in this sector in Kenya over the past 5 years

suggests virtual universal coverage is likely in the near future. This level of coverage could

potentially lead to mobile phones being a platform providing universal access to various services,

including agricultural extension services. To date, Kenya has already experienced significant use

43

of financial and social services (e.g., M*PESA and Twitter), indicating access is not a major

impediment to agricultural extension services delivered on a mobile platform.

The capabilities analysis of more targeted services now entering the market, such as

Awaaz De, FarmFox, and Community Knowledge Workers (CWK) revealed disparate strengths

amongst them. The ease of convenience and accessibility provided by Awaaz De and FarmFox

greatly outweigh the burden of contacting your local official in CWK, but also may limit the

quantity and quality of extension services which can be provided. The success of M*PESA and

Twitter have been based on ease of access and the relatively quick level of integration, indicating

that these are vital aspects to the success of agricultural extension applications (Aker and Mbiti

2011; Davis 1989) however intensity and quality of service are also vital to the effectiveness of

these services as a complement or eventual replacement of traditional delivery mechanisms

employed by extension services. Therefore, it is unclear as of yet which service best suits the

need of agricultural extension in Kenya.

5. Recommendations While a preliminary capabilities analysis has demonstrated the relative strengths and

weaknesses of each of the three services, it is still unclear as to which service would be best

implemented as a supplement, complement, or even substitute to existing agricultural knowledge

and technology transfer measures. This study recommends further research on:

• Overall effects of mobile based extension services, consisting of independent studies

or actual extension services trials of Awaaz De, FarmFox, and CWK

• Contributing factors to educational application adoption in East Africa, drawing from

the lessons of financial services such as M*PESA and social network microblog

services such as Twitter

44

• Optimal levels of information/educational quantity and quality for mobile phone

services

Once these factors are addressed, it may be more clear which approach mobile based

extension services should follow. For now it is clear that mobile based extension services can

address several of the pressing challenges facing agricultural extension in Kenya.

6. Conclusions As demonstrated, investment in agricultural research and technologies contribute

significantly and positively to agricultural gross domestic product. Furthermore, current efforts

to encourage associated knowledge and technology transfer are challenged by resource

limitations which in turn hinders effectiveness in supporting the Kenyan agricultural sector.

Several dominant challenges emerged during the case study analysis, the most prominent being

inequitable distribution of services, limited contact with extension officials, lack of flexibility of

services, and high costs of diffusion amongst providers. The emergence of the mobile phone

sector in Kenya provides great opportunities address these concerns. The relative rate of adoption

over the past 10 years suggests universal coverage is fast approaching and similar trends in

mobile financial and services indicate rapid adoption rates when implemented effectively. Three

services, Awaaz De, FarmFox, and Community Knowledge Workers, have the necessary

capabilities to address many of the issues facing current attempts at agricultural technology

transfer and diffusion, and hold great potential for a significant transformation of how knowledge

and technology can be delivered by agricultural extension services. These potential future

positive developments may help the extension services make the most out of agricultural

research, which in turn should support future agricultural productivity in Kenya.

45

7. Appendix: Case Study Analysis Study Area Challenges to Extension Recommendations

FFS_Kenya, 2010 Kenya -Extension agents have performance contracts, so they are under pressure to work where they will see the most results -Lack of time by farmers -Extension is too far from their homes

-Support farmer organizations as major vehicle for farmer development -Invest in infrastructure in rural areas

Evenson,1998 Kenya -Diminishing Returns to Extension -Farmers at the Middle points of the yield residuals gain less than at the two ends

-Target mid-range farmers -Make Extension Services more flexible

WBOED,1999 Kenya -Extension currently focused on men and large scale operation, missing majority of farmers -Expected effects of spread have been limited due to poor communication

-Improved information systems -Increased intensity in extension services -Give farmers a voice in the system -Extension services targeting areas where marginal impact is greatest

Birkaheuser,1991 -Cannot determine the exact specifications which promote extension effectiveness, need to be able to adapt locally

-Continuous flow of research-generated technology -Improved availability of complementary inputs

Nambiro, 2010 -80% of households did not access agricultural information in past two years -Delivery of extension services by non-governmental organizations viewed to be lower quality

-Strengthening of formal and informal agricultural extension services -Stronger linkage among agricultural research, agricultural extension, and farm level activities

Bindlish, 1997 Kenya and

Burkina Faso

-Discipline of Regular Extension Visits to Farmers -Inadequate proficiency on subject matter of specialists -Lack of Credit for Farmers

-Extend role of credit in Extensions

Muyanga, 2006 Kenya -Extension provision skewed towards well-endowed regions and high value crops -Public Extension is expensive compared to private alternatives -Public Resources Constrained

-Public Extension target remote areas, underserved populations, Private sector target elsewhere. -Division of Extension services, not overlapping.

46

Study Area Challenges to Extension Recommendations Anandajayasekeram,

2007 East &

Southern Africa

-Competition between donor groups -Lack of national coordination -Low level of participation and involvement from village to regional level -Use of complicated designs in field demos

-Substitute traditional extension systems with participatory pluralistic knowledge systems that are more gender sensitive and pro-poor -Higher incorporation of information and technology systems in extension

Anderson, 2004 -Scale and Complexity-diversity of local problems -Weak Accountability -Weak Political commitment and support -Fiscal constraints -Poor Use of Information Technology

-Training and Visit Extension -Decentralization -Fee for Service and Privatized Extension -Farmer Field Schools

Belay, 2004 -Extension work not participatory -Little consideration for farmer’s experience and knowledge -Extension workers lack practical skills -Bias towards farmers who are financially sound and interested in extension services

-Increase role of farmer in extension services -Timely and consistent services

Davis, 2008 -Lack of relevant technology -Failure by extension to involve clinetel in problem defining and solving -Lack of incentives for extension agents -Weak linkages between extension, research, and farmers

-Pluralistic Demand Driven models that incorporate use of information and communication technologies

Chapman, 2003 -Failure of extension to reach all populations -Poor turnout for extension services -Lack of flexibility

-Embrace a variety of extension modalities -Development of a new incentive system -Improve responsiveness of extension to farmer’s priorities

Leonard, 1972 -Skewed in favor of progressive farmers -Senior agriculture staff isolated from majority

-Design frameworks for targeting poorest farmers

Moock, 1976 -Male oriented services, failure to approach large portion of women

-Target women to diffuse innovations

47

8. References Cited Aker, J.C. (2011). “Dial ‘A’ for agriculture: a review of information and communication

technologies for agricultural extension in developing countries”. Journal of Agricultural Economics 42, 631-647.

Aker, J.C. and I.M. Mbiti. (2010). “Mobile Phones and Economic Development in Africa.” Journal of Economic Perspectives, 24(3): 207-232.

Alila P.O., and R. Atieno (2006). “Agricultural Policy in Kenya: Issues and Processes” Presentations at the Future Agricultures Consortium Workshop, Institute for Development Studies, Nairobi, March, 2006.

Alston, J. (2010), “The Benefits from Agricultural Research and Development, Innovation, and Productivity Growth”, OECD Food, Agriculture and Fisheries Working Papers No. 31.

Anandajayasekeram, P., K.E. Davis, and S. Workneh (2007) “Farmer Field Schools: An Alternative to Existing Extension Systems? Experience from Eastern and Southern Africa” Journal of International Agricultural and Extension Education 14(1):81-93.

Anderson, J.R. and G. Feder (2004) “Agricultural Extension: Good Intentions and Hard Realities” The World Bank Research Observer 19(1): 41-60.

BBC News (2008) “Kenya rivals agree to share power” 28 February, 2008.

Belay, K. and D. Abebaw (2004) “Challenges Facing Agricultural Extension Agents: A Case Study from South-western Ethiopia” African Development Bank.

Bewket, W. (2009) “Rainfall variability and crop production in Ethiopia: Case study in the Amhara region” Presented at Proceedings of the 16th International Conference of Ethiopian Studies, Addis Ababa, 2009.

Bigsten, A., and P. Collier (1995). “Linkages from agricultural growth in Kenya,” in J.W. Mellor (ed.) Agriculture on the Road to Industrialization, Johns Hopkins University Press for the International Food Policy Research Institute: Baltimore, MD., U.S.A.

Bindlish, V. and R.E. Evenson “The Impact of T&V Extension in Africa: The Experience of Kenya and Burkina Faso” The World Bank Research Observer, 12(2):183-201.

Birkhaeuser, D., R.E. Evenson, and G. Feder (1991) “The Economic Impact of Agricultural Extension: A Review” Economic Development and Cultural Change, 39(3): 607-650.

Boserup, E. (1965) The Conditions of Agricultural Growth: The Economics of Agrarian Change under Population Pressure. LondonL: G. Allen and Urwin.

Boserup, E. (1981) Population and Technological Change: A Study of Long Term Trends. Chicago: University of Chicago Press.

48

Brown, E. and R. Nooter (1992). “Successful Small-Scale Irrigation in the Sahel” World Bank Technical Paper No. 171, Washington, D.C.: World Bank.

Bruan, H.M.H. (1980) Agro-Climatic Zone Map of Kenya. Appendix 2 to Report No. E1. Republic of Kenya, Ministry of Agriculture Kenya Soil Survey, Nairobi.

Chapman, R. and R. Trip (2003) “Changing Incentives for Agricultural Extension- A Review of Privatised Extension in Practice” Agricultural Research and Extension Network, Network Paper No. 132.

Chowdhury, S.K. (2006) “Investments in ICT—Capital and economic performance of small and medium scale enterprises in east Africa” Journal of International Development 18(4):533–552.

Cobb, C.W., and P.H. Douglas (1928). “A theory of production” American Economic Review 18 (Supplement): 139–165.

Communications Commission of Kenya (2012). Sector Statistical Reports. Available online: http://www.cck.go.ke/resc/statcs.html. Accessed May 3, 2012.

Davis, F.D (1989), “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology” MIS Quarterly 13(3): 319-340.

Davis, K.E, E. Nkonya, E. Kato, D.A. Mekonnen, M. Odendo, R. Miiro, and J. Nkuba (2010) “Impact of Farmer Field Schools on Agricultural Productivity and Poverty in East Africa” IFPRI Discussion Paper 00992.

Davis, K.E. (2008) “Extension in Sub-Saharan Africa: Overview and Assessment of Past and Current Models, and Future Prospects” Journal of International Agricultural and Extension Education 15(3):15-28.

De Janvry, A. and E. Sadoulet (2002). “World Poverty and the role of agricultural technology: Direct and indirect effects” Journal of Development Studies 38(4): 1-26.

Dey, B., D. Newman, and R. P. Renee (2011). “Analysing appropriation and usability in social and occupational lives: An investigation of Bangladeshi farmers' use of mobile telephony” Information Technology & People 24(1):46-63.

Donner, J. (2008). “Research approaches to mobile use in the developing world: A review of the literature” Information Society 24(3): 140-159.

Duflo E., M. Kremer, and J. Robinson (2008) “How High Are Rates of Return to Fertilizer? Evidence from Field Experiments in Kenya” American Economic Review 98(2): 482-488.

The Economist (2012a) “Mobile Money in Africa: Press 1 for modernity” April 28, 2012

The Economist (2012b) “#AfricaTweets” April 28, 2012.

49

Evenson, R.E. and G. Mwabu (1998) “The Effects of Agricultural Extension on Farm Yields in Kenya” Economic Growth Center Discussion Paper No. 798.

Evenson, R.E., C.E. Pray, and M.W. Rosegrant (1999). “Agricultural Research and Productivity Growth in India” Research Report 109. International Food Policy Research Institute.

Evenson, R. E. (2001). Economic Impacts of Agricultural Research and Extension. In B. Gardner & G. Rausser (Eds.), Handbook of Agricultural Economics, Volume 1. Elsevier Science.

Evenson, R. E. and D. Gollin (2003). “Assessing the Impact of the Green Revolution, 1960 to 2000”, Science 300: 758-762.

FAO Aquastat (2006). Kenya geography. Food and Agriculture Organization of the United Nations (FAO), Rome. Available Online: http://www.fao.org/nr/water/aquastat/countries_regions/kenya/index.stm. Accessed: May 2, 2012.

FAO (1986). “Irrigation in Africa south of the Sahara” FAO Investment Centre Technical Paper 5, Rome : FAO.

Fan, S. and P.G. Pardey. (1997). “Research, productivity, and output growth in Chinese agriculture.” Journal of Development Economics, 53:115-117.

Foster, A. D. and M. R. Rosenzweig (1996). “Technical Change and Human-Capital Returns and Investments: Evidence from the Green Revolution” American Economic Review 86(4): 931-953.

Foster, A. D. and M. R. Rosenzweig (2004). “Agricultural Productivity Growth, Rural Economic Diversity, and Economic Reforms: India 1970-2000” Economic Development and Cultural Change 52(3): 509-542.

Furuholt B. and E. Matotay. (2011) “The Developmental Contribution From Mobile Phones Across the Agricultural Value Chain in Rural Africa.” Electronic Journal on Information Systems in Developing Countries 48(7):1-16.

Going, S. B., N. Hongu, B.J. Orr, D.J. Roe, M. Nichter, and K. Astroth. (2008). Stealth Health: youth innovation, mobile technology, online social networking and informal learning to promote physical activity. University of Arizona: USDA National Research Initiative under the Human Nutrition and Obesity program.

Goldstein, J. and J. Rotich, (2008) “Digitally Networked Technology in Kenya’s 2007-2008 Post-Election Crisis” Internet and Democracy Case Studies Series, Berkman Center Research Publication No. 2008-09.

Grameen Foundation (2009) “Grameen Foundation Expands Technology Program for Poor Farmers in Uganda” Grameen Foundation. Available Online: http://www.grameenfoundation.org/grameen-foundation-expands-technology-program-poor-farmers-uganda. Accessed: May 3, 2012.

50

Griliches, Z. (1964) “Research Expenditures, Education, and the Aggregate Agricultural Production Function” American Economic Review 54(6): 961-974.

Huffman, W.E. and R.E Evenson (2006). Science for Agriculture: A Long-Term Perspective. Ames, Iowa: Blackwell Publishing.

IFAD. (2011). Enabling poor rural people to overcome poverty in Kenya. International Fund for Agricultural Development (IFAD), Rome. Available Online: http://www.ifad.org/operations/projects/regions/pf/factsheets/kenya.pdf. Accessed: May 2, 2012.

IFPRI. (2012). Data. Agricultural Science and Technology Indicators. Available Online: www.asti.cgiar.org/kenya. Accessed: April 15, 2012.

International Telecommunications Union (2011) Data. ICT Indicators. Available Online: http://www.itu.int/ITU-D/ict/statistics/. Accessed : May 1 2012.

Inukai, I. (1974), “African Socialism and Agricultural Development Strategy: A Comparative Study of Kenya and Tanzania” The Developing Economics 12(1): 3-22.

Jensen, Robert T., (2007) “The Digital Provide: In-formation (Technology), Market Performance and Welfare in South Indian Fisheries Sector,” Quarterly Journal of Economics, 122(3): 879-924.

Jin, S., J. Huang, R. Hu, and S. Rozelle (2002) “The Creation and Spread of Technology and Total Factor Productivity in China’s Agriculture” American Journal of Agricultural Economics 84(4): 916-930.

Johnston, B. F. (1989) "The Political Economy of Agricultural Development in Kenya and Tanzania." Food Research Institute Studies 21.

Jones, W.I. 1995. The World Bank and Irrigation. Washington, D.C.: World Bank.

Kenya Central Bureau of Statistics (KCBS), 1997. National Development Plan 1997-2001. Nairobi, Government Printer.

Kriem, M.S. (2009) “Mobile telephony in Morocco: a changing sociality.” Media, Culture & Society 31(4):617-632.

Kyem P.A.K., and P.K. LeMaire (2006) “Transforming recent gains in the digital divide into digital opportunities: Africa and the boom in mobile phone subscription.” Electronic Journal on Information Systems in Developing Countries 28: 1-16.

Leonard, D.K. (1972) “The Social Structure of the Agricultural Extension Services in the Western Province of Kenya” Institute for Development Studies Discussion Paper No. 126.

51

Lofchie, M. F. (1989) The Policy Factor, Agricultural Performance in Kenya and Tanzania. Boulder, Colorado, Lynne Rienner.

Moock, P.R. (1976) “The Efficiency of Women as Farm Managers: Kenya” American Journal of Agricultural Economics, 58(5): 831-835.

Muyanga, M. and T.S. Jayne (2008) “Private Agricultural Extension Systems in Kenya: Practice and Policy Lessons” The Journal of Agricultural Education and Extension, 14(2): 111-124

Nambiro, E., J. Chianu, and A. Murage (2010) “The association of agricultural information services and technical efficiency among maize producers in Kakamega, Western Kenya” Presented at the Joint 3rd African Association of Agricultural Economists Conference, Cape Town, South Africa, September 19-23, 2010.

Okech, B. A., W. V. Mitullah, and M. L. Awiti, (1996) Agricultural Sector Management Reform and Policy Analysis, The Kenyan Case. Nairobi, Institute for Development Studies, The University of Nairobi, 1996.

Patel, N., D. Chittamuru, A. Jain, P. Dave, and T. Parikh, (2010) “Avaaj Otalo- A Field Study of an Interative Voice Forum for Small Farmers in Rural India” Presented at ACM Conference on Human Factors in Computing Systems, Atlanta, GA.

Republic of Kenya and KIPPRA (2009) Ministry of Agriculture: Kenya’s Agricultural Data Sector Compendium. Available online: www.kilimo.go.ke. Accessed April 5, 2012.

Republic of Kenya (2005). Review of the National Agricultural Extension Policy (NEAP) and it’s Implementation. Volume II-Main Report and Annexes. Ministry of Agriculture and Ministry of Livestock and Fisheries Development, Nairobi.

Republic of Kenya (1965) Sessional Paper Number 10 of 1965 on African Socialism and its Application to Planning in Kenya. Nairobi: Government Printer.

Rono, J. K. (2002) “The Impact of Structural Adjustment Programmes on Kenyan Society”, The Journal of Social Development in Africa 17(1).

Salia M., N.N.N. Nsowah-Nuamah, and W.F. Steel. 2011. “Effects of mobile phone use on artisanal fishing market efficiency and livelihoods in Ghana.” Electronic Journal on Information Systems in Developing Countries 47(6):1-26.

Simon, J.L. (1986) Theory of Population and Economic Growth. Oxford, UK, Basil Blackwell.

Schwartz, L. A. and J. Kampen (1992) Agricultural Extension in East Africa. Washington D.C: The International Bank for Reconstruction and Development.

Stockholm Environment Institute (2009) “Economics of Climate Change: Kenya.” Stockholm Environment Institute, Project Report 2009.

52

Sutherland, A.J., J.W. Irungu, J. Kang'ara, J. Muthamia, and J Ouma (1999) “Household food security in semi-arid Africa - the contribution of participatory adaptive research and development to rural livelihoods in Eastern Kenya.” Food Policy 24(4): 363-390.

Thirtle, C. X. Irz, L. Lin, V. McKenzie-Hill, and S. Wiggins (2001) “Relationship between changes in agricultural productivity and the incidence of poverty in developing countries” DFID Report No 7946.

Tilman, D., K. G. Cassman, P. A. Matson, R. Naylor, and S. Polasky (2002) “Agricultural sustainability and intensive production practices” Nature 418.

UNSD. 2009. UNdata: Kenya. United Nations Statistics Division (UNSD), New York. Available Online: http://data.un.org/CountryProfile.aspx?crName=kenya. Accesses May 4, 2012.

World Bank (2012a). Data. African Development Indicators. Available Online: http://data.worldbank.org/data-catalog/africa-development-indicators. Accessed April 5, 2012.

World Bank (2012b). Data. Population, total. Available Online: http://data.worldbank.org/indicator/SP.POP.TOTL Accessed April 30, 2012.

World Bank OED (1999) “Agricultural Extension: The Kenya Experience” Précis No. 198.

Zhu, Z.L. and D.L. Chen (2002). “Nitrogen fertilizer use in China- Contributions to food production, impacts on the environment and best management strategies” Nutrient Cycling in Agroecosystems 63(2-3):117-127.