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

net/publication/334637013

Small-scale irrigation scheme governance - poverty nexus: evidence

from Ethiopia

Article in Food Security · July 2019

DOI: 10.1007/s12571-019-00953-8

CITATIONS
READS
23
786

3 authors, including:

Fitsum Assefa Adela

Gumataw Kifle Abebe
Planning and development commission,
Ethiopia Dalhousie University

1 PUBLICATION 23 CITATIONS 34 PUBLICATIONS 604 CITATIONS

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

Drivers for the implementation of Food Safety Management Systems in emerging agro-food supply chains View project

Supermaketization in the MENA region View project

All content following this page was uploaded by Gumataw Kifle Abebe on 24 July 2019.

The user has requested enhancement of the downloaded file.

Food Security
https://doi.org/10.1007/s12571-019-00953-8

ORIGINAL PAPER

Small-scale irrigation scheme governance - poverty nexus:

evidence from Ethiopia
Fitsum Assefa Adela1 & Joachim Aurbacher 2
& Gumataw Kifle Abebe3

Received: 1 October 2018 / Accepted: 5 July 2019

# International Society for Plant Pathology and Springer Nature B.V. 2019

Abstract
Irrigated agriculture has been popularized as a key factor to improve crop yields and enhance food security in Africa.
However, empirical findings are mixed. This study analyzes determinants of small-scale irrigation adoption and the impact
this may have on food security in Ethiopia, where agricultural land is extremely fragmented and densely populated. Data
were collected from 240 farmers, and the findings from the survey were triangulated with focus group discussions and key
informant interviews. First, the Foster-Greer-Thorbecke indices indicated high poverty levels among farmers without
access to irrigation. Second, accounting for a self-selection bias by using the endogenous switching regression (ESR)
model, scheme governance, perceived water scarcity, and access to information were found to have significant effects on
the adoption of irrigation schemes. Model estimates further indicated that access to small-scale irrigation resulted in better
living conditions for both current users and non-users when compared to their counterfactual situations. Farm income of
the user households would decrease by 42% (birr 151,419 or US$ 5,500 per ha) had they not used irrigation. Similarly,
farm income of the non-users would increase by 149% had they used irrigation. Per adult equivalent consumption
expenditure has also shown a decrease of 35% for irrigation users and an increase of 40% for non-users compared to their
respective counterfactual situations. We conclude that much of the perceived water scarcity level is attributed to existing
governance regimes more than the physical scarcity of water. The study draws several implications for household food
security and local-based water use policies.

Keywords Small-scale irrigation . Ethiopia . Endogenous switching regression . Scheme governance . Poverty . Farm income

1 Introduction 2014) as the average landholding is generally less than two

hectares and continually declining (Vanlauwe et al. 2014).
Irrigated agriculture has been popularized as a key input to In fact, irrigation can have a multiplier effect through the
improve crop yields and enhance food security. It is en- hancement of employment opportunities, consumption,
expected to account for 53% of the projected increase in and dietary changes (Rosegrant et al. 2009), and through
the global cereal crop production between 2000 and 2050 the in- creased use of productivity-enhancing inputs such
(Rosegrant et al. 2009). Irrigated agriculture is particularly as fertilizer and improved varieties (Gebregziabher et al.
important for developing countries (Josephson et al. 2014; 2009; Smith 2004; Namara et al. 2005).
Vanlauwe et al. Although small-scale irrigation schemes have been pro-
moted in sub-Saharan Africa (SSA) to enhance the food
secu- rity situation in the region (Burney and Naylor 2012;
* Fitsum Assefa Adela
Dillon 2011; Xie et al. 2017), not more than 6% of the total
fitsumassefa@gmail.com
cultivated area is covered by irrigation (You et al. 2011)
and only 4% of renewable water is used for agriculture
1
Faculty of Business and Economics, Hawassa University, (Hanjra et al. 2009). Also, empirical studies are mixed
P.O.Box 05, Hawassa, Ethiopia
2
regarding the impact of investment on small-scale irrigated
Institute of Farm and Agribusiness Management, Justus- agriculture. Several stud- ies, particularly those related to
Liebig University of Giessen, Senckenbergstraße 3,
35390 Giessen, Germany Asia, have indicated a strong positive link between
3 irrigated agriculture and enhanced pro- ductivity and food
Faculty of Agricultural and Food Sciences, American University
of Beirut, Riad El Solh, Beirut, Beirut 1107 2020, Lebanon security (Bacha et al. 2011; Hussain and Hanjra 2004;
Smith 2004; Wichelns 2014; De Haen 2003).
 Adela F.A. et

Most of these studies have documented doubled or tripled (Awulachew et al. 2010), adoption of irrigation tech- nology
incomes of farm households that used small-scale irrigated and estimation of probability of adoption
agriculture. Much of the success in Asia is often associated (Gebregziabher 2012), optimal irrigation scheduling (Setegn
with the role of informal institutions in the governance of
irrigation water, implementation of bottom-up approaches
to select participants in irrigation schemes, and creation of
mar- ket linkages for irrigation users (Mutambara et al.
2016; Mukherji 2012; Mwendera and Chilonda 2013).
However, not much has been documented in SSA
regarding the impact of small-scale irrigation schemes in
improving rural liveli- hoods mainly due to factors related
to infrastructure, institu- tions, and socioeconomics
(Burney and Naylor 2012; Schuenemann et al. 2018;
Mwendera and Chilonda 2013). Furthermore, various
findings have indicated that farmers from households that
successfully adopted small-scale irriga- tion were
relatively better off than the ultra-poor ones (Mangisoni
2008; Namara et al. 2005).
Previous studies have generally indicated that small-
scale irrigation schemes have the potential to enhance crop
yields and rural livelihoods. However, the decision to
participate in small-scale irrigated agriculture and the
impact this may have on poverty reduction may be
influenced by several factors, including farm and
household characteristics, and institutional and context-
specific variables. This study focuses on Ethiopia, where
the government’s stake in the construction and gover- nance
of small scale irrigation systems is high. We explore the
factors influencing the adoption of small-scale irrigation
schemes and analyze whether access to small-scale
irrigation water enhances food security in areas where
agricultural land is extremely fragmented and densely
populated. Indeed, food security is a multidimensional
concept, mainly defined in terms of food availability,
access, utilization, and stability (FAO 2008). Several
studies suggest that access to small- scale irrigation has the
potential to increase incomes and re- duce poverty in
Africa where sustainable water access for smallholders is
difficult owing to recurrent drought. Our study directly
relates to the availability and access dimensions, which are
the prerequisites for household food security (Barrett
2010; Denny et al. 2018).
Ethiopia provides an interesting perspective for the
follow- ing reasons. First, Ethiopia has recently been
aggressively pursuing several small-scale irrigation
schemes aimed at im- proving rural livelihoods. Second,
this study was carried out in a region where annual rainfall
is relatively modest, and thus provides a better opportunity
(i.e., based on economic theory) to determine the factors
influencing the decision to enroll in irrigation schemes or
stay with rain-fed agriculture. Third, there are very few
impact studies related to small-scale irriga- tion schemes in
Ethiopia; available studies have focused on identification
and description of irrigation potentials and chal- lenges
 Small-scale irrigation scheme governance - poverty nexus: evidence from
et al. 2011), problems associated with small-scale A polycentric approach relies on the interplay among
irrigation intervention (Aberra 2004; Amede 2015), local, regional, and national actors to govern irrigation
water-based growth challenges (Hanjra et al. 2009; schemes (Baldwin et al. 2016; Lecoutere 2011). A key
Namara et al. 2010), and a few impact studies (Bacha et aspect of such a pragmatic approach is that institutions
al. 2011; Belay and Bewket 2013). Furthermore, none of emerge from
these few previous impact studies have paid attention to
institutional factors (or scheme governance). Existing
studies elsewhere also indicate mixed results regard- ing
the role of institutions in the adoption of small-scale irri-
gation and its impact on household food security.
The remainder of the paper proceeds as follows. The
next section presents a conceptual framework where the
role of institutions is discussed as related to irrigation
water access, governance, and management. This is
followed by the methods, results, and discussion sections.
The final section presents conclusions and policy
implications.

2 Role of institutions in the

governance of small-scale irrigation
schemes

The concept of institutions has become a popular

framework in natural resource economics in order to
understand and an- alyze the governance of natural
resources (Ostrom 1990, 2010). Ostrom provided
pioneering work in expanding our understanding of the
relationship between institutions and management of
natural resources. According to Ostrom, suc- cessful
management of common pool resources cannot be
generalized to specific institutional arrangements or set of
rules; it rather depends on the context under
consideration. A key feature of Ostrom’s 1990 seminal
work is the eight design principles that set the conditions
for effectively man- aging community-based natural
resources such as irrigation systems, pastures, forests, and
fisheries. Ostrom argued that such design principles are
best practices for the sustainable management of common
pool resources (see Ostrom 2010).
Studies focusing on the governance of natural
resources in Africa have markedly shifted from
centralized systems to more decentralized ones (Ribot et
al. 2010; Andersson and Ostrom 2008). In a decentralized
system, natural resource management is shaped by the
informal institutions or local authorities who can entrust
power in resource allocations. Although a decentralized
system of natural resource gover- nance is not new, the
outcomes are mixed (see for example Andersson and
Ostrom 2008). Andersson and Ostrom (2008) argue that
“a perfect institution does not exist” and suggest a
polycentric approach. A polycentric perspective relies on
the complementarity of community-based institutions and
poli- cies in the governance of complex natural resources
(Ostrom 2010; Andersson and Ostrom 2008).
 Adela F.A. et

structures and sets of rules but are constantly (re)shaped political rights, access to important services and
through individuals’ practices; hence, resource scarcity infrastructure, and vulnerability (Namara et al. 2010; Smith
would not necessarily result in conflicts but rather lead to 2004). The food security literature classifies poverty into
institutional evolution (Lecoutere 2011). A polycentric absolute and relative poverty. Absolute poverty refers to
approach may be preferable for bringing about sustainable the headcount of house- holds who are unable to afford a
use of irrigation wa- ter as it offers the option of combining certain standard of basic goods and services. Relative
government policies, collective action, societal norms, and poverty, on the other hand, mea- sures the relative shortfall
the market. Most coun- tries in SSA are challenged by of a household’s income from the economy’s average.
economic water scarcity due to mismanagement, water Another concept related to but wider than poverty is equity,
theft, and turn abuses (Belay and Bewket 2013), and this which refers to the level of equality in in- come and wealth
has become a key source of failure for several small-scale distribution. Poverty is also a dynamic con- cept, and thus,
irrigation schemes in SSA, including Ethiopia (Amede factors affecting poverty can change (Smith 2004).
2015; Yohannes et al. 2017). Commonly, the Foster-Greer-Thorbecke (FGT) indices
In addition to the conventional farm- and household- (Foster et al. 1984) are used in poverty analysis:
related characteristics, several factors are suggested for the Σ Σ∅
extremely 1 H z−e
low use of irrigation in SSA, including deficient institutions ð1Þ
(Hanjra et al. 2009; Yohannes et al. 2017), poor irrigation P ¼ ∑i¼1
N z
infrastructure (Mdemu et al. 2017), mismanagement and
cor-
ruption (Namara et al. 2010; Dudu and Chumi 2008), and Where Z is the poverty line, N is the number of observa-
context-specific factors (Yami 2016). Governance of small- tions, H is the number of households below the poverty
scale irrigation matters because it affects the efficiency of line Z,
water use, investment in irrigation infrastructure, and the e is the consumption expenditure per capita for the ith
de- cision to adopt irrigation practices (Baldwin et al. person and ø is a poverty aversion parameter. At ∅ =0 P
gives the headcount index: the number of people below the
2016). Although several government-sponsored small- set poverty line. At ∅ = 1, the resulting P is the poverty
scale irriga- tion schemes are present across Ethiopia, there gap index, which measures the aggregate shortfall of the
is little evi- dence for the impact of governance on consumption of the poor from the poverty line Z. At ∅
adoption and house- hold food security. We argue that the =2 P is the squared
low adoption of small- scale irrigation in Ethiopia is poverty gap and measures the severity of poverty. As ∅ gets
largely attributed to scheme gov- ernance; this can have a larger, the measure gives more emphasis to the poorest of the
significant impact on farm income and consumption poor (Foster et al. 1984).
expenditure, which are associated with the availability and
access dimensions of household food security (Barrett
2010; Denny et al. 2018). This relationship is depicted in 3.2 Theoretical model specification
Fig. 1.
The decision to adopt a technology can be modeled in a
ran- dom utility framework by expressing the unobservable
utility from adoption and non-adoption through observable
3 Methodology and data variables (Khonje et al. 2015). Accordingly, the adoption
decision is modeled, considering the assumption that
3.1 Poverty smallholder farmers choose (not) to irrigate. It is assumed
that farmers choose (not) to participate in irrigation
The concept of poverty has evolved from the original idea schemes considering the benefits from participation
of inadequacy of income, consumption and wealth (Watts through the farm income derived from crop production.
1969) to concepts of capabilities and functioning (Sen The following model specifies the selection eq. P*, where
1981) and further to include multidimensional aspects such P* is the latent variable which is not observed. P* can be
as socio- expressed as a function of some observed farm, house-
hold, and institutional characteristics.
Access to information, market, and technology
P* ¼ αZi þ ui
Farm income
ð2Þ
I i ¼ 1 if P >*0 and I i ¼ 0 if P* ≤ 0
Adoption of small- scale irrigation
Farm and Scheme governance
household Consumption expenditure
Fig. 1 Conceptual diagram
 Small-scale irrigation scheme governance - poverty nexus: evidence from
Ii is a binary
variable which takes
a value of 1 for
farmers who irrigate
and 0 for those who
do not irrigate. Zi
represents factors
that affect the
irrigation decision. α
denotes the vector of
parameters
indicating the
magnitude and
direction of each
explanatory variable
on the decision to
irrigate. The residual
ui captures the
unobserved factors
and measurement
errors.
 Adela F.A. et

The following two regression equations represent the the following specification gives the outcome regression
two regimes that the smallholder farmers fall into: equations for the two regimes:
Regime 1 : Y 1i ¼ β1X i þ ε1i if I i ¼ 1 ð3aÞ users : Y 1i ¼ β1X i þ ε1i if I i ¼ 1 ð4aÞ
Regime 2 : Y 2i ¼ β2X i þ ε2i if I i ¼ 0 ð3bÞ non−users : Y 2i ¼ β2X i þ ε2i if I i ¼ 0 ð4bÞ
Y1i and Y2i are the dependent outcome variables The error terms ε1i, ε2i,and ui are assumed to have a
determined by the exogenous variables. Xi, β1, and β2 are trivariate normal distribution, with mean vector zero and
parameters that show the direction and strength of the co- variance matrix (Lee et al. 1982),
relationship between the outcome variables and the
2 2
:2 : 3
independent variables. ε1iand ε2i σu
are error terms. 4ð Þ¼
Cov ui; ε1i; ε2i σε1iu σε1i : 5 ð5Þ
σε2iu : σ2
3.3 Empirical model specification ε

Farmers decide to irrigate if they assume that the net Where σ2 variance of the error term in the selection
benefits
from irrigating is higher than that of not irrigating. Several u
equation, σ2 1 and σ2 are variances of the error terms in
2
ε ε
types of observable and unobservable factors may affect the outcome equations. σε1iu and σε2iu are covariance of ui
the farmers’ decision to enroll in small-scale irrigation and ε1i and ε2i respectively. Since Y1i and Y2i are not
schemes. The latter may result in a selection bias if observed simultaneously, a covariance of the correspond-
unobservable factors affect both error terms in the ing error terms is not defined (Maddala 1986). This struc-
selection equation (ui) and the outcome equation (ε). This ture of the error terms indicates that the error terms of the
results in a correlation between the outcome equation and the error term of the selection equa-
error terms of the two equations: corr(ε,ui)= ρ ≠ 0. The tion are correlated which results in non-zero expected val-
corre- ue of ε1i and ε2i given ui- error term of the selection
lation between the error terms implies endogenous equation (Abdulai and Huffman 2014). Therefore, the ex-
switching pected values of the truncated error terms E(ε1i| I = 1) and
(Maddala 1986). E(ε2i| I = 0) are given below:
The critical issue in impact study is acknowledging
potential biases. Often, the literature mentions two sources Eðε1ij I ¼ 1Þ ¼ Eðε1ij u > −ZαÞ
.
Zα
of biases: observable (farm and household) char- acteristics φ
and unobservable factors. In such cases, stan- dard
regression models (e.g., OLS) are biased and incon-
sistent (Khandker et al. 2009). Several methods are ap- σ ð6aÞ
¼ σε1iu ≡σε1i uλ1
plied in impact studies, including Propensity Score . Σ
Matching (PSM), Endogenous Switching Regression Zα
Φ
(ESR), and Heckman’s selection model. The latter two σ
are preferred over the former to account for a potential And,
selection bias because PSM takes a strong assumption of
unconfoundedness; i.e., in the context of the study, farm Eðε2ij I ¼ 0Þ ¼ Eðε2ij u ≤−ZαÞ
households’ decision to participate in the irrigation .
Zα
−φ σ ð6bÞ
scheme is based on observed characteristics. However, this ¼ σε2iu .
≡σε uλ2
assumption cannot be tested, and if violated, PSM Σ 2
1 Zα
would not be appropriate (Khandker et al. 2009). There
σ
are, however, differences between ESR and Heckman’s
selection model. Heckman’s model is appropriate to ac- φ and Φ are the probability density and cumulative
count for selection bias due to selective sampling of po- distribu- tion function of the standard normal distribution,
tential outcomes (Khandker et al. 2009). Nonetheless, in respectively. The ratio of φ and Φ evaluated at Zα is
referred to as the
inverse Mills ratio λ1 and λ2 (selectivity terms). If the
estimat-
our study, the two outcome variables (farm income and
per adult equivalent consumption expenditure) are ob- ed covariance σ2 and σ2 are significantly different from 0,
1 2
ε ε
 Small-scale irrigation scheme governance - poverty nexus: evidence from
served for both the irrigation users and non-users. In such the decision to irrigate and the outcome variables are
a situation, ESR is the preferred method for impact as- correlat- ed, and indicates the presence of a selectivity bias
sessment (Abdulai and Huffman 2014). ESR both tests (Maddala and Nelson 1975; Maddala 1986).
and corrects any self-selection bias due to observable Full Information Maximum Likelihood method is sug-
and unobservable characteristics. gested as an efficient method for estimating the model.
Provided that different farm and farmer characteristics Following this argument and considering the assumption
de- termine whether the farm household decides to irrigate re- garding the distribution of the disturbance terms,
or not, the
 Adela F.A. et

logarithmic likelihood function for the system of eqs. (5a) Wondo Genet district was purposively selected for its
Σ
and (5b) is given below: Σ
ε1i N rela- tive long years of practice in small-scale irrigation.
lnL ¼ I lnφ −lnσ þ lnΦðρ Þ There are two major rivers (Worka and Wosha) currently
being used
irrigated for
agriculture. The Worka river irrigation scheme was
i ∑i
i¼1 ε1i 1
σε1i Σ selected for the study because it is the oldest modern
ε2i scheme in the area and covers a wide area, with the
þ ð1−IiÞ σ −lnσεε2i þ ð7Þ potential to irrigate
Σ
selected
about 272 ha of agricultural land. The

Where ρ1 and ρ2 are correlation coefficients between the this study is clustered into 33 farmer groups, and only nine
selection equation error term ui and the error terms of the farmers groups were users of the irrigation scheme. A
outcome equations ε1i and ε2i. farmer group includes 40 to 75 households. There are a
Furthermore, estimations of treatment effects were total of 680 (660 male- and 20 female-headed) households
made. Average Treatment Effect on the Treated and currently using the irrigation scheme.
Untreated (ATT and ATU) are computed using the results The study used both qualitative and survey methods to
for expected values of the dependent variable for users and gather data.
non-users in actual and counterfactual scenarios: For the qualitative study, we carried out key informant
interviews, focus group discussions (FGDs) and direct
obser-
EðY φðZαÞ vations to explore the irrigation infrastructure, scheme
j I ¼ 1; 1iÞ ¼ β1X þ ρ 1 ΦðZαÞ ð8Þ
1i gover- nance, and water use practices, and triangulated the
i X 1i σϵ1iu
survey
φðZαÞ results. Key informants were held with three experts from
EðY j I ¼ 0; X Þ ¼ β X −σ ρ ð9Þ the district agriculture and rural development office (i.e.,
2i i 2i 1 2i ϵ u
ð1−ΦðZα the heads of the district, irrigation, and crop production
work pro-
EðY φðZαÞ cesses). The questions were related to irrigation practices
j I ¼ 1; 1iÞ ¼ β2X þ ρ 2 ΦðZαÞ ð10Þ
2i and implementation modalities of the Worka river
i X 1i σϵ2iu
irrigation
φðZαÞ season. According to the Ethiopian Central Statistical
EðY j I ¼ 0; X Þ ¼ β X −σ ρ ð11Þ Authority (CSA), the study area has a population of
1i i 2i 2 2i ϵ u
ð1−ΦðZα 191,116, and 78.13% live in rural areas. It is one of the most
ATT is the difference between the expected value of the densely pop- ulated areas in the country. The average
outcome variable from eqs. (8) and (10) and measures the landholding of the sampled households is 0.40 ha versus the
difference in the expected value of the dependent variable national average of
for users with and without the use of irrigation. ATU is the 1.37 ha. Small-scale perennial crop production is the domi-
difference between eqs. (9) and (11) and estimates the
differ- ence in the expected value of the outcome variable
for non- users had they used irrigation.

3.4 Study area, study design, and data

Wondo Genet is located in the south-central rift valley of

Ethiopia, about 263 km south of Addis Ababa. It covers an
area with altitude ranging from 1600 to 2580 m above sea
level. The mean annual temperature is between 17 °C and
19 °C and the climate is characterized as sub-humid with
bi- modal rainfall distribution. The area receives an annual
aver- age rainfall of 1079.7 mm; February to April is a low
rainfall period, and June to September is the main rainy
 Small-scale irrigation scheme governance - poverty nexus: evidence from
scheme. Likewise, three FGDs were held with irrigation disputes (e.g., D’Exelle et al. 2012; Kalu et al. 1995).
users and non-users, and one with water users’ Accordingly, the farm households were catego- rized into
associations (WUA) committee members. The results of three categories: upstream, middle-stream, and
the FGDs and key infor- mant interviews were mainly downstream. For this purpose, a two-stage sampling proce-
used to analyze the irrigation infrastructure, scheme dure was performed. In the first stage, the households
governance, and water use practices as well as to farming in the upstream, middle stream, and downstream
triangulate the survey results. parts of the ‘Worka’ river were identified. In the second
Following the qualitative study, a large quantitative stage, 80 house- holds (each) were randomly drawn (using
survey was carried out using a structured questionnaire. a systematic ran- dom sampling technique) from the three
First, a sam- pling frame was prepared based on a list of categories. The two- stage approach increases variation and
farm households obtained from the district agriculture and ensures that house- holds from the three locations are
rural development office registry. It was necessary to represented. Furthermore, the application of ESR in our
stratify all farm households based on their relative study would account for a po- tential selection bias in our
location along the scheme (i.e., the ‘Worka’ river) because sample. Second, the survey ques- tionnaire was pre-tested
spatial variation in irrigation water use can lead to for potential ambiguity in the survey questions. Finally,
data were collected from 240 smallholder
nant farming system in the area with ‘enset’, khat, and sugar-
1
cane being the major crops (Dessie and Kinlund 2008). Peasant association, which is the lowest administrative unit in Ethiopia
 Adela F.A. et

farmers located in Wotera Kechema ‘kebele’ and based on awareness of water storage systems that could be used to
information about the 2014/15 cropping season. The harvest water during the rainy season. Supplemental
sample represents 35% of the irrigation users along the irrigation in the study
'Worka’ river irrigation scheme. Five enumerators
undertook the household survey following intensive
training and practice. The selected enumerators spoke the
local language (Sidamu Afo), which was used to conduct
the survey. The principal investigator also speaks the local
language, which made it easier to supervise and coordinate
the data collection process.

4 Results and Discussion

4.1 Analysis of the irrigation

infrastructure and scheme

The modern irrigation scheme was built in 1993 (with the

help of an NGO) with the potential to irrigate 272 ha of
land both in the upstream and downstream part of the river.
The headwork of the scheme on the river channels water
into two diversions (Wotera Kechema and Wosha
diversions), serves two ‘kebeles’, and currently covers a
total of 245 ha. The main canal used to divert the river to
‘Watera Kechema’ is 4.6 km long from the headwork.
Only part of the major canal (1.648 km), which is used to
divert the river from the head- work, is concrete and
cement lined. The rest of the waterway is an earthen canal
dug by the farmers themselves, leading to heavy water loss
during transportation. Generally, rainfall in excess of
evapotranspiration and groundwater recharge is re- quired
for a profitable investment in small-scale irrigation (You et
al. 2011). The study area has an average annual rain- fall
(i.e., 1079.7 mm), which is 35 mm above the African
average and only 71 mm less than the world average. This
problem is not unique to the study area. Water loss through
seepage and runoffs is also a common problem in other
irri- gation schemes in Ethiopia (Yohannes et al. 2017).
Records obtained from the district agriculture and rural
development office indicates an average conveyance
efficien- cy of canals at 65%, suggesting high water loss in
the study area. A similar study in India (Rath and Swain
2018) reported a conveyance efficiency of 90.5%
(Senhapali distributary) and 81.2% (Paramanpur
distributary). In the study context, the mode of irrigation is
only furrow, and modern irrigation tech- nology is
nonexistent. Furthermore, communal or private wa- ter
storage facility is almost zero in the area; only one farmer
reported having some kind of water storage system. Of the
farmers who did not have any water storage mechanism
(res- ervoir), 13.7% of them claimed to have either a small
farm, no storage space or a steep slope; 24% of the farmers
lacked surplus water to reserve; and 53% of the farmers
reported to have let the water go. The results imply limited
 Small-scale irrigation scheme governance - poverty nexus: evidence from
area is mostly practiced from December to March. During However, users were expected to contribute (1) free labor
these months, khat and sugarcane, on average, get 3 and 4 to clean and maintain the tertiary canals, mostly once a
days of irrigation per week, respectively. As per the year, just before the start of the irrigation
bylaws of the WUA, the allocation of water is based on
the size of the irrigable plot for all users. However, water
requirements of crops, distribution and use are not
considered.
Currently, the irrigation scheme is constrained by a
high level of ‘water scarcity’, according to the district
agriculture and rural development office. This was later
confirmed by 99% of the farmers in the survey; they
claimed to have per- ceived ever decreasing precipitation
in the area over the last 10 years. The allocation of
irrigation water has been carried out by the WUA. Before
the establishment of the modern scheme, irrigation use
was governed through informal institu- tions, called
‘water fathers’ – a well-respected and trusted group of
individuals in the community. Our FGDs revealed that
little or no problem was observed when the allocation of
irrigation water was administered under these traditional
leaders. However, disputes over the allocation of
irrigation water surfaced with the introduction of the
modern irrigation scheme. More importantly, ‘water
fathers’ were left out of governance of the scheme (i.e.,
allocation decisions, dispute resolution, etc.) and replaced
by the WUA.
Key informants, which involved only local government
officials, however, did not attribute the current irrigation
water scarcity to governance of the scheme but rather to
physical water scarcity and population pressure.
Approximately 92% of the households in the survey d that
irrigation water was not sufficient to meet their current
needs. The remaining 8% of the households, who
responded positively, were all from up- stream areas.
Despite differing views on the source of the problem,
both the FGDs and key informant interviews con- firmed
increasing irrigation water scarcity in the study area,
suggesting the necessity of more effective governance of
the scheme.

4.2 Analysis of the institutional environment

The study provides an overview of three types of

institutions influencing the use of irrigation water in the
study area: col- lective action (WUA), market
characteristics, and access to information.

4.2.1 Water user association (WUA)

Like most water use associations in Ethiopia, the WUA

was established in 1993 by the government in 1993,
which dis- mantled the traditional practices of irrigation
water use that had existed for many years (see section
4.1). Participants of the FGDs revealed that members paid
no fees for membership or the use of irrigation water.
 Adela F.A. et
2
season; (2) contribute money if the main canal was One US dollar is approximately equivalent to 27.3 Ethiopian birr.
3
One US dollar is approximately equivalent to 27.3 Ethiopian birr.
damaged and needed to be repaired; and (3) pay 20 to 50
birr2 to cover part of the expenses related to irrigation canal
operators and water guards.
The use of irrigation water is managed by a committee
whose members are elected by the WUA. Depending on the
number of members, each farmer group in the study area is
represented by one or two individuals of the committee. A
committee usually consists of 11 individuals: a chairperson,
deputy, and secretary; two controllers; three water distributors;
and three members of the conflict resolution sub-committee. A
committee member is supposed to serve for a year but can be
re-elected for an unlimited number of times. According to the
bylaws, a committee member could be suspended or
dismissed from the committee for violat- ing the stated rules.
According to the participants of the FGDs, local authorities
and development agents have undue influence on the
(re)election process. As a result, committee members often
align themselves with local authorities, further eroding the
users’ trust in the association. As per the bylaws, the
committee should meet once per week to discuss the
maintenance of the irrigation canals, watershed management,
water allocation, distribution schedules and other outstanding
issues related to the irrigation scheme. But participants during
the FGDs revealed that the com- mittee members do not meet
regularly.
There is no local water use policy in place to guide the
committee’s decision making. According to participants of
the FGDs, the current water use practices were flawed .
Over-abstraction of water by upstream users and water
theft were common despite the stated fine of 500 birr3 for
such acts and there were repeated complaints against
committee mem- bers and the current scheme of
governance. Most of the com- plaints were related to the
lack of established water use poli- cies and corrupt
practices. Some of the direct quotes from the FGDs related
to the current scheme governance include the following:

& “There should be fair distribution of water between the

rich and the poor, and the committee members should
have integrity.”
& “The committee should distribute the water fairly
between people in the upstream and downstream part of
the river.”
& “The committee should create awareness programs on
the use of irrigation water.”
& “They should create a forum for all the users to find
better solutions about current use of irrigation water.”
& The committee should work on improving relationships
among users in the community.
& “Though there is scarcity, there are ways to get more
wa- ter. If you can bribe the committee and water
guards, you can get enough water”.
 Small-scale irrigation scheme governance - poverty nexus: evidence from
& “They are deceivers, and highly corrupt; something en- hancement of food production and food security in
must be done about it.” rural com- munities. In the context of this study, the district
agriculture
The current conflicts over the use of irrigation water in
the study area appear systematic. The conflicts were
largely attrib- uted to governance rather than physical
water scarcity. In the words of one of the participants, “It
is the committee that is creating the scarcity. We are
trying to adapt to it by bribing its members”. Indeed, the
problem is expected to be worse in the future as the
physical water scarcity increases. Therefore, de- spite a
huge investment in small-scale irrigation projects in
Ethiopia, the existing practices of irrigation water use are
largely ill-defined (Amede 2015). Several studies in
Africa have documented the increasing importance of a
polycentric approach in the governance of small-scale
irrigation schemes (e.g., Lecoutere 2011; Baldwin et al.
2016). This approach has proved to be effective in
searching for new solutions (i.e., institutional
arrangements) as disputes emerge.

4.2.2 Market characteristics

Market characteristics are one aspect of the institutional

fac- tors considered important in this study for two main
reasons. First, the decision to grow crops during the off-
rainy season (and thus become a member of the WUA)
may be influenced by the households’ market orientation.
Second, the two most important crops frequently grown
using irrigation water (i.e., sugarcane and khat) are the
most important cash crops in the study area.
The local markets are located approximately within a
5.4 km radius of the study area. Considering a cut-off
travel time of five hours to the nearest market (You et al.
2011), the households in the study area have modest
access to market. Nonetheless, the marketing system in
that part of the country is underdeveloped; most farmers
sell their produce at the farm gate and often involve
brokers (Abebe et al. 2016). For exam- ple,
approximately 48% of sales from all crops and 91% of
sales related to sugarcane and khat involve brokers. There
are no well-functioning marketing cooperatives in the
area. Like many other places in Ethiopia, farmers in the
study area are dependent on the government for the
supply of modern inputs such as fertilizers and improved
seed varieties. Access to for- mal credit is very limited
even though two micro-finance in- stitutions – Sidama
and Omo– have branches located in the district; only
4.5% reported to have had access to credit from both
formal and informal sources.

4.2.3 Access to information

Access to information and extension services is critical to

understanding the importance of irrigation water and the
 Adela F.A. et

and rural development office is responsible for the

4.3 Descriptive statistics of variables used in
provision of extension services. These were provided by model estimation
four extension agents, specializing in crop production,
natural resource pro- tection, livestock husbandry and
Table 1 presents descriptive statistics of the survey data for
veterinary work. The exten- sion agents were primarily
the relevant variables included in the model estimation.
responsible for: (1) the promotion and dissemination of
Chi- square test and t-test were used to test for statistical
modern agricultural inputs; (2) the pro- vision of technical
differences between irrigation users and non-users on the
support and advice to farmers; (3) organiz- ing and training
mean values of the categorical and continuous variables,
farmers in new farming practices; and (4) providing up to
respectively.
date information related to weather and mar- ket
Irrigation users had a significantly higher mean value
conditions. The study area was strategically located close
for all the variables except for those related to gender,
to the regional research center (South Agricultural
education level, off-farm income, and extension services
Research Institute) and an agriculture-based academic
(Table 1). The number of households among the non-users
institution (Hawassa University College of Forestry).
who acknowledged high level of water scarcity was sig-
However, exten- sions services from such institutions were
nificantly higher than the user households. Mean values
fragmented and limited to their thematic areas and research
for ownership of radio and mobile phone, which are used
priorities.

Table 1 Summary statistics for irrigation users and non-users

Variable and variable definition Measure Users (n = 160) Non-users (n = 80) p value

Household characteristics
Head of the household (1 = male) Dummy 0.95 0.98 0.20[1.61]
Age of the head of the household (year) Continuous 46.95 43.58 0.038**
Household size in adult equivalent units Continuous 5.61 5.09 0.024***
Labor endowment in adult equivalent units Continuous 3.86 3.26 0.003***
Education level of the head (school years) Continuous 2.96 2.73 0.285
Highest education by an adult member of the household (school years) Continuous 4.22 3.37 0.038**
Farm characteristics
Landholding per plot (ha) Continuous 0.172 0.148 0.02***
Land covered by cash crop Continuous 0.780 0.613 0.000***
Asset
Landholding (ha) Continuous 0.436 0.323 0.007***
Household asset value (ETB) Continuous 66,709 26,928 0.000***
Income and Consumption
Farm income per ha (ETB) Continuous 432,507 101,689 0.000***
Non-farm income (ETB) Continuous 3,166 4,033 0.230
Total per adult equivalent food consumption expenditure (ETB) Continuous 3965 2848 0.000***
Access to market and modern technology
The distance of the nearest market (km) Continuous 5.29 5.74 0.060**
Chemical fertilizer use per ha (kg) Continuous 929.5 621.8 0.000***
Insect/herbicide use per ha (ETB) Continuous 21.5 12.7 0.002***
Institutional and information access related variables
Visit by extension agent during last 6 months (Yes = 1) Dummy 0.88 0.87 0.776[0.08]
Perceived water scarcity (highly scarce = 1) Dummy 0.50 0.97 0.000[52.41]***
Ownership of radio (Yes = 1) Dummy 0.63 0.46 0.013[6.21]***
Ownership of mobile phone (Yes = 1) Dummy 0.66 0.35 0.000[21.12]***
Extent of scheme governance problem (1 = little/no; 5 = high Ordinal 4.3 4.9 0.000[33.68]***
Measures taken to increase water share (1 = little/no effort 5 = high Ordinal 3.41 3.3 0.000[42.99]***
efforts including bribing)
Role of WUA committee in the prevailing scheme governance Ordinal 3.5 3.7 0.12[5.71]*
problem (1 = little/no, 4 = high)
 Small-scale irrigation scheme governance - poverty nexus: evidence from
***significant at 1% level, **significant at 5%level, significant at 10% level. Chi-square in square brackets
Source: own calculation from survey data
 Adela F.A. et

to access information, were significantly different be- indices. Based on the recommended daily energy
tween the users and non-users. requirement of the
In addition, sampled households were asked several
ques- tions regarding the governance of the irrigation
scheme in- cluding: (1) their level of
agreement/disagreement regarding the prevailing water
scarcity and governance scheme (2) ac- tions taken at the
household level to increase water share, and
(3) the role of the WUA committee in the prevailing water
scarcity and governance issues. The first variable was mea-
sured using a five-point Likert scale. A score of five to the
question “To what extent do you agree that the current water
scarcity problems are largely due to mismanagement rather
than physical scarcity of water” demonstrates the highest
lev- el of agreement. A limitation with the odd-number
response Likert scale is the central tendency bias (related to
interpreting a midpoint scale such as “indifferent”, “no
opinion”, etc.). In our study, this problem was minimized as
the survey was administered in person; respondents were
encouraged to ask for clarification and whether they
understood the questions before giving their answers. The
other two variables – ‘Measures taken to increase water
share’ and ‘Role of WUA committee in the prevailing
scheme governance problem’ – were constructed by
developing indices based on responses to several questions.
Scheme users and non-users generally agreed that the
pre- vailing water scarcity was due to poor governance;
however, the mean score of the latter was significantly
higher (Table 1). Tests for the indexed variables were also
significant. The mean value for the current irrigation
scheme participants was higher than the non-users,
meaning that the former tend to take some measures
(including bribing WUA committee members) to
maximize their water share (Table 1). The third variable
measured the role of WUA committee in the prevail- ing
governance problem. The mean values were close to four
for both users and non-users, indicating poor governance. A
similar result was found in a study undertaken in Southern
Ethiopia, where farmers rated fairness of water allocation
by WUA as ‘poor’ (Yami 2013).
The descriptive statistics, however, provides only the
ob- servable situation and mean differences cannot be used
to make conclusions about the factors that affect farmers’
deci- sions whether to irrigate or not. Furthermore, the
above anal- ysis does not account for unobservable
characteristics. Results of the econometric analysis are
presented and discussed below.

4.4 Econometric results and discussion

4.4.1 Poverty analysis

The level of poverty between irrigation water users and

non- users was tested using Foster-Greer-Thorbecke (FGT)
 Small-scale irrigation scheme governance - poverty nexus: evidence from
2100Kcal, Poverty line (Z) of Birr 3,329.27 (=USD Severity 0.008 0.06
123.30) per adult equivalent per year was used to estimate
Source: computation on own survey date
the FGT indices of poverty. The poverty line was
constructed using food and non-food per adult equivalent
consumption expen- diture of the households (Table 2).
The incidence of poverty was measured by the
headcount index and shows that 67% of the households
that did not have access to irrigation fell below the
consumption-based poverty line (Z) of ETB 3,329. On the
other hand, only 28% of the farmers who participated in
the irrigation scheme were below the poverty line. This
result reinforces the claim that preva- lence of poverty is
higher among households relying heavily on rain-fed
agriculture as opposed to those that irrigate (Bacha et al.
2011; Hanjra et al. 2009). The depth and severity of
poverty were also higher among the non-users. The
consump- tion expenditure of the non-users should be
increased by 18% of its current amount if they were to be
lifted out of poverty but only by 3% for users. These
results are in line with most micro level empirical studies
on poverty and irrigation link- ages. Bacha et al. (2011),
for instance, reported the depth of poverty among non-
irrigation users in the Ambo district of Western Ethiopia)
at 21% while it was only 10% among users.

4.4.2 Endogenous switching regression (ESR)

model estimation results and discussion

Consistent with the descriptive and poverty analysis

above, there is a significant difference between the
users and nonusers in several relevant variables and
welfare indicators. These differences could be due to
several observable and un- observable factors in addition
to differential access to irriga- tion. Two outcome variables
were used as proxies for food security: farm income per ha
and adult equivalent food con- sumption expenditure. Such
indicators have been frequently used in past studies
(Amare et al. 2012; Shiferaw et al. 2014). Table 3 presents
the estimation results for the model with farm income per
hectare as the outcome variable.
The second column of Table 3 reports the estimates for
the determinants of small-scale irrigation adoption. Education
and age have a positive non-significant association with
farmers’ decision to participate in the irrigation scheme
though other studies have indicated a positive relationship
(e.g., Deressa et al. 2009; Nhemachena and Hassan
2007). Huffman

Table 2 Foster-Greer-
Thorbecke indices on Poverty estimates Groups
consumption at α = 0, 1,
and 2 and Z = ETB Users Nonusers
3746.77
Incidence 0.28 0.67
Depth 0.03 0.18
 Adela F.A. et

Table 3 Estimates of the switching regression model for farm income per ha

Variables Model Estimates

Irrigation 1/0 Users Non-users

Highest education level of adult member 0.023(0.64) −0.018(1.58)* 0.004(0.21)

Age of the household 0.011(1.23) −0.003(0.89) 0.002(0.33)
Log of distance to the nearest market −0.001(0.03) −0.021(1.09) −0.040(1.21)
Visit by extension workers 0.078(0.24) 0.198(1.61)* 0.178(0.84)
Log of landholding 0.434(1.34) 0.415(3.73)*** 0.578(2.45)***
Log of landholding per plot −0.441(1.08) −0.267(1.95)** −0.652(2.33)***
Log of percentage of land covered by cash crop 1.27(3.48)*** 0.423(3.10)*** −0.295(1.39)
Log of time endowment in adult equivalent units −0.003(0.01) 0.019(0.19) 0.110(0.53)
Log of non-farm income −0.096(2.80)*** −0.001(0.09) 0.046(2.51)***
Log of chemical fertilizer applied 0.115(0.86) 0.202(3.81)*** 0.095(1.18)
Log of pesticide applied 0.092(1.09) 0.149(4.09)*** 0.142(2.61)***
Owned radio(Yes = 1) 0.594(2.50)***
Owned mobile phone (Yes = 1) 0.959(4.02)***
Perceived water scarcity (High = 1) −1.911(5.10)***
Extent of scheme governance problem (1 = little/no; 5 = high −0.846(3.35)***
Measures taken to increase water share 0.149(1.85)**
(1 = little/no effort 5 = high efforts including
bribing) −0.330(2.12)**
Role of WUA committee in the prevailing
scheme governance problem (1 = little/no, 4 =
high)
Constant 4.49(2.48)*** 11.40(27.15)*** 9.31(12.51)***
ρ1, ρ2 −0.822[0.115]*** −0.518[0.232]**
Model diagnosis
Wald x2 115.40**
Log-likelihood −255.09
LR test of independence 19.76***

Absolute value of z statistics in parenthesis standard errors in square brackets

*Significant at the 10% level, ** Significant at the 5% level, ***Significant at the 1% level
Source: Own calculation using survey data

(2001), however, argues that when an intervention has been results on relationship between plot size and the adoption of
there for a relatively long time, education and experience agricultural technology (Bradshaw et al. 2004; Deressa et al.
may not significantly affect the decision to participate. 2009; Khonje et al. 2015).
Thus, the positive but non-significant relationship between
education and age and adoption could be because irrigation
has been practiced for at least the last 30 years in the study
area.
Land covered by cash crops has a positive impact on the
decision to irrigate; households with a larger share of their
land covered by cash crops are more likely to irrigate than
others with smaller proportions of land covered by cash
crops. In the study area, irrigation was practiced in the
cultivation of the most important cash crops, sugarcane and
khat. Land size per plot, on the other hand, had a negative
effect on the decision to irrigate. This is possibly because
farmers with smaller plots increase their food production
through irrigation in addition to relying on the occurrence
of rain. Previous studies, however, documented mixed
 Small-scale irrigation scheme governance - poverty nexus: evidence from
Similarly, time endowment shows a negative
association with the decision to irrigate. This can be
attributed to the high population density and land
fragmentation in the study area. This result is consistent
with one study in Ethiopia (Tizale 2007) but different
from another study which showed a posi- tive association
between labor endowment and adoption deci- sion
(Deressa et al. 2009). In southern Ethiopia, however, the
rural youth is forced to search for alternative livelihood
op- tions because of the scarcity of agricultural land
(Bezu and Holden 2014). It is worth noting that the
average land holding of the surveyed households is 0.4
ha, which is significantly less than the national average of
1.37 ha (Central Statistical Agency and the World Bank
2013).
Non-farm income has a negative and significant impact
on the practice of irrigation. This result is consistent with
eco- nomic theory that participation in non-farm activities
may constrain the amount of labor available for farming.
As expected, intensive use of chemical fertilizer and
pesticide is positively correlated with the decision to
 Adela F.A. et

irrigate. This result is in keeping with previous studies water for both upstream and downstream farmers in the
(Gebregziabher et al. 2009; Namara et al. 2010; Smith area. However, the findings of the survey sug- gest
2004) and confirms the view that a stable supply of otherwise. These were consistent with the results of both
irrigation water would encourage farmers to invest in
productivity-enhancing inputs (Aberra 2004).
Variables for access to information (proxied by
ownership of radio and mobile phones), perceived level of
water scarcity, and scheme governance (proxied by three
variables) were used in the selection equation but not in the
outcome equation. This is because a correct specification
of the model requires the inclusion of at least one
explanatory variable in the former which directly affects
the irrigation decision but not the out- come variable
(Abdulai and Huffman 2014; Khonje et al. 2015).
Estimates for ownership of radio and mobile phone
variables were positive and significantly different from
zero. Farmers’ perception of the scarcity of water in their
area was negatively and significantly correlated with the
likelihood of irrigation water use. This is supported by
previous studies from developing countries, including
Ethiopia (Hanjra et al. 2009; Namara et al. 2010).
Scheme governance is the main focus of this study, and
was measured by three variables – ‘Extent of the scheme
gover- nance problem’, ‘Measures taken to increase water
share’, and ‘Role of WUA committee in the prevailing
scheme gover- nance’. As shown in Table 3, all three
variables are signifi- cantly correlated with farmers’
decision to participate in the irrigation scheme.
Accordingly, the existing scheme gover- nance in the study
area is affecting the irrigation water use in three ways.
First, it is creating additional scarcity unrelated to the
physical availability of water due to the poor management
of the physical infrastructure and corrupt practices.
Second, due to the perceived water scarcity, some farmers
are excluded from participation and forced to rely soley on
rain-fed agricul- ture. Finally, other farmers have to incur
additional expenses for bribing committee members and
water guards. About 78% of the respondents believe that
members of the WUA commit- tee were corrupt with
regard to the allocation of irrigation water. Farmers who
have an adaptation strategy (e.g., bribing committee
members and water guards) are more likely to be users of
the irrigation scheme. Of the households that current- ly
have access to the irrigation scheme, 34% stated that they
had bribed committee members or water guards to get
more water during the last irrigation season. Other
studies in Northern and Eastern Ethiopia have also
indicated the exis- tence of ‘bribery’, ‘corruption’, and ‘rent-
seeking’ behavior among WUA committee members
(Deneke et al. 2011; Yami 2013). Sampled farmers
perceived that the WUA com- mittee was responsible for
the prevailing governance problem of the scheme. Current
bylaws authorize theWUA committee to develop a system
that ensures fair distribution and efficient use of irrigation
 Small-scale irrigation scheme governance - poverty nexus: evidence from
the FGDs and past studies. These show that poor crops in the area, at least twice per year (mostly three
governance is a key factor in small-scale irrigation times), while non-users harvested the same crop only once
schemes in SSA (Belay and Bewket 2013; Dudu and per year (due to their reliance on rainwater). The strong
Chumi 2008). correlation between the share of
The model estimates of the variables against farm
income per ha for users and non-users are presented in the
third and fourth columns of Table 3. Correlation
coefficients ρ1 and ρ2 show that they were statistically
significant for both adopters and non-adopters. Indeed,
the use of ESR was proved to be appropriate in our study,
as the results indicated the existence of selection bias. The
estimate was also negative for both users and non-users
indicating positive selection bias such that farmers with
above-average farm income tended to irrigate. The
likelihood ratio test was also significant, indicating the
existence of joint dependence between the outcome and
selec- tion equation between users and non-users.
Education showed a negative and marginally
significant effect on farm income per ha for the adopters.
Di Falco et al. (2011) also found a similar result in
Ethiopia. Accordingly, households with a high education
level in the family tend to be involved in nonfarm
employment. This, however, may come at the expense of
the farm income because of less time being allocated to
farm activities. Distance to the nearby local markets also
negatively and significantly affected the farm income per
ha for both users and non-users. The proxy vari- able
measuring access to information (agricultural extension
agents) showed a positive association with farm income
for both users and non-users. This is in line with the
theory that farmers with better information and access to
extension agents are likely to have better productivity
(Abdulai and Huffman 2014). Landholding had a positive
and significant impact on the outcome variable for both
users and non-users. This is consistent with an earlier
study in Ethiopia (Belay and Bewket 2013). On the
contrary, landholding per plot had a negative and
significant impact on farm income for both users and non-
users. This may be due to an inverse farm size-
productivity relationship. Past studies have indicated that
small farms are more productive than big farms (Abdulai
and Huffman 2014). A study in Ethiopia also found a
strong association between land pressure, crop yield, and
income (Headey et al. 2014). The amount of non-farm
income was positively correlated with non-users. This
may suggest that failure to participate in irrigation
schemes may force farmers to look for alternative sources
of income. As expected, inten- sive use of chemical
fertilizers and insecticides/herbicides was positively
associated with farm income of both users and non- users;
this relationship was very strong for irrigation scheme
users. The proportion of farm covered by cash crops, such
as khat and sugarcane, had a significant and positive
impact for irrigation users. For example, scheme
beneficiaries harvested khat, one of the most traded cash
 Adela F.A. et

irrigable land and cash income was also confirmed by a positive and significant relationships reported between
sim- ilar study in other parts of Ethiopia (Amede 2015). non- farm income and increased consumption for non-
Model estimation was also carried out using total per users. This is expected as non-users tend to augment their
adult equivalent food consumption as the dependent income and con- sumption expenditure from non-farm
variable. Table 4 presents the model estimates. activities (Dorward et al. 2004). Similarly, in keeping with
The results of the selection equation with food earlier findings (Bacha et al. 2011), the value of household
consump- tion considered as outcome variable are assets was positively and sig- nificantly correlated with
presented in the sec- ond column of Table 4. The direction consumption expenditure for both groups of farmers.
of relationships reported in Tables 3 and 4 are similar, with Finally, the average treatment effect on the treated
some variation in the level of significance. New (ATT) and untreated (ATU) is presented in Tables 5 and 6.
explanatory variables such as farm in- come and the value As presented in Tables 5 and 6, access to irrigation
of household assets had a positive and highly significant signif- icantly affects both outcome variables for both
effect on the farmers’ decision to irrigate; these findings groups. For the current irrigation users, their farm income
are consistent with those of Bacha et al. (2011) and per hectare and annual per adult equivalent food
Deressa et al. (2009). consumption expenditure would have decreased by 42%
The factors that affect food consumption are reported in and 35%, respectively if they had not used irrigation. This
the third and fourth columns of Table 4. Estimates for suggests that current irrigation users would have lost farm
variables related to household were negative and income per ha of birr 151,419 (or US$ 5,500) and per adult
significantly different from zero for both users and non- equivalent consumption expenditure level of birr 1,338 (or
users. This may be because, other things kept constant, US$ 49) had they not used irrigation water, respectively.
large household size means less consumption per-head. Similarly, the farm income per ha and annual per adult
Similar results were reported by Khonje et al. (2015) and equivalent food consumption expenditure of non-users
Bacha et al. (2011) in the context of Eastern Zambia and would respectively increase by 149% and 40% if
Western Ethiopia, respectively with

Table 4 Estimates of the

switching regression model Variables Model Estimates
for
per adult equivalent food
consumption Irrigation 1/0 Users Non-users

Log of farm income 1.55(5.75)*** 0.075(2.16)** 0.061(0.87)

Highest education level of adult member −0.013(0.28) 0.001(0.23) 0.002(0.29)
Age of the household head 0. 013(0.91) 2.6 e-05 (0.02) 4.2–04(0.16)
Log of landholding −0.919(2.07)** 0. 033(0.65) 0. 060(0.83)
Log of land size allocated to cash crops 0. 394(1.20) 0. 032(0.93) 0. 056(1.15)
Log of adult equivalent household size −0.559(1.23) −0.365(7.26)*** −0.597(5.81)**
Log of productive adult equivalent labor to total −0.123(0.17) −0.218(2.69)*** −0.157(0.94)
adult equivalent household size
Log of non-farm income −0.036(0.90) −0.001(0.25) 0. 017(2.26)**
Log of value of household assets 0. 219(2.63)*** 0.034(3.13)*** 0. 029(1.73)*
Log of number of extension officers visit −0.254(1.68)*
Owned radio(Yes = 1) −0.208(0.65)
Owned mobile phone (Yes = 1) 0. 575(1.70)*
Perceived water scarcity (high = 1; Low = 0) −1.355(2.86)***
Mismanagement of the scheme −1.318(3.34)***
Measures/Actions to get more water 0.145(1.17)
Constant −13.609(3.58)*** 7.66(15.60)*** 8.06(9.92)***
ρ1, ρ2 0. 35[0.250]* 0. 34[0.459]*
Model diagnosis
Wald x2 81.59***
Log-likelihood −35.77
LR test of independence 2.03*

Absolute value of z statistics in parenthesis and standard errors in square brackets

*Significant at the 10% level, ** Significant at the 5% level, ***Significant at the 1%
Small-scale irrigation scheme governance - poverty nexus: evidence from
level Source: Own calculation using survey data
 Adela F.A. et

Table 5 Impact of irrigation

on farm income and Outcome variable Mean outcome ATT t-value
consumption- ATT
I= 1 I =0

Log of farm income per ha 356,407 204,916 151,419 42% 7.73***

Log of total per adult equivalent consumption 3,817 2,479 1,338 35% 18.47***

***significant at the 1% level, Source: own calculation from survey data

Source: own calculation from survey data

they had had access to irrigation. The better livelihoods of key determinants of participation in irrigation schemes and
beneficiaries of the irrigation scheme was also confirmed found scheme governance to be an important factor
through direct observations of the housing conditions by (Table 3). Both the FGDs and survey results confirm that
the principal investigator and during FGDs and key the governance of the existing irrigation scheme was highly
informant interviews. In one of the FGDs, for instance, one defi- cient, discouraging households from participating in
of the par- ticipants described the situation as follows“…It the scheme. Thus, if small-scale irrigation schemes are to
is a pity that our fellow farmers who are using irrigated contrib- ute to food security by means of increased farm
agriculture are making lots of money while those of us who income (i.e., food availability) and improved consumption
rely on rain-fed agriculture run out of cash off the rainy (i.e., accessibil- ity), the existing irrigation scheme of
season and sell our produce in seasons when everything is governance (i.e., the allocation of irrigation water via the
cheap…” Another par- ticipant added that “Lack of WUA committee) has to be improved. One way to do this
irrigation water has reduced my production potential by is to involve informal institutions such as ‘water fathers’ in
more than half.” This assessment is consistent with the decision-making process. Ironically, the current
previous studies that reported a positive direct link situation has created additional scarcity, excluding some
between irrigation use and farm income (Amede 2015; farmers from being users and leaving others to obtain water
Bacha et al. 2011; Belay and Bewket 2013; Smith 2004; through corrupt practices.
Wichelns 2014). According to De Haen (2003), the use of
irrigation has the potential to increase crop yields by 100 to
400% compared to the rainfed crop production system.
5 Conclusions
In sum, the findings confirm that access to a small-scale
irrigation scheme was significantly correlated with Although past studies generally highlight the potential
increased farm income and food consumption (Tables 5 impact of investment on small-scale irrigated agriculture in
and 6) and has addressed the availability and access enhancing productivity and food security, empirical
dimensions, which are prerequisites for household food evidence is varied and context-specific. Among the less
security (Barrett 2010; Denny et al. 2018). Irrigation explored but perhaps the most important in agricultural
allows year-round production of two of the locality’s most irrigation is the role of in- stitutions. Only a few studies
important cash crops, sugarcane and khat. Unsurprisingly, have documented the impor- tance of governance and
irrigation users had a lower inci- dence of poverty (28%) informal institutions in enhancing the contribution of
than the non-irrigation users (67%). This is consistent with small-scale irrigated agriculture to food security
several studies in Ethiopia (e.g., Tesfaye et al. 2008; (Mutambara et al. 2016; Mukherji 2012).
Bacha et al. 2011; Hussain and Hanjra 2004). For example,
This study has analyzed the determinants of adopting
Tesfaye et al. (2008) reported a house- hold food security
small-scale irrigation schemes and the impact these may
rate of 70% for small-scale irrigation users while the
have on food security in the case of Ethiopia, where
corresponding rate for non-users was only 20%. However,
agricultural land is extremely fragmented and densely
our study went a step further in order to identify
populated. Ethiopia provides an interesting perspective
as it has been

Table 6 Impact of irrigation

on farm income and Outcome variable Mean outcome ATU t-value
consumption- ATU
I= 1 I=0

Log of farm income per ha 197,404 79,09 118,307 149% 10.46***

7
Log of total per adult equivalent consumption 3779 2,690 1,089 40% 12.44***

***significant at 1% level
Source: own calculation from survey data
 Small-scale irrigation scheme governance - poverty nexus: evidence from

aggressively pursuing small-scale irrigation schemes irrigation use has severely affected participation and the
across the country. Empirical work was carried out in a potential for reducing poverty through primary food pro-
region where relatively modest annual rainfall is present, duction. These findings also provide several policy
providing a good opportunity to determine behavioral and
institutional factors influencing the decision whether or not
to enroll in irrigation schemes. Assessment of the physical
irrigation infrastructure revealed performance problems of
the canals, leading to sig- nificant water loss during
transportation. Irrigation technology was only a furrow, and
there were no water storage facilities. Poverty analysis was
made using the FGT indices based on an own constructed
consumption poverty line. The incidence, depth, and
severity of poverty were found to be higher among farmers
who did not have access to irrigation. Further analysis was
made, using ESR, to examine the determinants of
farmers’ decisions to irrigate and the impact of irrigation
on household food security. The results indicated selection
bias among users and non-users. Variables relating to
institutions – governance of the irrigation scheme, access
to information – and perceived water scarcity had a
significant impact on farmers’ decisions to irrigate. The
use of irrigation water was managed by the WUA
committee, which was described as untrustworthy, unfair,
corrupt and exacerbating water scarcity. Among the other
factors that affected farm income were in- tensive use of
chemical fertilizers and pesticides, total land- holding,
landholding per plot, and distance to nearest market.
Regarding per adult equivalent food consumption, farm in-
come, non-farm income, household assets, household size
and dependency ratio had significant impacts. The
treatment effects (ATT and ATU) were positive and
significant for both users and non-users, indicating that
access to irrigation has resulted in a significant positive
impact on farm income and consumption expenditures. Our
findings show that even with poor irrigation infrastructure
and technology and small plot sizes (0.4 ha on average),
participation in small scale irrigation schemes had
significantly improved farm income by 42% (or
approximately birr 151,419 or US$ 5,500 per ha) and con-
sumption by 35% for users compared to their counterfac-
tual situations. Controlling other variables related to as-
sets, farm and household characteristics, we found insti-
tutional factors had strong negative effects on the adop-
tion of small-scale irrigation schemes. Consequently,
much of the perceived water scarcity was attributed to
governance rather than scarcity of water.
In conclusion, the study has demonstrated that adoption
of small-scale irrigation schemes improves food security
through increased farm income (availability) and
consumption (ac- cess). However, access to small-scale
irrigation is significantly influenced by governance, in
addition to the conventional farm and household
characteristics associated with adoption of agricultural
technologies. In the study area, the mismanage- ment of
 Adela F.A. et
implications. We showed that access to irrigation water Abebe, G. K., Bijman, J., & Royer, A. (2016). Are middlemen
had a significant positive impact on rural livelihoods but facilitators or barriers to improve smallholders’ welfare in rural
economies?
required intervention in several aspects. Regarding
physical infrastruc- ture, renovating and improving the
canals is the best starting point for reducing wastage and
ensuring access of more peo- ple to irrigation water.
Additional water storage facilities could significantly
enhance the availability of water and its multiple uses
such as in fisheries.
These findings highlight the need to design appropriate
governance schemes to enhance the contribution of small-
scale irrigated agriculture for household food security.
Raising managerial skills and preventing corrupt practices
of the WUAs that are mushrooming across the country is
crucial for the improvement and welfare of rural
households. WUAs should be supported by sound policies
for water use with appropriate oversight by local
authorities to enhance the effi- ciency and equity of
irrigation water use. We suggest the in- volvement of the
traditional leaders (i.e., the so-called “water fathers”) to
build trust in the governance of irrigation schemes.
Several studies in Africa have documented the in-
creasing importance of “institutional pluralism” that
involves local communities (informal institutions) but
provides local governments with more authority in the
governance of small-scale irrigation schemes (e.g.,
Lecoutere 2011; Baldwin et al. 2016). This approach has
proved to be effective in searching for new solutions (i.e.,
institutional arrangements) as disputes emerge. Targeted
technical support from extension officers could play a
significant role in making effective use of irrigation
water. The findings also highlight the increasing reliance
of non-irrigation users on non-farm income, which may
aggravate rural-urban migration. Farmers in the study area
use irrigation water for growing khat and sugarcane.
Thus, it would be beneficial for farmers if they organized
themselves as producer or marketing cooperatives. With
the construction of several heavy sugar industries
underway, there would be a huge market for sugarcane
farmers. Creating ac- cess to small and medium credit
facilities could also improve farmers’ capacity to invest in
alternative irrigation water sources such as shallow well
and rainwater harvesting facilities.

Compliance with ethical standards

Conflict of interest This is to declare that there is no conflict of

interest to report.

References
Abdulai, A., & Huffman, W. (2014). The adoption and impact of
soil and water conservation technology: An endogenous
switching regres- sion application. Land Economics, 90(1), 26–
43.
 Small-scale irrigation scheme governance - poverty nexus: evidence from

Empirical evidence from Ethiopia. Journal of Rural Studies, 43, climate change in the Nile Basin of Ethiopia. Global
203–213. Environmental
Aberra, Y. (2004). Problems of the solution: Intervention into small-
scale irrigation for drought proofing in the Mekele Plateau of
northern Ethiopia. Geographical Journal, 170(3), 226–237.
https://doi.org/ 10.1111/j.0016-7398.2004.00122.x.
Amare, M., Asfaw, S., & Shiferaw, B. (2012). Welfare impacts of
maize- pigeonpea intensification in Tanzania. Agricultural
Economics, 43(1), 27–43. https://doi.org/10.1111/j.1574-
0862.2011.00563.x.
Amede, T. (2015). Technical and institutional attributes constraining
the performance of small-scale irrigation in Ethiopia. Water
Resources and Rural Development, 6, 78–91.
https://doi.org/10.1016/j.wrr.
2014.10.005.
Andersson, K. P., & Ostrom, E. (2008). Analyzing decentralized
resource regimes from a polycentric perspective. [journal article].
Policy Sciences, 41(1), 71–93. https://doi.org/10.1007/s11077-007-9055-
6. Awulachew, S. B., Erkossa, T., & Namara, R. E. (2010). Irrigation
poten- tial in Ethiopia: Constraints and opportunities for enhancing the
system. International Water Management Institute. Addis Ababa,
Ethiopia.
Bacha, D., Namara, R., Bogale, A., & Tesfaye, A. (2011). Impact of
small-scale irrigation on household poverty: empirical evidence
from the Ambo district in Ethiopia. Irrigation and Drainage,
60(1), 1–10. https://doi.org/10.1002/ird.550.
Baldwin, E., Washington-Ottombre, C., Dell’Angelo, J., Cole, D., &
Evans, T. (2016). Polycentric Governance and Irrigation Reform
in Kenya. Governance, 29(2), 207–225.
Barrett, C. B. (2010). Measuring food insecurity. Science, 327(5967),
825–828.
Belay, M., & Bewket, W. (2013). Traditional irrigation and water
man- agement practices in highland ethiopia: Case study in
dangila woreda. Irrigation and Drainage, 62(4), 435–448.
https://doi.org/ 10.1002/ird.1748.
Bezu, S., & Holden, S. (2014). Are rural youth in Ethiopia
abandoning agriculture? World Development, 64, 259–272.
https://doi.org/10. 1016/j.worlddev.2014.06.013.
Bradshaw, B., Dolan, H., & Smit, B. (2004). Farm-level adaptation to
climatic variability and change: Crop diversification in the
Canadian prairies. Climatic Change, 67(1), 119–141.
https://doi.org/10.1007/ s10584-004-0710-z.
Burney, J. A., & Naylor, R. L. (2012). Smallholder Irrigation as a
Poverty Alleviation Tool in Sub-Saharan Africa. World
Development, 40(1), 110–123.
https://doi.org/10.1016/j.worlddev.2011.05.007.
Central Statistical Agency and the World Bank. (2013). Ethiopia
Rural Socioeconomic Survey (ERSS). Addis Ababa, Ethiopia:
Survey Report.
D’Exelle, B., Lecoutere, E., & Van Campenhout, B. (2012). Equity-
Efficiency Trade-Offs in Irrigation Water Sharing: Evidence from
a Field Lab in Rural Tanzania. World Development, 40(12),
2537– 2551. https://doi.org/10.1016/j.worlddev.2012.05.026.
De Haen, H. (2003). The state of food insecurity in the world 2003:
monitoring progress towards the world food summit and
millennium development goals. Rome: FAO.
Deneke, T. T., Mapedza, E., & Amede, T. (2011). Institutional implica-
tions of governance of local common pool resources on
livestock water productivity in ethiopia. Experimental
Agriculture, 47(S1), 99–111.
https://doi.org/10.1017/S0014479710000864.
Denny, R. C. H., Marquart-Pyatt, S. T., Ligmann-Zielinska, A., Olabisi,
L. S., Rivers, L., Du, J., et al. (2018). Food security in Africa: a
cross-scale, empirical investigation using structural equation model-
ing. Environment Systems and Decisions, 38(1), 6–22.
https://doi. org/10.1007/s10669-017-9652-7.
Deressa, T. T., Hassan, R. M., Ringler, C., Alemu, T., & Yesuf, M.
(2009). Determinants of farmers’ choice of adaptation methods to
 Adela F.A. et
Change, 19(2), 248–255. https://doi.org/10.1016/j.gloenvcha.2009. Equations
01.002.
Dessie, G., & Kinlund, P. (2008). Khat expansion and forest decline
in wondo genet, Ethiopia. Geografiska Annaler, Series B:
Human Geography, 90(2), 187–203.
https://doi.org/10.1111/j.1468-0467. 2008.00286.x.
Di Falco, S., Veronesi, M., & Yesuf, M. (2011). Does adaptation to
cli- mate change provide food security? A micro-perspective
from Ethiopia. American Journal of Agricultural Economics,
93(3), 825–842. https://doi.org/10.1093/ajae/aar006.
Dillon, A. (2011). The Effect of Irrigation on Poverty Reduction,
Asset Accumulation, and Informal Insurance: Evidence from
Northern Mali. World Development, 39(12), 2165–2175.
https://doi.org/10. 1016/j.worlddev.2011.04.006.
Dorward, A., Fan, S., Kydd, J., Lofgren, H., Morrison, J., Poulton,
C., et al. (2004). Rethinking agricultural policies for pro-poor
growth. Natural Resource Perspectives, 94, 1–4.
Dudu, H., & Chumi, S. (2008). Economics of irrigation water
manage- ment: A literature survey with focus on partial and
general equilib- rium models. Policy Research Working Paper
4556. World Bank, Washington, DC.
FAO, A. (2008). An introduction to the basic concepts of food
security.
Rome, Italy: FAO.
Foster, J., And, J. G., & Thorbecke, E. (1984). A class of
decomposable poverty measures. Econometrica: journal of the
econometric socie- ty, 52(3), 761–766.
Gebregziabher, G. (2012). Water Lifting Irrigation Technology
Adoption in Ethiopia: Challenges and Opportunities.
International Water Management Institute (IWMI),
Ethiopia(September).
Gebregziabher, G., Namara, R. E., & Holden, S. (2009). Poverty
reduc- tion with irrigation investment: An empirical case study
from Tigray, Ethiopia. Agricultural Water Management,
96(12), 1837–1843. https://doi.org/10.1016/j.agwat.2009.08.004.
Hanjra, M. A., Ferede, T., & Gutta, D. G. (2009). Reducing poverty
in sub-Saharan Africa through investments in water and other
priori- ties. Agricultural Water Management, 96(7), 1062–
1070. https:// doi.org/10.1016/j.agwat.2009.03.001.
Headey, D., Dereje, M., & Taffesse, A. S. (2014). Land constraints
and agricultural intensification in Ethiopia: A village-level
analysis of high-potential areas. Food Policy, 48, 129–141.
https://doi.org/10. 1016/j.foodpol.2014.01.008.
Huffman, W. E. (2001). Agricultural Production. Handbook of
Agricultural Economics, 1, 333–381. https://doi.org/10.1016/
S1574-0072(01)10010-1.
Hussain, I., & Hanjra, M. (2004). Irrigation and poverty alleviation:
Review of the empirical evidence. Irrigation and Drainage,
53(1), 1–15. https://doi.org/10.1002/ird.114.
Josephson, A. L., Ricker-Gilbert, J., & Florax, R. J. G. M. (2014).
How does population density influence agricultural
intensification and productivity? Evidence from Ethiopia. Food
Policy, 48, 142–152.
https://doi.org/10.1016/j.foodpol.2014.03.004.
Kalu, I. L., Paudyal, G. N., & Gupta, A. D. (1995). Equity and
efficiency issues in irrigation water distribution.
Agricultural Water Management, 28(4), 335–348.
Khandker, S. B., Koolwal, G., & Samad, H. (2009). Handbook on
impact evaluation: quantitative methods and practices: The
World Bank.
Khonje, M., Manda, J., Alene, A. D., & Kassie, M. (2015). Analysis
of Adoption and Impacts of Improved Maize Varieties in
Eastern Zambia. World Development, 66, 695 706–695 706,
https://doi. org/10.1016/j.worlddev.2014.09.008.
Lecoutere, E. (2011). Institutions under construction: resolving
resource conflicts in Tanzanian irrigation schemes. Journal of
Eastern African Studies, 5(2), 252–273.
Lee, L.-F., Maddala, G., & Trost, R. P. (1982). Testing for
Structural Change by D-Methods in Switching Simultaneous
 Small-scale irrigation scheme governance - poverty nexus: evidence from

Models. CENTER FOR NAVAL ANALYSES ALEXANDRIA

impacts under uncertainty and labor constraints. Agricultural
VA NAVAL STUDIES GROUP.
Economics, 49(2), 237–250. https://doi.org/10.1111/agec.12412.
Maddala, G. S. (1986). Disequilibrium, self-selection, and switching
Sen, A. (1981). Poverty and famines: an essay on entitlement and
models. Handbook of econometrics, 3, 1633–1688.
deprivation. Oxford university press.
Maddala, G. S., & Nelson, F. D. (1975). Switching regression models
with exogenous and endogenous switching. Proceedings of the Setegn, S. G., Chowdary, V. M., Mal, B. C., Yohannes, F., & Kono, Y.
American Statistical Association, 5, 423–426. (2011). Water Balance Study and Irrigation Strategies for
Mangisoni, J. H. (2008). Impact of treadle pump irrigation Sustainable Management of a Tropical Ethiopian Lake: A Case
technology on smallholder poverty and food security in Malawi: Study of Lake Alemaya. Water Resources Management, 25(9),
A case study of Blantyre and Mchinji districts. International 2081–2107. https://doi.org/10.1007/s11269-011-9797-y.
Journal of Agricultural Sustainability, 6(4), 248–266. Shiferaw, B., Kassie, M., Jaleta, M., & Yirga, C. (2014). Adoption of
https://doi.org/10.3763/ijas.2008. improved wheat varieties and impacts on household food
0306. security in Ethiopia. Food Policy, 44, 272–284.
Mdemu, M. V., Mziray, N., Bjornlund, H., & Kashaigili, J. J. (2017). https://doi.org/10.1016/j. foodpol.2013.09.012.
Barriers to and opportunities for improving productivity and Smith, L. E. D. (2004). Assessment of the contribution of irrigation to
profit- ability of the Kiwere and Magozi irrigation schemes in poverty reduction and sustainable livelihoods. International
Tanzania. International Journal of Water Resources Journal of Water Resources Development, 20(2), 243–257.
Development, 33(5), 725– 739. https://doi.org/ 10.1080/0790062042000206084.
https://doi.org/10.1080/07900627.2016.1188267. Tesfaye, A., Bogale, A., Namara, R. E., & Bacha, D. (2008). The
Mukherji, A. (2012). Commentary-A five-pronged strategy to impact of small-scale irrigation on household food security: The
revitalizing Asia’s public irrigation systems. New Delhi: case of Filtino and Godino irrigation schemes in Ethiopia.
International Water Management Institute. [journal article]. Irrigation and Drainage Systems, 22(2), 145–
Mutambara, S., Darkoh, M. B. K., & Atlhopheng, J. R. (2016). A 158. https://doi.org/ 10.1007/s10795-008-9047-5.
com- parative review of water management sustainability Tizale, C. Y. (2007). The dynamics of soil degradation and incentives
challenges in smallholder irrigation schemes in Africa and Asia. for optimal management in the Central Highlands of
Agricultural Water Management, 171, 63–72. Ethiopia (Doctoral dissertation). University of Pretoria,
https://doi.org/10.1016/j.agwat. Vanlauwe, B., Wendt, J., Giller, K. E., Corbeels, M., Gerard, B., &
2016.03.010. Nolte,
Mwendera, E., & Chilonda, P. (2013). Conceptual framework for C. (2014). A fourth principle is required to define Conservation
revitalisation of small-scale irrigation schemes in southern Agriculture in sub-Saharan Africa: The appropriate use of
Africa. Irrigation and Drainage, 62(2), 208–220. fertilizer to enhance crop productivity. Field Crops Research,
Namara, R. E., Upadhyay, B., & Nagar, R. K. (2005). Adoption and 155, 10–13. https://doi.org/10.1016/j.fcr.2013.10.002.
impacts of microirrigation technologies: Empirical results from Watts, H. W. (1969). An economic definition of poverty (No. 1907–
se- lected localities of Maharashtra and Gujarat States of India 2017- 1986).
(Vol. 93): IWMI. Wichelns, D. (2014). Investing in small, private irrigation to increase
Namara, R. E., Hanjra, M. A., Castillo, G. E., Ravnborg, H. M., production and enhance livelihoods. Agricultural Water
Smith, L., & Van Koppen, B. (2010). Agricultural water Management, 131, 163–166. https://doi.org/10.1016/j.agwat.2013.
management and pov- erty linkages. Agricultural Water 09.003.
Management, 97(4), 520–527. Xie, H., You, L., & Takeshima, H. (2017). Invest in small-scale
https://doi.org/10.1016/j.agwat.2009.05.007. irrigated agriculture: A national assessment on potential to
Nhemachena, C., & Hassan, R. (2007). Micro-level analysis of expand small- scale irrigation in Nigeria. Agricultural Water
farmers adaption to climate change in Southern Africa. Intl Management, 193, 251–264.
Food Policy Res Inst. https://doi.org/10.1016/j.agwat.2017.08.020.
Ostrom, E. (1990). Governing the commons. Cambridge University Yami, M. (2013). Sustaining participation in irrigation systems of
Press. Ethiopia: What have we learned about water user associa-
Ostrom, E. (2010). Beyond markets and states: polycentric tions? Water Policy, 15(6), 961–984. https://doi.org/10.2166/
governance of complex economic systems. American economic wp.2013.031.
review, 100(3), 641–672. Yami, M. (2016). Irrigation projects in Ethiopia: what can be done to
Rath, A., & Swain, P. C. (2018). Evaluation of performance of enhance effectiveness under ‘challenging contexts’? International
irrigation canals using benchmarking techniques–a case study of Journal of Sustainable Development and World Ecology, 23(2),
Hirakud dam canal system, Odisha. India. ISH Journal of 132–142.
Hydraulic Engineering, 1–8. Yohannes, D. F., Ritsema, C. J., Solomon, H., Froebrich, J., & van Dam,
Ribot, J. C., Lund, J. F., & Treue, T. (2010). Democratic J. C. (2017). Irrigation water management: Farmers’ practices,
decentralization in sub-Saharan Africa: its contribution to forest per- ceptions and adaptations at Gumselassa irrigation scheme,
management, live- lihoods, and enfranchisement. Environmental North Ethiopia. Agricultural Water Management, 191, 16–28.
Conservation, 37(1), 35–44. https://doi. org/10.1016/j.agwat.2017.05.009.
Rosegrant, M. W., Ringler, C., & Zhu, T. (2009). Water for You, L., Ringler, C., Wood-Sichra, U., Robertson, R., Wood, S., Zhu,
Agriculture: Maintaining Food Security under Growing Scarcity. T., et al. (2011). What is the irrigation potential for Africa? A
Annual Review of Environment and Resources, 34(1), 205–222. combined biophysical and socioeconomic approach. Food Policy,
https://doi.org/10. 1146/annurev.environ.030308.090351. 36(6), 770– 782. https://doi.org/10.1016/j.foodpol.2011.09.001.
Schuenemann, F., Thurlow, J., Meyer, S., Robertson, R., &
Rodrigues, J. (2018). Evaluating irrigation investments in Malawi:
economy-wide
 Adela F.A. et
Mrs. Fitsum Assefa Adela is a
Gumtaw Kifle Abebe (PhD,
final year PhD student at Justus-
Wageningen University) is
Liebig University, Giessen at the
As s i s t a n t P r o f e s s o r of
I n s t i t u t e of F a r m a n d
Agribusiness Marketing and
Agribusiness Management. Her
Management at the Faculty of
research project focuses on Sub-
Agricultural and Food Sciences,
Saharan Africa and Ethiopia
American University of Beirut,
with multi-disciplinary thematic
Lebanon. He has several years
areas including impact of
of experience teaching business
technology adoption on food
and agribusiness courses at
security, factors affecting
under- graduate and
adoption of improved
postgraduate levels. He has
technology with emphasis on in-
experience in interdisci- plinary
stitutional and governance set-
research related to agri- food
tings, bio-economic modeling of
supply chains in sub Saharan
optimal water allocation among
Africa, the Middle East
upstream and downstream users, and scenario analysis of different
and North Africa. His research focuses on exploring the efficiency
policy interventions in order to trigger actions for creating a
and effectiveness of agri-food supply chains in response to recent
sustainable bio- economy.
trends for food safety and food quality; food retail expansion, its
drivers and impact on food security; and economics of agricultural
technology adoption in arid and semi-arid regions.

Joachim Aurbacher is
professor of Agricultural
Production Economics at the
Institute of F a r m a n d A g r i
b u s in e s s Management of
Justus-Liebig University,
Giessen. His research focus is on
the interplay of farm business
management and envi-
ronmental impacts. He is
involved in research projects
concerning adaptation of short
term farm management with
regard to cli- mate change,
evaluation of crop rotations
and novel crops concerning
the provision of
bioenergy and the role of agriculture in remote areas. His study area
focuses on West and East European countries.
 Small-scale irrigation scheme governance - poverty nexus: evidence from

View publication stats