Land Use Policy
Land Use Policy
Land Use Policy
A R T I C L E I N F O A B S T R A C T
Keywords: Sustainable land management is of utmost importance in Ethiopia and relies on Soil and Water Conservation
Sustainable land management (SWC) measures collectively implemented by smallholders through participatory processes. This paper con-
Land degradation tributes systematic evidence on how SWC strategies are implemented and how participation is operationalized.
Upper Blue Nile Basin Drawing upon inductive, qualitative research, we explore the design, implementation and evaluation of SWC
Commons
activities, as they relate to Ostrom’s Institutional Analysis and Development (IAD) framework, in order to de-
Drought
termine how the activities can be made more appropriate, effective and sustainable. Findings show that on all
Sustainability
levels of Ostrom’s framework, there are shortcomings in the SWC institutions, which have to be addressed with
more participatory approaches, a change from top-down to bottom-up measures, and economic incentives for
farmers to invest in SWC measures instead of e.g. compulsory labor, and the integration of so far neglected
groups like youth, women and the landless.
⁎
Corresponding author.
E-mail address: zeriye@gmail.com (Z. Nigussie).
https://doi.org/10.1016/j.landusepol.2017.11.039
Received 18 September 2017; Received in revised form 20 November 2017; Accepted 21 November 2017
Available online 28 November 2017
0264-8377/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
framework’s core is the ‘action arena’. The action arena is composed of sustainable land management issues into their national policies and
an action situation and actors. The action situation refers to a social frameworks (Akhtar-Schuster et al., 2011). We employ Ostrom’s IAD
space where the actors interact, solve the commons problem, and ex- framework as a means to evaluate SWC activities within this context of
change goods and services; the actors are those who participate in the land degradation, analyzing the broader environment that is influen-
situation (Ostrom, 2007; Ostrom et al., 1994). In the Ethiopian SWC cing the challenges related to land degradation and restoration as well
case, the action arena was assumed to shape the efforts towards sus- as the development of more sustainable land management practices.
tainable watershed management. Soil degradation due to water erosion from heavy rains (Ayele et al.,
By following the steps in the IAD framework (Fig. 1) and using the 2016), overgrazing (Alemayehu et al., 2013), conversion of marginal
action arena as the unit of analysis, the analysis systematically follows lands to croplands (Bewket and Sterk, 2002), and inappropriate farming
the path of decision making from pre-planning to planning, on to ex- practices (Astatke et al., 2003) remain major threats to sustaining
ecution and also ensuring the sustainability of a project. When the ac- agricultural yields and soil fertility. Hurni et al. (2010), for example,
tion arena and its associated rules are evaluated against the background estimated that soil loss due to water erosion of cultivated fields in
of watershed development projects in terms of their structure, man- Ethiopia amounts to about 42 Mg ha−1 year−1. Considering that sub-
agement, and outcomes or performance in community engagement, the stantial efforts to promote soil conservation and environmental re-
results can provide useful guidelines for practitioners regarding how habilitation have been on-going for four decades (Bayabil et al., 2010;
and where to act to improve the broad societal value of ongoing SWC Dessie et al., 2012), the continued loss of this amount of soil due to
projects. water erosion suggests these efforts have not sufficiently and/or ap-
In the action arena, interests of the different stakeholders confront propriately addressed the causes. Recent national strategies and policy
and planned initiatives are (re)shaped. Therefore our approach is first documents in Ethiopia have also considered combating land degrada-
to analyze, what was planned by official interventions and which ad- tion as one of the most important development priorities (MoANR,
ministrative set up has been installed, then to compare the planned 2017; NPC, 2016).
changes and the real ones. The region chosen for the investigation is the However, soil conservation and environmental rehabilitation in-
UBNB where planned change with the “community participation” terventions have had little success in bringing about the voluntary
strategy has been implemented for more than five years. As set out in uptake of improved SWC technologies by smallholder farmers to tackle
government documents (e.g., MoANR, 2017; MoFED, 2010; NPC, soil degradation problems in the drought-prone highlands of the UBNB
2016), SWC development works have been and are going to be widely (Dessie et al., 2012; Smit et al., 2017; Tesfaye et al., 2014). The lack of
executed across degraded watersheds through community participation integration from the different disciplines and sectors (German et al.,
to achieve sustainable land management. The ongoing initiative was 2007), limited stakeholder participation (Bewket and Sterk, 2002; Smit
launched in 2010, following the release of the country’s 5-year Growth et al., 2017), inappropriate incentives such as food-for-work programs
and Transformation Plan (i.e., GTP-I (2010/11-2014/15), followed by (Amsalu and de Graaff, 2006), rigid technical packages, unmanageable
the GTP-II (2015/16-2019/20)) (MoFED, 2010; NPC, 2016). Therefore planning units (Desta et al., 2005), and top-down extension systems
we were able to examine the pre-planning and planning processes, (Amsalu and de Graaff, 2006; Dessie et al., 2012; Smit et al., 2017) have
implementation, participation of marginalized and disadvantaged been reported as limiting factors to success.
groups, and monitoring and evaluation (M&E) systems. Evidence suggests that without localized participatory initiatives,
there will be limited widespread adoption of exogenous SWC technol-
3. Contextualizing land degradation ogies by smallholder farmers (Desta et al., 2005; German et al., 2007).
These factors influenced the Ethiopian government, with the support of
Land degradation, a process that involves a decline in the processes the FAO, to pilot community-based participatory watershed develop-
and productivity of ecosystem goods and services provided by land ment approaches from 1988 to 1991 (Desta et al., 2005). Following this
(e.g., soils, water, vegetation) (Vu et al., 2014), poses enormous chal- initiative, various international agencies (e.g., WFP (United Nations
lenges to both humanity and ecological systems. This challenge is ex- World Food Program), GTZ (Deutsche Gesellschaft für Technische Zu-
perienced across all regions in the world (Vu et al., 2014), but parti- sammenarbeit)) have adopted similar, but scattered, watershed devel-
cularly in sub-Saharan Africa, which has the highest rate of land opment approaches (Desta et al., 2005) to support the government’s
degradation (Tully et al., 2015). In the United Nations Convention to efforts to improve the land resource base. Community-based partici-
Combat Desertification (UNCCD) conference, poor management of land patory watershed development guidelines were formulated in 2005, the
resources was identified, among other factors, as an important driver of intent of which was to provide adaptable planning and implementation
land degradation (Kust et al., 2016; UNCCD, 2009). As a result, sus- tools (Desta et al., 2005).
tainable land management practices offer synergistic solutions in pro- Recently, a growing body of work (e.g., Amare et al., 2014; Amsalu
tecting land from being degraded and in restoring degraded land and de Graaff, 2006; Haregeweyn et al., 2012) has shown some positive
(Akhtar-Schuster et al., 2011; Kust et al., 2016). Hence, parties in de- outcomes in sustainable land management in Ethiopia in general and
veloping countries (e.g., Ethiopia) that have been affected by land de- the UBNB in particular. However, intervention programs still lack a
gradation have been receiving support (e.g., technical guidance, fi- concerted focus on sustainability issues from their early diagnostic
nance, and knowledge transfer) to mainstream land degradation and phases and instead start considering sustainability during the phase-out
2
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
stage as an exit strategy. Moreover, they seem to degrade rather than 5. Soil and water conservation activities
sustain local collective organizations and management capacities as
well as indigenous people’s capacities for conservation (Amsalu and de SWC activities refer to technologies which are aimed to preserve
Graaff, 2006; Nigussie et al., 2017). and enhance the productive capabilities of a land, or they embrace
those practices that significantly reduce soil losses and their con-
sequences. In the UBNB, various SWC technologies have been promoted
4. Methods among smallholder farmers to control soil erosion through campaign-
based SWC programs. Common physical technologies being promoted
This research utilized a comparative, case study based approach. In across the study sites include soil bund, fanya juu (a ditch excavated
order to better analyze changes and their causes, we compare arenas in along a contour with the soil placed on the uphill side to form a ridge),
three different agroecological contexts within the UBNB. It was ex- stone-faced soil bund, and trench. The modality of this program is that
pected that the more acute the land degradation, the more prone to farmers are expected to provide compulsory free labor under close su-
collective action and common norms stakeholders may be. The com- pervision of DAs and local authorities. In addition, district-, zonal- and
parative analysis allows for an assessment if the findings are location regional-agriculture offices, and the Ministry of Agriculture and Natural
specific (socio-cultural or political) or linked to agroecology. Resources are key players under the ongoing SWC program. Currently,
Consistency of findings across the case studies suggests more general- different externally funded programs are also involved in SWC activ-
ized trends, while also recognizing the limitations of what general- ities. Concerning our study, Water and Land Resource Center (WLRC)
izations can be made from a study based on a comparative analysis of project is involved in the Aba Gerima watershed funded by the Swiss
three cases. Agency for Development and Cooperation (SDC) and Sustainable Land
The fact that this research has been conducted within a research Management Programme (SLMP) project in Guder watershed supported
project with a life span of 2013–2022, with field assistants who are by the World Bank, whereas, the Dibatie watershed is not under any
residents of the study watersheds, allowed the researchers to establish external support for SWC activities. Although it requires further study,
trust with the community. This enabled: (1) to avoid, or at least mini- we hypothesize that the existence of external projects could have di-
mize, suspicious view of interviewees regarding the identity of the re- verse impacts. In addition to the government’s SWC programs, external
searchers, (2) to know the communities more (who is who? affiliations activities could be helpful in enhancing awareness regarding land de-
with State? economic status? etc.) and spot respondents easily, and (3) gradation problem and possible ways to rehabilitate degraded water-
to verify information that we were soliciting from agricultural exten- sheds, such activity may also serve as an important source of additional
sion agents [also known as development agents or DAs] and local au- pressure on local actors for SWC success.
thorities.
The study is based on the qualitative research conducted in three 6. Study areas in the upper Blue Nile Basin
periods of time: February to March 2015, October to November 2015,
and February to March 2017. Individual interviews and focus group The Ethiopian administrative system is composed of (from lowest
discussions were conducted on current outcomes in ongoing collective level): sub-Districts, Districts, Zones, and Regions. The UBNB is located
SWC initiatives, the different factors impacting these outcomes and within three regional administrative boundaries: Amhara, Benishangul
their envisaged sustainability. In sum, 60 farmers and local authorities Gumuz, and Oromia. The study was undertaken in three watersheds
from the 3 study sites, 20 agricultural experts (2 regional, 4 zonal, 6 within the UBNB: the Guder watershed from the Fagita Lekoma District
district, and 8 DAs) from the Bureau of Agriculture, and 4 university (population 161,002, Amhara region); the Aba Gerima watershed from
professors were individually interviewed as subject-matter experts. As a the Bahir Dar Zuria District (population 222,304, Amhara region); and
qualitative study, the participants are not a statistically representative the Dibatie watershed from the Dibatie District (population 90,577,
sample of the population. As our aim is to provide insights and in-depth Benishangul Gumuz region) (see Fig. 2). Livelihoods of the three wa-
understanding of common perceptions and experiences rather than tershed communities are dominated by rain-fed cereal based mixed
empirical generalizations, the respondents for individual interviews farming systems. These watersheds were selected because of their
were chosen using a purposeful sampling technique, as suggested by ability to capture a variety of bio-physical and socio-economic char-
Patton (2005), so as to select a specific segment of the community that acteristics and to represent higher, medium, and lower elevation wa-
is believed to represent the range of variation expected in a population. tersheds within the highlands of the basin. The climate in the basin
Additionally, a total of 15 focus groups—consisting of watershed changes from humid in the higher elevation zones (Guder) to semi-arid
development committee (WDC) members and sub-groups of men, in the lower zones (Dibatie). Detailed descriptions of the study water-
women, young adults, and mixed farmers, each composed of approxi- sheds are beyond the scope of this article, but have been elaborated by
mately 8–10 participants—were carried out at the 3 study sites. Three Nigussie et al. (2017).
meetings (i.e., one meeting per study site) for each sub-group were Following the 1975 land reform, all land in Ethiopia is owned by the
held. The participants were recruited in consultation with the project government. Individual land holders in rural areas have rights to use
field assistants and DAs, in order to take into consideration differences land, which have recently (post-1998) been protected via a land certi-
in age, gender, education, and wealth within the focus groups. The fication system. Rural land holders are not allowed to sell or exchange
principal researcher and his field assistant were present at each focus their land, rentals are only permitted for temporary periods of time.
group discussion to facilitate deliberations and take notes. Land can only be transferred in the form of inheritance. All unoccupied
Approximately, the study respondents represented not more than 10% land is owned by the government, plots of which are maintained in a
of the watersheds community. Moreover, participant observation and ‘land bank’ that can be distributed to community members, such as the
field notes were also utilized to augment findings from the other data landless. Unused land is returned to government control in the land
sources. The information obtained from any given source of data was bank. The land tenure system has direct implications for SWC activities.
also triangulated with other qualitative data to ensure adequate re- For example, while land security has relatively improved, it remains
presentation from multiple viewpoints, to increase synchronic relia- precarious as land redistributions were conducted as recently as the
bility of data, and to uncover any deeper meaning in the data. 1990s in Amhara region.
Transcripts and emerging themes were categorized and (re)coded
seeking for possible patterns in the texts, finally making the way to 7. Action arenas
interpretation of results and analysis. Analysis was manual rather than
computer-assisted. The following assessment of action arenas, which follows the
3
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
contextualization as outlined in Ostrom’s IAD framework, is organized activities in rehabilitating their degraded landscapes. In addition, they
around the stakeholders involved with SWC activities in the case study have started witnessing some positive environmental outcomes in their
locations, namely: community members, community-based government immediate landscape from SWC works. Respondents from Aba Gerima
staff, experts, and regional stakeholders. and Guder sites mentioned that some wild animals are returning, some
springs are reviving in downstream areas, and flood risk in downstream
farmlands is decreasing. However, there was a shared sentiment that
7.1. Community members farmers were not participants in designing and planning the activities,
but were rather the implementers. For example, one farmer explained
Not all community members have equal interest in rehabilitation, that when he tried to express his opinions regarding the prioritization of
nor do they have equal ability to engage and influence what the com- needs he was portrayed as being opposed to positive change:
mittees decide to do. Of the community members, landholders have the Soil erosion is an important problem, but the shortage of feed,
most power, but not the same interest as others. Those without land water, roads, firewood, land, and off-farm and non-farm employments
may support SWC activities in hopes of obtaining rehabilitated land. are even more important. When you bring up these issues, sometimes
While the landholders may have an interest in expanding their land you are categorized as ‘anti-development’ instead of being given the
holdings, they also face competing labor priorities – maintaining their opportunity to state your point of view.
own agricultural livelihoods as opposed to investing their labor in SWC The lack of needs assessment, and thus the inability to align local
activities. Due to a lack of land, youth prefer to pursue work options priorities with plans, is also reflective of the different options and op-
outside of the community (Bezu and Holden, 2014; Smit et al., 2017), portunities afforded to different members of society. SWC activities may
and only reluctantly join SWC activities with few expectations of ben- not have the same level of urgency for livestock owners as they do for
efiting by such action – often these youth are landless and earn their landless farmers. Rather than be a participatory process – as designed –
daily needs, thus are less incentivized to participate in voluntary SWC the plans were developed outside of the community, and ‘participation’
activities. In theory, labor contributions are to be equal. In actuality the equated with following the pre-determined plans. Government autho-
contributions are not equal nor are the burdens equally borne. Most rities claimed that in the current SWC initiative, communities have
importantly, as outlined below, it is unclear who will benefit from the been involved in every activity. However, the government has sought
SWC activities. As a result, individuals with less ability to participate, consent while the community members express dislike from a lack of
and who are less likely to benefit from the process, are less likely to participation and ownership. Thus, a key issue is the notion (or use of
engage in communal SWC activities (e.g. youth, women, those without the notion) of ‘participation’. As outlined elsewhere by Cochrane and
land). Skjerdal (2015), the State appears to view participation as equating
In setting up the SWC activities at the community level, local gov- with consent, not as involvement in decision making.
ernment staff are to involve community members. The majority of Socio-cultural factors also play a role in influencing broad com-
landholders in all three sites confirmed that they attended a presenta- munity participation in SWC activities. For example, women partici-
tion by the local government staff, which presented the SWC plans. pation was noted as being low by community members. However, their
Fewer individuals without land, as well as youth and women reported involvement was strongly resisted by men, particularly married men
having been involved in such a meeting. This is reflective of commu- refusing to allow their spouses to participate in these public activities.
nity-level power dynamics as well as socio-cultural norms. The product, The main reason given by the male focus group members for this was
however, is that landholders have had a far greater ability to influence their belief that women should be taking care of small children, tending
the SWC activities than other members of society have. livestock, and doing household chores. To exemplify this point, one
The majority of interviewed farmers had a high level of knowledge district interviewee stated, “if you find a woman in the interventions,
about the extent of soil degradation and likely outcomes of SWC
4
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
she must be widowed or divorced.” In the focus group discussions with in large communities. However, according to experienced experts, the
women, it was highlighted that these challenges have deeper roots; they definition and diagnosis processes were fully run by government offi-
stem from the women’s spouses’ negative views of women participating cials, although the expectation was that local residents would be part of
in the public sphere. Existing cultural practices and taboos regulating the process through consensus building.
women’s relationships with their environment have the potential to A related concern of the external experts is that the lead agency (the
deny them of their rights and roles in watershed management systems Bureau of Agriculture) has inappropriately took up the SWC work as
and to institutionalize their traditional roles (e.g., farming, cooking, part of its annual planned activities. The Bureau is expected to co-
and collecting firewood), which have often been characterized by un- ordinate actions and resolve conflicts among actors within the similar
equal resource and power distributions. A further reason given by most areas of the organization (e.g., crop, livestock, natural resources, and
respondents for the limited involvement of women is that local autho- extension) and also across different tiers of the organization (e.g., roads,
rities did not take the lead in having their own spouses participate in education, cooperatives, NGOs, and land administration). Contrary to
such initiatives as examples to others. Lubell (2004) noted that the the design of the ongoing SWC program, it is still following a campaign
community’s lack of trust in local leaders to have their own spouses style approach, with little integration and without any holistic views of
participate in collective SWC works can engender resistance among rural landscapes. In addition, little consultation among the actors in-
other farmers to do so. volved in the various activities under the supervision of the lead agency
seems to occur, even though watershed rehabilitation involves multiple
7.2. Local government staff stakeholders and interdependent collective actions.
Community chairpersons and DAs briefed community members on 7.4. Regional stakeholders
the SWC activities, inventoried tools and identified available labor.
These personnel determine the specific nature of the implementation, Regional informants confirmed that few provisions were made for
based upon the watershed area, nature of landscape, and available re- active community participation in the process. They attributed the re-
sources. However, these decisions fall within the parameters of the SWC luctance to involve the community in planning at the lower adminis-
guidelines and direction from higher authorities (zonal and regional trative or community levels to the communities’ low capacities, lack of
levels). A draft document developed by local government staff is sub- knowledge, and lack of resources and information for reliable planning
mitted to the District Agricultural Office for approval. There is an op- activities. As a result, plans were formulated at higher levels and sent
portunity in the drafting of these plans to involve community members down to the lower levels. This also led to a mismatch between the
in an inclusive and participatory process. While this is ideal, local planned physical SWC structures, which were seemingly based on a
government staff have limited resources, and are particularly con- quota system, and watershed realities, which in turn led to difficulties
strained in having few personnel, they often have to cover very large in subsequent processes. The regional government personnel have,
areas and populations, have no vehicles for transportation, and have no therefore, discounted the value and usefulness of local, inclusive and
budget for convening community members for such a process. As is participatory processes. This is exactly what community members and
typical for rural participatory programs in Ethiopia (Berhanu and local government personnel have experienced, and is a key reason why
Poulton, 2014; Cochrane and Tamiru, 2016), all these activities are there is a lack of participation and ownership in the SWC activities.
expected to be voluntary.
Another reason why local government staff do not undertake such 8. Patterns of interaction
an effort is because they report that the approved plans rarely resemble
the submitted draft. DAs felt that the District officers did not appreciate An analysis of design, implementation and perception of SWC
their reports, nor believe the information to be accurate. As lower level measures allows for insight into the assumed and actual outcomes of
administrative staff, these personnel understand they have little to no the interactions and therefore for understanding the patterns of inter-
option to refuse or return these re-made plans. In many instances, the action for SWC activities. For Ostrom et al. (1994), this included an
result is that DAs felt the plans they were called to implement were not assessment of market structures, information flows and political parti-
reflective of the realities of the watershed within which they work. cipation, which we have adjusted to the SWC scope of this paper. The
design of the SWC program was to be participatory and community
7.3. Experts driven, but in practice it is not. While each actor has different experi-
ences (see Fig. 3), all are in agreement that decision making occurs at
External researchers and academics with a focus on agriculture higher administrative levels (district level and above) and those based
provide another perspective into the SWC activities. In the assessment within the community (government staff and community members
of such experts, one good part of the process has been the political alike) have little ability to influence what they are required to do in the
commitment that puts the initiative as ‘a matter of life and death’. SWC activities.
However, they questioned if the manner in which the initiative is being According to Tippett et al. (2007), sustained participation and
implemented will lead to its intended outcomes. This concern is sup- support are only possible when members in local communities feel that
ported by their past experiences with similar initiatives, where rural they have had a say, see their concerns reflected in the process, and
communities destroyed everything accomplished by community labor when they foresee how the end product is going to benefit them. None
within a short time during periods of political neglect (Harrison, 2002). of these factors was a characteristic of the current process, which was
In other words, the experts doubt that the prior experiences of stake- limited to community members’ mere physical presence at community
holders (i.e., farmers) in earlier similar initiatives have been in- fora and to labor provision. Broad public participation is viewed as
corporated into the design of the current project. crucial by the government for successful implementation of watershed
The experts also shared concerns regarding the preplanning activ- development initiatives (Desta et al., 2005; MoANR, 2017; MoFED,
ities performed at community level. Among their concerns, the most 2010; NPC, 2016). In this regard, the issue is not a matter of poor de-
important were how watershed problems were identified and prior- sign, but of implementation diverging from the design.
itized and how related tasks are performed. For these tasks, the ap-
proach is supposed to fully rely on community fora. This method, as 8.1. Input and benefit sharing
pointed out by German et al. (2007), is a common practice when de-
fining watershed problems in the eastern African highlands (including As outlined at the outset, there is an annual labor contribution of
UBNB), and it incurs lower transaction costs by generating agreement 40–60 days, which is mandatory. SWC is labor intensive and often
5
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
conflicts directly with agricultural livelihoods (i.e. time spend in SWC holders often prefer to lease their land to adult farmers. Younger people
activities competes with tending to one’s own field or laboring on an- have suggested redistribution as a solution but land holders have re-
other person’s land). The government has not neglected this concern, jected that idea. Local authorities also agreed that land redistribution is
and has struggled to find a solution, setting up the time for compulsory impossible in the foreseeable future because it would lead to further
labor during the dry season. In particular, farmers in Aba Gerima and fragmentation of already meagre per capita land holdings.
Dibatie sites associated this problem with the timing of the SWC in- An equally important issue concerns communal lands. Landless in-
itiative – January to March – because these months are part of the long dividuals want to gain access to such lands for the development of in-
dry season, the soil moisture is fairly low; this makes the soil digging, come-generating activities such as farming, livestock fattening, bee-
excavating and embanking activities labor demanding. This confirms keeping, and agroforestry, whereas land holders are not willing to offer
what Kassahun and Jacobsen (2015) noted about farmers in Debre- such rights. These conflicts of interest have remained unresolved, which
kelem, Yetsed and Gelila sub-districts—all along the Choke Mountain has become an important source of dissatisfaction among landless in-
ranges of the UBNB—not being in favor of the current timing of com- dividuals and has limited the expected labor contribution from this
munity labor (mid-January to end of March), which was originally in- group. Generally, it seems that land holders’ claims are likely to prevail
tended to avoid labor competition. In addition, many social ceremonies over those of others within the existing social norms, for example, in
(especially weddings) take place during this period of the year. Farmers managing communal lands, which means there is a need for additional
instead prefer an arrangement that distributes the labor load equally community-level negotiations. At the time of our interviews, these is-
between each month of the year. sues were more contentious in the Guder and Aba Gerima watersheds
One of the greatest issues of contention within the communities is than in the Dibatie watershed, possibly because Dibatie has better ac-
who will benefit from the SWC activities. Community members stated cess to farmland in nearby, and less populated, lowland areas.
that the main reason for low levels of participation by youth was that Most respondents noted that the issue of ‘how the disadvantaged
their primary concerns, access to land and non-farm employment op- groups within the watershed communities would benefit’ was not dis-
portunities, were not addressed. In existing rural settings, the only way cussed among stakeholders. Negotiations about how future accrued
young farmers can get access to farmland is through informal land benefits (e.g., property rights) would flow to various actors within
markets, primarily through share-cropping arrangements. The problem communities were not conducted in a transparent or timely manner,
with such arrangements, according to the interviewees, is that land which could have arisen from government’s incorrect assumptions of
6
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
homogeneity of interests arising from the soil erosion management Watershed Development Committees (WDCs) were established in al-
regimes. These issues are critically important for disadvantaged people most all lower administrative units and were expected to create direct
within communities, in particular youth, the landless, and women. engagement and interactions between local governments and commu-
The question of how land holders are going to equitably share future nities. The WDCs are supposed to be fully involved in all watershed
benefits also did not receive much attention and was left vague during development activities and all governance processes. However, all three
community fora, as noted by almost all respondents. This has the po- case study areas experienced challenges in establishing these WDCs.
tential to be a source of conflict among watershed users. This conflict is Primarily, these committees were not inclusive of all social categories,
likely to arise due to distributional issues within the resource users’ and thus not representative of their interests, priorities and concerns. As
community and the inequities created by excluding some groups from a result, community members who had pre-existing power utilized
receiving benefits, as also already noted by Natcher and Hickey (2002). these opportunities to maintain and entrench their position in society,
while entrenching the exclusion of others.
9. Evaluative criteria Even if WDCs were representative of their constituencies, committee
representatives sat in between officials’ and their principals’ expecta-
The IAD model itself did not develop the evaluative criteria in sig- tions. In the Aba Gerima watershed, for example, even though ob-
nificant detail, but included potential areas of investigation such as servable positive bio-physical changes in the environment have oc-
efficiency, finance, equity, accountability, conformance to values and curred, power dynamics in the community is a very important concern.
sustainability. For the purpose of this paper, we focus upon aspects of This may be related to the fact that the watershed’s close proximity to
evaluative criteria that relate to (1) the relevance of SWC measures as Bahir Dar—the city where the lead agency’s office is located—has en-
affecting the program efficiency, (2) the management of the activities as hanced the purview of the agency on local leadership (the DAs and local
influencing equity, (3) monitoring, evaluation and learning processes as leaders), which has in turn enabled the latter to exercise substantial
major aspects of accountability, and (4) sustainability of achievements authority. In addition, the Aba Gerima watershed serves as a learning
as the main expected value. site in the UBNB. This may also have attracted more attention from
external stakeholders (e.g., NGOs and government agencies), thereby
9.1. Relevance of the SWC measures imposing a relatively high psychological burden on local leaders to
push decisions (e.g., compulsory mass mobilization) to their limits so as
UBNB covers different agro-ecological zones characterized by dif- to meet expectations. In relation to this point, one interviewee stated,
ferent farming systems coupled with highly variable biophysical factors “Who is going to stop them when they penalize us for not attending in
(e.g., soil, rainfall, vegetation, land use types) both in terms of location collective SWC works? When they [local administrators] fine us
and seasonal distribution (Alemseged and Tom, 2015). This naturally 100–300 Ethiopian Birr [≈5–15 USD] for non-attendance, they do not
calls for variation in land management interventions. Recognizing this, listen to our reasons, and they do not even give us a receipt as evidence
albeit without sufficient scientific evidence, the Ministry of Agriculture of our payment.”
and Natural Resources suggested different improved SWC measures for
possible implementation across different agro-ecologies (Hurni et al., 9.3. Monitoring, evaluation, learning and accountability
2016). However, our field observations revealed few variations in im-
plemented SWC technologies. Interventions were restricted to the use of Regional informants confirmed that there was no formal M&E
soil bunds, stone-faced soil bunds, trenches, and fanya juu, an issue system during the initial stage of the ongoing SWC initiative. However,
which is also confirmed by others (Nigussie et al., 2017; Teshome et al., during the implementation process, the government came to the view
2013). Furthermore, in all three locations, none of the community that installing such an M&E system would be crucial for program
members who attended the planning meetings recall any discussion management. In the original design, teams were created in which one
regarding issues about how to fit traditional SWC practices (e.g., drai- ‘model’ farmer would lead five others (a ‘one-to-five’ team). Each team
nage channel, traditional stone bund, grass strip) into the proposed was expected to report its achievements (i.e., team members involved
plans. As outlined also elsewhere by Amsalu and de Graaff (2006) and and daily work accomplished) to its respective development group
Assefa and Hans-Rudolf (2017), such local practices and technologies (DG), which was composed of 4–6 one-to-five teams, or a total of 20–30
have been neglected by plans. households. Each DG was supposed to consolidate its respective report
Farmers noted that the newly introduced SWC techniques were not and pass it to the WDC. After approving the reports, the WDCs were to
resulting in productivity improvements in the short term. While this is deliver them to their respective DAs. The DAs then compiled the lower
concerning, many SWC activities are geared to improve the sustain- administrative (i.e., sub-district) level reports and passed them along to
ability of agricultural practices rather than result in immediate yield their respective district agriculture offices. The reporting procedure was
increases (as a newly introduced seed or crop might). This is suggestive to continue up the line according to formal organizational arrange-
of a communication issue during the implementation. A more sys- ments until it finally would reach the regional government.
tematic challenge is the poor selection of techniques being advocated, However, regional informants noted that the WDC establishment
which may not contribute to the aims of the program. For example, process was not uniform throughout the region. As a result, WDCs were
about 55% of farmers in the Chemoga watershed, a watershed in the not established in all areas. Even in areas where committees were es-
UBNB, believed that “fanya juu construction was not for the sake of tablished, they often had no clear idea, for example, of how to collect,
conserving farmers’ land, but to meet government’s development pro- manage, and report information; undertake periodical committee level
gram” (Bewket and Sterk, 2002: 194). The use of fanya juu can even be meetings; or meet with the general community. Data that was collected
counter-productive in high rainfall areas due to its limited resistance to was limited, and no information was collected about the maintenance
intense rainfall, especially during the months of July and August (Kassie of SWC investments. Although our aim was not to provide conclusive
and Holden, 2006; Nigussie et al., 2014). The yield penalty from reasons for these problems, it could be at least partly associated with
adopting such physical structures is high for small farmers because of the absence of a tailor-made participatory watershed M&E working
the loss of productive land and the associated higher investment costs manual and the low focus given to empowering WDCs by the concerned
(Shiferaw and Holden, 2001). bodies. Therefore, the M&E system largely bypassed any active in-
volvement of the WDCs. The failure of this system, and the WDCs in
9.2. SWC management and equity general, was acknowledged by government staff at the Bureau of
Agriculture, stating, “we failed to functionalize WDCs, in general.”
As part of the plan for the current SWC program in the region, Because of the poorly functioning monitoring and evaluation, the
7
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
quality of the available data obtained via the reporting systems is intervention outcomes. More importantly, it encourages long-term co-
questionable. In effect, therefore, there is no feedback loop to ensure operation in cases where short-term cooperation is predicted to be
that monitoring data enables learning and improvements, because the unsuccessful (Lubell, 2004). However, the country’s mainstream nat-
data is inaccurate. In relation to this, one DA stated, “Sometimes I hate ural resource management policy and its institutional designs at work
myself when I prepare reports. I write them with my bosses [district are very far from this ideal vision, and little attention has been given to
agriculture officers] in my mind, questioning myself, ‘are they going to such potential outcomes from participation (Dessie et al., 2012); as a
be satisfied if I say this or that…”'. The experts we interviewed said that result, outcomes are often unsustainable. This finding is not surprising
nobody at the higher administrative levels was in a position to dis- in this region (Cochrane and Skjerdal, 2015; Harrison, 2002; Smit et al.,
courage such behavior given the structure of the M&E system. One 2017). These types of processes have traditionally operated in such a
potential reason for this, as pointed out by Faravelli et al. (2014), could way that communities have felt alienated from the process.
be related to the ‘competition effect’ of reporting falsified progress, Many stakeholders involved directly or indirectly recognize that in
where an individual’s misreporting improves his or her outcomes at the the current collective SWC initiative, State bodies still appear to be
expense of others. Those who are honest may be overtaken by their reluctant to share decision-making with local communities. The State is
dishonest colleagues and then begin to misreport results themselves, trying to rehabilitate degraded watersheds mainly by introducing
which in turn, is likely to affect the whole M&E system. physical SWC measures, but it is paying little attention to public in-
Furthermore, the lack of participation has limited social learning volvement in decision-making processes, including problem diagnosis,
processes, which were expected to develop and sustain the outcomes of planning, implementation, and monitoring and evaluation activities. In
the initiatives. Dessie et al. (2012) pointed out that the components of particular, the problems lie in the following areas: (1) planning pro-
social learning from such processes include strengthening the interplay cesses are top-down, not interactive, and (2) involvement of stake-
between local and outside knowledge, building mutual trust, collabor- holders, in particular farmers, seems to be poorly managed and some-
ating in land management activities, and promoting leadership within times appears to exist only on paper. Stakeholders recognize that such
local communities. problems could be addressed by actively including them and ensuring
that their needs and suggestions are integrated and that their partici-
9.4. Sustainability pation should be voluntary, not compulsory. These actions should then
lead to more and better participation, in particular, more differentiated
The lack of participation in design and planning has serious im- and improved plans. Divergent farmers’ needs and resource endow-
plication for participation, as it does for community ownership of the ments should be respected, which could lead to a higher acceptance of
activities, and thus for the long term sustainability. Many farmers ex- SWC measures. Plans and involvement would have to be adjusted to
pressed disinterest in the existing governance approaches and felt that different stakeholders’ circumstances and trade-offs made explicit.
their concerns went unheard or were quickly overruled. They only
provided their labor contribution to avoid penalties and to satisfy local 11. Recommendations for reform
administrative bodies. In some instances, not participating in SWC ac-
tivities and being forced to pay fees may be more financially rewarding The findings from the analysis based on the IAD framework result in
(i.e. the returns of following own economic activities instead of parti- some recommendations, which are intended to improve all IAD levels in
cipating in the activities are greater than the costs involved in paying particular the institutional context, the action arena, as well as the
the fees), such as those who opt to grow high-value crops such as khat evaluation and learning processes.
(chata edulis) in Aba Gerima. These feelings exacerbated governance Instead of compulsory measures, which should be abandoned, SWC
problems within the current land management systems because farmers programs should have selective provisions (e.g., formal credits) to in-
believed that their mere physical presence by itself was an end, and centivize all groups’ full engagement in collective endeavors. Such kind
they mainly did so to avoid punishment. In its current implementation of provisions could, for example, be done through investment subsidies
modality, the program will likely cease to exist if the governmental for SWC measures that on the one hand provide means to overcome
pressure, fees and (dis)incentives (e.g. patronage and exclusion) end. short term financial bottlenecks for farmers, and on the other hand
The SWC activities, in their current modality, are not serving the needs support disadvantaged target groups like youth, landless farmers and
of community members and thus the program is unsustainable in its less resource endowed women. This would also increase gender sensi-
present form. tivity of SWC programs. Landless groups could also be integrated by
other labor programs to help establish SWC in less advantaged areas, or
10. Discussion where the ecological benefits of SWC exceed the economic returns.
For successful collective watershed management, due consideration
Returning to Ostrom’s IAD framework, a summarized version of the to different gender-related needs, responsibilities, and endowments,
UBNB results is proposed in Fig. 4. The study has assessed the institu- and in particular the gender division of labor, should be given by SWC
tional settings for SWC watershed development programs, stakeholders’ programs to enhance women’s collaboration. As a result, for example,
perceptions of these, and of the impediments to efficient collective ef- introducing gender-sensitive watershed management strategies and
forts to develop within watershed communities. capacity building could contribute to women’s effective participation in
Crucial to the notion of people’s involvement in watershed devel- general, as well as to better incentivize labor contributions from all
opment is to ensure the inclusion of voices of all segments of the women, not only from female household heads.
community to decision making. However, researchers have suggested A proactive conflict management scheme should be introduced into
that participation should not be an end in itself but a means to an end ongoing SWC programs for ensuring effective participation of various
both to facilitate deliberation among actors (Rist et al., 2007) and segments of the community, sustaining watershed management out-
harness local community resources and support for the program comes, and also solving the prevailing and potential conflicts of interest
(Parfitt, 2004). In practical terms, promoting inclusive governance, for (e.g., clear and transparent system for contribution and benefit alloca-
example, through engaging the WDCs could provide synergy to increase tions) among watershed users. For instance, an entity that could be
collective investments in watershed development. All else being equal, involved in such a scheme is the WDC, however it needs to be endowed
inclusive governance can support local values, trust, customs, and with legitimacy by local authorities.
ownership; strengthen the enforcement of rules; address power rela- There should be a stringent and independent strategizing and
tions; and thereby improve local implementation capacity, monitoring planning process, including evaluation criteria and monitoring pro-
and sanctioning efforts, and long-term sustainability of the desired cesses. Independent, maybe even privatized bodies are needed to
8
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
organize and facilitate the strategizing and planning processes and also Acknowledgements
to monitor and evaluate SWC activities and institutional performance of
the SWC implementing organization. In particular, implementing and The authors are grateful to all respondents for their willingness to
monitoring/evaluating institutions should be separated to align the provide data, and to Anteneh Wubet and Getachew Dagnew for their
reward system with respect to results obtained. Moreover, local deci- field assistance. We are also thankful to the International Platform for
sion making about SWC activities should actively involve all watershed Dryland Research and Education (IPDRE) of Tottori University for of-
users, with inputs from government and DAs expertise. This would fering an overseas travel grant for the principal author. This work was
ensure and foster bottom-up processes. The processes should also be supported by JST/JICA, SATERPS.
evaluated by the target groups, in order to avoid suppression of criti-
cism and a bias of reporting to fulfilling plans on paper rather than in References
reality on the ground.
Akhtar-Schuster, M., Thomas, R.J., Stringer, L.C., Chasek, P., Seely, M., 2011. Improving
the enabling environment to combat land degradation: institutional, financial, legal
and science-policy challenges and solutions. Land Degrad. Dev. 22, 299–312.
12. Conclusion Alemayehu, M., Amede, T., Böhme, M., Peters, K., 2013. Collective management on
communal grazing lands: its impact on vegetation attributes and soil erosion in the
Unless addressed in a timely fashion, the above issues will jeo- upper Blue Nile basin, northwestern Ethiopia. Livest. Sci. 157, 271–279.
Alemseged, T.H., Tom, R., 2015. Evaluation of regional climate model simulations of
pardize current gains and potential future outcomes from SWC mea- rainfall over the Upper Blue Nile Basin. Atmos. Res. 161, 57–64.
sures in watershed communities and pose continuing problems for re- Amare, T., Zegeye, A.D., Yitaferu, B., Steenhuis, T.S., Hurni, H., Zeleke, G., 2014.
spective programs’ sustainability. Specifically, the findings from this Combined effect of soil bund with biological soil and water conservation measures in
the northwestern Ethiopian highlands. Ecohydrol. Hydrobiol. 14, 192–199.
study demonstrate that the established ‘participatory’ approaches are Amsalu, A., de Graaff, J., 2006. Farmers’ views of soil erosion problems and their con-
subject to criticisms that they are dominated by top-down processes, servation knowledge at Beressa watershed, central highlands of Ethiopia. Agric. Hum.
which lack stringent monitoring and reporting systems, and raise the Values 23, 99–108.
Assefa, E., Hans-Rudolf, B., 2017. Indigenous resource management practices in the Gamo
risk of conflicts between farmers, local authorities or extension workers.
Highland of Ethiopia: challenges and prospects for sustainable resource management.
As a consequence, approaches to solve soil and water management Sustainability Sci. 1–15.
problems in the region need to cover a wider range of options, which Astatke, A., Jabbar, M., Tanner, D., 2003. Participatory conservation tillage research: an
include: making programs inclusive of the broad public starting from experience with minimum tillage on an Ethiopian highland Vertisol. Agric. Ecosyst.
Environ. 95, 401–415.
their early stages, i.e. applying bottom-up approaches; ensuring effec- Ayele, G.K., Gessess, A.A., Addisie, M.B., Tilahun, S.A., Tebebu, T.Y., Tenessa, D.B.,
tive incentives for community participation (instead of e.g. applying Langendoen, E.J., Nicholson, C.F., Steenhuis, T.S., 2016. A biophysical and economic
penalties), and introducing holistic watershed development ap- assessment of a community-based rehabilitated gully in the Ethiopian highlands.
Land Degrad. Dev. 27, 270–280.
proaches, such as extending the programs to reach disadvantaged Bayabil, H.K., Tilahun, S.A., Collick, A.S., Yitaferu, B., Steenhuis, T.S., 2010. Are runoff
groups, taking closer looks at economic incentives and disincentives, processes ecologically or topographically driven in the (sub) humid Ethiopian high-
and promoting employment opportunities and integration of local lands? The case of the Maybar watershed. Ecohydrology 3, 457–466.
Benson, D., Jordan, A., Cook, H., Smith, L., 2013. Collaborative environmental govern-
technologies from and in SWC. As one major bottleneck in conducting ance: are watershed partnerships swimming or are they sinking? Land Use Policy 30,
SWCs, we have identified bureaucratic structures, like the rigid re- 748–757.
porting systems and the above mentioned penalty systems. This brings Berhanu, K., Poulton, C., 2014. The political economy of agricultural extension policy in
Ethiopia: economic growth and political control. Dev. Policy Rev. 32, s199–s216.
about new research needs for best practices of collaboration between Bewket, W., Sterk, G., 2002. Farmers’ participation in soil and water conservation ac-
State agencies and local–formal and informal–institutions in charge of tivities in the Chemoga watershed Blue Nile basin, Ethiopia. Land Degrad. Dev. 13,
managing commons as SWCs and their constituencies. 189–200.
Bezu, S., Holden, S., 2014. Are rural youth in Ethiopia abandoning agriculture? World
9
Z. Nigussie et al. Land Use Policy 71 (2018) 1–10
10