aid effectiveness to infrastructure: a comparative study of east … · 2016. 4. 19. · kenya case...
TRANSCRIPT
JBICI Research Paper No. 36-3
July 2008
ISSN 1347-5703
4-1, Ohtemachi 1-chome, Chiyoda-ku, Tokyo 100-8144, Japan
Tel: 03-5218-9720 (JBIC Institute)Internet: http://www.jbic.go.jp/ Recycled paper
JBICI R
esearch Paper No. 36-3
JBIC InstituteJapan Bank for International Cooperation
July 2008
Aid Effectiveness to Infrastructure:A Comparative Study of East Asia and
Sub-Saharan Africa
Case Studies of Sub-Saharan Africa
Aid Effectiveness to Infrastructure: A
Com
parative Study of East Asia and Sub-Saharan A
frica, Case Studies of Sub-Saharan A
frica
Francis M. MwegaUniversity of Nairobi
Aid Effectiveness to Infrastructure:a Comparative Study of East Asia And
Sub-saharan Africa
Kenya Case Study
i
i
Table of CONTENTS
1. OVERVIEW OF THE STUDY …………………………………………………………… 11.1 Background …………………………………………………………………………… 11.2 Objectives of the overall study ……………………………………………………… 11.3 Evolution and composition of foreign aid to Kenya ……………………………… 21.4 Foreign aid and economic growth in Kenya ……………………………………… 31.5 Choice and justification of Kenya project cases ………………………………… 5
2. THE BURA IRRIGATION AND SETTLEMENT SCHEME (BISS) ……………… 82.1 Background …………………………………………………………………………… 82.2 Conceptual and economic problems ……………………………………………… 102.3 Technical problems …………………………………………………………………… 112.4 Institutional and management problems ………………………………………… 132.5 Efforts at rehabilitating BISS ……………………………………………………… 14
3. COMPARISON OF BISS WITH THE TANA RIVER DELTA IRRIGATION PROJECT (TDIP) ………………………………………………………………………… 183.1 Summary of the JBIC (2001) evaluation of TDIP ……………………………… 183.2 Lessons from TDIP …………………………………………………………………… 233.3 Some findings from the fieldwork on TDIP ……………………………………… 27
4. THE NAIROBI WATER SUPPLY PROJECT (NWSP) ……………………………… 294.1 Summary of the JBIC (2002) evaluation of NWSP ……………………………… 294.2 Situation analysis of the NWSP …………………………………………………… 324.3 Some findings from the fieldwork on NWSP ……………………………………… 38
5. COMPARISON OF NWSP WITH THE NYERI TOWN WATER SUPPLY SYSTEM …………………………………………………………………………………… 40
6. FINDINGS AND RECOMMENDATIONS …………………………………………… 456.1 Sustainability of the infrastructural services …………………………………… 456.2 Institutional spillovers and explanatory factors ………………………………… 496.3 The role of the aid relationship in generating sustainability and spillovers … 52
RREFERENCES ……………………………………………………………………………… 54APPENDIX …………………………………………………………………………………… 57
ii iii
List of Tables and Box
Tables
Table 4.1: Water tariff rates in Nairobi …………………………………………………… 34Table A1.1: Japanese supported projects in Kenya ……………………………………… 57Table A2.1: Performance of Kenya’s major irrigation schemes, 1985/86 - 2004/05* … 58
Box
Box 2.1: From uncertainty to great expectations in Bura scheme …………………… 16
ii iii
Acronyms and Abbreviations
AfDB African Development Bank
AWSP Athi Water Services Board
BADEA Arab Bank for Economic Development in Africa
BISS Bura Irrigation and Settlement Scheme
EDF European Development Fund
EIB European Investment Bank
EIRR Economic Internal Rate of Return
FINIDA Finnish International Development Agency
FIRR Financial Internal Rate of Return
GDP Gross Domestic Product
GTZ Germany Agency for Technical Cooperation/ Deutsche Gesellschaft für Technische Zusammenarbeit
JBIC Japanese Bank for International Cooperation
JICA Japan International Cooperation Agency
KfW German Bank / Kreditanstalt für Wiederaufbau
NCC Nairobi City Council
NIB National Irrigation Board
NWSC Nairobi Water and Sewerage Company
NWSP Nairobi Water Supply Project
NYEWASCO Nyeri Water and Sewerage Company
ODA Overseas Development Assistance
OECD-DAC Organization for Economic Cooperation and Development – Development Assistance Committee
OECF Overseas Economic Cooperation Fund
OFID The OPEC Fund for International Development
OPEC Organization of Petroleum Exporting Countries
SIDA Swedish International Development Agency
TARDA Tana and Athi Rivers Development Authority
TDIP Tana River Delta Irrigation Project
TDIPO Tana Delta Irrigation Project Office
UK United Kingdom
UFW Unaccounted-for-water
WSD Water and Sewerage Department
WSRB Water Services Regulatory Board
1
1
1. OVERVIEW OF THE STUDY1
1.1 Background
In 1975, the average Gross Domestic Product (GDP) per capita gap between Africa
and East Asia and Pacific was $3942. Almost thirty years later in 2004, the GDP
per capita gap had increased almost ten times to $3,889. Similarly, in 1977, the life
expectancy gap at birth between the two regions was 16 years; in 2004, the gap was
24 years. What contributed to the difference in development trajectories? There have
been many attempts to answer this puzzle. The conclusions that these literatures
draw, however, vary widely indicating lack of consensus. Some studies point to
geographical factors (Sachs and Warner 1997), demographic factors (Kremer 1993),
while others focus on the differences in governance and policies (Burnside and Dollar
1997; Sachs and Warner 1997; World Bank 1998), ethnic fractionalization (Easterly
and Levine 1997) and institutional factors (Sachs and Warner 1997).
In the period between 1975 and 2003, Africa received $394 billion of ODA while
Asia (Far East and Oceania) received $228 billion (OECD-DAC 2005)3. It is natural for
us to ask, what happened in between those years? What has been the role of aid and
aid effectiveness in explaining the differences between Africa and East Asia in terms
of economic growth and poverty reduction? To borrow the words of the very people
who have contributed much to this debate, “while producing interesting results and
raising fascinating questions, cross-sectional macroeconomic studies cannot be the
last word on the topic of aid effectiveness” (Dollar and Levine 2005). In other words,
we know a whole lot about correlations and possible causalities between patterns
of aid allocations and changes in development indicators, based on macroeconomic
analyses. What is missing in these studies, however, is a careful and detailed analysis
of the differences in the transformation mechanism of foreign aid into growth in Asia
and in Africa. Without this careful analysis, the way out of “Africa’s Growth Tragedy”
(Easterly and Levine 1997) will remain unimaginable.
1.2 Objectives of the overall study
The wider study undertakes a comparative analysis of aid projects focusing on
ways in which these projects may have contributed to transforming growth processes,
1 Sections 1.1 and 1.2 draw heavily on the study’s Terms of Reference (February 12, 2007, version).2 In this paper, unless otherwise noted, Africa refers to Sub-Saharan Africa and Asia refers to East
Asia.3 Kenya received about US$18.7 billion during this period.
2 3
beyond the economic impacts of planned physical outputs. The study is interested in
impacts related to ideas influencing policy, transfer of knowledge and lessons learned,
organizational capacity to plan, implement and operate, and human resources
development in general. The main hypothesis is that the Asia-Africa contrast, outlined
above, and the role of foreign aid, to a large extent can be explained by differences
in the institutional linkages between aid projects and wider systems at the national
level.
It has been argued that infrastructure aid projects in Asia have contributed to
institutional reform, capacity building and human capital development of the recipient
countries. For example, reflecting on the infrastructure development projects of Japan
in Asia, Arakawa and Wakabayashi (2006) states that “The experiences at the project
level have led to capacity building and the establishment of comprehensive country
systems at the national level.” Because this said contribution of infrastructure aid to
institutions building - which may be one of the factors for realizing project impact - is
often not recognized and therefore less documented, it is the main focus of the study.
Thus, in the wider study, processes of institutional changes associated with specific
infrastructure projects and relationships between those institutional changes to
the overall impact of the infrastructure projects are examined in both East Asian
countries and African countries. Then comparing the differences and similarities of
the above processes draws out policy implications for aid in Africa.
1.3 Evolution and composition of foreign aid to Kenya
Various studies have analyzed the pattern and evolution of foreign aid in Kenya
(Mwega 2003). The country experienced a dramatic build-up in nominal aid flows
during the 1970s and 1980s, with a slackening of donor support in the 1990s. Gross
ODA inflows increased from an annual average of $582 million in the 1970s to $673
million in the 1980s and to $857 million in the 1990s. During this period, foreign
aid averaged about 9% of GDP, accounting for about 20% of the annual government
budget and financing slightly over 80% of development expenditures (Njeru 2004).
Donor support however reached a peak of $1.6 billion in 1990, generally declining
thereafter as the government reneged on its commitment to donors, leading to aid
freezes in 1992, 1997 and 2000. At its peak, net ODA inflows were equivalent to about
45% of the government budget, with Kenya’s share of total development aid in Sub-
Saharan Africa remaining remarkably stable at 5.2%4. Kenya also benefited from the
4 Following the December 2002 elections, government revenues have increased dramatically so that ODA currently (2007) accounts for less that 5% of the total government budget.
2 3
fact that the terms of its ODA softened over the period. The share of grants increased,
rising from 47% in the 1970s to 51% in the 1980s and to more than 69% in the 1990s.
ODA takes many forms. Kenya, for example, received approximately three-quarters
of its total aid from bilateral donors, with no distinct trend toward greater reliance on
either multilateral or bilateral aid. The share of multilateral aid increased moderately
in the 1980s, primarily due to the disbursement of the World Bank adjustment loans,
but the bilateral share rose again in the 1990s with the decline in new adjustment
lending after 1991. Bilateral aid has been mainly in the form of grants (72% of the
total), with the share of grants increasing in recent years, whereas multilateral aid
has mainly been in the form of loans (86%). The principal source of multilateral loans
has been the World Bank group, accounting for almost 80 percent of total loans in the
study period.
A large proportion of bilateral grant aid (58 percent over the 1970s) was given for
technical assistance. During the 1980s, the share of technical assistance in total grant
assistance began to decline, to 39% in the 1980s and 36% in the 1990s, as bilateral
donors shifted an increasing proportion of their project assistance to grant terms.
However, the absolute amounts expended for technical assistance have remained high,
averaging $140 million per year in the 1980s and $225 million per year in the 1990s.
Much of this is money that the country does not receive, since it is paid directly to the
resident advisors, or to education and training institutions outside of Kenya, by the
aid-giving country or international agency.
Bilateral donors have also provided debt forgiveness of prior ODA debt of $700
million between 1986 and 1992 and $26 million in 1997. The principal sources of ODA
debt relief were the United States, Germany, Canada, the Netherlands and the United
Kingdom (UK). The Japanese government has not provided debt forgiveness, but has
offset debt repayments with supplemental grant aid.
1.4 Foreign aid and economic growth in Kenya
There are various ways to analyze the linkage between foreign aid and economic
growth. The most common approach is through cross-country regressions. A major
but controversial finding of this literature is that foreign aid can only raise economic
growth in a good policy environment (Burnside and Dollar 1997, 2000; World Bank
1998). A related finding is that foreign aid cannot ‘buy’ good policies through ex
ante conditionality. This is because economic reforms induced by foreign aid are
‘time-inconsistent’ and therefore lack credibility. This is not changed by repeated
donor-recipient interactions, since the threat to punish those who reverse policies by
4 5
denying them future lending is itself not credible, due to the pressure on donor staff to
continue lending.
These empirical findings have had a major influence on the debate on foreign aid,
for if ex ante conditionality does not work, donors then have two options: (i) redesign
foreign aid contracts to improve the policies pursued by recipients; or (ii) switch to ex
post conditionality or selectivity. Donors are increasingly moving towards the second
option, while aid recipients are wary about this shift in the foreign aid policy.
Another way of analyzing the relationship between foreign aid and economic growth
is to utilize the two-gap model. This model is derived from the national accounts
identity:
I-S= M-X,
where I is investment, S savings, M imports and X exports. While these two gaps
are equal ex post, they may differ ex ante, as the expected savings gap (I-S) deviates
from the expected foreign exchange gap (M-X). In the absence of private financing,
donors fill the binding gap with foreign aid, to achieve a target growth rate, with a
transmission mechanism from aid to investment to growth. The role of foreign aid is
therefore to fill the smaller of these gaps, hence supplementing domestic savings or
foreign exchange earnings from exports depending on whether investment or imports
are the binding constraints to economic growth, respectively.
One simple way to assess whether it is the foreign exchange or savings gap which is
the binding constraint to economic growth in Kenya since the 1980s is to analyze the
relationship between (i) foreign aid and imports; and (ii) foreign aid and investment
(see Easterly 2001). The foreign exchange (savings) gap is binding on growth if the
first (second) relationship is positive and statistically stronger. Ideally, a 1% of AID/
GDP should be reflected in a 1% increase in imports (investment) GDP ratio if the
foreign exchange (savings) gap is the binding constraint. These variables could also
be expected to rise by even more than one for one because of increased savings (export
earnings) by aid recipients. The trade gap is expected to be less influential following
the economic liberalization of the 1990s with foreign exchange shortages in the
country becoming less likely. Kenya has witnessed a substantial liberalization of its
foreign exchange market since the early 1990s.
An analysis by Mwega (2004) found the relationship between ODA and investment
to be positive and much stronger than that between ODA and imports, although the
coefficient was much smaller than one (0.374). From these results, it was quite clear
4 5
that it is the savings gap that has been the binding constraint to growth in Kenya
since the 1980s.
The study also found the relationship between foreign aid and public gross fixed
capital formation highly significant since the 1980s (t-value=3.313), with the decline
in ODA accompanied by a drastic decline in public investment during the study
period. Njeru (2004) also found a strong statistical relationship between ODA and
government development expenditures, with a shilling increase in ODA leading to 57
cents increase in government development spending over 1970-99. ODA was therefore
almost equally shared between recurrent and development expenditures, even though
all ODA in Kenya is recorded on the development expenditure vote. This is consistent
with the results above that only a fraction of ODA is spent on investment. The study
by Mwega (2004) also showed a positive but non-significant relationship between
foreign aid and private gross fixed capital formation since the 1980s as both also
generally declined during the study period.
1.5 Choice and justification of Kenya project cases
We have selected two projects for detailed study. These major projects are:
1. The Bura Irrigation and Settlement Scheme (BISS); and
2. The Nairobi Water Supply Project (NWSP).
The two projects satisfy the following criteria set out in the study’s Terms of
Reference.
Each country study will include one case with funding from Japan. Other donors •
should preferably fund the other case, although for some countries there can be
valid reasons for selecting two Japanese-funded projects.
The World Bank and the European Development Bank (EDF) mainly funded BISS.
Bilateral finance in the form of grants and soft loans was also provided by Britain,
Finland, Holland and Japan (Howells 1985). The main actor was however the World
Bank, with the other donors taking little part in the monitoring of the project. On the
other hand, NWSP was co-financed by Overseas Economic Cooperation Fund (OECF)
(former name of the present Japan Bank for International Cooperation (JBIC)), the
World Bank, the African Development Bank (AfDB) and the European Investment
Bank (EIB). While the project was commissioned in 1994, the tunnels started to be
used in 1992.
6 7
The selected projects should be large infrastructure projects, considered as having •
national significance at the time of planning or even in retrospect. Large projects
means “challenging” and “pioneering” projects (i.e., projects that were beyond
the capacity of the recipient countries at the time of implementation). Integrated
programmes consisting of many smaller investments may be selected.
BISS was a large project, projected to cost about KSh 766 million (about US$ 98.4
million) in 1977, equivalent to Ksh 148,700 (about US$ 18,000) per each family settled
(Ruigu 1988). There were however major delays from the onset of the implementation
of the scheme in practically all administrative areas as well as major revisions on
scheme design. As a result, costs soared, rising to about US$ 121.7 million at 2000
prices (Inocencio et al. 2005)5. According to Ruigu, the World Bank pumping and
gravitational projects cost US$ 145 and US$ 31, respectively, compared to over
US$1,500 in Bura, making it one of the most capital-intensive schemes in the world.
As seen in Table A1.1 in the appendix, NWSP is one of Japan’s largest projects in
Kenya’s social sector (Japanese Yen 5.342 billion, about US$ 38.321 million)6.
The selected projects should be economic infrastructures such as transport (road, •
ports, etc), energy/power, urban water and sewerage, telecommunications, or
irrigation.
The two selected projects cover irrigation, floods control and settlement; as well as
urban water supply, respectively.
Each country study will have to cover 2 sectors, and as a minimum select one project •
for in-depth study in each sector. This will be complemented with background
information on the sectors selected, allowing a comparison of the project case with a
general pattern of the sector possibly with examples of reference projects.
Our reference projects are the Tana River Delta Irrigation Project (TDIP); and the
Nyeri Town Water Supply System, respectively.
The Japanese Government has devoted one of the large amounts of foreign aid
resources in Kenya to the TDIP and related investment (Japanese Yen 6,619 billion,).
5 According to the World Bank Statement of Loans for February 2007, its component (approved in 1977) was US$ 34.00 million, of which US$ 28.85 million was disbursed and US$ 5.05 million was cancelled. The loan has been fully repaid by the government.
6 Based on an average exchange rate of Yen 137.96 per US dollar in 1989. Earlier TOR (e.g. the November 1, 2006, draft) explicitly required analysis of one economic infrastructure project and another project on basic service delivery such as water and education.
6 7
The Nyeri Town Water Supply System, mainly supported and funded by German
development agencies (Germany Agency for Technical Cooperation / Deutsche
Gesellschaft für Technische Zusammenarbeit(GTZ) - capacity building and German
Bank/ Kreditanstalt für Wiederaufbau (KfW) - investment), is considered an example
of successful water commercialization in the country. Commercialization has mainly
involved formation of water companies with private-sector-like management, allowing
them sufficient autonomy in decision making while ensuring that asset ownership
remains within the public domain. Water supply and sanitation has substantially
improved in Nyeri town, with mainly local funding arising from a systematic
expansion of the revenue base. The approach has been used as a model for the country
as a whole. The Nyeri case illustrates how successful and national institutions are
created to include greater private sector participation and to ensure their financial
involvement in urban service provision (RTI International, 2005)
To ensure that institutional effects can be identified, only projects completed a •
minimum of 5 years back will be selected.
The two main selected projects were completed in the late 1970s and the late
1990s, respectively.
8 9
2. THE BURA IRRIGATION AND SETTLEMENT SCHEME (BISS)
2.1 Background7
Many irrigation schemes have been initiated along the Tana River (the longest and
biggest river in Kenya). Beside the TDIP (which is discussed below), Tana River had
two major irrigation projects (Hola and Bura) all of which failed. As seen in Table A2.1
in the Appendix, both irrigation schemes were non-operational by the mid-1980s and
late-1990s, respectively8. The schemes were badly mismanaged, eventually resulting
in collapse. The BISS, at the insistence of the World Bank, had relied on pumping
water down large irrigation trenches. According to a former employee, a wrong choice
of the pumps was made whereby components and spare parts came from different
continents. Siltation (mainly sand) either clogged or destroyed these pumps and the
dredgers were rendered useless. Hence, when the pumps broke down, the waterways
silted up; while prices for the main crop planted in the area (cotton) fell drastically
on the open international market9. By the late 1980s, BISS was an open wasteland
of invader bushes, unpalatable by animals10. The failure of the project to meet
expectations caused the Kenya Government to shy away from similar schemes. TDIP
(discussed below) was once considered for settlement possibilities, but now proceeded
on the basis of commercial estates with a few outgrowers. This type of development is
unlikely to have the same impact on unemployment and landlessness as settlement,
although it is more likely to produce an economic return. There were also plans to
run the semi-completed part of the BISS as an estate in an attempt to reduce the
operating subsidy.
7 Some information in this sub-section is based on Howells (1985) and Adams (1990). 8 In the 1980s, irrigation activities were mainly concentrated in seven major schemes. These are
(i) Mwea, (ii) Ahero, (iii) West Kano, (iv) Bunyala, (v) Tana, (vi) Bura, and (vii) Perkerra. Mwea is the biggest scheme, which produces more than three-quarters of the rice consumed in the country.
9 According to the former employee, other acts of mismanagement included undue influence of politicians in the award of contracts; dubious criteria in the selection of tenants; hiring of consultants who did not hand over reports, were there for too long and had nothing to show for it; uncontrolled and irregular recruitment of workers, some ghost; conflict of interest in the supply of inputs; the insurance arrangements were irregularly procured with the result that the expensive guest house was not properly covered and burnt without compensation; main canals without lining were filled with sand and soil after rains; and so on.
10 Former President Moi once described the Scheme as “a failure, a disgrace and the height of mismanagement” (Kenya Times, January 22, 1986). Horta (1994) called it an “an unmitigated disaster, the result of remarkably poor planning”. According to Horta, the project owners spent $55,000 for every settler on the project site, in a country where the per capita income is only $350, and yet “these settlers and their families suffer abject poverty and drought and famine…. Malnutrition and disease are rampant. The Bura Project initially planned to build 20 village health units and various health centres, but these were cancelled”.
8 9
The main objectives of the BISS were seen as settling landless people, providing
them with worthwhile employment and increasing agricultural output. Settlers were
to be selected from each of Kenya’s Provinces on a quota basis. The aim was to settle
5,200 families in 23 villages and the population was expected to ultimately reach
65,000. Infrastructure in the form of roads, treated water supplies, sewerage, schools
and clinics were to be provided to accommodate this population.
Each settler family was to be provided with a house, a 0.5 hectares vegetable plot
and 1.25 hectares of land on which they were required to grow cotton and maize.
Cotton was to be the major cash crop, with maize grown for subsistence and local
resale. Services such as land preparation, crop protection, and provision of water, seed
and fertilizer were to be provided by the scheme authorities and charges deducted
from the proceeds from the sale of cotton. Cotton was to be ginned on site and sold for
export, despite Kenya being a net importer, since its superior quality was expected to
attract a price premium. Irrigated fuel-wood plantations were included in the plan to
provide fuel for the settled population and to prevent the destruction of the riverine
forest strip (which is considered to be biologically unique).
The executing agency was the National Irrigation Board (NIB), a parastatal body
that was established in 1966 through an Act of Parliament (Cap 347). Construction
started in 1979 and the first settlers moved in 1981. However, about that time the
project began to run into financial difficulties. Inflation, devaluation, underestimations
and unforeseen expenditures led to rapidly rising costs. With a fixed aid package,
Kenya’s share of the costs rose from KSh160 million (20%) in 1977 to KSh l billion
(45%) by 1982. The government was unable to meet its obligations on many contracts
that led to substantial delays and costly claims for damages.
In 1983, a decision was made to curtail the scope of the project by about half,
reducing the number of settler families to 2,500 and the cost to KSh 1.50 billion. This
was despite the fact that irrigation and water treatment works had already been
substantially completed for the full population, the main outstanding works being
houses, schools and the fuel-wood plantations. The aid agencies were asked and
agreed to provide 100% financing for the remaining construction to ease the cash flow
problems.
As well as the severe financial problems, the project suffered from conceptual,
technical and institutional deficiencies.
10 11
2.2 Conceptual and economic problems11
i. Settlement issues
The concept of settling people from all over Kenya on a quota basis is dubious.
This is because settlers from the Kenya highlands, where the pressure on land is the
greatest, are unused to the inhospitable climate and lack resistance to the endemic
diseases such as malaria and bilharzia. Desertion and death rates amongst settlers
was therefore high and many household heads chose to leave their families at home.
The result was that the settlement objectives and the relief of pressure on the land
were not met. In addition, the settlers were unable to manage the holdings single-
handedly and production consequently suffered.
Many of the settlers consequently saw the project as providing only temporary
employment, and intended to return home if they could save enough to buy land there.
In contrast, ‘settlers’ from the local Coast Province who were more used to the climate
and had more resistance to the endemic diseases adapted well to the Scheme, and
farmed reasonably successfully.
Thus the project as a solution to landlessness and urban drift may have been
misconceived from the outset.
ii. Incentives for production
Ownership of land is an important issue in Kenya, yet settlers were given land
on one-year renewable leases. This led to a sense of insecurity and destroyed any
incentive for the tenants to develop the land or settle permanently, unlike settlers in
rain-fed settlement schemes who got freehold titles as a matter of policy. The provision
of all inputs (seed, fertilizer, and so on) by the authorities and the authoritarian
management style also led the settlers to view themselves as labourers, rather than
self-sustaining farmers. The result was that the settlers came to expect to be spoon-
fed and showed few signs of initiative.
The management position was that provision of these inputs, like the tenancy
arrangements, is necessary to retain control and to ensure that the crop calendar is
fulfilled. It is also argued that these inputs had to be provided since the settlers had no
resources of their own. In practice, inputs were not being provided on time and yields
were depressed. Yet the tenants had no recourse, since charges were automatically
11 This sub-section is mainly based on Howells (1985) and Adams (1990).
10 11
deducted from the proceeds of the sale of their cotton, which was managed by the
authorities. Co-operative arrangements might have provided a better approach to
managing inputs and sales and would have encouraged the settlers to become more
involved with the scheme.
One further major disincentive was the producer price of cotton, which was
controlled by the Cotton Board and estimated to be about 60% of a realistic price that
might exist in a free market. Low returns hardly provided much incentive for far
mers to put effort into growing cotton, and the more enterprising farmers were able
to make more money (and get a better cash flow) by selling tomatoes grown on their
vegetable plots, to the detriment of cotton production.
iii. Economies of scale
Early studies advocated a minimum project of 15,000 hectares to achieve economies
of scale in the provision of infrastructure and abstraction of water from the river by
gravity. However, the surveys failed to find more than 4,000 hectares of suitable soils
(despite a lowering of the criteria). After a World Bank appraisal, a decision was made
to proceed with a project comprising 4,000 hectares of suitable soils and 3,000 hectares
of marginal soils. This decision was taken in the expectation that soils would prove
suitable and that additional areas of marginal soils could be developed later to give
scale economies.
At the time of appraisal, returns on investment were projected at 13%. However,
these were based on allocation of certain costs over a wider (future) project base.
Realistically, returns were probably well below 10% before the rapid cost escalations
started and static world cotton price quickly led to the project becoming uneconomic.
The Bank raised the possibility of abandonment in 1979, but construction was already
underway. By then, the loan and credit agreements had been signed.
2.3 Technical problems12
The project also faced many technical problems, the most detrimental of which are
discussed briefly below.
i. Crop yields
These proved to be well below expectations with cotton yielding 70% and maize
12 Ibid.
12 13
below 50% of the lowest projections, resulting in low settler incomes (Howell 1985,
Adams 1990). Reasons included poor soils, poor crop varieties, labour shortages,
inexperienced farmers, poor reliability of water supplies and late inputs.
One concession to the economies of scale problem was that gravity abstraction of
water was postponed and a diesel pump station built instead. The remote location,
pump availability (below 50%) and poor road access resulted in erratic fuel supplies.
Procurement of spares from overseas was difficult and slow because of foreign
exchange problems.
Given the location and the poor wages paid by the sector, it was not possible to
obtain the highly skilled mechanics needed to maintain the station. The station was
the lynchpin of the project and success depended on its proper functioning, yet it was
inoperable for much of the time. Problems of maintenance, spare parts, fuel supply
and lack of trained operators and mechanics led to repeated breakdowns in 1983 and
1984, with water available for only 25% of the time.
ii. Housing
Settler housing also proved a particular problem in that it was extremely difficult
to find low cost housing solutions in an area devoid of most natural materials. The
construction of tenant houses fell behind schedule. The solution finally adopted, at
a cost of KSh 20,000 per house, was a mud and mangrove pole affair with rendered
walls and a corrugated iron roof, with many houses collapsing within two to three
years of construction. Toilets were also a challenge as they easily collapsed due to poor
soils. Tenants were expected to buy these houses over 20 years through deductions
made from their cotton sales. Not surprisingly, this created a great deal of resentment.
iii. Choice of technology
In many cases, the choice of technology was over-sophisticated in design and
concept, and this resulted in subsequent operating problems. These included problems
with an expensive (KSh 120 million) and complex water treatment and distribution
system (never operated); the river pump station; a dredger to remove silt from the
main canal (never operated); and an over-sophisticated irrigation system with many
controls (most of which were never used).
12 13
2.4 Institutional and management problems13
i. Co-ordination with other organizations
One of the major problems was obtaining support from other government bodies,
particularly in getting them to take over completed infrastructure. The Ministry of
Water Development refused to accept the new treatment works on the grounds that it
did not have a budget line or the staff to run it.
Another was increased health risk to settlers. Disease incidence was high among
settlers and their children, particularly because of malaria. Most settlers used canal
water for supplies with a commensurate increase in the incidence of water-borne
diseases, decreasing community productivity. Similarly, the Ministry of Health was
three years late in taking over a clinic. In the interim period, the Catholic Sisters
provided healthcare but this proved inadequate given the number and condition of the
settlers. Delay of donated medical equipment at the port on grounds of taxation, also
delayed the opening of the health clinic.
ii. Staff development
The World Bank appraisal report recognized the weakness of the NIB as an
implementing agency and recommended that it should be strengthened. Competent
staffing levels were not met due to poor government scales and regulations. The report
also recommended the appointment of a team of expatriate managers for the first five
years of operation. These recommendations were included as covenants in the loan
agreement, but were ignored until the project got out of hand.
By that time, the EDF had decided to withdraw their support. The project staff were
poorly trained, poorly motivated and in deficient numbers. Qualified and experienced
staff were difficult to obtain, given the project’s location and the salaries offered.
iii. Over-centralization of decisions
Staff on site had little authority, with all major and procurements being made in
Nairobi (Howell 1985). Given the location and the poor communications, this resulted
in unnecessary delays and lowered the morale of the site staff to the detriment of the
project and the settlers, besides the high costs of transport for inputs and Nairobi- site
visits
13 Ibid.
14 15
In 1985, the World Bank conducted a review of the project that was highly critical
of the way in which it was conceived, designed and implemented. As a result of this
review, the project was taken from the NIB and set up as a separate parastatal under
the Ministry of Agriculture, the intention being to decentralize decision-making and
give more autonomy to the site staff.
2.5 Efforts at rehabilitating BISS
The discussion above shows that besides mismanagement, BISS was not well
designed (irrigation water for example should not be used for maize cultivation) and
was therefore not sustained. Our analysis therefore mainly focuses on the lessons
learned from the project for the Kenyan policy-makers, and to a lesser extent, the
World Bank. The most interesting “institutional spillovers” in this case is probably the
lessons drawn from the problems encountered by the project.
As seen in Box 2.1 below, there are recent efforts to revive the scheme and policy-
makers seem to have learned some lessons from past mistakes. According to a
government official, the key problem was that BISS was scaled back, leading to
limited exploitation of scale economies, increasing the operations and maintenance
costs of the project. According to the NIB, which came back to manage BISS in 2005,
the challenges facing BISS included:
Farmers have not planted a cash crop for 15 years (from 1990-2005). Similarly, •
they have not planted subsistence crops for 9 years (1994-2002). This has resulted
in famine, increased poverty levels and unemployment for the scheme farmers and
community.
Frequent breakdown of the pumping station.•
Silted canals are ringed by • Prosopis Juliflora (Mathenge) bush.
Weak cooperative and other farmers’ organizations.•
Change of approach to management in public irrigation projects in line with the new •
Government and NIB policy of Participatory Irrigation Management.
Proliferation of • Prosopis Juliflora bush in farming areas and in water conveyance
and storage.
When NIB took over, it started the rehabilitation process, especially of existing
14 15
pumps, with financing from the Kenya government14. The objective was the immediate
resumption of irrigated agricultural activities in the area, with only 500 hectares
under irrigation at that time. It was anticipated that the water irrigation supply
system would be stabilized, with 2,500 hectares put under irrigation; this increasing
to 5,500 hectares once the pump-gravity system was fully operational.
According to a NIB official, the beneficiaries paying for the operations and
maintenance costs of the scheme will secure the future sustainability of the project.
NIB was in early 2008 in the process of hiring consultants to determine how much
farmers should pay, estimated to be about KSh 5-10,000 per hectare per annum.
Farmers would be advised on what crops to grow, based on their relative returns, with
NIB facilitating farmers’ connection to markets. Commercial farmers would also be
encouraged to lease land as part of the efforts to make the scheme self-sustaining. The
enhanced use of the gravity system would also lower the operation and maintenance
costs15. According to the official, the loans from the Arab Funding Institutions do
not have policy conditionality, although they have guidelines on how to manage the
funds16.
According to the official, NIB has accumulated a lot of experience in managing
irrigation schemes and BISS is not different from the other schemes that it manages
such as Ahero, Bunyala and Hola, which are all currently under rehabilitation.
The performance of schemes declined with the general decline of the agricultural
sector and the economy in general in the 1980s and 90s due to poor governance in
the country. There have also been attempts to streamline the irrigation policy in
the country, to bring about harmony in the sector. A cabinet paper has already been
prepared for discussion.
14 The management of the scheme changed hands from NIB to the Ministry of Agriculture in 1985; to the Ministry of Regional Development in 1988; to the Ministry of Reclamation, Regional and Water Development in 1993; to the Ministry of Agriculture in 2001; and to the Ministry of Water and Irrigation in 2003. A NIB official claimed that the taking away of the management of BISS to the Ministry of Agriculture in 1985 was driven by politics rather than lack of capacity at NIB.
15 Some stakeholders pointed out that use of the gravity system limits the area that can be irrigated, so that this may not be a panacea to the problems facing the scheme.
16 Government officials have however complained that the negotiations take too long to conclude. The negotiations for BISS for example that started in mid-1990 were concluded in early 2007 when the Kuwait Fund carried out a successful appraisal of the project.
16 17
Box 2.1:
From uncertainty to great expectations in Bura scheme
Daily Nation, November 27, 2007
For decades, farmers in Bura irrigation scheme have lived each day at a time, their agriculture lifeline severely threatened. The back and forth shift in the scheme’s management over the years had left them disillusioned and bitter with the Government. The scheme has been under seven different management teams since its inception in 1978, with the Ministry of Agriculture moving in on three occasions. However, a new lease of life has been unleashed in the area following the successful negotiation for a KSh 4 billion funding from Gulf financiers to improve the water supply in the area through gravity.
When the National Irrigation Board took over the management of the scheme mid-2005, the farmers resented. This was the second time the board was returning to the scheme. The farmers, with knowledge of hindsight, were apprehensive since the board did not prove its worth between 1979 and 1985 when it was managing the scheme. Currently, 2,500 hectares are under irrigation, with 2,200 resident farmers. Each household has been allocated 1.25 hectares for main crop farming and 0.05 hectares for vegetables. Cotton farming is the dominant activity while maize, cowpeas and groundnuts are planted between October and December.
On taking over, the NIB put a raft of measures to ensure that the scheme realized its potential. Among them were repairs to enable efficient irrigation water supply and expansion of the water supply to cover 5,500 hectares in three years time. Changing of the irrigation water supply system from pumping to gravity also featured prominently in the programme. In the gravity system, water is held in a reservoir then released into the field systematically. With soaring cost of fuel, extensive use of pumps to drive water through irrigation channels is uneconomical. Bura is not connected to the national power grid. Operations at the scheme solely rely on diesel.
The KSh 4 billion funding component was signed in Vienna, Austria, and will go towards the implementation of the gravity programme as well as other expansion programmes.
The loan has been secured from three groups namely; Kuwait Fund for International Development, Arab Bank for Economic Development in Africa (BADEA) and OPEC (Organization of Petroleum Exporting Countries) Fund for International Development (OFID). Last year, a socio-economic study was carried out in the scheme. The study proposed new cropping patterns while laying emphasis on commercial farming. High value crops such as soya beans, maize, groundnuts, passion fruit and sunflower were recommended.
The 500 acres research farm serves as a demonstration field to the farmers. Contract farming has also been successfully negotiated with Kenya Seed Company. Faida Seeds
16 17
and Bidco Oil companies are other firms that have expressed interest in dealing with scheme farmers. This kind of farming helps in assuring the farmers of a ready market once they harvest the produce.
Information from the board indicates that the scheme will resume production in two phases. In the first phase, 2,500 hectares will be put under cultivation before the lapse of the 2007/2008 financial year. The next phase, to incorporate the intended expansion of the area under cultivation to 5,500 hectares, will go on up to 2012. The pumps were handed over to the farmers last week at the NIB headquarters in Nairobi. The farmers’ representative was optimistic that their production would improve and help in sustaining their households. Purchased at KSh 96 million, the pumps are to be installed in the next three weeks. The operation of the new pumps will begin before the end of the year.
According to permanent secretary Mahboub Maalim, this is part of an ongoing rehabilitation process for irrigation schemes. Already, the Government has committed Sh300 million towards the rehabilitation. This will include both private and public schemes across the country. The Ministry of Water and Irrigation has also finalised the irrigation and drainage policy, to support the processes.
18 19
3. COMPARISON OF THE BISS WITH THE TANA RIVER DELTA IRRIGATION PROJECT (TDIP)
3.1 Summary of the JBIC (2001) evaluation of TDIP
3.1.1 Objective of TDIP
The objective of the project was to utilize the land and water resources of the lower
Tana delta for irrigation to increase rice production to meet the growth in consumption
and to contribute to greater import substitution and self-sufficiency (JBIC 2001).
Unlike BISS, TDIP utilized the “estate system“, under which all agricultural
development activities, including agricultural infrastructure, production, quality
control, marketing and sales, administration, operations and maintenance were
to be managed and administered in a unified manner. The execution agency of the
project was the Tana and Athi Rivers Development Authority (TARDA), a parastatal.
One of the postulated advantages of the estate system is that it enables efficient
and profitable agricultural operations with a small investment when compared to a
settlement scheme.
3.1.2 Project scope
The ODA loan covered 85% of the total project cost. Specifically, it was allocated
to procure materials and equipment as well as services required for the construction
of irrigated farmland, in addition to irrigation and farming facilities (such as staff
members’ accommodations and rice-processing mills). The loan was also allocated to
procure most of the consulting services.
The amounts, terms and the timing of the ODA disbursements were to be as follows:
Loan amount/Loan disbursed amount
¥6.031 billion/¥6.025 billion
Exchange of notes/Loan agreement
March 1990/March 1990
Terms and conditions repayment period
(grace period):
Interest rate:; 2.5%, 30 years (10
years), generally untied
Final disbursement date December 1997
3.1.3 Project innovations
The project had four major innovations, which were pilots for the agricultural
18 19
sector in Kenya (JBIC 2001)17. The first innovation was the introduction of the large-
scale state management system. The Kenyan government already had experience
in implementing large-scale irrigation projects, but the previous irrigation projects
employed the settlement (tenant) system in which land rights were granted to
individual farmers (see Table A2.1 in the appendix). The estate management system
was introduced in the Tana delta project wherein the management and execution of
all processes from production to harvesting to sales were to be conducted by the Tana
Delta Irrigation Project Office (TDIPO). Irrigated rice production under this system
was the first attempt of its kind to be made in Kenya.
The second innovation was the introduction of large-scale mechanization technology
to the project. Major agricultural works, including land cultivation, rice planting
and harvesting, were to be executed through the introduction of large agricultural
machines rather than the previously utilized labor-intensive systems. Similarly, as
part of the rice planting techniques, a less labor-intensive system of directly sowing
seed in irrigated rice fields was to be used for the first time.
The third innovation was the development of high-yield rice varieties suitable for
the coastal region. The project aimed to develop unique high-yield rice varieties that
suit the geographical and climatic conditions of the Kenyan coastal region where the
Tana delta is located. The final innovation was double-cropping of rice, another first in
Kenya.
3.1.4 Results and evaluation
a. Relevance and efficiency
The project aimed to increase rice production and reduce rice imports, thereby
raising the ratio of self-sufficiency. A look at the changes in domestic rice production
during the period from 1990 to 1998 shows that annual production remained equal
or less than 40,000 tons with no marked fluctuations. Immediately after the project
was completed, however, the site sustained enormous damage from floods caused by
El Nino rains in 1997, and the initially intended effects failed to be delivered. The site
was still under recovery at the time of the JBIC (2001) evaluation.
Initial plans called for the project to be completed by October 1996, but it was
actually completed in December 1997, approximately 14 months behind schedule. The
17 Some respondents commented that these were too many innovations to be undertaken at the same time, undermining the feasibility of the project.
20 21
major reasons for the delay were (i) the need to partially reassess the design of the
intake ports and other facilities, and (ii) the undeveloped state of the access road from
Malindi.
b. Effectiveness
i. Effects on irrigation area and rice production
According to JBIC (2001), the project was completed in December 1997. However,
because the site sustained enormous damage in the same month from floods caused
by the El Nino phenomenon, it was not possible to attain the planned levels of
effectiveness in terms of irrigation area (1,840 hectares) and rice yield (18,400 tons on
an unhulled-rice basis and 12,000 tons on a polished-rice basis). In terms of irrigation
area, however, targets had effectively been met in 1997, when work was completed.
The El Nino phenomenon caused extraordinarily heavy rains, and the resulting
flooding in the lower Tana caused flood protection banks that had been constructed by
the Kenyan government and TARDA in 1989 to the east and west of the project site
to collapse at various locations. In addition, the irrigation channels, drainage, control
roads and farmland facilities were heavily damaged, rendering most of the irrigation
facilities that had just been completed virtually unusable. This flood was of a scale
comparable to those of a once-in-50-year probability and exceeded the design scope for
the protection banks. Therefore TARDA was forced to suspend all project activities.
Under these circumstances, OECF/JBIC undertook a survey on “Special Assistance
for Project Sustainability of the Tana Delta Irrigation Project (I)” in July 1998 and
proposed recovery plans and temporary measures to the Kenyan government and
TARDA. In 1999, based on these proposals, TARDA began temporary construction,
allowing production to be restarted in certain areas that had sustained relatively less
damage.
The Kenyan government meanwhile made a request to the Japanese government
for an ODA loan to fund the Tana River Flood Damage Recovery Plan, which included
the rehabilitation of major structures and other full-scale recovery construction. Such
ODA loan had not yet been granted by the time of the JBIC (2001) report. The Kenya
government continued to promote recovery work and allocated KSh 155 million to
TARDA for recovery plans from its budget in 2000/2001. Lack of finances and possible
20 21
financiers has limited the pace of the rehabilitation.18 19
ii. Effects of the project innovations
With respect to the introduction of the large-scale estate management method,
the TDIPO confirmed that the estate administration and management system was
established during and after the project. Fully-fledged large-scale mechanization
technology was also introduced. In order to develop high-yield rice varieties suitable
for the coastal geographical and climatic conditions, the number of varieties was
narrowed down from 300 to five varieties that were selected for their (a) resistance to
insects, (b) fast growth, (c) ability to meet the tastes and preferences of Kenyans, and
(d) abundant yield. Furthermore, double-cropping and direct-sowing technologies were
firmly established. Moreover, the project was promoting activities to disseminate new
rice production technologies to local communities. One example was to provide free
seeds of new rice varieties to farmers living in the project site, though this was limited
to certain areas.
iii. Employment creation effects
In order to cope with the sharp reduction in operating income due to flood damage,
TARDA reduced the number of personnel at the TDIPO from 233 in December 1997
to 68. Although the number of personnel directly hired by the executing agency to
work on the project was expected to return to former levels and to increase in the
future as recovery work progressed, it would take some time to return to the level
planned at appraisal. TARDA was also implementing organization-wide plans to
reduce employees from 405 (in 2000) to 208 in 2002. Thus the executing agency was
working hard to achieve greater organizational efficiency. Overall, new employment
opportunities were created in an area in which there were no such opportunities
before.
18 According to the 2006/07 Medium Term Expenditure Framework, the Tana Delta Rice Project was 30% complete, with financial support expected from donors to complete the rehabilitation of the project, with other estimates indicating this would require about KSh 2 billion (Coastal Express, Jan 14, 2003). Central government grants to TARDA amounted to KSh 132.6 million in 2006/07 for recurrent expenditures; and KSh 91 million for development expenditures.
19 According to JBIC officials, the rehabilitation of TDIP following its destruction by the 1997 El Nino rains was not done because of a shortage of funds. JBIC was in that period involved in the Tana Delta Road Construction Project as well as the Sondu Miriu Project. There were also concerns about Kenya’s external debt sustainability as well as its worsened governance and policy environment. Consequently, JBIC did not even give an envisaged second loan to Kenya for the Sondu Miriu Project during this period.
22 23
iv. Calculation of the economic and financial internal rates of return
At appraisal, the Economic Internal Rate of Return (EIRR) for the project based on
international rice prices and the Financial Internal Rate of Return (FIRR) based on
domestic rice prices were predicted at 8.4% and 11.9%, respectively. The assumptions
underlying the calculations were: (a) a project life of 30 years after project completion;
(b) the benefits were from income from the sale of polished rice, rice bran and crushed
rice; and (c) the costs were civil engineering and construction works (including
equipment procurement), consulting, operations and management.
Due to the flood damage, an accurate comparison of EIRR and FIRR was not
possible at the time of the JBIC (2001) evaluation.
3.1.5 The environmental impact
Forests in the Tana delta host the red colobus, a rare primate. The primate (Colobus
badius rufomitratus) is registered with the Washington Convention and the African
Convention on the Conservation of Nature and Natural Resources. The red colobus are
indigenous to Kenya and are legally protected because of their rarity and importance
as a national heritage. Measures were therefore adopted to prevent decreases in its
population and to protect the species during the project implementation. Measures
undertaken included minimizing deforestation for the development of farmland and
planting trees. At completion, trees had been planted in a 323-hectares area.
In addition, an office was established within the TDIPO to monitor changes in
the red colobus population in the project area and its vicinity. The environmental
monitoring office was established to focus on environmental conservation in the Tana
delta and its sustainable development. It was to monitor vegetation, wild animals and
birds, livestock breeding and grazing lands, agricultural production and population,
health and sanitation, water quality and fish, hydrological and climatic conditions,
and other environmental elements according to environmental monitoring indicators.
Based on the project’s monitoring records, there was no marked environmental
degradation during the period from 1991 to 1997, the project completion year. No
particular negative effects on rare animals had been reported by the time of the JBIC
(2001) review.
3.1.6 Sustainability
As described above, TARDA initiated emergency recovery work for destroyed
irrigation channels, banks and other major structures on the project site. In addition,
22 23
repairs were made to damaged construction equipment, agricultural machinery, rice-
polishing mills and other facilities to prepare for full-scale resumption of the project.
These self-help efforts for recovery deserve to be evaluated. The major prerequisite
for effective and sustainable development of the project, however, was to advance the
recovery work in earnest20.
3.2 Lessons from TDIP
As in the case of BISS, our analysis mainly focuses on the lessons learned from
TDIP for the Kenyan policy-makers, and to a lesser extent, the project, according to
JBIC, was completed successfully and its subsequent destruction by the 1997 El Nino
floods was an Act of God. Hence the most interesting “institutional spillovers” in this
case are probably the lessons drawn from the problems encountered by the project.
In a review article, Oosterbaan (1988) concludes that many new irrigation projects
have disappointing results. The disappointing results stem from overestimates of the
benefits and underestimates of the costs, losses, and damages, in the light of the ever-
decreasing availability of good quality land and water resources. The benefits are often
less than expected because many projects are initiated, implemented and operated
“top-down” without regard for farmers needs and motivations; the discrepancy
between national (urban) and rural development targets; and the scarcity of well
qualified and motivated managerial technical staff. The costs are often higher than
expected because not all direct costs are recognised or foreseen; while the prevention
of negative secondary effects requires unforeseen additional measures, which increase
the direct costs. The secondary (environmental) costs, losses, and the damages are
higher than expected because they are not foreseen or taken into account.
Many of these issues were discussed in the feasibility study of TDIP (Ecosystems
1983). The decision to undertake the project was for example made top-down, starting
from the national economy and government policies from which it was concluded that
rice production in Kenya had to be increased, that the Tana delta offered possibilities
for this national target, and that the project should be of the state-enterprise type
with only one proprietor of both land and the water under hierarchical management
and with contracted labour. While the top-down approach does not explain the
failure of TDIP, with this approach shared by many successful projects in Africa and
elsewhere, the top-down approach undermines the “ownership” of the project by the
20 According to one official, TARDA did manage to salvage 200 hectares from its own saving, to bring the total area in use to 380 hectares. There was however another mild flooding in 2004 and another one in 2007, which reduced the area under use to only 60 hectares. It requires at least 600 hectares to breakeven, with a crop every season (2 seasons per year).
24 25
local communities, and hence its sustainability in the context of a country where good
land is in short supply and the land question is a very sensitive political issue. These
TDIP communities are some of the poorest in the country.
Following extensive damage resulting from the 1997 El Nino rains, it has been
proposed that TDIP be rehabilitated21. This provides an opportunity to improve on the
design and implementation of the earlier collapsed project. We therefore draw on an
impact assessment of such a rehabilitation done by Luke et al. (2005).
According to the study by Luke et al. (2005), significant levels of poverty and
vulnerability characterize the TDIP-communities, with the greatest constraint
being lack of rainfall. The irrigated rice scheme – if targeted correctly - therefore
holds the potential to contribute significantly to livelihood improvements and, by
extension, sustained use of the environment. The proposed rehabilitation should
attempt to redress the TDIP communities’ livelihoods vulnerability and lack of
development options, which will also serve to reduce pressure on the surrounding
natural habitats. In order for sustainable outcomes to be achieved, appropriate design
and implementation must necessarily engage the communities as partners, and be
characterized by information sharing, consultation and collaboration. This requires a
through understanding of the stakeholders involved in the project.
3.2.1 The TDIP communities
According to the study, the Polder 1 TDIP is commonly associated with six villages,
therefore considered to be the legitimate stakeholders of the project. Land falling
within the traditionally-demarcated boundaries of three of these – Kulesa, Wema
and Hewani – were incorporated into the project, while the other three villages
– Bfumbwe, Sailoni and Baandi – border the project, and have traditionally used
common property resources within the project, and continue to do so – typically the
floodplain forests and available grazing areas.
The traditional Pokomo cultivators, with the exception of Baandi, which is inhabited
by traditional Orma pastoralists, inhabit all the six villages. The Orma residents of
Baandi distinguish themselves amongst pastoralist as being “permanent” within the
Tana River delta, rather than “nomadic”22.
21 See the statement by the Kenya Minster for Regional Development in the Daily Nation, November 1, 2006
22 The utilization of the waters of the Tana River has been in the middle of a conflict pitting the two communities against each other. The Pokomo claim the land along the river and the Orma claim the waters of the river.
24 25
Assessment by the Luke et al. study of key informants approximated that about 61%
of the population in the TDIP villages were poor, 30% middle-class and 9% rich23. In
an index of vulnerability that ranged from 1 to 4, they were vulnerable with respect to
housing (2.1), food supply (2.4), access to incomes (2.6); and access to education (2.5).
On average, households live in non-permanent houses, struggle to produce sufficient
food, are either reliant on the land for income, or on selling labour in the absence
of a viable land base; and either have enrolled some of their children in secondary
schools or not at all. From village focus groups, the principal constraint to livelihoods
and development was (i) poverty, caused by lack of cultivatable land; (ii) low-per-acre
productivity; (iii) lack of infrastructure to access markets; and (iv) insecurity of land
tenure and property.
Overall, the study concludes that subsistence farming results in a lack of community
capacity to accumulate needed capital, in order to precipitate investment into
strategies that break the poverty cycles. These underlying challenges are exacerbated
by insecurity of land tenure, banditry and loss of a primary resource to TDIP. If village
farmers were allowed to cultivate rice on their traditional land within the (improved)
TDIP and selling to TDIP (as has been suggested in the past, and indeed expected at
the project’s inception), the rehabilitation of the TDIP directly and indirectly has the
potential to contribute positively towards diminishing all four basic causes of local
poverty, as identified above by the communities.
3.2.2 Institutional linkages with TARDA
From the focal group perceptions of existing linkages between their villages and
both internal and external institutions, the dominating institutional relationship for
the six villages is the proximal and negative impact of TARDA (Luke et al. 2005). The
negative impacts of TARDA fall into three general categories. First is loss of resources
to the TDIP. The effects include loss of both use and ownership of prime agricultural
land; loss of prime dry-season grazing land (for Orma); and increased illness and
diseases, especially malaria due to TDIP rice paddy flooding
The second is non-delivery of promised benefits by TARDA, in return for Hewani
and Kulesa villages agreeing to transfer traditional ownership of prime village land
to the TDIP. The non-delivered benefits are principally (i) rice cultivation by the
communities on the portion of lands annexed by the project, accompanied by the
sale of rice to TARDA at a fair price; (ii) building of schools and clinics; (iii) building
23 Tana River is one of the districts with the highest levels of food poverty (70.6%) and overall poverty (71.8%) based on the 1994 Kenya Welfare Monitoring Survey.
26 27
of a bridge over Tana River connecting upper TDIP (Sailoni) with Garsen town, to
assist with transport and market access; and (iv) construction of an all weather road
connecting the villages to main Malindi-Lamu highway, to assist with transport and
market access.
The third is that TARDA neither consults with villages before taking action, nor
informs villages when/ after taking action. TARDA also does not pay casual workers in
a timely manner, or always in cash as agreed. TARDA only engages the villages when
problems are encountered on TARDA’s part and does not take the subsequent views
expressed by the community seriously24.
A benevolent attitude by TARDA towards communities would be expected, given
that TDIP represented a development-focused project in a poverty-stricken area.
The only aspect in which TARDA has been seen to deliver benefits during the more
than 15-year course of the project was the pre-1997 El Nino period, when the TDIP
provided significant casual employment to villages, paid in a timely manner. One
of the problems contributing to continued disaffection is the (apparent) lack of a
“written” contract or memorandum of understanding between TARDA and the
communities outlining the nature and details of the relationship surrounding TDIP.
The communities have not been aware of the existence of such a document.
Luke et al. (2005) conclude the TDIP holds significant potential to contribute to
resolving a significant number of the underlying conditions that maintain villages
in a cycle of poverty. Potential interventions include engagement of the communities
as development partners, through shared consultation, information, planning and
implementation; TARDA to consider delivering on the original expectations that
communities would be able to cultivate rice within those traditional lands that fall
within the TDIP, and sell to the rice to TARDA at fair prices; continue providing
employment opportunities, with payment executed at a timely manners; and consider
the feasibility of assisting communities with periodic water supplies for crops.
3.3 Some findings from the fieldwork on TDIP
The stakeholders talked to agree on the need for TARDA to enhance the ownership
24 Some respondents pointed out that this behaviour is not unique to TARDA in estate-type irrigation projects. In some cases, even though the staff of the executing agency do not consult and communicate closely with the local people, the projects have been successful. Therefore, it cannot be concluded that lack of consultation always brings out failure. Institutions should be evaluated in the context of the relationships between the government and the local people, and government and donor agencies.
26 27
of the project by the local communities. There was also a major problem of marketing.
Tomatoes, watermelons and other products are grown in the area, but they go to waste
due to poor marketing so that there were high post-harvest losses. There was thus
need to develop marketing linkages and to enhance the capacity of TARDA in this
area.
Some stakeholders suggested that instead of looking for funding to rehabilitate
TDIP, mechanisms should be found to mitigate against the impact of events such as
the 1997 El Nino floods. The country could get external resources, but a similar thing
could happen again, hence the need to look for other ways for intervention, including
project re-design. One suggested possibility was to look for a strategic investor to work
with TARDA, to alleviate management problems at the institution. They suggested it
was not the business of government to farm; this should be left to the private sector.
A private company such as Dominion (farming in Yala) was interested in TDIP. An
Investment Conference on Rice had also been suggested. It was also possible to form a
cooperative society to take over the project, including sub-division of the land among
members. They therefore argued that the institutional arrangement for the project
was not appropriate, with its focus on a commercial enterprise. TARDA did not have a
comparative advantage in this situation, and was unlikely to efficiently operate such a
commercial venture. Hence there was need to improve the institutional arrangements
for undertaking such projects
In terms of sustainability, stakeholders also recommended the building of dams
upstream (e.g. at Mutonga and Grand falls) to reduce the probability of flooding
downstream for both BISS and TDIP, given that the flow pattern of the river and
the precipitation seems to have changed due to climate change. Thus the whole river
basin management and strategy is a crucial issue for the Kenya government. Another
possible solution was moving from rice to sugar cane farming, as the latter is less
prone to floods, consistent with TARDA’s proposed sugar project in the Tana delta25 26.
In terms of policy reforms, stakeholders suggested that an irrigation policy paper
had been proposed with irrigation stranded across various ministries: Agriculture;
Regional Development; Water and Irrigation. TDIP for example is under TARDA in
25 According to the TARDA website, subsequent phases of the planning and development will investigate other commercial crops such as cotton, palm oil, bananas, and so on or a settlement programme under a small-holders model of commercial farming.
26 TARDA has proposed to convert 12,400 hectares of land in the Tana Delta to sugar cane growing and production, generating a large controversy (see The Kenya Wetlands Forum 2005). According to TARDA, the Mumias Sugar Company was being offered mainly land, while the output of the project is clear. To mitigate the top-down nature of the TDIP, active participation of all stakeholders is being facilitated, with sensitization workshops held across the country.
28 29
the Ministry of Regional Development; the Bura Scheme on the other hand is under
the NIB in the Ministry of Water and Irrigation. The policy would separate production
from marketing and the provision of infrastructure.
On the issue of spillovers, stakeholders were agreed it is difficult to get lessons
from TDIP as the project collapsed before it could generate any experiences. It would
be difficult to tell a priori whether the project could have been a failure or a success.
Moreover, TARDA is involved in many other projects, not just TDIP, so that the
performance of the institution should be assessed in a more holistic manner27.
It was suggested that TARDA staff were now very well trained and are adequate
for the implementation of the project if and when it is rehabilitated even though the
Kenya government has increasingly become unwilling to guarantee and to provide
financial security to state corporations, especially non-strategic ones like TARDA.
There has also been a discernible shift by donors from financing large projects to small
projects, although this cannot be attributed to TDIP alone.
27 Since its inception over 30 years ago, TARDA has planned and implemented many projects, including the Masinga and the Kiambere Reservoir and Irrigation Projects. Proposed projects by TARDA include rehabilitation of TDIP; promotion of eco-tourism in the Tana Delta as well as undertaking diverse environment conservation and water management activities.
28 29
4. THE NAIROBI WATER SUPPLY PROJECT (NWSP)
4.1 Summary of the JBIC (2002) evaluation of NWSP
4.1.1 Objectives of NWSP
The main objective of the project was to expand the water supply system in the
project area (Thika River), from 194,000m3 per day at the time of appraisal, to
492,000m3 per day as of 2002, in order to meet the growing demand for water in
Nairobi. The specific objectives of the whole project included improvement of water
supply and sewage treatment capacity, expansion of water supply to low-income
communities, and institutional development of the Water and Sewerage Department
(WSD) of the Nairobi City Council (NCC).
The NWSP was co-financed by the World Bank, the AfDB, the EIB and OECF/JBIC.
The three funding agencies were represented by, and their funding channeled through
the World Bank (Syagga 1999). It is therefore difficult to delineate the influence and
contribution of OECF/JBIC alone.
4.1.2 Project scope
The total scope of the project consisted of (1) construction of transmission pipelines;
(2) construction of a dam; (3) construction of a water intake system; (4) construction
of a water treatment system; (5) construction of a water distribution system; (6)
construction of a sewage water treatment system; (7) consultant services; and
(8) technical assistance (JBIC 2002). The OECF/JBIC ODA loan portion covered
construction of transmission pipelines (Item 1 above). The Government of Kenya and
the NCC (formally called Nairobi City Commission) were the borrowing and executing
agencies.
4.1.3 Results and evaluation
i. Relevance
In government policy documents, the importance of water as both a basic need and
an input in the economic and social development process is emphasized. In addition,
providing a sufficient quantity of good quality water is listed as an overall goal of the
national water development policy. In this light, it can be concluded that this project
is still relevant.
30 31
ii. Efficiency
The OECF/JBIC ODA loan portion was implemented with no substantive change
in scope, despite nine months’ delay in project completion mainly due to delays in the
tender evaluation process (JBIC 2002). Of the total loan amount of 5,342 million yen,
92.9% was disbursed. This decrease is largely attributable to the depreciation of the
Kenya shilling against the Japanese yen during the construction period. It can be
concluded that the Japanese ODA loan portion was efficiently implemented.
iii. Effectiveness
a. Physical effectiveness
With the completion of the project, water production capacity was expanded from
194,000 m3 per day to 455,000m3 per day, slightly lower than the target of 492,000 m3
per day (JBIC 2002). At the time of appraisal, it was expected that water production
through this project would meet the water demand until 2000. In spite of the increase
in water production, however, the total water sale was apparently almost unchanged
from 1985, before the project. This may be attributed to the high rate of unaccounted-
for-water (UFW). In fact, the rate of UFW was getting worse. According to the NCC,
(1) the number of unmetered estates and illegal connections was quite high; (2) even
if metered, the quality of meter readings was inadequate, and (3) water pipelines
that are not repaired leaked and burst frequently. It can be concluded that while this
project helped cope with growing water demand in Nairobi, UFW rates would require
to be improved in the future.
b. Interview survey
In order to assess the effectiveness and impact of the project from the stakeholders’
perspective, an interview survey of a sample of the population (size=100 people)
was conducted in high, medium and low-income areas in Nairobi (sample sizes of
30, 30, and 40, respectively, representing the population ratio). On average, 91% of
respondents were connected to the Nairobi water supply source, although a correlation
between income and the connection ratio was apparent (JBIC 2002). However a
significant proportion of those connected did not pay their water bills at all.
c. Financial effectiveness
According to JBIC (2002), 52% of water was unaccounted for as of 2000. This figure
represents a sharp increase from the estimate made at the time of appraisal in 1989,
30 31
and is much higher than the target figure of 20%. In order to re-calculate the FIRR,
UFW was assumed to be 50% after 1997. The re-calculated FIRR was 5%. The original
FIRR at the time of appraisal was 7%. This discrepancy is due to the high rate of UFW
(about 50%) compared with the original plan (20%).
4.1.4 Impacts
i. Decrease in water-borne diseases
Half of the sample population interviewed for the survey responded favorably when
questioned about the impact of the project on the incidence of water-borne diseases;
63% of the higher income group, 50% of the middle-income group, and 43% of the
lower income group replied positively, respectively (JBIC 2002).
ii. Environmental and social impacts
The Kenya government undertook a study on the environmental impact of the
project28. Although the Japanese portion of the project (i.e., transmission pipes) did not
involve any resettlement or environmental externalities, other co-financed portions,
especially the dam construction, involved an inundation of land and the resettlement
of 335 families. They received a compensation package for relocation.
4.1.5 Sustainability
i. Operation and maintenance
The NCC’s WSD, with a total staff of 506 as of 2001, was directly responsible
for the operation and maintenance of the project. The division responsible for
water transmission was then under the authority of the Deputy General Manager,
Operations and Maintenance (Water).
There was a severe lack of financial and managerial sustainability at the WSD.
With more than 50% UFW rate, the financial situation of the WSD deteriorated. The
extent of the financial distress was not totally clear, as no financial audit of the WSD
had been conducted in the previous several years. According to the investigation,
payment of WDS staff salaries had been delayed for months. This made it difficult to
offer much-needed services such as meter reading to the Nairobi population.
28 Nairobi City Council (1998), Third Nairobi Water Project: Environmental Appraisal Report.
32 33
In order to arrest the deteriorating situation, with the guidance of the World Bank,
the WSD was at the time studying the possibility of privatization in the form of
divestiture, concession, leasing or management contracting.
ii. Conditions of facilities and materials
All Japanese financed pipelines remained functional and operational, and WSD
inspectors were routinely inspecting them for cracks, leakage and any other technical
problems (JBIC 2002). However, according to the WSD, no major maintenance work
had been carried out since the completion of the project. Maintenance works such as
protection works on sections damaged by erosion and general maintenance of access
roads were needed. To keep the whole project sustainable, it was essential to conduct
this maintenance work.
4.1.6 Overall recommendations for enhancing the sustainability of NWSP
To keep the whole project sustainable, it was essential to conduct the maintenance
work mentioned above and examine countermeasures to reduce UFW29.
4.2 Situation analysis of the NWSP
4.2.1 Nairobi’s water supply situation
Since Nairobi became the administrative capital of Kenya, a number of water-supply
projects have been undertaken. Nairobi’s first water supply project was commissioned
in 1899 based on the Nairobi River in the Athi River catchment area (Wambua 2004).
By 1986, Kikuyu Springs, Ruiru Dam, Sasumua Dam and Chania Dam provided
Nairobi with a total of 192,000m3 per day. The supply was insufficient, as the city’s
population had increased tremendously, placing severe pressure on the water supply
and the sewerage system.
NWSP helped provide relief to the situation. According to Syagga (1999), when
completed, the project was expected to displace 500 households and flood 350 hectares
of land used for small-scale tea farming. In addition, the water was to be pumped to
29 As noted in the Japanese Ministry of Finance Evaluation of the project (FY 1996), even though the Japanese portion of the project was completed without problems, it was essential for the donors to cooperate with one another in monitoring the NCC fiscal management. The Evaluation noted that while the project did achieve the goal of alleviating water shortages and improving sanitary standards, the sustainability of the project would depend on whether water fees are charged and adequate revenues raised. There was a concern whether there would be enough effort on the Kenyan side to improve collection rates and reduce leakage.
32 33
Nairobi in 4-metre-diameter pipes, constructed on 24-metre-wide wayleaves acquired
through private parcels of land, over a distance of some 60 kilometers. Although
the project site was in a rural district, and hence the affected people were in a rural
settlement, the ultimate beneficiaries were to be the residents of Nairobi.
Among the positive spillovers from the project, it created direct employment at the
construction site and in small-scale businesses that served the construction workers
(Syagga 1999). Labour for future dam maintenance would also be required. The project
also created a number of facilities for the benefit of the surrounding community,
notably a primary school, a health clinic, an improved road network, and a power-
supply station. Above all, the project would supply enough water to sustain Nairobi’s
growing needs to 2005.
However, the project caused considerable amount of suffering for displaced
households (Syagga 1999, Ndung’u 2003). Extended families were split up,
compensation money was inequitably distributed within households, the total
compensation was insufficient, and families suffered a loss of earning capacity. The
dam altered the microclimate in the vicinity, and more mosquitoes were breeding and
causing sickness in the surrounding areas. Farmers also had to deal with soil erosion
arising from the unprotected high embankments.
Many studies have subsequently been done on the water situation in Nairobi.
According to the World Bank (2005), water supply in Nairobi has been plagued for
many years with inefficiency and complex management and logistical problems. This
has led to inadequate water services, with the poor suffering the most. The study for
example estimates that, of those that were served by the utility, 40 percent did not
receive a 24-hour supply; some 30 percent received water once in two days; while 10
percent received water only once a week. A household survey by Gulyani et al. (2005)
revealed that poor households in Nairobi spent 45 minutes on average collecting
water every day; the non-poor spent only 18 minutes while households with private
connections spent about 5 minutes. Nairobi utility’s water revenue collection efficiency
was below 30%, so that out of every 100 metered water users in Nairobi, less than 30
users paid their bills. The utility billed KSh 3 billion per year but collected only 30%
or KSh 900 million (Wambua 2004).
As noted above, UFW was over 50 percent of the total volume of treated water
produced. Much of the UFW resulted from physical leakage, and the rest was due to
water theft and failure to bill, essentially as a result of mismanagement. According to
the World Bank (2005) study, only about 187,000 or 42 percent of the total households
in Nairobi had legal water connections. Nearly all others, largely poor households,
34 35
obtained water from kiosks, water delivery services and illegal connections, with
the water purchased from vendors usually originally sourced from the network. The
new Nairobi Water and Sewerage Company (NWSC) formed in 2004 was tasked with
improving this situation.
In the household survey conducted by Gulyani et al. (2005), the utility coverage,
or the proportion of households with access to piped water supply, appears to be
relatively high in Nairobi compared to many other capital cities in Africa. Specifically,
the survey found that 71 percent of the households in Nairobi in the sample had access
to piped water supply, either through a private in-house connection or a yard tap. By
comparison, a study of water supply and independent providers in 10 African capital
cities (including Nairobi) estimates that in six of these cities, only 27–49 percent of the
households have access to piped supplies, with the rest of the households relying on
independent providers or traditional sources (Collignon and Vézina 2000).
Although a relatively high proportion of households in Nairobi have access to
piped supplies, water usage levels is low (about 42 litres consumption per day)
when compared with that of other countries as well as when compared with earlier
consumption and use levels in Kenya itself. Thomson et al. (2000) argue that there has
been a dramatic decline in domestic water use in urban Kenya: it has fallen from 105
lcd in 1967 to 45 lcd in 1997 (and to 40 lcd in 2000). The explanation for the low levels
of consumption and decline over time is that piped households have been forced to cut
back because of failing municipal supplies but also from a combination of price and
quantity factors.
The tariff structure in force in Nairobi is an increasing block tariff (see Table 4.1).
According to the World Bank (2001), the average tariff charged by Kenyan water
utilities, from all categories of customers combined, was about US$0.4/m3, sufficient
to cover capital plus operations and maintenance costs if the water utilities were run
efficiently.
Table 4.1: Water tariff rates in Nairobi
Block 1 Block 2 Block 3 Block 4
Consumption – m3 0-10 10-30 30-60 > 60
Tariff – KSh/ m3 12 18 28 35
Tariff – US$/ m3 0.16 0.24 0.37 0.47Source: World Bank (2005). 1 $ is equivalent to KSh 75.
The official water tariff provides little indication of what people were actually
paying. Despite low average water use, households pay remarkably high unit prices
34 35
for water. Gulyani et al. (2005) estimated the average cost to be KSh 260 per cubic
meter (US$3.50). The main reason behind the high unit cost of water reported by
households is that many of them, including those with private connections, buy
water from expensive sources such as kiosks and vendors. Households in the sampled
population were paying, on average, KSh 205/m3 (or US$2.7/m3) for water from kiosks,
and KSh 630/m3 (or US$8.4/m3) for water from vendors who deliver at home (including
tankers).
These prices for vended water are similar to those reported for Nairobi by Collignon
and Vézina (2000) in their study of African cities; their data also show that vended
water costs more in Nairobi than in most of the other capital cities included in their
10-country sample. Hence the average cost of water from kiosks is remarkably high,
given that utilities in Kenya usually supply water to kiosks at a “social” or bulk rate of
about KSh 11/m3 (US$0.15/m3).
Despite the Nairobi utility’s attempts to deliver a subsidy through the tariff, there
is evidence that the poor, who are more likely to rely on water sold by third parties,
pay more per unit of water. In an attempt to partially address the problem, the utility
established a flat rate of KSh 10 per cubic meter for bulk supply to water kiosks
serving informal settlements. However, this has not been effective in bringing down
costs to consumers as few kiosk operators are actually billed at this rate as they
often end up being charged the regular domestic tariff. These costs, as well as the
investment costs and overheads incurred by the kiosk operators, translate into very
high prices at kiosks.
In other words, the kiosk owners are charging 18 times the price that they pay for
the water. Even after taking into account that kiosk owners have to incur initial costs
for installation of kiosks, as well as some recurrent overhead costs (including illegal
payments), the difference between the price paid to the utility and the price charged
for water by kiosk owners is large. In fact, Collignon and Vézina (2000) calculated
the minimum gross profit margin of “standpipe operators” (kiosks) in Nairobi to be
80 percent and the maximum to be 90 percent, the highest profit margins in their
10-country sample. This suggests that much of the subsidy provided by the utility was
not accruing to the poor for whom it is intended; rather, these subsidies were negated
by the high costs of installing and running kiosks, or they accrued to kiosk owners,
those collecting illegal payments from them, or both.
It is estimated that 55% of the Nairobi’s population reside in informal settlements
that lack adequate network infrastructure. In order to access the services they need,
the less poor among them have taken their own initiative, extending pipe work for
36 37
several kilometers to a single dwelling, or combining efforts for mutual benefit.
However for the poorest households, the problem is more difficult to solve. It is
unlikely that they can afford the options currently open to them and it is likely that a
number of other constraints will stand in their way.
The first step in a strategy to improve access to private connections should be to
facilitate the extension of network infrastructure into those informal or unplanned
areas. A good example is Kibera, a settlement in Nairobi of up to 500,000 people. More
than 1,000 private connections have been installed (Lamba and Memon 2005). These
pipelines stretch up to 1 kilometer from the nearest utility main. Increasing security of
tenure is a key step that the government can take to avoid inefficiencies and improve
access to water supply to poor households.
Some of the measures that can be considered to increase the access of low-income
households to private connections include:
extending the piped water supply networks into informal and unplanned settlements;•
enabling low-income households to afford the upfront costs of a connection;•
removing administrative and legal barriers;•
setting the price of water at a level that is affordable to low-income households; and•
developing appropriate mechanisms for managing payment.•
The Nairobi water company is reported in January 2005 to have started
collaborating with small-scale water vendors in Kibera to improve water supply
service for low-income consumers. Maji Bora Kibera, an association of 500 small-scale
water vendors serving approximately 500,000 Kibera inhabitants, had entered into
a partnership with the NWSC. They agreed to form a task force that would address
concerns of UFW, revenue collection and rent-seeking behaviour.
4.2.2 Institutional reforms
A push for good governance in the last decade or so has also had an impact in the
water sector (World Bank 2005). A new water policy in Kenya came into effect in
1999, redefining the role of the government so as to focus on regulatory and enabling
functions rather than direct service provision. The government plans to emphasize
supporting private sector participation and community management of services,
rather than continue subsidizing inefficient utilities with public funds.
The Water Act of 2002 (enacted in March 2003) laid the legal framework for
implementing the policy and set up the institutions required (World Bank 2005). The
36 37
institutional framework for service delivery under the Water Act (2002) is supported
by all the donors active in Kenya’s water and sanitation sector. The influence of donors
– including the World Bank, Swedish International Development Agency (SIDA),
KfW, Finnish International Development Agency (FINIDA), the Netherlands, Belgian
Administration for Development, GTZ, Austria, AfDB, JICA, among others – has been
most profound throughout the reform process (Lamba and Memon 2005). Besides
providing financial support, the donors have played a key role in shaping the reform
agenda, refining policy objectives and the methods for pursuing them, and influencing
the pace of the reforms.
The World Bank in particular has played a key role in the global water sector
(Lamba and Memon 2005). It has crucially influenced the policies of the recipient
countries as well as those of the other multilateral and bilateral donors. It shapes
national and international water policy both via its linking the award of loans to strict
conditionalities and by its leading role in the formation of opinion in the water debate.
The central aspect of the World Bank’s water policy since the beginning of the 1990s
is the notion of water as an economic good. It is focused on a comprehensive reform of
the water infrastructure sector, which so far has been largely in public hands.
For a considerable period, the NCC’s WSD was regarded as ill equipped to deliver
quality service mainly due to corruption, with NCC ranked the fifth most corrupt
public sector organization in a recent survey by Transparency International (Lamba
and Memon 2005). The solution to this problem was the proposal that water service
functions of the NCC should be relocated elsewhere, preferably to an independent
entity. This was not possible until the enactment of the supporting legal framework in
the form of the 2002 Water Act.
The NWSC, recently weaned from the WSD of the City Council, is the principal
services provider to the city’s three million people. It was officially launched on August
19, 2004. It operates along the lines of a private enterprise, with an autonomous board
of directors. The company’s prime object is to provide water services within the City
of Nairobi pursuant to the provisions of the Water Act (2002). Pursuant to the said
Act, the company operates as an agent of the Nairobi Water Services Board. It is the
Board that holds the license for the provision of water within Nairobi. The benefits
of this license have been made available to NWSC by way of an agency agreement
that established a framework for delegation of the license. In consideration for the
benefits of the license, NWSC assumed a number of responsibilities, which include
its covenants to ensure adequate, and quality potable supply of water at affordable
tariffs and an obligation to improve water and sewerage infrastructure for the City of
Nairobi.
38 39
By May 2004, NWSC had taken over the functions of the Water and Sewerage
Services Department of NCC. It also inherited 2,200 staff and the operational
structures of the NCC Water Services Department. In August 2004, NWSC recruited
a team of 6 professionals whose overall responsibility was the reformation of water
service provision to improve its accessibility, availability and affordability. With the
said objectives, the professionals were mandated to run NWSC in an efficient and
profitable manner and to correct the shortcomings of the NCC Water and Sewerage
Services Department.
4.3 Some findings from the fieldwork on NWSP
4.3.1 Sustainability of the NWSP
Discussions with stakeholders pointed out that no major water project has been
implemented since 1994, with additional support by OECF/JBIC limited by a
shortage of resources. The NWSP (‘Chania 3’) was expected to be followed by another
project (‘Chania 4’) to bring more water into Ndaka-ini Dam. This however was not
implemented. The project was expected to cost $11 billion, with Phase I costing $5
billion. The donors were discouraged by poor governance in the country. The election
of a new government in December 2002 had not changed the situation30.
Officials of the NWSC say they have managed to reduce the UFW from about 52%
in 2001 to about 37-40% in 2007 by attending to physical water leaks, reducing illegal
connections, controlling carwash and improving the water metering-billing cycle. The
ultimate target is 20-25% (some countries have managed to reduce this to even less
than 10%). The challenge is that much of the water supply to informal settlements
in Nairobi is operated by illegal cartels that have proved difficult to dismantle. The
cartels also use cheap connection materials that have undermined the quality of the
water supplied to these areas. Kisumu had succeeded in getting rid of these cartels, so
there was hope that Nairobi could do the same.
The water company was currently collecting 60% of potential revenue, compared
to 30% when it started operations in 2004. The NWSC currently handled 230,000
accounts, out of a potential 400,000.
30 According to estimates by the Athi Water Services Board (AWSP) which manages water provision in Nairobi and surrounding areas, demand for water currently (2007) stands at 337,487 cubic metres, while only 248,000 cubic metres is reaching consumers. Demand is set to increase to 574,000 cubic metres by 2015, and over one million cubic metres by 2030.
38 39
4.3.2 Spillovers from the NWSP
According to the NWSC officials:
The company has trained people who had moved to other water projects. An •
example was given of the Managing Director of the Kisumu Water Supply Company,
who came from the NWSC.
The NWSC inherited 2,300 workers, of which only about 40 were university •
graduates. The number of workers had now reduced to 2,100, and the company
would like to reduce the number even further, to enable the hiring of more skilled
workers.
As noted above, the NWSP constructed a school and a health centre at the dam.•
There had been planting of trees around the dam (more than 40,000), by a Ndaka-•
ini conservation group (Ndaka-ini Environmental Association).
Ndaka-ini dam has become a recreation and tourism resort that is within the •
driving distance of Nairobi, with the annual Ndaka-ini Marathon held there.
Reforms in the sector were partially driven by the NWSP. In the 1980s, a large •
number of bilateral donors withdrew from the water sector complaining of poor
governance in the country, with NWSP being the only major water project during
that period. Much of the funding was going to NGOs. The World Bank then started
to put pressure for water reforms to be implemented if it was to increase funding
to the sector. These reforms included increasing the managerial autonomy of the
NCC’s WSD. The government then decided to look at reforms for entire water sector,
not just the NWSP.
The private sector has benefited from supplying the NWSC especially with •
chemicals, pipes, and so on.
There had also been transfer of technology from firms/ contractors that have worked •
with the NWSC transferring those skills to other water utilities. There are also
more that 160 companies (some unregistered) that provide bottled water, competing
with the water company. The formation of these bottled water companies was
driven by (a) the perception and propaganda that the NWSC water was not safe for
drinking (there were stories in the past that the NCC used chalk instead of chlorine
to treat the water); and (b) the portability of bottled water. Some of these companies
(5 out of the 16 large ones, including Coca Cola’s Dasani) used the company’s piped
water.
Another externality is the possibility of generating power from the Ndaka-ini Dam •
(about 2 megawatts). Discussions with Kenya Electricity Generating Company were
ongoing.
40 41
5. COMPARISON OF NWSP WITH THE NYERI TOWN WATER SUPPLY SYSTEM
The Nyeri Municipal Council took over the provision of water services from the
Central Government in 1982, and admittedly provided poor water and sewerage
services to this small central Kenya town until the water utility was commercialized
(Wambua 2004). The formation of WSD in the Council in 1995 did not make matters
any better because water revenues went to the Nyeri Municipal Council Treasury
and were often diverted to non-water areas. This meant that burst pipes could not be
repaired in time. It also meant that the water supply could not be expanded to match
population growth as the water revenues were not ploughed back to develop the
sub-sector. The desired result of service improvement and sustainability had failed
and adoption of commercialization as an alternative management approach became
inevitable. In June 1996, the Nyeri Municipal Council put a request to the GTZ to
support the privatization of Council’s WSD. GTZ came in handy and has been involved
in advising several other Kenyan local governments on the commercialization of their
water services.
According to a GTZ official, decision-makers in the 1970s and 1980s were not taking
into account operation and maintenance costs (the recurrent cost problem), which
is very crucial for sustainability of projects. By the early 1980s, most of the water
utilities were managed as departments in the local authorities. This made it quite
difficult to manage water utilities efficiently within such a structure. Repairing a
burst pipe for example could take a long time because of bureaucracy and regulations.
At that time, one could not for example spend more that KSh 5,000 without a council
resolution. A maintenance activity costing more than this amount would therefore be
unduly delayed.
In the mid-1990s, GTZ started promoting commercialization of water services with
pilot schemes in Nyeri, Eldoret and Kericho. Policymakers were initially skeptical
about the policy, saying this would not work. There were for example delays in
commercialization in Eldoret and Kericho as the local authorities attempted to
reverse the council decisions to commercialize. Due to these delays, Nyeri Water and
Sewerage Company (NYEWASCO) absorbed the bulk of the support provided by GTZ
for the three pilot commercial companies. The objective of commercialization was to
promote the sustainability of the water utilities, by shielding them from politics in the
local authorities. It has been a policy of GTZ since the 1980s to build local capacity
first before investment in water projects. The GTZ principles of capacity building for
sustained development have consisted of (i) transfer of knowledge by formal training
40 41
and training on the job (learning by doing); (ii) promotion of behaviour change; (iii)
introduction of incentive schemes; and (iv) improvement of work processes and
management tools (WSRB 2007).
Commercialization entailed making the water department autonomous, permitting
recruitment of management and other staff through a competitive process to work
alongside a board of directors. This was crucial as the main stakeholders in the
created water company (the management and the Board) do not directly contribute
their own resources to the company. One then requires to recruit individuals who are
likely to pursue the “public interest”. If the right people are not recruited, things could
go very wrong.
Commercialization of course should take into account equity issues by using block
tariffs adjusted for inflation, with cross-subsidization across water users such as
through providing cheaper water through water kiosks in informal settlements.
Unlike Nairobi, Nyeri town does not have cartels that control the supply in informal
settlements. One problem with cartels in Nairobi is that they can influence water
supply in their areas and hence the tariffs that they charge for the water.
The NYEWASCO was incorporated as an independent company owned by the Nyeri
Municipal Council in 1997, after the council realized it could not offer efficient water
services to the residents of Nyeri town (Wambua 2004). Through GTZ’s technical
inputs and the willingness of Nyeri Municipal Council to free its grip on the water
department, NYEWASCO started its operations in 1998.
The commercialization of the Nyeri Town water supply worked very well, after
the new water company got financial and managerial autonomy. In its board sat a
Chairman and a corporate management team comprising of a Managing Director,
Commercial Manager and Technical Manager. Legally, NYEWASCO is an agent of
the Nyeri Municipal Council. Under the agency agreement signed in 1999, to regulate
the relations between the Municipal Council and NYEWASCO, water infrastructure
valued at KSh 509.7 million was passed on to NYEWASCO. NYEWASCO also
absorbed all staff from the Municipal Council’s water department (Wambua 2004).
NYEWASCO is regarded as a success story on commercialization of water services
in Kenya and has become a “learning centre” for many local water utilities, including
the NWSC. The Nairobi City councilors went to Nyeri to study the modus operandi
of the water company, before the reforms were undertaken in Nairobi. The Nyeri
project therefore gave courage to the NCC to proceed with the water reforms. This was
however a two-way process, with Nyeri councilors for example also visiting the NWSC
42 43
to study its sewerage management. GTZ has also learned from the success of Nyeri.
Taking water as a business, it has promoted syndication of water supply in several
small towns in Western Kenya, whereby water to these towns are supplied by a single
company to exploit economies of scale.
The success of the Nyeri water company was due to various factors. According
to some stakeholders, one reason for the ease of the commercialization is that the
Nyeri Council kept good books of their accounts. The local authority could therefore
tell that it had been subsidizing the water department, and hence making losses.
Consequently, it was easy to persuade the authority to let go of the water department
in order to stop this subsidy. This happened even before the countrywide water sector
reforms were undertaken. Another factor is the quality of staff, management and
leaders involved in the water utility. Unlike the other pilots, these people were more
educated, with some councilors university graduates. The good performance is also
attributed to the fact that it is easier to manage the water supply of a small town
(50-60,000 people). Unlike NWSC, NYEWASCO was able to fund expansion of water
supplies even without donor support, increasing its water capacity from 6000 m3 to
9000m3 per day (see below).
NYEWASCO is run on strict corporate lines with water revenues being ploughed
back into improving water and sewerage provision. According to Wambua (2004):
Its Kamakwa water treatment works production capacity had increased by 50% •
from 6,000m3 / day to 9,000m3 per day.
Complaints of water turbidity have been eliminated.•
Reported water pipe bursts are attended to promptly 24 hours a day.•
It was continuously employing new workers to improve on service delivery.•
The number of new registered connections increased from 6,586 in 1999 to 8,318 in •
March 2003.
Over the same period, metered connections jumped from 6,721 to 9,271.•
It had established retail water kiosks for the poor sections of the municipality while •
slum residents paid KSh 12 for a cubic meter of water. On the other hand, rich
neighbourhoods and corporations e.g. Mt. Kenya Bottlers (a Coca-Cola subsidiary)
and tourist lodges using more than 100m3 of water daily paid KSh 75 per cubic
meter or 50 per cent above the domestic tariffs.
It had created employment without taxing the poor by selling water to kiosk •
operators who sell it to poor residents at affordable rates.
It had ensured lower rates for public institutions e.g. local schools and medical •
facilities.
It had pursued needs-driven commercial policies, for instance, interrupting supply •
42 43
to the local golf club in times of scarcity.
It changed its main source of water from Nairobi River that was drying because of •
farming activities upstream to Itooni River, hence saving the river that flows all the
way to the Kenya capital.
It had cut down water losses from 55% to 36-40%, one of the lowest UFW rates the •
country.
Because of these practices, NYEWASCO was boasting of a water surplus of 1000m• 3
per day and was negotiating for a loan from KfW for expansion31.
It was collecting over KSh 8 million per month and serving 50- 60,000 people.•
91.6% of its clients expressed satisfaction with its front desk attendance. More •
significantly, 81.1% were satisfied with the quality of water supplied. While 41.6%
were not happy with its prices, 76.8% got their monthly bills promptly.
In 1990, the Municipality could produce only 2,171,309 m• 3 of water, selling
1,388,687m3 meaning it was losing over 700,000m3 of water while earning KSh
8,864,109. In 2003, NYEWASCO produced 908,469m3 of water and sold 516,442m3
for KSh 24,692,305, indicating improved efficiency.
According to WSRB (2007), NYEWASCO has achieved all the major targets set in
1998. It doubled the annual revenue, expected to reach KSh 120 million in 2007. The
revenue collection efficiency is about 98% on average, with 10-25% of the revenues re-
invested to improve service provision, especially to the poor, through water kiosks in
densely populated low-income areas and informal settlements. Other achievements
include (WSRB 2007):
The use of alum powder instead of lumps reduced the alum usage by over 50%, •
leading to cost savings on chemicals.
The recirculation of wash water resulted in energy savings of more than 30%.•
UFW dropped from 46% in 1999 to 30% in 2006.•
Staff per thousand connections down to 7 in 2006 from 24 in 1998.•
Cost coverage increasing/ operation and maintenance cost coverage achieved.•
Water production capacity has increased to 27,000m• 3 per day, increasing service
hours from an average of 10 hours to 24 hours.
In a 2002 survey, 76-81% of the consumers expressed satisfaction with the service •
delivery; quality of the supplied water; receipt of (correct) bills; and consumer
31 In 2003, this German agency agreed to fund the US$10.5 million rehabilitation of the Nyeri town water system (Water & Water International, May 2003). The objective of the project was to expand water treatment plants and water intake, and rehabilitation of the sewerage treatment works. The project would also develop sewer infrastructure to homes and businesses. The project would increase water intake from 8,000m3 per day to 26,000m3 per day. In 2006, the interest rate on the loan was reduced from 6.5% to 2.5% to enable the financial sustainability of the project after completion. The grace period was also increased to 10 years.
44 45
service. However, only 53% expressed satisfaction with sewerage services, indicating
that more requires to be done in this area.
44 45
6. FINDINGS AND CONCLUSIONS
In this study, we selected two main projects: the BISS and the NWSP. These are
compared to the TDIP; and the Nyeri Town Water Supply System, respectively. The
study focused on the following issues:
Assessment of the individual project’s sustainability. •
Identification of important spillovers (or lack of) in terms of human resources •
development; capacity building; and institutional and policy reforms. The spillovers
may also be in the form of new technology spread to other projects; new forms of
revenue generation; stimulation of local private sector (e.g. construction); impacts of
donor policy and practice; impacts on government policy and regulation, and so on.
Factors explaining such institutional impacts.•
Assessment of the influence of donor policy and practice on such impacts (possibly •
related to policy development, ownership and capacity development).
Comparison of BISS with TDIP. What accounted for the poor record of irrigation •
projects in the Kenyan coastal region? Both settlement and estate irrigation schemes
on the Tana River are quite vulnerable to flooding. The interviewed stakeholders
also cited lack of infrastructure in the affected areas as a major problem.
Comparison of the NWSP with the Nyeri Town Water Supply System, with the •
latter often taken to be a success case of water services commercialization. What
lessons did the NWSP learn?
In these section, we pay particular attention to the two main hypotheses regarding
sustainability of aid-financed infrastructural services, namely (i) institutional effects
(spillovers) are important; and (ii) that the nature of the aid relationship matters,
which reflect the analytical framework as captured especially by the first four bullets
above.
6.1 Sustainability of the infrastructural services
6.1.1 Irrigation
The main objective of BISS was to alleviate landlessness in the overpopulated
agricultural areas of the country; provide employment; and increase the output of
cotton for export. The project failed to meet its objectives (as described in the paper).
BISS proved to be costly. The project’s settlement component has been described
as the most expensive scheme in the world, with families settled at a cost of about
$55,000 per household. The project suffered from conceptual and economic problems
46 47
including settlement issues; poor incentives for production; inability to exploit
economies of scale; poor crop yields; housing problems; inappropriate technology; and
institutional and management problems. A whole range of factors therefore explain
the project’s non-sustainability, As documented in the paper, these include poor
planning, design and administration of the project by the World Bank, the government
and NIB. The institutional framework was not geared to foster a pioneer spirit
amongst the settlers. The project faced and continued to face technical problems that
escalated costs, depressed returns and made operations difficult. Under pressure from
the World Bank, the government attempted to dismantle some of the deficiencies,
including making the scheme more autonomous and abolishing the Cotton Board
(Howells 1985).
According to Howells (1985), many of the problems that led BISS to under-
perform are unique to the project and it is perhaps an overreaction to dismiss
settlement schemes out of hand. However, the feasibility of moving people en masse
into inhospitable areas and expecting them to settle permanently and perform
economically remains in doubt. While the “estate” type developments are one method
of tackling the joint problems of unemployment and agricultural output, they do little
to satisfy the Kenyans’ hunger for land. Shortage fertile land is a big issue for Kenya’s
development, with about eighty seven percent of the country’s landmass arid or semi-
arid, not suitable for arable farming.
The main objective of TDIP, on the other hand, was to increase rice production to
enhance self-sufficiency in Kenya. The project had four major innovations, which were
pilots for the agricultural sector in Kenya: (i) the introduction of a large-scale estate
management system; (ii) the utilization of large-scale mechanization technology in the
project; (iii) the development of high-yield rice varieties suitable for the coastal region;
and (iv) double-cropping of rice, another first in Kenya. TARDA was the executing
agency.
The objective of the project was also not achieved. Immediately after the project was
completed, the site sustained enormous damage from floods caused by El Nino in 1997.
The site is still under rehabilitation efforts, with only 30% complete (in 2007).
What explains the non-sustainability of the project? The project, according to JBIC,
was completed successfully and its subsequent destruction by the 1997 El Nino floods
was an Act of God. TDIP was not fully rehabilitated due to an alleged lack of resources
both by donors and the government.
46 47
6.1.2 Urban water supply
The main objective of the NWSP was to expand the water supply system in the
project area from 194,000m3 per day at the time of appraisal, to 492,000m3 per day as
of 2002, in order to meet the growing demand for water in Nairobi. The total scope of
the project consisted of seven components. The Japanese ODA loan portion covered
one of these: the construction of transmission pipelines. The government and the NCC
were the borrowing and executing agencies.
As warned in the Japanese Ministry of Finance Evaluation of the project (Fiscal
Year 1996), even though the Japanese portion of the project was completed without
problems, it was essential for the donors to cooperate with one another in monitoring
the NCC’s fiscal management. The Evaluation noted that while the project did
achieve the goal of alleviating water shortages and improving sanitary standards,
the sustainability of the project would depend on whether water fees are charged and
adequate revenues raised. There was a concern whether there would be enough effort
on the Kenyan side to improve collection rates and reduce UFW.
A push for good governance in the last decade or so has also had an impact in the
water sector including NWSP. A new water policy in Kenya came into effect in 1999,
redefining the role of the government as to focus on regulatory and enabling functions
rather than direct service provision. The institutional framework for service delivery
under the Water Act (2002) is supported by all the donors active in Kenya’s water
and sanitation sector. The influence of donors has been most profound throughout
the reforms process. The World Bank in particular has played a key role in the global
water sector (Lamba and Memon 2005).
The NWSC, weaned from the WSD of the City Council, is the principal services
provider to the city’s more than three million people. It was officially launched
on August 19, 2004. It operates along the lines of a private enterprise, with an
autonomous board of directors. The company operates as an agent of the Nairobi
Water Services Board that holds the license for the provision of water within the City
of Nairobi.
Discussions with stakeholders pointed out that no major water project has been
implemented since 1994, with additional investments limited by a shortage of
resources, undermining the sustainability of the NWSP. The NWSP (‘Chania 3’) was
expected to be followed by another project (‘Chania 4’) to bring more water into Ndaka-
ini Dam. This however was not implemented. The donors were discouraged by poor
governance in the country during the regime of President Moi.
48 49
Officials of the NWSC say they have enhanced the sustainability of the project by
managing to reduce the UFW from about 52% in 2001 to about 37-40% in 2007 by
attending to physical water leaks, reducing illegal connections, controlling carwash
and improving the water metering-billing cycle. The ultimate target is 20-25%. The
main challenge is that much of the water supply to informal settlements in Nairobi is
operated by illegal cartels that have proved difficult to dismantle. The cartels also use
cheap connection materials that have undermined the quality of the water supplied
to these areas. The water company was currently collecting 60% of potential revenue,
compared to 30% when it started operations in 2004. The NWSC currently handled
230,000 accounts, out of a potential 400,000.
The Nyeri Municipal Council took over the provision of water services from the
Central Government in 1982 (Wambua 2004). The formation of WSD in the Council in
1995 did not improve water service delivery because water revenues went to the Nyeri
Municipal Council Treasury and were often diverted to non-water areas. In June 1996,
the Nyeri Municipal Council put a request to the GTZ to support the privatization of
Council’s WSD. GTZ has been involved in advising several Kenyan local governments
on the commercialization of their water services.
The NYEWASCO was incorporated as an independent company owned by the
Nyeri Municipal Council in 1997. Through GTZ’s technical inputs and the willingness
of Nyeri Municipal Council to free its grip on the water department, NYEWASCO
started its operations in 1998. The commercialization of the Nyeri Town water supply
worked very well and stakeholders are agreed that NYEWASCO is a success story on
commercialization of water services in Kenya.
The success of NYEWASCO was due to several factors. According to some
stakeholders, one reason for the ease of the commercialization is that the Nyeri
Council kept good books of their accounts. The local authority could therefore
tell that it had been subsidizing the water department, and hence making losses.
Consequently, it was easy to persuade the authority to let go of the water department
in order to stop this subsidy. This happened even before the countrywide water sector
reforms in early 2000s. Another factor is the quality of staff, management and leaders.
Unlike in the other pilots, these people were more educated, with some councilors
university graduates. The good performance is also due to the fact that it is easier to
manage the water supply of a small town, with NYEWASCO serving only 50-60,000
people, compared to the NWSC’s over 3 million people.
Unlike NWSC, NYEWASCO has been able to fund expansion of water supplies from
its own resources; increasing its water capacity from 6000 m3 to 9000m3 per day, even
48 49
though it has received financial support from a German development agency, KfW, to
expand its water supply capacity.
6.2 Institutional spillovers and explanatory factors
6.2.1 Irrigation
The most interesting “institutional spillover” in both the BISS and TDIP cases is
probably the lessons drawn from the problems encountered by the projects. According
to Howells (1985), BISS provides useful insights into the practical problems of
settlement and the lessons learned could be instrumental in determining a framework
for future development. In the case of TDIP, respondents are agreed that it is difficult
to determine the project’s spillovers as it collapsed before it could generate any
experiences and it would be difficult to tell a priori whether the project would have
been a failure or a success.
From early on, the failure of BISS to meet expectations caused the Kenya
Government to shy away from similar schemes (Howells 1985). It was from this lesson
that TDIP proceeded on the basis of commercial estates with a few outgrowers. There
were also plans to run the semi-completed part of the BISS as an estate in an attempt
to reduce the operating subsidy. However, such an estate-type irrigation project is
unlikely to have the same impact on unemployment and landlessness as settlement,
although it is more likely to produce an economic return. It would also do little to
satisfy the Kenyans’ hunger for land, given the shortage of good land in the country.
As discussed in the paper, there are recent efforts to revive the scheme and policy-
makers seem to have learned some lessons from past mistakes. According to an NIB
official, the key problem is that BISS was scaled back, leading to limited exploitation
of scale economies, hence increasing the operation and maintenance costs of the
project. NIB proposes to enhance the future sustainability of the project by asking the
beneficiaries to pay for the operations and maintenance costs of the scheme. Farmers
would be advised on what crops to grow, based on the crops’ relative returns, with NIB
facilitating access to markets. Commercial farmers would also be encouraged to lease
land as part of the efforts to make the scheme self-sustaining. The enhanced use of the
gravity system would also lower the operation and maintenance costs.
BISS was managed by NIB before 1984 and since 2005. According to its officials,
NIB has accumulated a lot of experience in managing irrigation schemes and BISS
is not different from the other schemes that NIB manages such as Ahero, Bunyala
and Hola, which were all undergoing rehabilitation. According to these officials, the
50 51
performance of the schemes declined with the general decline of the agricultural sector
and the economy in the 1980s and 90s due to poor governance in the country.
Stakeholders indicated that an irrigation policy paper had been proposed with
irrigation stranded across various ministries: Agriculture; Regional Development;
Water and Irrigation. The Bura Scheme for example is under the NIB in the Ministry
of Water and Irrigation; while TDIP is under TARDA in the Ministry of Regional
Development. The policy would separate production from marketing and the provision
of infrastructure. A cabinet paper had already been prepared on this. If implemented,
it would bring about institutional harmony in the sector.
As with BISS, there are plans to rehabilitate the TDIP. According to Luke et al.
(2005), the rehabilitation efforts of the project should attempt to be more bottom-up by
redressing the TDIP communities’ livelihoods vulnerability and lack of development
options. This should include addressing the negative effects of TARDA on these
communities that fall into three general categories. First is loss of both use and
ownership of prime agricultural land; loss of prime dry-season grazing (for Orma); and
increased illness and diseases, especially malaria due to TDIP rice paddy flooding.
The second is non-delivery of promised benefits from TARDA, in return for Hewani
and Kulesa villages agreeing to transfer traditional ownership of prime village land
to the TDIP. This is principally rice cultivation by the communities on the portion of
lands annexed by the project, accompanied by the sale of rice to TARDA at a fair price;
building of schools and clinics; and building of a bridge over Tana River connecting
upper TDIP (Sailoni) with Garsen town, to assist with transport and market access;
and the construction of an all weather road connecting villages to main Malindi-Lamu
highway, to assist with transport and market access. The third is that TARDA neither
consults with villages before taking action, nor informs villages when/ after taking
action.
Some stakeholders suggested that instead of looking for funding to rehabilitate
TDIP, mechanisms should be found to mitigate against the impact of events such as
the 1997 El Nino floods. The country could get external resources, but a similar thing
could happen again, hence the need to look for other ways for intervention, including
project re-design. One suggested possibility was to look for a strategic investor to work
with TARDA, to alleviate management problems at the institution. They suggested it
was not the business of government to farm; this should be left to the private sector.
A private company such as Dominion (farming in Yala) was interested in TDIP. An
Investment Conference on Rice had also been suggested. It was also possible to form a
cooperative society to take over the project, including sub-division of the land among
members. They therefore argued that the institutional arrangement for the project
50 51
was not appropriate, with its focus on a commercial enterprise. TARDA did not have a
comparative advantage in this situation, and was unlikely to efficiently operate such a
commercial venture. Hence there was need to improve the institutional arrangements
for undertaking such projects
Stakeholders also recommended the building of dams upstream (e.g. at Mutonga
and Grand falls) to reduce the probability of flooding downstream for both BISS and
TDIP, given that the flow pattern of the river and the precipitation seems to have
changed due to climate change. Another possible solution was moving from rice to
sugar cane farming, as the latter is less prone to floods, consistent with TARDA’s
proposed sugar project in the Tana delta.
6.2.2 Urban water supply
The paper has identified several institutional and other spillovers from the NWSP
of differing levels of importance. According to the NWSC officials:
Reforms in the water sector were at least partially driven by the NWSP. In •
the 1980s, a large number of bilateral donors withdrew from the water sector
complaining of poor governance in the country. Much of the funding was going
to NGOs. The World Bank then started to put pressure for water reforms to be
implemented if it was to increase funding to the sector. These reforms included
increasing the managerial autonomy of the NCC’s WSD. The government then
decided to look at reforms for the entire water sector, not just for Nairobi.
The company has trained people who had moved to other water projects.•
The NWSP constructed a school and a health centre at the dam and provided •
employment to the local community.
There has been planting of trees around the dam (more than 40,000) by a Ndaka-ini •
conservation group.
Ndaka-ini dam has become a recreation and tourism resort that is within the •
driving distance of Nairobi, with the annual Ndaka-ini Marathon held there.
The private sector has benefited from supplying the NWSC especially with •
chemicals, pipes, and so on.
There has also been transfer of technology with firms/ contractors that have worked •
with the Nairobi water company transferring those skills to other water utilities.
There are also more that 160 companies (some unregistered) that provide bottled
water, competing with the water company. Some of these companies (5 out of the 16
large ones, including Coca Cola’s Dasani) used the company’s piped water.
There is a possibility of generating power from the Ndaka-ini Dam (about 2 •
megawatts). Discussions with Kenya Electricity Generating Company were ongoing
52 53
(in 2008).
The main institutional spillover effect from the commercialization of the Nyeri
Water and Supply System (and the creation of NYEWASCO) is that it has become
a “learning centre” for many local water utilities, including the NWSC. The Nyeri-
GTZ water commercialization approach has been used as a model for the country
as a whole. The Nyeri case illustrates how successful and national institutions are
created to include greater private sector participation and to ensure their financial
involvement in urban service provision (RTI International, 2005)
The Nairobi City councilors for example went to Nyeri to study the modus operandi
of the water company, before the reforms were undertaken in Nairobi. The Nyeri
project therefore gave courage to the NCC to proceed with the water reforms. This was
however a two-way process, with Nyeri councilors for example also visiting the NWSC
to study its sewerage management. GTZ has also learned from the success of Nyeri.
Taking water as a business, it has promoted syndication of water supply in several
small towns in Western Kenya, whereby water to these towns is supplied by a single
company to exploit economies of scale.
6.3 The role of the aid relationship in generating sustainability and spillovers
6.3.1 Irrigation
BISS’s multilateral lenders were the World Bank and the EDF. Bilateral finance
in the form of grants and soft loans was provided by Britain, Finland, Holland and
Japan. BISS was therefore financed by a consortium of donors. The World Bank was
however the main actor, with the other donors taking little part in the monitoring of
the project. There is consensus that this is one example of the World Bank’s failed
projects. According to Horta (1994), the “World Bank’s 1990 Project Performance
Audit Report…indicates that project managers should have been aware of problems
and halted the project early on. Technical studies on the lack of suitable soils for
irrigation in the area existed but were not taken seriously. Project managers, rushing
to get the ill-prepared project approved by the Bank’s Board of Directors, downplayed
the risks and vastly underestimated costs”. In an analysis of 314 irrigation projects
implemented from 1967 to 2003 in 50 countries in six regions, Inocencio et al. (2005)
lists Bura among the relatively costly projects largely because of the reduction in
the actual irrigated area versus the planned area. At the World Bank, “aggressive
agricultural and irrigation expansion and establishing and attempting to meet lending
targets resulted in downgrading of technical and economic screening of projects”.
52 53
TDIP on the other hand was largely financed by OECF/JBIC,with TARDA, a
parastatal, the executing agency. TDIP utilized the “estate system“, under which all
agricultural development activities were to be managed and administered together
under one agency. One of the postulated advantages of the estate system is that it
enables efficient and profitable agricultural operations with a small investment when
compared to a settlement scheme. While this top-down approach in the design and
implementation of the project, of course, did not cause its collapse, some commentators
(e.g. Luke et al. 2005) call for a more bottoms-up approach in a future rehabilitation of
the project if it is to significantly improve the welfare of the stakeholder communities
in the Tana delta.
6.3.2 Urban water supply
NWSP was financed by a consortium of donor including OECF/JBIC, the World
Bank, the AfDB and EIB. The three funding agencies were represented by, and their
funding channeled through the World Bank (Syagga 1999). It is therefore difficult
to delineate the influence and contribution of OECF/JBIC alone, even though it
maintained its financial autonomy, by financing one component of the project. The
World Bank was a major influence on the water sector reforms adopted in the country
that saw the commercialization of water utilities, enhancing their sustainability.
The Nyeri Town Water Supply System has mainly been supported and funded
by German development agencies. According to an official, GTZ started promoting
commercialization of water services in mid-1990s with pilot schemes in Nyeri, Eldoret
and Kericho. The objective of commercialization was to promote the sustainability of
the water utilities, by shielding them from politics in the local authorities. It has been
a policy of GTZ since the 1980s to build local capacity first before seeking investment
in water projects. Investment support for example came much later in 2003, when the
German agency, KfW, agreed to fund the US$10.5 million rehabilitation of the Nyeri
town water system (Water & Water International, May 2003). The GTZ principles of
capacity building for sustained development have consisted of (i) transfer of knowledge
by formal training and training on the job (learning by doing); (ii) promotion of
behaviour change; (iii) introduction of incentive schemes; and (iv) improvement of
work processes and management tools (WSRB 2007). This approach seems to have
worked very well in Nyeri.
54 55
REFERENCES
Adams W. M. (1990), “How Beautiful is Small? Scale, Control and Success in Kenyan Irrigation”. World Development, Vol 18, No. 10, October.
Arakawa, Hiroto and Jin Wakabayashi (2006), “Budget Support and Aid Effectiveness: Experience in East Asia”. JBIC Review, No. 14.
Barro, Robert J. (1991), “Economic Growth in a Cross Section of Countries”. The Quarterly Journal of Economics, Vol. 106, No. 2, pp. 407-443.
Burnside, Craig and David Dollar (1997), “Aid, Policies, and Growth”. World Bank Policy Research Paper.
Burnside, Craig and David Dollar (2000), “Aid, Policies, and Growth”. The American Economic Review, Vol.90, No.4, pp 847-868.
Collignon, Bernard and M. Vézina (2000), “Independent Water and Sanitation Providers in African Cities: Full Report of a Ten-Country Study”. World Bank, Water and Sanitation Program.
Dollar, David and Victoria Levin (2005), “Sowing and Reaping: Institutional Quality and Project Outcomes in Developing Countries”. World Bank Policy Research Working Paper 3524.
Easterly, William (2001), The Elusive Quest for Economic Growth. MIT Press, Cambridge.
Easterly, William and Ross Levine (1997), “Africa’s Growth Tragedy: Policies and Ethnic Divisions”. The Quarterly Journal of Economics, Vol. 112, No. 4, pp. 1203-1250.
Ecosystems Ltd (1983), “Tana Delta Ecological Impact Study”. Nairobi, Kenya.
Gulyani S., D. Talukdar and R.K. Kariuki (2005). “Water for the Urban Poor: Water Markets, Household Demand and Service Preferences”. World Bank Water Supply and Sanitation Sector Board Discussion Paper No. 5.
Horta K. (1994), “Troubled Waters: World Bank Disaster along Kenya’s Tana River”. Multinational Monitor, July/ August.
Howells, Keith (1985), “Settlement Projects as a Solution to Landlessness and Urban Drift in Kenya”. http:/ dspace. lib. cranfield. ac. uk: 8080/ bitsream/ 1826/ 535/ 2/ SWP. 126. pdf.
Inocencio A., M. Kikuchi, D. Merry, M. Tonosaki, A. Maruyama, I. de Jong, H. Sally and F. Penning de Vries (2005), “Lessons from Irrigation Investment Experiences: Cost-reduction and Performance-enhancing Options for sub-Saharan Africa”. Paper submitted by the International Water Management Institute to the African Development Bank.
JBIC (2001), Tana River Delta Irrigation Project: An Evaluation.
54 55
JBIC (2002), Nairobi Water Supply Project: An Evaluation.
Kremer, Michael (1993), “Population Growth and Technological Change: One Million B.C. to 1990”. Quarterly Journal of Economics, Vol.108, No.3, pp.681-716.
Lamba, Davinder and Ali Memon (2005), “Water Sector Reforms in Urban Africa: A Road to Mercantilization?”. Paper Presented at a UCLA Conference on “Global Cities: Water, Infrastructure and Environment”, July.
Luke, Quentin, Richard Hartfield, Pamela Cunneyworth (2005), “Rehabilitation of the Tana Delta Irrigation Project Kenya: An Environmental Assessment”. Report Prepared for Critical Ecosystem Partnership Fund, July.
Mwega, Francis M. (2003), “Financing International Cooperation in a Globalizing World: A Case Study of Kenya”. Paper prepared for the UNDP Project on Public Finance in a Globalizing World: Innovations in Theory and Practice.
Mwega Francis. M (2004), “Foreign Aid, Monetary Policy and Economic Growth in Kenya”. Paper prepared for Kenya Institute for Public Policy Research and Analysis (KIPPRA).
Ndun’gu A.G. (2003), “Ndaka-ini Dam and Drainage Basin: Impact Within, Without and Beyond”. Mimeo.
Njeru, James (2004), “The Impact of Foreign Aid on Public Expenditure: The Case of Kenya”. AERC Research Paper No. 135, March.
OECD-DAC (2005), Promoting Pro-Poor Growth: Infrastructure.
Oosterban R. J.(1988), “Effectiveness and Social/ Environment Impacts of Irrigation Projects: A Critical Review”. International Institute for Land Reclamation and Improvement (ILRI) Annual Report, Wageningen, The Netherlands.
RTI International (2005), “Mobilization of Private Sector Resources for Urban Water Services in sub-Saharan Africa: Case Studies of Water Provision in the City of Dakar, Senegal and the Cities of Nyeri and Eldoret, Kenya”. Paper Submitted to the Municipal Finance Task Force.
Ruigu, George M. (1988), “Large Scale Irrigation Development in Kenya”. Report No. 23/1988 for the Food and Nutritional Planning Unit, Ministry of Planning and National Development, Nairobi, Kenya.
Sachs, J.D., Warner, A. (1997), “Sources of slow growth in African economies”. Journal of African Economies, Vol 6, pp 335– 376.
Syagga P. M.(1999), “The Third Nairobi Water Supply Project evaluation”. Paper Prepared for IDRC.
The Kenya Wetlands Forum (2005), “Development, Conservation and People’s Livelihoods at Cross-roads: The Proposed Sugar-cane Project in the Tana Delta, Tana River District, Coast Province, Kenya”. March.
56 57
Thomson J., I.T. Porros, E. Wood, J.K. Tumwine, M.R. Mujwahuzi, M. KatuiKatua and N. Johnson (2000b), “Waiting at the Top: Changes in Water Use in East Africa over Three Decades”. Environmental and Urbanization, Vol 12, No. 2.
Wambua, Sammy (2004), “Water Privatization in Kenya”. Paper Prepared for the Heinrich Böll Foundation.
World Bank (1984), World Bank Development Report.
World Bank (1998), Assessing Aid. Oxford University Press.
World Bank (2001), “Kenya: Review of the Water Supply and Sanitation Sector. Report 22182”. Washington D.C.
World Bank (2004), “Kenya: Nairobi Water and Sewerage Restructuring Project: Project Appraisal Document 28825-KE”. Washington D.C.
World Bank (2005) Water and Sanitation Program, “Rogues No More? Water Kiosk Operators Achieve Credibility In Kibera”. June.
WSRB (2007), “Socially Responsible Commercialization”. Nairobi, Kenya.
56 57
APPENDIX
Table A1.1: Japanese supported projects in Kenya
Name of the Project SectorDate of
approvalAmount JY
million
Kenya Broadcasting Corporation Modernization Project Telecommunications 28/06/1989 16,198
Mombasa Diesel Generating Power Plant Project Electric power and gas 31/03/1995 10,716
Mombasa Airport Improvement Project Transportation 30/03/1990 9,010
Telecommunication Expansion Project Telecommunications 03/03/1997 8,724
Telecommunications Modernization Project Telecommunications 07/02/1980 7,878
Kilifi Bridge Construction Project Transportation 20/01/1986 7,840
Cement Plant Rehabilitation Project mining and manufacturing 30/03/1990 7,674
Sondu Miriu HydropowerProject Electric power and gas 03/03/1997 6,933
Tana River Basin Road Construction Project (II) Roads 30/3/1990 6,523
Telecommunications Modernization Project (II) Telecommunications 27/07/1983 6,450
Tana River Basin Road Construction Project (I) Roads 2/4/1982 6,100
Tana River Delta Irrigation Project Irrigation and flood
control30/03/1990 6,031
Grain Silo Construction Project Agriculture, forestry and Fisheries 18/07/1985 5,521
Nairobi Water Supply Project Social services 17/03/1989 5,342
Greater Nakuru Water Supply Project Social services 30/03/1987 5,017
New Nyali Bridge Project Transportation 18/12/1975 4,900
Mombasa Airport Project Transportation 09/05/1973 4,086
Rural Road Project Transportation 15/08/1978 3,381
Horticultural Produce Handling Facilities Project Agriculture, forestry and Fisheries 28/10/1993 2,016
New Mtwapa Bridge Construction Project Transportation 15/07/1977 750
Engineering Services for Sondu Miriu Hydropower Project Electric power and gas 17/10/1989 668
Engineering Services for Tana River Delta Irrigation Project
Irrigation and Flood Control
30/3/1987 588
Engineering Services for Mwea Irrigation Project Project Irrigation and floor control 28/10/1993 572
Engineering Services for Grain Silo Construction Project
Agriculture, forestry and Fisheries 13/02/1984 391
Source: JBIC
58
Table A2.1: Performance of Kenya’s major irrigation schemes, 1985/86 - 2004/05*
1985/86 1990/91 1995/96 2000/1 2004/05*Mwea
Hectares cropped (ha) 8,271 5,802 5,901 10,590 10,000Number of plots-holders 3234 3240 3243 3,381 5,400Paddy yields (Million tons) 26407 25504 25887 45,810 59,520Gross value of crop (KSh millions) 84.24 113.36 352.8 1,238 1,786Payments to Plot-holders (KSh million) 29.16 61.98 230.8 - 1,066Ahero
Paddy yields (Million tons) 4378 2986 2054 1,222 741Perkerra
Onions yields (Million tons) 587 1630 889 102 -Chillies yields (Million tons) 234 368 0 32 -Water Melons (Million tons) - - - -Paw paw (Million tons) - - 220 - -Cotton - - 208 - -Bunyala
Paddy yields (Million tons) 1259 403 920 491 1,068West Kano
Paddy yields (Million tons) 2650 3890 1645 1,742 1,348Sugar cane - - - - -Tana
Cotton yields (Million tons) 1839 - - - -Bura
Cotton yields (Million tons) 5182 1678 - - -All (mainly 7) Schemes
Hectares cropped 13950 9365 9039 12,431 10,832Number of plots-holders 7376 7688 7243 5,348 6,660Gross value of crop (KSh millions) 149.06 165.12 444.94 1,291.3 1,880.6Payments to Plot-holders (KSh millions) 57.24 88.26 277.92 26.4 1,115.0
Source: Kenya, Economic Survey, Various Issues, * Provisional
JBICI Research Paper No. 36-1
July 2008
ISSN 1347-5703
4-1, Ohtemachi 1-chome, Chiyoda-ku, Tokyo 100-8144, Japan
Tel: 03-5218-9720 (JBIC Institute)Internet: http://www.jbic.go.jp/ Recycled paper
JBICI R
esearch Paper No. 36-1
JBIC InstituteJapan Bank for International Cooperation
July 2008
Aid Effectiveness to Infrastructure:A Comparative Study of East Asia and
Sub-Saharan Africa
Framework Paper
Aid Effectiveness to Infrastructure: A
Com
parative Study of East Asia and Sub-Saharan A
frica, Framew
ork Paper