Title Page
 Table of Contents
 Key to abbreviations
 Chapter 1: Problem identification...
 Chapter 2: Conceptual framework...
 Chapter 3: Periurban dairying in...
 Chapter 4: Analysis of policy impacts...
 Chapter 5: Kenya-Ethiopia dairy...
 Chapter 6: Discussion and...
 Appendix A: Calculation of milk/concentrate...
 Appendix B: Sources of feed...
 Appendix C: Farm and post-farm...
 Appendix D: Example of structure-of-activity...
 Appendix E: Addis Ababa farm-level...
 Appendix F: Calculated effective...
 Appendix G: Dairy processing...
 Appendix H: Expanded Kenya farm-level...
 Appendix I: Dairy producer...
 Biographical sketch

Group Title: Periurban dairying and public policy in Ethiopia and Kenya : : a comparative economic and institutional analysis
Title: Periurban dairying and public policy in Ethiopia and Kenya
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00056230/00001
 Material Information
Title: Periurban dairying and public policy in Ethiopia and Kenya a comparative economic and institutional analysis
Physical Description: x, 275 leaves : ill. ; 29 cm.
Language: English
Creator: Staal, Steven J., 1957-
Publication Date: 1996
Subject: Dairy farming -- Ethiopia   ( lcsh )
Dairy farming -- Kenya   ( lcsh )
Dairy farming -- Economic aspects -- Ethiopia   ( lcsh )
Dairy farming -- Economic aspects -- Kenya   ( lcsh )
Dairy farming -- Government policy   ( lcsh )
Food and Resource Economics thesis, Ph. D
Dissertations, Academic -- Food and Resource Economics -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (Ph. D.)--University of Florida, 1996.
Bibliography: Includes bibliographical references (leaves 226-234).
Statement of Responsibility: by Steven J. Staal.
General Note: Typescript.
General Note: Vita.
Funding: Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
 Record Information
Bibliographic ID: UF00056230
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 002100704
oclc - 35122404
notis - AKT9733

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
    Key to abbreviations
        Page vii
        Page viii
        Page ix
        Page x
    Chapter 1: Problem identification and research objectives
        Page 1
        Overview of the study
            Page 1
            Page 2
        The potential for dairy development
            Page 3
            Page 4
            Page 5
            Page 6
            Page 7
            Page 8
            Page 9
            Page 10
            Page 11
        Identifying the problem
            Page 12
            Page 13
            Page 14
            Page 15
            Page 16
    Chapter 2: Conceptual framework and methodological approach
        Page 17
        Factors in dairy development
            Page 17
            Page 18
            Page 19
            Page 20
            Page 21
            Page 22
        Public policy and dairying
            Page 23
            Page 24
            Page 25
            Page 26
        Policy effects on levels of protection and competitiveness
            Page 27
            Page 28
            Page 29
            Page 30
            Page 31
            Page 32
            Page 33
        The policy analysis matrix
            Page 34
            Page 35
            Page 36
            Page 37
            Page 38
            Page 39
            Page 40
            Page 41
            Page 42
            Page 43
            Page 44
            Page 45
            Page 46
            Page 47
        Data sources
            Page 48
            Page 49
            Page 50
            Page 51
            Page 52
            Page 53
    Chapter 3: Periurban dairying in Ethiopia
        Page 54
        Evolution of the dairy industry and dairy policy in Ethiopia
            Page 54
            Page 55
            Page 56
            Page 57
            Page 58
            Page 59
            Page 60
            Page 61
            Page 62
            Page 63
            Page 64
            Page 65
            Page 66
            Page 67
            Page 68
            Page 69
        Other public policies affecting dairy markets
            Page 70
            Page 71
            Page 72
            Page 73
            Page 74
            Page 75
        Dairy production in the Addis Ababa milkshed
            Page 76
            Page 77
            Page 78
            Page 79
            Page 80
            Page 81
            Page 82
            Page 83
            Page 84
            Page 85
            Page 86
        Feed usage and feed markets
            Page 87
            Page 88
            Page 89
            Page 90
            Page 91
            Page 92
            Page 93
            Page 94
            Page 95
        Animal health and reproductive services
            Page 96
            Page 97
            Page 98
            Page 99
            Page 100
            Page 101
            Page 102
            Page 103
            Page 104
            Page 105
            Page 106
        Dairy markets in the Addis Ababa milkshed
            Page 107
            Page 108
            Page 109
            Page 110
            Page 111
            Page 112
            Page 113
            Page 114
            Page 115
            Page 116
            Page 117
    Chapter 4: Analysis of policy impacts on Ethiopian Periurban dairying
        Page 118
        Farm-level budgets
            Page 118
            Page 119
            Page 120
            Page 121
            Page 122
        Social valuation at the farm-level
            Page 123
            Page 124
            Page 125
            Page 126
            Page 127
            Page 128
            Page 129
            Page 130
            Page 131
        Farm-level PAM results
            Page 132
            Page 133
            Page 134
            Page 135
            Page 136
            Page 137
            Page 138
            Page 139
        Dairy processing PAM results
            Page 140
            Page 141
            Page 142
            Page 143
            Page 144
            Page 145
        System PAM results
            Page 146
            Page 147
        Policy and comparative advantage indicators
            Page 148
            Page 149
            Page 150
            Page 151
            Page 152
            Page 153
        Sources of milk market failure
            Page 154
            Page 155
            Page 156
            Page 157
            Page 158
            Page 159
            Page 160
            Page 161
            Page 162
    Chapter 5: Kenya-Ethiopia dairy comparisons
        Page 163
        Dairying in Kenya
            Page 163
        Evolution of the dairy industry in Kenya
            Page 164
            Page 165
            Page 166
            Page 167
            Page 168
            Page 169
            Page 170
            Page 171
            Page 172
        Kenya PAM analysis
            Page 173
            Page 174
            Page 175
            Page 176
            Page 177
            Page 178
            Page 179
            Page 180
            Page 181
            Page 182
            Page 183
            Page 184
            Page 185
        PAM comparisons between Kenya and Ethiopia
            Page 186
            Page 187
            Page 188
        Institutional comparisons between Ethiopia and Kenya
            Page 189
            Page 190
            Page 191
            Page 192
            Page 193
            Page 194
            Page 195
            Page 196
            Page 197
            Page 198
            Page 199
            Page 200
        Summary of Ethiopian and Kenyan dairy comparisons
            Page 201
            Page 202
            Page 203
            Page 204
            Page 205
            Page 206
            Page 207
            Page 208
    Chapter 6: Discussion and conclusions
        Page 209
        Economic analysis and diary policy
            Page 209
            Page 210
            Page 211
            Page 212
            Page 213
            Page 214
            Page 215
        Institutional analysis and dairy policy
            Page 216
            Page 217
            Page 218
        Roles of participants in dairying and dairy policy
            Page 219
            Page 220
            Page 221
            Page 222
            Page 223
            Page 224
            Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
    Appendix A: Calculation of milk/concentrate price ratios
        Page 235
        Page 236
        Page 237
        Page 238
    Appendix B: Sources of feed purchases
        Page 239
    Appendix C: Farm and post-farm dairy activity budgets
        Page 240
        Page 241
    Appendix D: Example of structure-of-activity budgets
        Page 242
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
    Appendix E: Addis Ababa farm-level PAM results
        Page 248
        Page 249
    Appendix F: Calculated effective tax on producer profits
        Page 250
    Appendix G: Dairy processing budgets
        Page 251
        Page 252
    Appendix H: Expanded Kenya farm-level PAM results
        Page 253
        Page 254
        Page 255
    Appendix I: Dairy producer questionnaires
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
    Biographical sketch
        Page 275
        Page 276
        Page 277
Full Text








I would like to acknowledge first of all the financial and research support of the

International Livestock Centre for Africa (ILCA, now ILRI), and the support of Dr.

Barry Shapiro and Dr. Simeon Ehui. Without their support this work would not have

been accomplished. Ato Gemachu played an important role in supervising the field

research. Also, Ato Solomon and Ato Berhanu were very effective enumerators and

research assistants.

At the Ministry of Agriculture offices in Ethiopia, Ato Getachew Felleke and Ato

Hassan Ali of the Dairy Development and Rehabilitation Project were extremely helpful

in assisting me to establish contacts in the dairy sub-sector. Ato Muluget Getu,

Ato Eshetu Gebra Selassie, and Wza. Nigat Tessema were effective enumerators. Dr.

Zerihun of the veterinary department was very helpful. I am grateful to Ato Belachew

Hurissa and Wzo. Tsehainish at the Dairy Development Enterprise for thier time and

assistance. Ato Degu Yoseph and Wzo. Meselu Gebra Selassie of the Dairy Producers'

Association played a critical role in establishing contact with dairy producers. Unnamed

but thanked for their assistance are a large number of Ethiopian government officials

from organizations ranging from oilseed mills to the Central Statistical Authority. The

help of a number of officers of the Dairy Producers Association is gratefully


In Kenya, Richard Ephanto and Gem Ardwings-Kohdek of the USAID/Egerton

University PAM Project were generous with both time and assistance. Samuel Njiri

Ndirangu, Chrales Okoth Kenyanito, and Patrick Karanja were all effective enumerators.

At the Kenyan Ministry of Livestock Development, Mr. B.M.Onyango, Mrs. Ouma, Mr.

Paul Odhiambo and Mrs. Mary Mwambia were very helpful in establishing field

contacts. At the KCC Mr. Peter Arap Bii, and Mr. Murithi assisted greatly. At the

KDB Mr. Lenaronkoita and Mr. Evans Gatharia were helpful in explaining Kenyan dairy


Finally, my greatest thanks are to the Ethiopian and Kenyan dairy producers,

women and men, who took to the time to teach me about their important enterprise.



KEY TO ABBREVIATIONS ................................... vii

ABSTRACT .......................................... ix


OBJECTIVES .................................. 1

Overview of the Study .............. .................... 1
The Potential for Dairy Development .................... 3
Identifying the Problem ............... .. ........... .. 12

APPROACH .................................... 17

Factors in Dairy Development ............................ 17
Public Policy and Dairying ............................ 23
Policy Effects on Levels of Protection and Competitiveness ........ 27
The Policy Analysis Matrix ....................... ...... 34
Data Sources ................... ................. 48

3 PERIURBAN DAIRYING IN ETHIOPIA .................... 54

Evolution of The Dairy Industry and Dairy Policy In Ethiopia ....... 54
Other Public Policies Affecting Dairy Markets ................ 70
Dairy Production in the Addis Ababa Milkshed ............... 76
Feed Usage and Feed Markets ............................ 87
Animal Health and Reproductive Services .... .............. 96
Dairy Markets in the Addis Ababa Milkshed ................. 107
Discussion ...................................... 115

PERIURBAN DAIRYING ............................ 118

Farm-Level Budgets ........................ .......... 118
Social Valuation at the Farm-Level ................ .... 123
Farm-Level PAM Results ............................. 132
Dairy Processing PAM Results ......................... 140
System PAM Results ................ .............. 146
Policy and Comparative Advantage Indicators .. ............. 148
Sources of Milk Market Failure ................ ........ 154
D discussion ...... .... ..... .... .... .... ... .. ... 159


Dairying in Kenya ................................. 163
Evolution of the Dairy Industry in Kenya ....... ......... 164
Kenya PAM Analysis ......... .................... 173
PAM Comparisons between Kenya and Ethiopia ............. 186
Institutional Comparisons between Ethiopia and Kenya .. ........ 189
Summary of Ethiopian and Kenyan Dairy Comparisons ........... 201

6 DISCUSSION AND CONCLUSIONS ..................... 209

Economic Analysis and Dairy Policy .... .. .. ........ 209
Institutional Analysis and Dairy Policy .............. .... 216
Roles of Participants in Dairying and Dairy Policy ............. 219
Conclusions .................................... 223

REFERENCES ........................................ 226



B SOURCES OF FEED PURCHASES ...................... 239





G DAIRY PROCESSING BUDGETS ........................ 251



BIOGRAPHICAL SKETCH ............................... 275




Addis Ababa Dairy Development Project
Addis Ababa Dairy Industry
Addis Ababa Dairy Producers Association
Artificial Insemination
Agricultural and Industrial Development Bank (Ethiopia)
Agricultural Inputs Supply Corporation (Ethiopia)
Commercial Bank of Ethiopia
Computable General Equilibrium
Central Statistical Authority (Ethiopia)
Danish International Development Agency
Dairy Development Agency (Ethiopia)
Dairy Development Enterprise (Ethiopia)
Domestic Resource Cost
Domestic Resource Cost Ratio
Dairy Rehabilitation and Development Project (Ethiopia)
Ethiopian Birr (approximately 6.15/US$ in 1995)
Effective Protection Coefficient
Ethiopian People's Revolutionary Democratic Front
Finnish Development Agency
Government of Ethiopia
Institute for Agricultural Research (Ethiopia)
International Fund for Agricultural Development
International Livestock Centre for Africa, Addis Ababa (now the
International Livestock Research Institute (ILRI), Nairobi)
Integrated Livestock Services Association (Ethiopia)
Inland Revenue Service (Ethiopia)
International Value Added
Kenya Cooperative Creameries
Kenya Dairy Board
Kenyan Shilling (approximately 50/US$ in 1995)
Milk/Concentrate (price) Ratio
Metabolizable Energy
Ministry of Agriculture (Ethiopia)
Ministry of State Farms Development (Ethiopia)
National Dairy Development Project (Kenya)

NEP Net Economic Profitability
NPC Nominal Protection Coefficient
PA Peasant Association (Ethiopia)
PAM Policy Analysis Matrix
PC Producer Cooperative (Ethiopia)
PRCR Private Resource Cost Ratio
PUD Periurban dairy or dairying
Q Quintal (100 kilos)
RTAPAP Research and Training in Agricultural Policy Analysis Project (Kenya)
SAM Social Accounting Matrix
SC Service Cooperative (Ethiopia)
SSA Sub-Saharan Africa
TLU Tropical Livestock Unit
WFP World Food Program

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Steven J. Staal

May 1996

Chairperson: Uma Lele
Major Department: Food and Resource Economics

Periurban dairying offers important income opportunities for smallholders in

Ethiopia. World milk production is declining, nominal world milk prices continue their

upward trend, and urbanization in Africa promises growing domestic demand. Periurban

dairy producers can potentially supply both the urban and export markets. This potential

has been illustrated in neighboring Kenya, with which Ethiopia shares agro-climatic

conditions favorable to dairying. Unlike Kenya, however, Ethiopia has not successfully

developed a formal dairy system. Some 88% of urban milk supply continues to pass

through informal market channels. This can be attributed largely to government policy

directed generally towards exchange rates, agricultural markets and production, and

specifically towards dairying.

This study examines the impacts of Ethiopian government policy and programs

on dairy producers and marketers. The Policy Analysis Matrix (PAM) methodology is

used to identify and quantify price distortions in the dairy market. Domestic market milk

prices are found to be lower than export-parity prices, indicating potential milk market

failure. This market failure results mainly from a previously-overvalued currency, which

precluded dairy exports at official exchange rates. Recent currency reforms, however,

have greatly improved the potential of dairy exports. Policy changes are now needed to

enable the formal market to capture more of the informal market through producer price

increases. Even under current market conditions, however, periurban dairying in the

Addis Ababa area exhibits above-normal returns, and so offer important income

opportunities for small resource-poor households.

Institutional analysis reveals additional non-price policy factors which have

impeded dairy development. Policies in the 1970s and 1980s aimed towards socialized

agriculture eliminated the private commercial dairy producers who had led dairy

development. Land tenure policies helped insure that milk production remained in small

backyard urban units selling to the informal market. Policies towards cooperatives

precluded the development of the strong role that cooperatives played in Kenyan dairy


Comparisons between Kenyan and Ethiopian dairy policies are made using the

PAM, and show that while Kenyan dairy is more competitive, both have a comparative

advantage with respect to world dairy markets due to low opportunity costs of domestic



Overview of the Study

This overview outlines the conceptual structure of the research. Each of the areas

outlined are discussed in more detail later in the study. The overview serves simply to

familiarize the reader from the outset with the issue being addressed and the general

approach being used.

Introduction to the Problem

This study examines the impact of public policy on dairy development in Ethiopia,

with comparisons to Kenya. It does so using both quantitative economic analysis and

institutional analysis tools. Although the quantitative analysis is limited to a cross-

sectional, point-in-time study, the institutional analysis attempts to examine dairy policy

over time.

The premise which drives the research is that dairying has not lived up to its

potential in Ethiopia and that this failure is closely linked to public policy. The broad

question addressed is thus: which specific policy interventions have had the largest effect,

positive or negative, cn dairy development in Ethiopia? Additionally, how have policy

failures resulted from conflicts between differing policy interventions, or from conflicts

with the interests of actors in the dairy sub-sector? How have the contributing factors


Overview of the Study

This overview outlines the conceptual structure of the research. Each of the areas

outlined are discussed in more detail later in the study. The overview serves simply to

familiarize the reader from the outset with the issue being addressed and the general

approach being used.

Introduction to the Problem

This study examines the impact of public policy on dairy development in Ethiopia,

with comparisons to Kenya. It does so using both quantitative economic analysis and

institutional analysis tools. Although the quantitative analysis is limited to a cross-

sectional, point-in-time study, the institutional analysis attempts to examine dairy policy

over time.

The premise which drives the research is that dairying has not lived up to its

potential in Ethiopia and that this failure is closely linked to public policy. The broad

question addressed is thus: which specific policy interventions have had the largest effect,

positive or negative, cn dairy development in Ethiopia? Additionally, how have policy

failures resulted from conflicts between differing policy interventions, or from conflicts

with the interests of actors in the dairy sub-sector? How have the contributing factors


of institutional constraints, infrastructure and technology affected policy success or


Introduction to the Conceptual Framework

The concept of public policy is defined in this study as having two broad

components. The first is policy as the process of priority setting and formulation of a

course of action. The second is policy as the implementation of that course of action,

in the form of specific policy interventions. These interventions can be characterized as

macroeconomic, regulatory, investment, and institutional. These interventions impact

the behavior of dairy producers and marketers in some economically-quantifiable ways

as well as in ways not easily quantified. A methodological approach is thus necessary

which can assess the relative importance of both types of impacts resulting from each

type of policy intervention.

Introduction to the Methodological Approach

An approach which uses two complementary tools of analysis was chosen to

examine both quantifiable and less-quantifiable policy impacts.

The Policy Analysis Matrix (PAM) approach was used to quantify, in terms of

economic incentives, the economic impacts of dairy policy interventions. This tool is

well suited to the point-in-time nature of the data gathered. It can also be used to make

inferences about economic impacts not directly quantifiable. The PAM is also applied

to Kenyan secondary data for the purpose of comparison of economic policy impact.


The institutional analysis, focused on impacts not identifiable by the PAM

approach, is based on careful observation of the evolution of dairy development over

time. A conceptual framework identifying critical factors in policy formulation and

implementation is used. Comparisons between Ethiopian and Kenyan dairy development

are particularly important in identifying these non-economic or institutional impacts of


Structure of the Study

The rest of this chapter more clearly defines the study problem and objectives.

Chapter 2 details the conceptual and methodological approaches introduced above.

Chapters 3 and 4 describe the dairy system in Ethiopia and present the results of the

PAM analysis. In Chapter 5 the results of economic and institutional comparisons

between Kenya and Ethiopia are presented. Conclusions and recommendations are made

in Chapter 6.

The Potential for Dairy Development

The 1990s have brought sweeping economic changes to many economies around

the world. Former centrally-planned economies are struggling to adjust to create

working markets, and many Latin American nations have loosened restrictions on

economic activity. Asian nations continue to grow economically and close the income

gap with the traditionally rich countries. Africa, however, lags behind in both the

economic reform which is now generally embraced and in economic growth. Further,

the process of industrialization remains slow, and African nations continue to rely heavily


on agriculture for economic activity and export earnings. Although prices of some

traditional African exports, notably coffee, have recently risen, the outlook for growth

in these income-inelastic commodities remains uncertain. Alternative activities are

needed which a) offer higher returns to land and labor and the expectation of future

growth, which b) utilize the large agricultural sector, and which c) are suitable for

adoption by the resource and technology-poor agriculturalists who continue to dominate

African production. Periurban dairy (PUD) production may be just such an activity.'

The Outlook for the World Milk Market

World production of liquid milk is steadily declining. From an average of 536

million metric tons in 1988-90, it fell to 518 million tons in 1993. The proportion of the

production occurring in less developing countries (LDCs), however, is rising. During

the same period, production in LDCs rose from 155 to 173 million tons, while that in

developed countries fell from 381 to 345 million tons (FAO 1994). Some of the overall

decline is due to sharp falls in production in former centralized economies in Eastern

Europe. Production has also declined in the industrialized countries, however, due to

programs to cull dairy herds and lower production quotas. In Africa, production fell

from 13 million metric tons in 1988-90 to 11 million tons in 1993, largely due to drought

in Southern Africa, the largest production area (FAO 1994).

1 In this study periurban dairying is considered to be that dairy production and
marketing which occurs in the milkshed of the urban area, where demand is expected
to be highest. Thus in the case of Addis Ababa, it includes producers within the
urban area, those near the urban area who supply the informal urban area, and those
further away who supply the formal collection centers. The marketing systems that
link these producers to the urban market are included in the periurban dairy system.

$ per Metric Ton of Milk Powder
4000 -

3500 .

2000 -
1500 .. .. -
500 ..
1983 1984 1985 1988 1987 1988 1989 1990 1991 1992 1993

Whole Milk Powder Skim Milk Powder

Source: FAO
Figure 1-1 Annual Average World Market Price of Whole and Skim Milk Powder,

World prices for milk powder remain highly volatile, as consumption patterns

shift. A trend towards lower consumption of milk fat in developed countries is

accompanied by rising consumption of milk products in developing countries. The

overall trend, however, is towards generally rising nominal prices. Figure 1-1 shows

average annual world prices in US$ for whole and skim milk powder from 1983 to 1993.

Prices rose sharply in the late 1980s, and peaked briefly in 1992. The FAO expects that

milk production levels will continue to fall, and that nominal prices will continue to trend

upwards (FAO 1994).

As world prices rise, dairy imports to Sub-Saharan Africa (SSA) may be

increasingly difficult to sustain financially. The effect of nominal world price rises on


Africa will be determined by exchange rates and domestic inflation. If exchange rates

devalue at the rate of domestic inflation over time, nominal world price increases will

translate into real increases in border prices. The fall in world production and

subsequent increase in the nominal world milk price is occurring at the same time that

demand for dairy products in Africa is increasing. Continued urbanization of the

continent, income growth among certain segments of the population, and changes in diets

are factors associated with increased dairy consumption (Shapouri and Rosen 1991).

Opportunities for African dairy producers are favorable and may grow.

The Potential Benefits of Periurban Dairy Development

In theoretical models of dualistic development, the economy is represented with

two complementary sectors.2 The first is a large primarily rural, traditional sector

engaged mainly in agricultural activities. The second is a small, modem capitalist sector

which is more industrialized. The agricultural sector is characterized by low wages and

labor surpluses, with varying degrees of marginal productivity of labor.

Commercialization of the economy occurs with the transfer of labor from the rural to the

capitalist sector, driven initially by a higher wage rate in urban areas, associated with

higher capital utilization. This is followed later by higher wages and productivity in the

rural sector as labor supply becomes more constrained. Lele and Mellor point out that,

assuming positive marginal productivity of labor in the agricultural sector,

industrialization must be accompanied by productivity growth in the rural sector (Lele

2 These models are typified by Lewis (1954), Fei and Ranis (1964) and Jorgenson


and Mellor, 1981). Otherwise, as labor transfers reduce food production, the resulting

increases in the price of wage goods would increase wage rates in the capitalist sector,

constraining further growth. Common to all these models is an emphasis on strong

linkages primarily through labor, but variously through capital and wage goods.

Although the scale of dairying activity may be small relative to other agricultural

sub-sectors, periurban dairying is an excellent example of an agricultural activity which

offers employment and income, and which also develops strong links between rural and

urban areas.3 It may also contribute to the higher labor productivity needed in the rural

sector through higher returns to labor. Given also the labor-intensive nature of dairying,

the employment potential of PUD may be important, at least relative to the small overall

scale of the activity.

Most important perhaps, is the potential which is embodied in the strong linkages

between PUD and the urban sector, on one hand, and between PUD and other

agricultural sub-sectors on the other. The outlet markets for PUD are found in urban

areas, which in Africa are characterized by rapid population growth. Most rural-urban

migrants turn to the market for their demands even though they may have produced milk

before migrating. Further, dairy products make up a small part of household

expenditures, and are characterized by relatively high income-elasticity of demand,

estimated at greater than one in some West African studies (Jabbar and di Domenico

3 Only a small proportion of any periurban population is typically involved in
PUD production and marketing. Of possibly some 500,000 households in the Addis
Ababa area, it is estimated that about 8,000 are involved in production for the milk
market (AADPA).


1992). Assuming future growth in urban household incomes, the prospects for long-term

dairy demand growth are good.

Further, growth in PUD implies increased demand for other agricultural products.

Most PUD depends heavily on agricultural products for primary inputs. A large

proportion of PUD input expenditures in Ethiopia is for food by-products for use as feed

materials. Food by-product markets in Ethiopia, however, are already experiencing

shortages in supply. As the value of dairy products increases, the derived demand for

feed can be expected to shift to food crops themselves, particularly grains. The resulting

increase in grain prices would benefit producers in other agricultural sub-sectors. In this

way the relatively small PUD sub-sector could positively effect the larger agricultural

sector. PUD thus offers a beginning step toward a more commercialized agricultural

system in which most primary agricultural output is consumed by livestock.

The rural-urban linkages offered by PUD can contribute to rural economic

development through the transfer of capital. Since PUD can potentially be carried out

by very resource-poor households, it enables capital accumulation and further investment

by those households. It thus enables the shifting of capital to those households to whom

it may be otherwise unavailable through traditional financial markets. Combined with

the close urban proximity of the producing households, which allows them some diversity

of economic activities for investment, PUD may help introduce capital in a variety of

small-scale income and employment generating activities.

Finally, PUD offers an excellent example of how public policy can affect the

welfare of both rural and urban areas. Policy which enhances PUD offers the potential


of both lowering prices to consumers and raising the incomes of producers. The effect

of policy interventions, however, will be influenced by other contributing factors,

particularly institutional constraints, infrastructure and technology. In spite of their

recognizable potential, PUD and dairying in general remain poorly studied in Africa

compared to other agricultural activities such as traditional food and cash crops.4

Overview of Dairy in Sub-Saharan Africa

Sub-Saharan Africa's dairy production has not kept up with consumption levels

(von Massow 1989). As a result, dairy imports to SSA substantially increased during

the 1980s, some in the form of donor aid, but mostly as commercial imports (von

Massow 1984). During the period 1961-88, dairy imports to SSA increased six-fold,

growing at an annual rate of 6.5%. However, between 1976 and 1988 the growth of

imports slowed to less than 1% annually, mainly due to foreign exchange constraints

(Shapouri and Rosen 1991). Since 1988, however, there has been a resurgence of SSA

imports mainly attributable to demand shifting faster than domestic supply.

In a study of the determinants of commercial dairy imports to SSA, von Massow

showed that most of the increase (almost 2/3) in commercial dairy imports since 1960

can be explained by increases in population. Increases in per capital incomes have also

been a factor, but since 1980 per capital income have been declining in many SSA

countries. He attributes the rest of the increase to government policies and import

4 As described by Nell (1990), the range of dairy systems currently found in SSA
include: pastoralism, agro-pastoralism, mixed farming, intensive dairying and
periurban dairying.


price/exchange rate polices (von Massow 1989). Thus public policies appear to have

encouraged imports.

The potential for increased dairy production in SSA is substantial. Both

mixed farming and periurban dairying complement the smallholder production systems

which are most prevalent in SSA. Input and technology requirements are generally

within the means of existing producers and markets, and production is compatible with

sustainable use of land resources (Nell 1990, Mbogoh 1987, Shapiro et al. 1990).

Already, smallholders play the largest role in East African milk production. In Kenya

smallholders now produce 80% of all dairy products (Sellen et al. 1990). In Ethiopia,

smallholders account for 75% of commercial liquid milk production, and even a higher

proportion of national milk production (survey results).

The potential for income-generation for smallholders through dairying is great.

Research by ILCA north of Addis Ababa showed returns to cross-bred dairy production

several times those available to traditional crops (Haile 1995). Furthermore, significant

opportunities exist for small-scale dairy processing (Shapiro et al. 1990, Mbogoh 1987),

which is essential for dealing with low density milk production. Such processing, given

limited requirement levels of energy, skill, or maintenance, fits similarly into existing

market and technology capabilities (Nell 1990).

Spatial and temporal factors affecting collection and marketing limit the economic

viability of small-scale milk production and processing. Such problems as small volume

per producer, seasonal variability of supply, dispersed/low-income retail markets, high

temperatures, poor transport systems, and the seasonality of rainfall constrain existing

systems (Nell 1990). Further, poor genetic quality of dairy animals limits productivity.

Overcoming these conditions not only requires improved infrastructure, but also a policy

environment that creates producer incentives and encourages investment.

Periurban Dairying in Ethiopia

Ethiopia's dependence on dairy imports has grown. During the 1977-88 period

the estimated annual growth rate of milk consumption in Ethiopia was 2.5%, while

growth in production during the same period averaged only 1.4%. Partially as a

consequence, the initially low import dependence grew from 4.1% to 12.8% during that

period (Shapouri and Rosen 1991). Although most of these imports occurred through

food aid programs now curtailed, the consequence of maintaining this trend would be a

growing financial burden during a time of rising world prices.

With Africa's largest livestock herd, the potential for dairy in Ethiopia is

substantial (FAO 1993a). Yet the inability to maintain per capital production levels points

to significant weaknesses in the milk production and marketing system. These

weaknesses do not reflect the potential that exists (Debrah and Anteneh 1991).

Neighboring Kenya, which shares some of Ethiopia's highland characteristics,5 is largely

self-sufficient in simple net quantity terms.

Unlike in Kenya, milk production from exotic and cross-bred animals is

insignificant in Ethiopia, without which large increases in milk production cannot be

5 The highland climates of Ethiopia and Kenya allow the use of higher-yielding
grade and cross-bred dairy cattle, which are bred for temperate conditions and can
withstand only with difficulty the higher temperatures and disease incidence of
lowland African conditions.


expected. PUD in Ethiopia does rely on cross-bred cattle, however, most of which are

concentrated in the periurban production area around Addis Ababa (MoA 1983). Most

are held by smallholder private producers in both periurban and urban locations.

Only 5% of Ethiopian production is marketed as liquid milk6, mainly due to the

rural location of most production, and extremely limited transportation infrastructure

(MoA 1983). Milk marketing in any volume is restricted to the periurban area and

primarily to Addis Ababa. Most of it occurs through the informal market in the form

of raw milk sold directly to consumers. The formal milk market, run by a parastatal,

supplies only 13% of the liquid milk to the capital (study results). The large role of the

informal market appears to result from low formal milk prices and institutional

constraints. Further, the evolution of dairy cooperatives which could assist producers

to supply the formal market has been precluded by public policy. Producers also

complain of regulations restricting feed markets and poor access to livestock services.

In the context of extensive national economic reforms, Ethiopian dairy policy is

in a transitional phase. The privatization of parastatals, the emergence of private

financial services, and the liberalizing of markets are occurring. However, clear dairy

policy decisions have not been made.

Identifying the Problem

The general trend towards the liberalization of both border and domestic market

controls may allow higher nominal world prices to raise domestic milk prices to

6 The remainder is consumed by the producing households.


producers and processors. Further, trends towards increased urbanization are raising

domestic milk market demand from urban non-producers. If economic growth improves

under current economic reforms, increased incomes will also raise domestic demand.

These factors suggest increased opportunities for dairy producers. Appropriate policies

could contribute to the realization of these opportunities. Kenyan policy has to some

degree succeeded in stimulating smallholder dairy development. Ethiopia, with some of

the same naturally advantageous conditions, has not matched the Kenyan success.

The reasons for Ethiopia's relative failure in dairy have not been clearly

identified. Although Kenyan dairy has been extensively studied, the various policy

factors affecting the Ethiopian dairy system are not well known and the specific impacts

of their effects on dairy producers and marketers are not well understood.7

Problem Statement

Although apparently sharing some of the natural advantages of its successful

dairy-producing neighbor Kenya, Ethiopia has not succeeded in effectively developing

its dairy system. Public policy interventions, as manifested in both direct economic

interventions and institutional factors, are believed to have contributed to this lack of

success. The primary question addressed by this research is thus:

7 Although prior research has described the dairy production and marketing
systems that serve the Addis Ababa urban market, little work has been done to
analyze the policy factors contributing to the poor performance of PUD in Ethiopia
(Debrah and Anteneh 1991, Debrah 1990). In Kenya, more substantial work has been
carried out regarding policy impact on dairy (Sellen et al 1990, Danida 1991, Ruigu
1976, Raikes 1981).

a) Which specific policy interventions have had the largest effect, positive or

negative, on dairy development in Ethiopia?

Further, the research to date indicates that some policy interventions have not had

their desired effect. This may be due to conflicting impacts resulting from multiple

policy interventions, or from other factors such as institutional constraints, infrastructure,

and technology. This may also be due to conflicts between policy and the interests of

the various actors involved in the dairy system. Thus two secondary questions can be


b) How have the interactions between specific policy interventions, and with

other contributing factors, compromised their impact on dairying?

c) How have the interests of various dairy system participants (public, private,

cooperative, and donor) compromised not only dairy policy interventions, but the policy

process itself?

Although Kenya has exhibited its dairy potential through the activity in that sub-

sector, and Ethiopia shares some of its agro-climatic conditions, dairy potential in

Ethiopia may be overstated. A further question can thus be posed:

d) Can the economic potential of periurban dairying in Ethiopia be demonstrated?


Based on the overview of Ethiopian dairy, and on the problem statement, some

hypotheses can be formulated:

1. If the economic potential of dairying in Ethiopia could be demonstrated, then

the failure to realize fhat potential resulted from poor policy in the broad sense and


inappropriate policy interventions, in combination with the contributing factors related

to institutions, infrastructure, and technology.

2. Multiple policy interventions have had conflicting and sometimes off-setting

impacts on the process of dairy development in Ethiopia.

3. Policies which have contributed to dairy development have been those in

which producers have had a part in forming, and which have supported the role of

producers' cooperative associations.


The general objective is to identify the most important impacts of public policy

on the Ethiopian periurban dairy systems, with comparisons to Kenya. Specific

objectives include the following:

1. Calculate the levels of comparative advantage at this point in time in periurban

dairying in Ethiopia and Kenya, and using economic protection indicators, evaluate the

impact of the overall policy environment on those levels.

2. For those policy interventions whose effects on economic incentives are

currently measurable in Ethiopia, quantify the differing levels of those incentives

resulting from various interventions.

3. For those policy interventions whose economic impact are not readily

quantifiable, evaluate their effects through institutional analysis over time using the

different patterns of dairy development in Ethiopia and Kenya as a basis for evaluation.


4. Identify and evaluate how the roles of the public sector, private actors,

cooperative organizations, and donors have interacted with dairy policy and markets to

affect dairy development in Ethiopia and Kenya.


This chapter first identifies some of the primary factors related to dairy

development in East Africa as a context within which to consider policy analysis. A

conceptual framework for policy analysis is then presented. The literature regarding

quantitative policy analysis is reviewed and then the quantitative approach is described.

Finally, data sources and survey methods are described.

Factors in Dairy Development

The emphasis in the 1980s in African agricultural development on "getting prices

right" has not been supported by all. A variety of observers have supported the

alternative view which stresses the importance of structural issues such as research, input

distribution systems and infrastructure (Lele 1981, Delgado and Mellor 1984). Such

institutional and infrastructural factors are likely to be particularly important in Africa

where modern (i.e. Western) institutions to support technological change are

undeveloped, and where transport and communication infrastructures often remain

rudimentary. Where these factors do not enter (at least in a remotely measurable way)

into production and marketing costs, they can be considered "non-price" or "non-

marginal" factors. Quantitative economic tools generally fail to adequately address these


This chapter first identifies some of the primary factors related to dairy

development in East Africa as a context within which to consider policy analysis. A

conceptual framework for policy analysis is then presented. The literature regarding

quantitative policy analysis is reviewed and then the quantitative approach is described.

Finally, data sources and survey methods are described.

Factors in Dairy Development

The emphasis in the 1980s in African agricultural development on "getting prices

right" has not been supported by all. A variety of observers have supported the

alternative view which stresses the importance of structural issues such as research, input

distribution systems and infrastructure (Lele 1981, Delgado and Mellor 1984). Such

institutional and infrastructural factors are likely to be particularly important in Africa

where modern (i.e. Western) institutions to support technological change are

undeveloped, and where transport and communication infrastructures often remain

rudimentary. Where these factors do not enter (at least in a remotely measurable way)

into production and marketing costs, they can be considered "non-price" or "non-

marginal" factors. Quantitative economic tools generally fail to adequately address these


factors. It is useful, therefore to address the range and types of factors which may affect


In discussing these factors, it is also useful to identify the roles of various actors

in the development process. Four sets of actors can be described: 1) those in the public

sector, 2) those in the private sector, 3) cooperative organizations and institutions, and

4) international donors.


Infrastructural factors affect dairy development in two ways; through milk or

output markets and through input markets, mainly those supplying feed. On the output

side, the direct infrastructural component of milk markets is the collection and processing

system.' Formal milk collection systems can be regarded as those supplying processed,

packaged milk to the market, while informal milk collection systems supply raw milk.

These designations are used because processed milk markets are regulated, while raw

milk markets are generally not.

The existence and extent of formal milk collection systems is closely linked to the

structure of production. Where they operate effectively, they provide an outlet of-last-

resort to producers who may sell first to the informal market. Producers are thereby able

to reduce the risks of not finding market outlets. Combined with the larger milkshed

made available by formal collection systems, this allows dairy production to take place

SIndirect infrastructural factors include feeder road systems and electrical grids
within the milkshed. These facilitate the operation of milk collection and storage


in periurban areas often well away from consumption centers. Such production is mainly

based on producers' own land resources, and relies heavily on grazing and forage crops.

This direct reliance on land resources could translate into a more economically-efficient

production system. A reliable outlet for milk also allows for larger production units,

which may have the potential to capture economies of scale, further improving


Conversely, the absence of an effective formal collection system forces production

into the urban areas, where direct producer-consumer marketing can take place. Under

such conditions, producers must rely on feed resource markets, distancing them from the

land resource base and raising their costs of production. Further, the generally more

risky and atomized nature of informal market outlets precludes the development of large

production units.

The link between types of collection systems and the location and scale of

production insures that land tenure policies and institutions play a critical role in

successful development of formal milk collection. Milk pricing policies in the formal

market will determine incentives for producers to either participate or to divert their

production to the informal market, where prices are by definition market-driven.

Further, the public sector generally plays a large role in building the formal market

system, often with the help of international donors. Cooperative societies may play a

large role in either formal or informal milk collection systems.

It should be noted that the milk being supplied by these two markets is not a

homogeneous product. It is not certain, however, that the relative prices that consumers


place on raw and processed milk reflect the additional costs inherent in processed milk.

Since the price of processed milk in Ethiopia is set administratively, relative open-market

prices for raw and processed milk are unknown. Consumers may prefer the higher fat

content of raw milk, in spite of variability in its quality due to common adulteration with

water by producers. Indeed some studies have suggested that a premium of some 15%

should be placed on raw milk relative to processed milk in East Africa (Mbogoh 1987).

The degree to which these are differentiated products, however, is unknown. The effect

of infrastructural constraints on milk markets will to some extent depend also on relative

consumer demand for these two types of milk.

On the input side, infrastructural factors also affect the markets for feed materials,

which are the most important input markets to dairy development. These provide

primarily grain milling by-products and oilseed cake to intensive periurban or urban dairy

producers. If these markets are inadequate to the demands of a growing dairy sub-sector,

long-term sub-sector viability will be impaired. These markets are supplied by flour

mills and oil processing plants, whose processing capacity and through-put will determine

the quantity of feed materials available. Public policies towards these activities and

direct public sector investment in food processing will naturally impact these markets.

Depending on policies, the private sector may or may not also have a prominent role in

feed supply. Further, as will be seen in the Ethiopian case, donor food-aid activities may

impact feed supplies.

Finally, transportation infrastructure can affect both output and input markets.

Milk collection may be limited to areas served by feeder roads, in spite of close


proximity to consumers. This is due to the high perishability of milk. Feed material

markets are unlikely to be affected to the same extent. Livestock services, however,

requiring the timely intervention of veterinary and artificial insemination (AI) technicians,

can be severely affected by poor transportation infrastructure. Although the public sector

is generally considered responsible for transportation infrastructure, in Ethiopia some

feeder roads are built and maintained by mutual self-help organizations (mehaber).


Levels of technology in East African dairy are measured in several ways: a) in

the genetic potential of the dairy cattle, b) in the types of feeds used, and particularly in

the degree of stall-feeding or grazing used, and c) in the level of reliance on modem

veterinary and reproductive services.

The genetic potential of the dairy stock will naturally affect the productivity and

potential of the system, primarily through milk yields, morbidity and mortality. Exotic

or cross-bred cattle yield more milk than indigenous Zebu cattle, but are generally more

susceptible to diseases and high temperatures. Nearly all PUD in both Kenya and

Ethiopia relies heavily on cross-bred cattle of various levels of grade. The ability of

producers to control or improve the genetic potential of their cattle depends, however,

on their use of artificial insemination. Similarly, the availability and quality of veterinary

services will affect producers' ability to raise high grade cattle. The availability and

effectiveness of these usually public sector services will depend largely on public policy.

Their may also be roles for both the private sector and cooperative organizations for

providing some of these services.

Periurban dairy production systems in East Africa are categorized as either open-

grazing, semi-zero grazing, or zero grazing systems. Open grazing systems rely

primarily on grazing to supply the maintenance needs of the cattle, with some use of

concentrate feeds for energy needs. Semi-zero grazing systems use a mix of grazing, cut

fodder, and concentrate feeds. Zero-grazing systems rely almost entirely on cut fodder

and concentrates. Zero-grazing systems allow the highest levels of control over diseases

and reproduction and so are considered the most desirable technology. These systems

are not necessarily linked to scale, but depend instead on land availability and value, and

relative prices of feed materials. They are thus affected by land tenure policies, and all

factors affecting input markets.

Institutional Factors

Institutional factors are those which fall outside of economic or technical

constraints. They generally.impact the behavior of dairy producers or processors through

public, private or community organizations and institutions.2 They include such things

as security of land tenure, access to government, and freedom to organize. Rental of

grazing land in Ethiopia, for example, is affected by traditional limits on its use. Direct

producer-consumer milk marketing in Addis Ababa occurs through traditional contractual

agreements, shaping the functioning of the market. It should be noted, however, that

2 While organizations define sets of individuals into groups, institutions define
sets of behaviors. A cooperative society is thus an organization, while a set of
customary land tenure practices may be considered an institution. An organization
such as a cooperative, however, may in time assume an institutional quality if the
behaviors driving the organization become pervasive.


institutional factors may come about through public policy, through government

regulations which allocate authority or constrain behavior. Institutional factors are thus

linked both to policy and to traditional social structures and organizations.

Public Policy and Dairying

Defining Public Policy

Webster's Dictionary includes among the definitions of the word 'policy' a) the

science/process of public administration, b) the objectives or course of action selected

by that process, c) the specific set of decisions and programs designed to carry out that

chosen course.3 This definition being overly broad, it is useful to focus on b) and c)

from above.

Policy objectives. These define the broad course of action chosen, including

social and economic priorities, the relative roles of the public/private sectors, and the

level of participation of various actors.'

Policy interventions. These are the specific instruments meant to implement the

policy objectives. They range from taxes to the choice of administrative structure.

3 This is a paraphrase of some of the definitions listed. Webster's 3rd
International Dictionary, 1971. Springfield: G & C Merriam Company.

4 The policy 'environment' is not necessarily defined by policy objectives. The
policy environment is the condition brought about by the sum of specific interventions
as implemented. Their implementation may fall short of objectives.


The distinction between these two policy levels raises the possibility that policy

interventions may not fulfill policy objectives if not well designed or properly

implemented. This distinction is further explored in Chapter 5.

Policy Categories

For our purposes, public policies can be grouped into four general and perhaps

simplistic categories: 1) macroeconomic policy, 2) regulatory and pricing policy, 3)

investment policy, and 4) institutional policy. Within each of these are objectives and

interventions. Policy objectives may also encompass some or all of these categories.

Macroeconomic policy. This includes exchange rates and interest rates, among

other interventions, von Massow (1989) determined that exchange rate policy was the

main factor influencing levels of dairy imports in Africa.

Regulatory and pricing policy. Under this broad grouping falls taxation, price

controls, market and trade regulations, licensing, etc. Licensing and tax requirements

can create incentives for dairy producers and processors to operate only through the

informal market.

Investment policy. This relates primarily to the allocation of government

resources, including dairy parastatal and infrastructure development, and expenditures

towards livestock services and dairy research.

Institutional policy. This category is arbitrarily defined to include policies

regarding a) the allocation of authority, b) the structure of administration, c) property

rights. The authority of food plant managers to control sales of livestock feed materials,

for example, can create rent-seeking activity. The structure of extension administration


can impact the effectiveness of livestock services. Land tenure policies can alter dairy

producers' choice of location and production technology.

Policy Impacts

All of the above policy types can be considered to affect actors in the dairy

system, and ultimately the behavior of those actors, through two basic ways: 1) purely

economic incentives which are measurable in the revenue and cost structures of dairy

system activities and 2) less quantifiable incentives which may be institutional in nature

or which have economic effects not measurable in revenue and cost structures.

Both of these types of incentives may result from any type of policy intervention.

Macroeconomic policies, however, are likely to affect producers through purely

economic incentives. Similarly, institutional policies are likely to affect producers

through purely institutional incentives. Some policies, however, lead to both types of

incentives to varying degrees. A tax on milk production will produce a measurable

economic impact, but the threat of collection of the same tax may additionally lead to

institutional incentives that alter producer behavior.

It should be made clear that economic incentives and impacts as described here

are not equivalent to economic effects in terms of net economic losses or welfare gains

to consumer and producers. Measurement of these effects even in the simplest single-

market analysis requires data regarding not only revenue and cost structures, but also

aggregate production and consumption data over time, and estimates of supply and

demand elasticities (Tsakok 1990). Historic production and consumption data for

Ethiopian dairy are not available. This study is limited to a point-in-time approach and


thus focuses on the economic incentives currently measurable, rather than attempting to

quantify the effects of those incentives over time. Quantifying these incentives, however,

will not only help to explain observed behavior by dairy producers and marketers, but

will enable inferences to be made as to expected behavior over time.

Given the distinction between institutional and economic impacts of public policy,

a methodological approach was chosen which combines quantitative economic analysis

and institutional analysis tools. The Policy Analysis Matrix (PAM) compares revenue

and costs structures under financial and economic valuation to quantify economic

incentives. The institutional analysis is based on careful observation of the evolution of

dairy development over time and relies heavily on comparisons between Ethiopia and

Kenya. Further, a conceptual framework identifying critical factors in policy formulation

and implementation is used. This latter analysis is particularly useful in evaluating the

interaction between dairy market actors and the first stage of policy, that of policy


The focus of this study will be on the production side of the dairy system.

Besides the necessity to limit the scope of any research, this comes from the already-

stated recognition of the potential importance of dairy production as a means of

agricultural income-generation. In Africa, there is some evidence to suggest that urban

consumers tend to be relatively better-off than typically poor producers (Timmer 1986).

Policy Effects on Levels of Protection and Competitiveness5

Protection Coefficients

Because this study is focused on the production side, it is necessary to consider

policy impact measures which are aimed particularly at production. Among these are the

Nominal and Effective Protection Coefficients (NPC and EPC) and Domestic Resource

Cost measures, all of which, it will be shown, are closely related. These can be traced

to early work by Barber (1955), Balassa (1965) and Corden (1966).

Price policy studies often rely on an aggregate commodity approach which uses

NPCs to assess the effects of direct or indirect price interventions on producer and

consumer welfare, by identifying the differences between domestic and international

prices (Bale and Lutz 1981, Jaeger and Humphreys 1988). These directly assess price

distortion, exchange rate distortion and tariffs. An NPC measure of protection for a

particular commodity produced domestically is achieved by the simple calculation:

NPC = d (1)

where Pd is the domestic market price per unit of output, and Pe is the border equivalent

price per unit of output. The NPC is thus a measure of the implicit tax or subsidy that

policy is imposing on a commodity. An NPC of less than one implies a tax, while an

NPC of greater than one implies a subsidy. Policies which are reflected in the NPC

5 Some parts of the methodology discussion follows closely the sources and
arguments as presented in Masters (1991), who provides a comprehensive theoretical
background to the PAM method.


include explicit tariffs and subsidies and exchange rate policies. Over time, overvalued

exchange rates will inflate NPC measures if they are not adjusted to reflect real exchange

rates. For this reason, Westlake stresses the importance of using shadow exchange rates,

allowing the NPC measure to incorporate the effects of exchange rate policy (Westlake


Bale and Lutz compared price distortions in a group of six developing countries

by this means (Bale and Lutz 1981). Producer welfare, however, was measured in their

study at wholesale prices which reflect combined changes to the welfare of distributors

and processors, as well as to producers. High domestic marketing costs can distort such

an NPC-based price policy evaluation. For instance, if a few wholesalers, processors

and retailers enjoy market power, the benefits of price liberalization may be captured at

the post-farm level, rather than farm level.

It is thus critical that border prices are adjusted to specific market levels by

adjusting for marketing costs. An NPC or other similar measure, to be meaningful, can

only be defined for a particular, physical, market location. Westlake maintains that the

identification of market level and location has been ignored in many studies using NPCs.

He suggests that this is often done because the additional effort involved is thought to

outweigh the benefit. For a specific market level and location, the NPC is defined as,

in the case of an exportable:

NPC, =d d (2)
P -C -C P
*w 'm t ^be


where C. is the unit market margin between market level and border market level, C, is

the unit cost of transport between the market location and the border, and P, is the

world market price of one unit of output. Similarly, in the case of an importable, NPC


NPC = (3)
Pw + Cm + C,

Williams examined livestock pricing policies in a group of five sub-Saharan African

countries by tracking changes in Nominal Protection Coefficients based on farm-gate

rather than wholesale prices (Williams 1993). Although Williams' results show that

product price incentives to producers have been improving in the countries studied, since

input costs are not considered it is difficult to know if producers are benefiting as much

as Williams' results suggest. Westlake addresses some of these limitations of the NPC

approach, and stresses the importance of incorporating input costs, thus using Effective

Protection Coefficients (EPC) to measure the effects of policy (Westlake 1987).

Westlake also points out that NPC studies have often ignored the need to adjust producer

prices by marketing costs to a particular market location.

Incorporating into the NPC the effects of inputs costs results in the Effective

Protection Coefficient (EPC), which is thus:

EPC = Fd (4)

where Fd are tradable inputs costs per unit of output at domestic market prices. Again,

all border equivalent prices should be adjusted for a particular market level and location.


The EPC examines the implicit tax or subsidy imposed by policy on the value added to

the commodity. Policy effects which are reflected, additional to those considered by the

NPC, include explicit taxes and subsidies on inputs. The effectiveness of the EPC is still

subject to the use of appropriate exchange rates, and to adjusting domestic prices for

marketing costs. Westlake further suggests that, by calculating EPC measures for

various points in the marketing chain, price distortions within the marketing system can

be identified. These are too often overlooked by use of an aggregated protection

coefficient which considers only the national market as a whole, as in the case of


Domestic Resource Cost Analysis

At about the same time that protection coefficients were being developed,

measures of comparative advantage based on domestic resource costs were being

formulated, particularly by Chenery (1961) and Krueger (1966). These have been used

as both an ex post measure of the costs of distortion, and as an ex ante criterion for the

allocation of resources, particularly in project planning (Pearson 1976). These were able

to incorporate policy analysis through measuring distortions, reflecting policy effects on

the international and domestic competitiveness of domestic production of a particular


Following Chenery (1961), comparative advantage exists if the opportunity cost

of producing a unit of the commodity is less than the commodity's import price.


F *E + D < P,*E (5)

where: F, = direct and indirect tradable inputs (at border equivalent prices, in

foreign currency) per unit of output

E = exchange rate (domestic/foreign currency)

D = direct and indirect domestic factor costs per unit of output

Pb = border equivalent price of the import per unit

These can be rearranged to evaluate the cost of domestic resources saved through

domestic production per unit of output. Domestic Resource Cost is thus:

DRC = D < E (6)
Pb Fb

Expressed this way, a country has a comparative advantage in the production of an

importable commodity if the ratio of the opportunity costs of total domestic factors used

per unit to the net foreign exchange saved per unit is less than the exchange rate. In this

form DRC also can be used to measure the relative comparative advantage of competing

activities within a country. It was first employed in this way in the 1960s by Israeli

economic planners (Bruno 1972). Given planned changes in exchange rates or tariffs,

the DRC can in this way be used as an ex ante measure of the impact of policy change.

An adaptation of the DRC concept allows comparisons to be made for levels of

competitiveness among different countries. If each side of equation 2 is divided by E,

the criterion for comparative advantage becomes:

DRC <1 (7)

This is the Domestic Resource Cost Ratio (DRCR), and can be used to assess the

'relative' comparative advantage of one country versus another in a single commodity,

at either overall or individual market levels. A DRCR of less than 1 indicates a

comparative advantage in the given activity with respect to both the world market and

alternative domestic activities. In cross country comparisons, the country with a smaller

DRCR will have a greater comparative advantage in production of the good (or

processing/marketing, depending on market level).

If economic or 'social' prices are used in calculating the DRCR, then the full

opportunity costs of domestic resources would be reflected.6 Social valuation would not

necessarily use market prices, but would attempt to value factors under competing

alternative uses. The resulting DRCR would be a closer assessment of underlying, and

even medium-term competitiveness.

Social prices can also be adjusted to counter the effects of policies such as

exchange rate distortions, tariffs, price controls, etc. A close examination of opportunity

costs in inputs and factors can even capture some effects of domestic market failure. A

comparison of DRCRs under social and 'private' (i.e. under existing market and policy

conditions) values would provide some meaningful, although static, ex post assessment

of the impact of existing policies and possibly of market failures. Therefore, besides

6 The terms 'social profitability' and 'social costs' are first used by Chenery
(1961). These value inputs and outputs under their full domestic and international
opportunity value, under optimal market and policy conditions.


being used as an ex ante measure of comparative advantage, the DRC method can also

be applied, using input-output analysis, as a systematic ex post measure of the cost of a

restrictive trade system or as a measure of effective protection of various goods. In this

form it was used by Krueger for Turkey (1966).

Chenery points out that comparative advantage arguments must be tempered by

a consideration of non-market intersectoral effects related to economies of scope. Under

this argument, development of an industry without comparative advantage can still have

benefits in terms of new skills and industrial base development. Since such factors may

be difficult to quantify, however, essential comparative advantage measures must lie at

the base of policy decisions, with consideration of these other factors particularly when

no clear advantage or disadvantage exists. Further, comparative advantage provides a

good basis for policy analysis, because not only does it consider the impact of policy on

the international competitiveness, important in this era of falling trade barriers and export

reliance, but it also evaluates the long-term viability of the economic activity, even in a

domestic context.

Krueger (1972) and Bruno (1972) show that the DRC and EPC are closely related

measures of policy impact. The EPC will be identical to the DRC in the extreme case

where all importables are actually traded, there is perfect competition in domestic factor

markets, and all prices reflect marginal rates of transformation among goods (Bruno

1972).7 A DRC calculated in this way will reflect the import-opportunity cost of

7 See Pearson (1976) for a rigorous comparison of the EPC and DRC measures.


domestic production. Further, both of the measures can be examined over time, by using

time series data deflated by the appropriate indices.

The Policy Analysis Matrix

Building on this earlier work, and recognizing the close relationship between DRC

and EPC measures of comparative advantage and policy impact, Monke and Pearson

(1989) developed a framework in which these measures can not only be compared, but

also decomposed by market level and technology type.8

The Policy Analysis Matrix (PAM) methodology developed by Monke and

Pearson provides a systematic framework to identify patterns of incentives for economic

agents under alternative technologies at each level of the commodity system. It thus

enables the analysis of the impact of policy on these incentives for alternative

technologies at each level in the marketing chain. At each market level, and for each

technology, the PAM compares private (existing market) prices with social prices (those

potentially available in an open market).

Structure of the PAM

PAM uses the concept of profit as its main point of analysis. While revenues

may be straightforward, costs must be broken down into two categories, tradables and

domestic factors. Tradables are those inputs which are available on the world market,

which serves as a basis for their valuation. Primary domestic factors, such as land, labor

I For recent applications of the PAM method, see Nelson and Panggabean
(1991), Masters (1991), or Gumaa et al. (1994).


and capital are non-tradable and thus are treated separately. These cost and revenue

structures are presented in the form of a matrix (the PAM), which allows for easy

presentation and interpretation of results (Table 2-1). The focus of the matrix and its

analysis is the identification of divergences between existing market (private) values and

socially-desirable (in terms of more efficient use of resources) values.

Table 2-1 The Policy Analysis Matrix (PAM)

Value/Ton of Costs
Commodity Revenues Profits
Tradable Domestic
Inputs Factors

Private Values A B C D'

Social Values E F G H2

Effects of I3 J4 K5 L6

Private profits (D) = A (B+C)
SSocial profits (H) = E (F+G)
SOutput divergences (I) = A E
Input divergences (J) = B F
SFactor divergences (K) = C G
Net divergences (L) = D H = I (J+K)
Source: Monke and Pearson 1989

Private values are the observed quantities and then market costs and returns.

These implicitly include the effects of all policy interventions, in both direct and indirect

subsidies and taxes, and all market distortions and failures. They are identical to the

'financial values' of cost-benefit analysis.

Social values, in contrast, provide a benchmark policy environment for

comparison. Social values can generally be considered those which would hypothetically


occur in a free market without policy intervention. Masters (1991) refers to them as

'national opportunity values, to underline their opportunity cost nature. They are

equivalent to the economic, as opposed to financial, value of resources and products in

the context of their value on the world market and the underlying scarcity and

opportunities for use of resources domestically. The difference between social and

private values, shown in row 3 of the matrix, represents the divergence between

theoretically optimal policy results and the observed policy results. These divergences

are the result of a variety of factors, the most important of which are direct and indirect

policy interventions, but also include the effects of rent-seeking behavior, macroeconomic

factors, and market failure. The examination of the extent and source of these

divergences is the focus of the PAM analysis, as the divergences reveal the impact of

government policy on the economic activities being studied. Policy interventions include

all those identified earlier in this chapter, both deliberate and unintentional.

Removing distortionary policy, however, may incur some cost. Corden states that

removing a policy restriction is optimal only to the point where marginal benefits equal

marginal costs (Corden 1971, Masters 1991). Masters thus declines to equate them to

free market results, preferring instead to accept the realistic constraints imposed by the

costs of policy change, and considering them the best available alternative.

Social values represent then the standard by which to measure the policy effects

inherent in the private values, and are shown in the second row of the matrix. The

resulting social profits illustrate the relative competitiveness or comparative advantage

of the economic activity. Such comparative advantage is an indication of the long-term


viability of the economic activity, particularly if policy support of the market is seen as


Indicators of Comparative Advantage. Protection and Policy Impact

Based on the domestic resource cost and effective protection methodologies which

are implicit in the PAM, DRCRs, NPCs, and EPCs can be calculated directly from the

matrix. Since these are ratios, or 'pure numbers', they can be used as a basis of

comparison between either different commodity systems or between similar systems in

different countries or settings. Monke and Pearson call them Ratio Indicators for

Comparison of Unlike Outputs, and based on Table 2-1, they can be derived as the


1) Domestic Resource Cost Ratio = DRCR = G/(E-F). Because exchange rates

have been incorporated into the PAM results, this measure is equivalent to that in

equation 7 on page 32. This is the relative cost, in terms of domestic resources, of

producing one unit of output instead of importing it. Any DRCR less than 1 indicates

an international comparative advantage in production of the commodity under study. A

DRCR smaller than the DRCR, similarly measured, of an alternative domestic activity,

shows a relatively greater international comparative advantage. In cross-country

comparison, this indicator can serve as a measure of the relative efficiency of domestic

resource use (Fox et al. 1990).

2) Nominal Protection Coefficient (NPC) on tradable outputs = NPCO = A/E,

and on tradable inputs = NPCI = B/F. These are equivalent to either equation 2 or 3

of page 28, depending of whether the good is an importable or exportable. The NPCs


compare domestic and international prices, adjusted for marketing costs and exchange

rates. Any differences in these prices will result in an NPC not equal to 1, and can be

attributed to the policy regime and market efficiency. NPC less than 1 indicates an

implicit tax on production (subsidy in case of inputs), and NPC greater than 1 indicates

implicit subsidy on production (tax in case of inputs).

3) Effective Protection Coefficient = EPC = (A-B)/(E-F). This is equivalent

to equation 4 on page 29. The EPC combines the above two measures to assess the

overall effect of implicit tax and subsidy through both output and inputs. The result is

a ratio of value added at private prices to value added at social prices, indicating the

effect of protection on value added. It is thus a more meaningful measure of policy and

market impact than either of the NPCs.

Non-ratio indicators include International Value Added = IVA = E-F which is

the value, net of tradables, of domestic factors and social profit per unit of output.

Social Profitability (E-(F+G)), measures social return per unit of commodity beyond

normal returns, which are implicit in the value of domestic factors.

All of the above measures can be used for both cross-system and cross-country

comparisons. For example, if alternative marketing channels exist for a commodity, the

indicators can be used to evaluate not only the relative desirability of each system, but

also the relative impacts of existing and potential policy interventions. Similarly, the

relative impacts of different policies between national systems can be evaluated.

Sensitivity Analysis

The comparison indicators can also be used to determine the sensitivity of private

and social incentives to changes in parameters, in this case those affected by policy

intervention. Of most use in this case are the EPC and the DRCR, which can assess the

policy impact on social protection and returns to the use of domestic resources. The

ability of policy intervention to improve these profit incentives can be measured by

sensitivity analysis. The effects of changes in such parameters as tradable input costs

(due to changes in import duties), output price (such as producer milk price), can be

examined ex ante, assuming that in the short run substitution effects will not be

significant. This would be done by increasing the cost coefficients in the PAM matrix

in turn by 1 percentage point and measuring the percent change in EPC and DRCR.

Identifying Policy and Market Divergences

An important characteristic of the PAM, and one which will be taken further in

this study than in any known previous study, is its ability to disaggregate and quantify

multiple policy and market divergences. Where the data render this possible, the specific

impact of each policy intervention or other market factor can be assessed. However,

whereas direct policy intervention may usually be quantifiable, the effects of market

failure may be less so.

Market failure can take a variety of forms, and be due to any of a number of

factors, including public policy interventions, particularly with regard to regulatory

policies. Complete market failure may exclude the possibility of some good being

transacted in a given area. On a lesser scale, partial market failure may allow some


actors in the market to earn excess profits, while others incur undue costs. Such failures

can thus be conceptualized as of three types: 1) partial market failure due to private rent-

seeking, 2) partial market failure due to public rent-seeking, and 3) complete market


Partial failures due to private rent-seeking are caused by the traditional

circumstances of oligopoly, monopoly, and monopsony, which in most cases are in turn

caused by some underlying institutional factors such as licensing, property laws and

absence of litigation. They may in some cases result from structural factors, such as

natural monopolies due to cost structure, or imperfect information in markets. The costs

to end-users, however, result from the rent-seeking; the underlying factors simply allow

the rent-seeking to occur.

Partial failure due to public rent-seeking takes the form of corruption by

individuals/institutions in the public sector who can charge unofficial payment for

services or resources which they control. This is caused by institutional factors in the

form of inadequate control of official behavior, or tacit approval of such behavior.

Complete market failure occurs when the effect of the institutional or structural

factor is so great that exchange of the good or service ceases completely for all or some

agents in the market. These are caused by "non-marginal" factors in that they do not

affect marginally the price of a good, but rather simply halt the exchange of that

good/service. An example would be smallholder non-access to credit because of lack of

collateral, perhaps required in the form of titled land.


Policy may contribute to the incidence of these market failures by creating the

conditions necessary for them to occur. A system of market information, for example,

may be a public good which is unlikely to be supplied by the private sector. Inactivity

of the government in supplying market information may lead to market failures as

described above. Market power may be inadvertently bestowed on some market actors

through bureaucratic regulations. Mid-level bureaucrats with the authority to control

sales of agricultural inputs, for example, and who are not considered to be market actors,

nevertheless collect rents through their market power. The result is price discrimination

and market failure.

Policy analysts are presented therefore with a potentially complex combination of

multiple and conflicting policy interventions, qualified by various degrees of effective

implementation, and further obscured by either contributing or detracting market failures.

Using this understanding of various market failure effects, as well as identification

of specific direct policy interventions, an expanded PAM matrix can be obtained. This

expanded PAM, illustrated in Table 2-2, breaks overall divergence down into those

resulting from specific policies and those market failure effects which are quantifiable.

The specific group of divergences chosen for identification in this study are

administrative prices, taxes and subsidies, public rent-seeking (corruption) and

identifiable market failure. Corruption takes the form of unofficial payments made to

public officials. Market failure divergences are those differences between social and

private prices which cannot be attributed to other factors.

Table 2-2 The Expanded Policy Analysis Matrix

Value/Unit of Commodity Costs
Revenues Profits
Tradable Domestic
Inputs Factors

Private Values A B C D'

Social Values E F G WI

Effects of Administrative Prices Ml N1 01 P1

Effects of Direct Taxes M2 N2 02 P2

Effects of Direct-Subsidies M3 N3 03 P3

Effects of Corruption7 M4 N4 04 P4

Effects of Market Failures' M5 N5 05 P5

Effects of All Divergences I3 J' K5 L6

Private profits (D) = A (B+C)
2 Social profits (H) = E (F+G)
3 Output divergences (I) = A E = Ml + M2 + M3 + M4
4 Input divergences (J) = B F = NI + N2 + N3 + N4
SFactor divergences (K) = C G = 01 + 02 + 03 + 04
6 Net divergences (L) = D H = I (J+K) = P1 + P2 + P3 + P4
7 Corruption is due to public rent-seeking, as described above.
s As defined here, market failure is a residual which captures all divergences not
otherwise identified.
Adapted from Monke and Pearson

It should be noted that other divergences could be identified if they were

quantifiable and considered important. These could include the effects of exchange rate

distortions, costs of inadequate infrastructure, etc.

PAM Analysis of Systems

The matrix as described above can be applied to any productive activity which

entails costs and revenues. PAM matrices for various activities within a commodity


system, such as a periurban dairy milk production, processing and marketing system, can

also be combined. PAMs for various technologies or market channels can be combined

horizontally across a market level, such as milk processing, create a PAM for the entire

processing level. These would be combined weighted on the basis of their volume share

of that market level. In Figure 2-1, this is illustrated in the first column, where various

production alternatives are combined into one production level PAM at the bottom of the


Activities carried out vertically within one marketing channel could also be

expressed in a PAM which combined the vertical elements of the channel. An example

of vertical aggregation is the sequence of activities which take place in the production

and marketing of periurban dairy: farm production, farm-to-processing, processing, and

further marketing and first row is expressed in the last PAM on the right.

Finally, PAM matrices can be estimated for each of these activities and combined

both vertically and horizontally to form a entire system analysis (bottom left PAM in

Figure 2-1). Such an analysis allows insights into the aggregate policy effects on an

entire commodity system, from producer to consumer. Further, horizontal aggregate

analysis can be carried out at any activity level in the system, such as at the production

level, so that policy effects on producers as whole, and separate from other actors in the

market, can be evaluated.

The combination of commodity-chain and activity alternative analysis allows study

of the various combinations of production and marketing which exist in the periurban

dairy system. Further, it allows an examination of policy effects on each commodity

Production Collection Processing Marketing Commodity

Alternative I A B, C D, A BCD, A, B, C, D, A4 B4 C4D4 ABCD
E, F, ,H, 4 E, F,20H, E, F, G, H, E4 F4 0. H, EFGH
I, JK,L, ,J2K2 L K KL, J, K4 L4 IJKL

Alternative II

Alternative III

Activity Level Entire Dairy

Adapted from Monke and Pearson

Figure 2-1 Illustration of Method of Combining PAM's for Activity Levels, Market Channels and
the Commodity System


chain, and on each activity level, as well as those impacts on each individual element of

the periurban dairy industry.

Estimation of PAM Coefficients

In order to make use of the PAM methodology, detailed budgets must be

estimated for not only the marketing and production of the main commodity being

studied, but also for the most important inputs. This is needed in order to break down

inputs into their tradable and non-tradable components.9 Once all components have been

identified, they must be valued in private and social terms. Private values require

measuring market prices of all inputs and outputs. Social values require an estimation

of long-term world prices for inputs and outputs, knowledge of all explicit taxes and

subsidies, and some valuation of domestic resources under alternative uses. They may

also require estimation of the costs due to market distortions and rent-seeking.

Calculation of social values may at times be problematic. In the case of many

tradable inputs, market prices are adjusted by identifiable distortions, such as taxes,

subsidies, and public corruption, to obtain social values. This is only justifiable in cases

of inputs for which there is no evidence that market failures exist. Otherwise, import-

parity prices would be calculated in the case of importables, and export-parity values in

the case of exportables. Social valuation of domestic factors is usually based on their

value under the most common alternative activity, which is considered to establish their

opportunity cost. This shadow value of factors is the basis of social profitability

9 See Appendix D for an illustration of the Structure of Activity Budgets,
showing the process of decomposition of inputs into tradables and factors.


evaluation, as defined by Chenery (1961). Using the value of factors under the highest-

valued alternative activity is not acceptable, because of capital or technology constraints

to their use under such an activity.

Strengths and Weaknesses of the PAM Method

The issues that PAM considers tend to be the ones central to agricultural policy

questions; the issues of comparative advantage, relative profitability and cost, and how

those are affected by policy interventions (Monke and Pearson 1989). Further, PAM

identifies these features and presents them in a simple and understandable manner,

allowing policy-makers convenient access to the most important facts to be considered

in their decision-making. Governments often institute policies that are overlapping and

contradictory. Policy makers may thus be unaware of the relative magnitudes of various

policy effects on individual commodities and levels in the commodity chain. The simple

and convenient manner in which policy effects, multiple and overlapping, can be

presented to policy-makers is one of the strengths of and principal motivations for the

development of the PAM (Monke and Pearson 1989). Furthermore, the PAM uses

simple budget data from representative farms instead of requiring time-series data of

prices and marketed quantities, often difficult to obtain in a developing country setting.

Further, by considering the value of tradables on world markets, the PAM

incorporates the protection coefficient approach. The inclusion of production costs at

each level allows Effective Protection Coefficient estimation. Furthermore, since the

PAM evaluates each level of a commodity marketing chain, comparisons between farm

and post-farm welfare changes can be made. This is equivalent to measuring EPC at


multiple market points to identify internal price distortions, as recommended by

Westlake. DRCRs can thus also be calculated at multiple market points, extending that

analysis further than it is usually taken.

Finally, while the standard EPC considers all inputs to be tradables, the PAM

produces EPC measures which recognize non-tradables such as domestic factors, and

values them at their domestic opportunity costs. Bruno (1972) declares this "social" EPC

to be a more robust use of the concept in the face of problems caused by factor

substitution, when non-tradables are assumed to be tradable.

The primary shortcomings to the PAM method are centered on the static nature

of the analysis. Not being a behavioral model, and limited to fixed input-output

coefficients, it is unable to predict the market changes which could be expected to occur

under social pricing. It thus is only able to assess incentives for behavioral change,

without predicting the extent of change (Masters 1991). Substitution or supply response

effects accompanying price changes are ignored and unknown, as are any economies or

costs associated with changes in scale of the productive activity. If policy changes which

would lead to changes in these levels are considered, these limit the PAM's implications.

Further, even under existing market conditions, the social prices used do not necessarily

represent an actual equilibrium (Masters 1991). This is due to the nature of the

calculation of social prices independent of any general equilibrium model. Masters

suggests that that alternative, predicting multi-sectoral price changes and resource

reallocation, is empirically very difficult. Under these circumstances, given limited


resources at hand, the PAM provides some measure of divergences in individual markets

which may have some relevance in the short run.

A further weakness of the PAM is its inability to adequately address institutional

factors which have a ion-price affect on producer or marketer behavior. Since PAM

deals only with issues that affect prices or economic returns at the margin, supplementary

analysis is necessary to consider non-marginal or non-price factors (Monke and Pearson

1989). The use of complementary institutional analysis in this study will aid in

overcoming this weakness.

While the PAM analysis remains essentially a static, ex post examination of

measurable policy intervention, application of sensitivity analysis allows some ex ante

assessment of potential policy change. The divergences as measured by the PAM can

only be considered as incentives, however, not economic impacts in a welfare sense.

Without an understanding of product and price relationship over time, sensitivity analyses

based on these incentives can only be considered appropriate within narrow ranges and

for the short-term.

Data Sources

Farm-Level Data

Data at the farm level were gathered principally by means of surveys of dairy

producers. Nine survey sites were selected, 5 of them urban (within the city limits of

Addis Ababa) and 4 periurban (within 60 kms of the capital). The urban sites were

Weredas (local administrative units) numbers 10,12,17,23 and 24, and were selected so


as to represent different socio-economic and altitudinal zones of the city. The periurban

sites were the areas of Sendafa, Chencha, Holleta, Sabeta, located along 4 of the major

roads leading out of the capital, and recognized as important supply areas within the

Addis Ababa milkshed. They also represent different altitudinal and climatic zones of

the milkshed. In this way, the two main technology types were represented in the

survey; open grazing, found in the periurban area, and zero grazing, practiced by urban


Since most producers, especially those in urban areas, operate on the informal

market they have little contact with extension services or regulatory agencies. Thus, at

the time of the survey (1992/1993) no baseline data were available with which to define

the survey population."o In this case, a grouped-accidental survey method was

employed. Using this method, dairy producer households were located on an accidental

basis within the target zone. A dairy producer household supplying milk to the periurban

milkshed was considered to be one which fulfilled the following criteria: 1) possessed at

least one cow, either local or cross-bred and 2) marketed at least some production as

milk or milk products. An attempt was made to locate and survey large producers in

each zone.

The survey instrument was comprised of two parts, administered several weeks

apart, and relied on respondent recall regarding all milk production and marketing

10 Subsequent to the survey, data on nearly the entire dairy producer population
became available from the Addis Ababa Dairy Producers Association (AADPA).
These data were used for estimation of the numbers of each producer type in
characterizing the dairy system.

activities for the last 12 months. After the first part of the survey was completed in a

given area, those producers judged as reliable were revisited for the second part. Some

questions, particularly regarding milk production and sales, were repeated in part 2 to

verify accuracy. The issue of respondent reliability is discussed in detail below. The

enumerators consisted of 3 extension officials from the Ministry of Agriculture (MoA),

with which a cooperative agreement had been signed, 2 agriculture MS students, and an

ILCA research assistant. All were trained and supervised by the principal researcher.

Use of Subjective Reliability Indicators

The primary source of potential inaccuracies in the survey was judged by this

researcher to be false statements by respondents. These could be either deliberate or

accidental. Accidental errors could occur through errors in recall, in response to

questions regarding previous milk.sales or feed purchases, for example. These could be

expected to fall randomly around the population mean and so should not effect the results

of a large.sample. Deliberately false statements, however, might be biased to one side

of the mean, and so could not be rectified by sample size. In this survey, the

enumerators were often viewed by the respondents as either potential sources of material

assistance, or potential reporters of informal and so unregistered activity. Thus they

were prone to under-report milk yields and sales, out of fear of paying the taxes they

were officially subject to but had never paid. They also had some incentive to over-

report the costs of feed and other input purchases, in an attempt to solicit assistance.

These tendencies were observed during the testing of the survey instrument, and so great

effort was made to ensure reliability to the extent possible. Reliability indicators were

thus included in the questionnaire. The enumerators were trained to look for signs of

false responses, such as contradictory responses, differences between feed purchase

responses and feed materials observed in the dairy shed, lack of milk-measuring

containers, etc. The point was stressed, however, that respondent illiteracy, simple

production technology, etc., were not equated with non-reliability. At the end of the

feed and milk production sections of the questionnaire, the enumerator subjectively

assigned a simple score of 1 to 3 as to the reliability of the responses, and reported the

reasons for the score. An overall reliability score was also given for the rest of the

questionnaire. These scores were averaged and formed a reliability score for the

respondent. A statistical clustering method was then applied to identify a natural division

between the upper two-thirds and lower third of the scores. In this way, about 1/3 of

the respondents were rejected. Although subjective, this method was considered to

enhance the accuracy of the survey by removing, to the extent possible, non-random

variation in the producer responses.

Using these methods, 231 respondents were interviewed. Some producers were

rejected before the questionnaires were completed because of unwillingness to cooperate.

Of those contacted, interviews were completed for 196 producer households. On the

basis of the reliability scores, the data for 138 respondents were used in the analysis.

Data Grouping

Besides being segregated into periurban and urban producers simply by location,

the data showed differences in their resource base, as periurban producers have

agricultural land, while urban producers do not. In order to capture differences in scale


economies or management/technology, the respondents were further differentiated into

large and small producers. This was done on the basis of herd size. Total herd size was

converted into Tropical Livestock Units (TLUs)," and these figures were subjected to

a statistical clustering method, which divided the respondents into two natural groups.

By this means, large producers were determined to be those with cattle amounting to

more than 9.45 TLUs while small producers had less than this number. Dividing the

sample this way, 19 respondents were deemed large and 114 small.

Subsequent to the survey, data became available and were obtained from the

recently-formed Addis Ababa Dairy Producer Association (AADPA). These comprised

herd composition data for all members of the AADPA, which is thought to include the

large majority of dairy producers in the Addis Ababa region who market milk. These

data were combined with the survey data to make estimates of total milkshed production.

Post-Farm Level Data

Data on formal post-farm activities were obtained from DDE officials and

accountants, and comprised mainly of annual budgets dating from 1991. These data

include collection, processing and marketing budgets. These budgets also supplied the

data for state farm dairy production activities. Two state dairy farms were visited.

Informal post-farm activity data were gathered by means of interviews with

several urban informal dairy processors, and three periurban dairy processors engaged

1 Standard assumed conversion factors used in calculating TLUs were based on
Gittinger, where breeding cow=l, bull= 1.2, and calves from 0.3 to 0.8 depending
on age (Gittinger 1982).


in butter and soured milk production. Data were also obtained from those producers

engaged in dairy processing on-farm.

Other Data

Data were gathered from various and numerous Ethiopian government agencies,

including the Central Statistical Office, Agricultural and Industrial Bank, Railway

Transport Authority, Ethiopian Freight Transport Corp, Revenue Authority, and the

official gazette, the Negarit Gazeta.

To gather data on the input supply market, interviews were held with managers

of major government flour and oil mills, whose by-products make up an important source

of feed. Owners of private oil mills were also interviewed. Interviews were also held

with managers of government feed mills, and the single operating private feed mill. The

head of the government Agricultural Input Supply Corp was interviewed, as were private

importers of veterinary supplies. Government veterinarians were interviewed, as well

as the sole recognized private veterinarian. The head of the Government of Ethiopia

(GoE) artificial insemination service was interviewed.

At the policy-making and implementing level, data on official policy were

gathered from many of the above officials, as well as others from the Dairy Development

Enterprise (DDE), the Dairy Rehabilitation and Development Project (DRDP), and the

World Food Program (WFP). During the survey procedure, a number of agricultural

extension agents were interviewed, providing insight into the policy-implementation



This chapter describes periurban dairying in Ethiopia, beginning with a historical

account of its evolution. It goes on describe dairy production and marketing in the Addis

Ababa milkshed, as well as feed markets and livestock services.

Evolution of The Dairy Industry and Dairy Policy In Ethiopia

Ethiopia displays an ancient tradition of dairy production in both highland and

pastoral areas, for consumption and for trade in traditional soured butter. Commercial

liquid milk production, however, only began in the 1950s with the introduction of exotic

animals, mainly from Kenya. This commercial production was begun by indigenous

Ethiopian landowners who established large farms in the Addis Ababa and Asmara areas,

and began marketing milk informally.

Early Dairy Policy

In 1960, a pilot processing plant, started at Shola on the outskirts of Addis Ababa

by UNICEF, began to process the milk being produced by the large farmers. In a short

time the plant was expanded, and collection centers were established along roads leading

out of the city. Besides allowing for the establishment of large farms further from the

city, this collection system began buying milk from smallholder producers with


This chapter describes periurban dairying in Ethiopia, beginning with a historical

account of its evolution. It goes on describe dairy production and marketing in the Addis

Ababa milkshed, as well as feed markets and livestock services.

Evolution of The Dairy Industry and Dairy Policy In Ethiopia

Ethiopia displays an ancient tradition of dairy production in both highland and

pastoral areas, for consumption and for trade in traditional soured butter. Commercial

liquid milk production, however, only began in the 1950s with the introduction of exotic

animals, mainly from Kenya. This commercial production was begun by indigenous

Ethiopian landowners who established large farms in the Addis Ababa and Asmara areas,

and began marketing milk informally.

Early Dairy Policy

In 1960, a pilot processing plant, started at Shola on the outskirts of Addis Ababa

by UNICEF, began to process the milk being produced by the large farmers. In a short

time the plant was expanded, and collection centers were established along roads leading

out of the city. Besides allowing for the establishment of large farms further from the

city, this collection system began buying milk from smallholder producers with


indigenous cattle. During the rest of the 1960s dairy production in the Addis Ababa area

developed rapidly, both through this new formal channel and informally. Although some

dairy cooperatives were established around this time, they were not to assume the

important role that cooperatives did in Kenya (FAO 1972).

The Addis Ababa Dairy Industry (AADI) was established by the government in

1966 to control and organize the collection, processing, and distribution of locally

produced milk. By 1969, the Shola plant had been expanded with help from UNICEF

to a daily capacity of 30,000 liters. During this period, several government owned dairy

farms were also established, with the main aim of supplying the formal market and as

serving as demonstration farms for the training of commercial farmers.

In 1971 a new parastatal body, the Dairy Development Agency (DDA) was

established. Besides taking over the AADI's responsibilities for milk collection and

processing, DDA was given the further mandate of providing producer assistance to

increase milk production, and expanding the formal milk market to urban areas outside

of Addis Ababa. The DDA was overseen by a Dairy Development Board consisting of

senior official from various ministries (FAO 1974).

Up to this point in time, dairy policy had not been clearly articulated. The AADI

had been designed with the rather simple goal of collecting and processing milk which

was being produced by a spontaneously emerging dairy industry. The launching of the

World Bank's Addis Ababa Dairy Development Project (AADDP) in 1971 required,

however, a more carefully planned and articulated dairy program. Of major concern to

policy makers was the large cost of milk imports. At this time, unlike neighboring


Kenya, Ethiopia was heavily deficient in dairy products. Between 1969-71, for example,

the value of dairy products imports was greater than any other imported food ;uff (FAO

1972). In line with the prevailing development philosophy of the day, import-substitution

was the major objective, targeted towards the periurban market. The potential for

exports was recognized, based on the relatively low costs of production (FAO 1974).

The dairy policy which sought to meet these objectives was heavily geared

towards large commercial dairying in the periurban area. This fit in with the 3rd Five

Year Plan in effect then, which had as one of its goals the development of commercial

agriculture. This policy also suited the social structure of the period, which saw all

industry, including the emerging dairy, controlled by the traditional ruling class (FAO

1972). The public formal processing system supported the private dairy production of

politically-connected entrepreneurs.

The AADDP, therefore, was aimed largely at developing commercial dairy,

although some provisions were made to assist smallholders to develop alongside the

larger producers. The primary objective was to increase production of milk for the urban

market by assisting the development of small and medium size commercial farms.'

Assistance would be provided in the form of credit, imported cattle, and technical

services. Breeding ranches would be established, collection would be extended to a

radius of 120 kms, and processing capacity at the DDA plant would be increased to

60,000 liters/day (FAO 1972).

Small farms were defined as those with an average of 10 milking cows each
and medium size farms with 40 milking cows each.


By 1972 the DDA was receiving about 21,000 liters/day, of which some 57%

came from 65 large farms, the remainder coming from smallholders through some 30

collection centers. These large commercial farms possessed herds of 10-250 cross-bred

Friesian cows, while smallholders typically held only a few indigenous Zebu cattle. At

that point in time there was no evidence of widespread use of cross-bred cattle by

smallholders (FAO 1972). The price paid to producers was set administratively by the


The milk and dairy products were sold through DDA kiosks and shops, and to

institutions. At the same time, large dairy farms supplied milk directly and informally

to institutional customers and retailers. A large informal market developed through

hawkers buying milk from the large farms and an estimated 1500 smallholders on the city

margins and selling directly to consumers. Surveys by the FAO in 1971 estimated that

of some 30,000 liters of liquid milk marketed daily, approximately one third passed

through informal channels (FAO 1972). While the DDA paid an average in 1971/72 of

Ethiopian Birr (EB) 0.35/1, a price of 0.4-0.45/1 was available on the informal market.

The DDA also promoted the creation of dairy cooperatives, each established

around a membership of 50 dairy producers. Unlike the milk collection objectives of

cooperatives in Kenya, these served mainly to assist with the extension of technical

services, and the provision of credit (FAO 1974).

By the early 1970s the dairy industry in the Addis Ababa area was well

established and growing. Surplus intake was even occurring at the Shola plant (FAO

1974). In 1972-73 there was a 30% increase in milk to Shola, due mainly to a 170%


increase in intake from large farms, as these enterprises developed (Purcell and Stotz

1980). Artificial insemination (AI) services were established in Shola, and the genetic

quality of even small producer cattle gradually improved. The existing informal market

was being accompanied by a growing and viable formal market. These developments

were not to last.

A Change in Regime and Policy

In 1974 the imperial government of Haile Selassie was overthrown and a period

of policy uncertainty began. All land officially returned to the state, and continued

tenure by dairy producers became uncertain. Most of the large farms were either

nationalized or annexed by the new Peasant Associations (PAs), and some of the owners

arrested or killed. Some large farmers were able to retain usufruct rights to their land

(Purcell and Stotz 1980).

Loans under the World Bank project to establish new commercial farms were

halted. The change in land tenure essentially precluded the granting of additional loans,

since land was the primary form of collateral required by the project. In late 1974,

recognizing the shift in circumstances, the project decided to concentrate on smaller

farms using cattle procured locally. Loans were instead given to the new Producer

Cooperatives (PCs) associated with the PAs (Purcell and Stotz 1980).

Technical services, characterized as good in 1974, declined rapidly as senior

personnel were transferred repeatedly or arrested. Labor difficulties plagued operations

at the Shola dairy plant. Farms which had been using AI reverted to bull services, or

some use of AI with bull back-up. Further, AI services became targeted at serving the


national farms. By 1978/79, only 127 of total 2583 inseminations were made on private

farms. Veterinary visits to private farms also declined (Purcell and Stotz 1980).

Predictably, the adverse affects of these changes on milk production became

apparent. Milk delivery from collection centers peaked in 1974 and then declined.

Prices paid to producers were held at nominal 1972 levels. By 1978, milk intake at the

Shola plant had fallen to 3.5 million liters/year from 5.8 million liters in 1974 (Purcell

and Stotz 1980).

Further, by the late 1970s dairy policies had been effectively altered to comply

with the by then established philosophy of the regime. At the scheduled end of the

World Bank project in 1979, DDA and MoA officials understood, in the absence of any

written proclamation, that it was against government policy to encourage any individual

ownership of dairy cows. The promotion program ceased altogether, and any further

encouragement of smallholder production would only occur through collectives (Purcell

and Stotz 1980). Official assistance was further narrowed in scope by the creation of

Producer and Service Cooperatives, which over time became a focus for dairy

development efforts.2

Another development which occurred during the 1970s was an increase in the

price of concentrate feed. In milk production systems such as the Addis Ababa periurban

system, which do not rely heavily on pasture grazing or fodder crops to provide animal

2 Producer Cooperatives (PCs) were formed by the pooling of peasant resources,
including land, which were then run collectively. Service Cooperatives (SCs)
marketed output, distributed inputs, and provided warehousing, credit, etc. By 1990
when some 5 million farmers were members of PAs, 6.5% also belonged to PCs, and
nearly all were members of SCs (Alemayehu 1992)


nutrition, the milk price/concentrate price ratio (MCR) is critical to economic viability.

In Ethiopia the MCR was traditionally low, due to an extensive cattle industry and

resulting low demand for concentrate feed. In the mid-1970s Ethiopia began exporting

oilseed cake to Eastern European countries, and concentrate feed prices began to rise and

the MCR to fall (Purcell and Stotz 1980).

The rise in feed price was likely exacerbated by a shift in milk production from

periurban land-based peasants to urban back-yard producers. This occurred as, during

the period after the change in regime, the structure of the periurban milk market shifted.

The role of the informal market in urban milk supply increased dramatically. By 1984

informal supply was estimated by WFP to be more than double the formal milk supply.

By that time, state farms were supplying 70% of the DDA milk, as commercial and

smallholder supply shifted to the informal market. The major suppliers became urban

back-yard producers, relying on only purchased feed, rather than periurban producers

with land resources. Driving this process was insecure tenure of land, under which PAs

no longer allowed farmers complete freedom of crop choice, thereby hindering the

planting of fodder. A large market for hay developed (Purcell and Stotz 1980). Milk

production thus shifted away from the rural feed base and closer to urban consumers, and

bypassed the formal milk collection and marketing system which remained geared

towards the rural areas.

Another structural change which occurred during the 1970s was an increase in the

number of cross-bred cattle held by smallholder producers. When large farms were


appropriated and given to PAs, many cross-bred cattle entered the local market, and were

bought by urban producers.

State Farm Emphasis

Beginning in 1982, the WFP launched a project to fund dairy development and

simultaneously better utilize the existing dairy processing capacity. This was done by

supplying milk powder and butter oil to be reconstituted into liquid milk by the Shola

plant, and marketed by the DDE. Some of the resulting revenues would then be

channeled into dairy development projects, as selected and supervised by a Dairy

Development Advisory Board set up specifically for the purpose.3 The funds were

projected to go 80% to dairy development, 10% for milk to vulnerable groups, and 10%

to cover the cost of recombination. At the same time, the Finnish international

development agency (Finnida) started a project to improve and expand the milk collection

system by building chilling centers, and to provide cross-bred heifers. Simultaneously,

the African Development Bank began the Dairy Rehabilitation and Development Project

(DRDP), aimed at distributing cross-bred cattle and improving forage use. Between

these projects, milk supply and capacity utilization were expected to increase (Empson

and Westerduin 1984).

Under the Mengistu regime, policies were inexorably directed towards the support

of state institutions. The DDA was replaced in 1979 by the Dairy Development

Enterprise (DDE), which took over its operations. Significantly, the DDE was placed

3 The former Dairy Development Board was discontinued under the Mengistu


under the administration of the Ministry for State Farm Development (MSFD). The

DDE was given responsibility for both dairy processing and collection, and the

management of the state dairy farms. The objectives for dairy development under the

DDE were altered, and to the existing goals of expanding milk collection and supply,

were added the goal of expanding the role and number of state dairy farms (Empson and

Westerduin 1984).

By 1988 a WFP report showed that official attention was heavily directed in favor

of state farms, to the detriment of small producers. Of the WFP funds that had been

allocated, 67% went to state dairy farms even though it was clearly stated in the project

plan of operations that the target groups were smallholders and cooperatives.

Expenditures towards small farmers and PAs only amounted to 13%, the remainder going

towards administration (Pronk and Tuinenburg 1988).

This emphasis on state farms was linked to the fact that of the 5 members on the

Dairy Development Advisory Board, 4 were representatives of the MSFD. More

fundamentally, however, were government policies which, between 1985 and 1989, led

to 63% of all public agricultural expenditures being directed towards state farms (Gizaw

and Amare 1992).

In spite of these investments in state dairy farms, their operations remained

inefficient, incurring heavy losses. Calf mortality rates on state farms were as high as

38%. Poor management contributed heavily to this as administration was highly

centralized, with managers required to refer to headquarters for even herd management

decisions (Dairy Development Policy Committee 1986). Production from state farms


remained stagnant or declined, from a high of some 6 million liters in 1983/84 to less

than 5 million liters in 1989/90.

Collection of milk from small producers declined until 1985/86, when the new

collection and chilling centers became operational. A small increase in 1984 in the price

paid to producers, from 0.45 to 0.5 EB/1 may have contributed to this rise. Collections

from small producers continued their decline after 1986/87. Beginning as early as 1980,

reports by WFP and others repeatedly highlighted the contribution of low producer prices

to the problem of poor supply (Purcell and Stotz 1980, Empson and Westerduin 1984,

Dairy Development Policy Committee 1986, Pronk and Abiy 1990). These concerns

were largely ignored by authorities. Prices paid to producers only increased by 43% in

nominal terms during the 20 years from 1972 and 1992. Even when prices were raised

in 1984 the stated objective was lowering unit costs of processing through increased

capacity, rather than an interest in producer welfare (Dairy Development Policy

Committee 1986).

Figure 3-1 shows milk supply volumes to the DDE during 1979 to 1993. DDE

processing levels were only maintained from 1987 to 1990 by utilization of the milk

powder supplied by the WFP, as can be seen by the large proportion of WFP powder

supply in the Figure 3-1 during that period. Capacity utilization never exceeded 60%

during any year during the period graphed.

Annual Liters of Milk (Millions

12 h ...




Collection Centers Large Farms -- State Farms
--- WFP Powder E Total

Source: Pronk and Ably, 1990 and DDE

Figure 3-1 DDE Milk Supply Sources and Quantity, 1979-1993.

Even those dairy development efforts geared towards rural producers became

narrowly targeted towards members of Producer Cooperatives, who only constituted

6.5% of PA members (Alemayehu 1992). For example, two thirds of the crossbred

cattle distributed by the DRDP project by 1992, the only form of dairy credit provided

then, went to PC members (Gizaw and Amare 1992). Similarly, the World Bank-funded

Fourth Livestock Development project, which began in 1988, identified as its 'front-line

implementation institutions' the producer and service cooperatives and PAs (World Bank


One result of the focus on state farms and cooperatives was the official

abandonment of urban producers. This took place in spite of 1988 estimates that showed


informal urban producers to be supplying 70% of the urban liquid milk demand (Pronk

and Tuinenburg, 1988). Urban areas contained no PAs or cooperatives to channel

assistance to dairy producers. Indeed, informal urban producers were not officially

recognized to exist. As a result, such producers had no access to the AI services which

were being extended to rural areas, and access to veterinary services only through the

Shola laboratory. Further, feed quotas extended to rural PAs were not available to urban


Dairy policy in the 1980s can thus be characterized as a severe misdirection of

effort. The focus of substantial resources on parastatal institutions yielded little benefit

to consumers or producers. Attempts to develop market-oriented dairying in rural PAs

were hampered by low producer prices and narrow attention on cooperatives. These

same attempts also led to a complete neglect of the most important producers, those

operating informally in the urban area, who were forced to seek the inputs and services

they needed without institutional support.

The Unravelling of Socialized Agriculture

By the end of the 1980s agricultural development in Ethiopia was stagnating. The

growth of agricultural production had fallen from 3.2% annually in 1973/74 to 1.9% in

1988 (Gizaw and Amare 1992). Recognizing, perhaps too late, that the system of

socialized agriculture under the PAs was in general not working, in 1990 the Mengistu

regime announced a plan to reorganize the Producer Cooperatives (PCs). Under the

reforms, PCs would be operated in a more democratic manner, allowing even the

dissolution of PCs if so voted by their members. Within 3 months of the announcement,


95% of PCs across the country had disintegrated, reflecting their unpopularity

(Alemayehu 1992). Collective property was either divided between members or sold,

and in this way, a large number of cross-bred dairy cattle came into the hands of small

private producers (Gizaw and Amare 1992).

In 1991 the Mengistu regime was forced out by a new Ethiopian People's

Revolutionary Democratic Front (EPRDF) government, which espoused economic

reforms in line with those favored by international donors. A period of popular

reaction against some public organizations followed the establishment of the EPRDF

government. Service cooperatives, which had been in decline after the unravelling of

PCs, became the target of looting. This was seen as a reaction against the authoritarian

attitudes of the former regime in its rural policies. Communal property which was

perceived as having been expropriated by the state was forcibly repossessed by peasants.

Cattle breeding ranches had large parts of their land repossessed by neighboring PAs

(Alemayehu 1992). The rapid collapse of milk supply from 1990 to 1992 may be

attributed to the increased insecurity during this period, culminating in the collapse of the

regime, and subsequent paralysis and uncertainty among official institutions.

The Current Policy Environment

The change in government of 1991 brought dairy policy changes only gradually.

The DDE has retained its role as the primary actor in the dairy market, although changes

have been proposed which may diminish that role over time. The direction of dairy

policy in Ethiopia is presently being defined, but will likely lead to a greater degree of

private participation, and some attempt for greater incorporation of the informal sector


into the formal. The only official body dealing with dairy policies is the Dairy

Development Advisory Board, whose sole task is the allocation of WFP milk powder-

generated funds towards dairy development. While no dairy policy is stated by the

Board, effective policy regarding areas of importance may be implied by its choice of

projects. WFP funds have to date gone to projects aimed at forage development,

expansion of veterinary and AI services, and the supply of feeds and veterinary inputs.

The disbursement and effective use of these funds, however, have been very limited, an

outcome which has been blamed on poor institutional management and an over-emphasis

on state farms (Pronk and Abiy 1990).

Official producer prices paid by the DDE were finally raised in May 1993, from

0.65 EB/1 to 1.0 EB/1. Supply to the Shola plant at that time was very low, about 1/6

of its capacity. The price change only occurred after a general policy of increased

autonomy to state enterprises was put into place. This granted to parastatal managers the

freedom to set prices and choose markets for suppliers and buyers. Milk price changes

up until then had required approval from the supervising Ministry of State Farms

Development, and even then that authority was not certain, at times requiring signals

from higher authorities.4

By mid-1994 there was speculation among dairy officials that the loss-making

state dairy farms would soon be privatized, but no public decision has been made at the

time of this writing. The responsibility of the DDE for the state farms has been the

largest source of losses in the enterprise. In early 1995, however, it was announced that

4 Chapter 5 discusses the problem of institutional authority in greater detail.


privatization in general of state institutions would begin, and it can be assumed that the

dairy farms would in time be subjected to this process.

Policies regarding livestock services have shifted somewhat towards recognizing

the vital role of urban producers in supplying the periurban market. The MoA office

responsible for the urban area now recognizes urban dairy producers, and will register

them on request. Artificial insemination services are now available to urban producers,

although at higher costs than in rural areas. Similarly, veterinary services are officially

available, although problems of accessibility remain.

The general demise of cooperatives has been replaced by a policy of encouraging

'user groups' (Gizaw and Aware 1992). In line with this, a group of prominent dairy

producers in late 1991 formed the Addis Ababa Dairy Producers Association (AADPA)

(mehaber). By the end of 1992 this group was thought to have enlisted 90% of all urban

dairy producers, and a large proportion of market dairy producers within a radius of 100

kms of Addis Ababa (AADPA 1992). It has been welcomed by dairy officials as a

means of approaching the urban dairy producers who were neglected and consequently

unknown to them. The driving objective of the AADPA is procurement of cattle feed

rather than milk collection.

A nation-wide project being planned at this writing, the Livestock Sector

Development Project, includes a dairy component with the stated objectives of increasing

milk production in areas with market access [periurban areas], increasing small farmer

income, and improving nutrition of rural and urban populations. Central means to these

are providing improved cattle to establish new smallholder dairy producers and to


assisting in forage development. Of 3,350 production units proposed, some 65% are

projected to be one-cow units, implying a strong orientation towards the smallest

producers. For the first time in Ethiopian dairy policy, an effort would also be made to

assist the establishment of small private processing (FAO 1993a and 1993b). Family-

based processing units would be set up in project areas to produce butter for local

markets. The establishment of user groups of producers would be encouraged to assist

these processing units, and also to organize the marketing of liquid milk. This project,

however, has been designed by a donor (the FAO) and although it is under review by

GoE authorities, may reflect mainly donor priorities.

Active policy development does appear to be occurring among GoE

agricultural/dairy officials. A general policy outline under consideration by the Ministry

of Agriculture proposes to:

1) Set up a dairy board with a broader mandate to oversee all aspects of dairy

production and marketing, including regulation of both producer and consumer prices,

with more frequent reviews of prices.

2) Increase aggregate milking herd size.

3) Increase milk productivity, with the provision for more research towards this


4) Establish quotas and import duties on dairy imports.

5) Develop the marketing and processing infrastructure, including establishing

more collection centers and small processing facilities.


6) Consider land use policy reform in order to provide greater incentives for

investment in dairy.'

Unlike the FAO proposal, these proposals suggest that not a great deal has

changed in official views towards dairy development. Beginning with a new and more

powerful regulatory body, they depend on a continued central role of official institutions.

They may also neglect the more basic function of enabling the dairy markets, from the

majority producers up, to operate more profitably and with less risk. It appears likely,

however, given the importance of donor priorities, that more market-oriented policies

will develop as officials re-learn their roles in an increasingly liberalized economy.

Other Public Policies Affecting Dairy Markets

Registration Policies

Dairy producers are required to register with two agencies: the Inland Revenue

Service (IRS), and the Ministry of Agriculture (MoA). Although registration with the

revenue service rarely occurs, producers generally are registered with the MoA because

of the potential benefits, in terms of feed quotas, that in the past could be obtained.

Those benefits have now largely disappeared. Dairy marketers and processors must be

registered with the Inland Revenue Service (IRS) and the Ministry of Trade.

5 The document under consideration is confidential and was not made available.
The objectives as described were reported by dairy officials.

General Tax Policies

By law, all producers are required to pay sales taxes of 5 % of gross sales of milk

and 12% of butter.6 Such taxes, however, are rarely collected. Table 3-1 shows

reported non-payment and payment of sales taxes by the producers surveyed. Large

producers in both periurban and urban areas are more likely than small producers to pay

sales taxes, with 17% and 22% reporting payment, respectively. This is likely to be due

to the greater visibility of their operations with regard to the IRS, thus reducing their

ability to avoid taxes. The high proportion of large producers who do not pay sales

taxes, however, results in overall effective tax rates of less than 1.5%.

Most small producers are able to avoid paying sales taxes. Only 7% of periurban

and 5% of urban small producers reported paying taxes. The resulting effective tax rates

on producers overall are very low, 0.7% in the case of small periurban producers, and

a negligible 0.1 % for small urban producers.

In spite of low effective taxation, for all types of producers the incentives to avoid

taxes are high. For those large producers who paid tax, payments average over EB

1,000 annually. Those small periurban producers who paid reported an average payment

of EB 706/year. This compares to agricultural shadow wages of EB 340/year. These

taxes form a strong inducement to remain in the/informal sector, as will be discussed

further under the PAM results.

6 A sales tax of 5% is also paid on purchases of inputs, including minerals,
processed feeds, and feed materials. Although hay sales are officially subject to this
tax, no producers or hay merchants reported the payment of tax. Veterinary materials
are subject to the 5% sales tax. General sales taxes of 12% are paid on the purchase
of all equipment, including construction and dairy materials (GoE 1993).

Table 3-1 Rates of Sales Tax and Tax Payment Compliance Reported by Ethiopian
Dairy Producers, by Producer Type

Producer Type Percent not Average sales Effective Effective
paying sales tax paid/year by average sales average % sales
taxes compilers tax paid/year tax

Large Periurban 83% 1500 250 1.4%

Small Periurban 93% 706 48 1.1 %

Large Urban (n=9) 78% 1048 233 0.7%

Small Urban 95% 80 4 0.1%

Source: Producer Survey, 1992/93

Import Duties

Import duties and excise taxes do not have great direct impact on producers,

although effects on collection and processing are more important. Most significant is an

excise tax on imported fuel of 30% of value, which affects tradable costs of all inputs,

as well as marketing and processing costs.

Materials directly related to dairy production, however, face few import duties.

Specialty dairy machinery can be imported duty-free, although milk cans are subject to

15% import duty. Veterinary products are not subject to any import duty, although this

policy is contradicted by reports from drugs importers. On the protection side, butter

is subject to 75% import duty, and dried milk to 10% import duty (GoE 1990).

Macroeconomic Policies

Exchange rate policies were recently dramatically altered by the GoE. For 30

years the Ethiopian Birr (EB) has been fixed officially against the USS at the rate of 2.05

EB/$. This represented an overvaluation of several hundred percent, evident in the rates

of exchange available on the informal market, which ranged from 6 to 8 EB/$. In

October 1992 a devaluation was announced to 6 EB/$. Subsequent to the initial

devaluation, a biweekly US$ auction has been established by the central bank, fetching

prices of 6 to 6.5 EB/$.

All banks in Ethiopia are in the public sector, and interest rates are fixed

administratively. Long-term agricultural loans in 1993 were charged a rate of 12%

annually, with short-term loans of less than 12 months granted at 11%. Before the

currency devaluation of 1992, all loan interest rates were 4% less than the rates

subsequently. Although such loans are officially available to smallholder dairy

producers, regulatory requirements generally put them out of reach. Strict requirements

are imposed on lenders in the form of formal project proposals, documents for land

rights, building and equipment plans, and proforma invoices for all purchases of material

and equipment. Most dairy producers do not have or prefer not to expend the resources

to meet these requirements.

Land Tenure Policy

All land in Ethiopia is considered to be public land, to which the government

grants usufruct rights in various forms. Although land tenure procedures have changed


somewhat under the new regime, the government's control of land has not been altered


Under the former regime, a highly centralized system of land allocation applied.

A person seeking land would present his needs and reasons to the local administrative

authority (wereda) and PA, which have authority to grant usufruct rights to specific

parcels of land; the size depending on the purposes. The allocation required confirmation

by the Ministry of Urban Development in the case of urban land, or the Ministry of

Agriculture in the case of rural land (GoE 1975). A flat rate of EB 50/HA/year was paid

as a 'land service fee'. Under current rules, the procedure remains the same, except that

central confirmation is no longer required. Local authorities thus have a great deal of

power in the allocation of land.

Land transactions officially pertain only to the property built on the land, although

usufruct rights accompany the property in such a transaction. By the trading of fixed

assets, usufruct rights to land can thus be effectively traded. A sales tax of 4% of the

value of the sale is paid to the local wereda. This transfer of usufruct land rights

in fixed asset transactions has allowed an informal land market to operate, even for

vacant land. In such cases, buyer and seller agree informally on a price for the piece of

land, after which the buyer will build a small shack, often only a framework of poles,

on the land to be bought. This structure is then traded officially at a price which reflects

the full value of the land on which it is located. This can apply even to parcels of land

many hectares in size. In prime dairy-producing areas in the Addis Ababa periurban

area, the land traded in this way fetches prices of EB 10,000-20,000/HA.78

The important role of local authorities in land allocations has allowed them to

collect informal rents, and has allowed land transactions to occur in ways not prescribed

by law. Wereda officials are reportedly bribed frequently in order to obtain agreement

for the allocation of un-allocated land. Such unofficial costs are reported to be in the

range of EB 5,000/HA of land.

Land rights are leased in a more direct manner. Traditionally, land is rented for

agricultural purposes in return for one third of the production. Cash rental is also

observed in rural areas, with current rates of some EB 1,600/HA/year in the Addis

Ababa periurban area. The only use of land rented in these ways allowed by the renter

is the cultivation of annual crops. Pasture land is reported to be rarely rented. Such

7 At an exchange rate of EB 6/US$, these are $1600-$3333/HA.

8 In April 1994, the administration of the Addis Ababa urban area issued new
regulations stipulating lease payments for landholders in the urban area (GoE 1994a).
Under these regulations, new lease payments would be required from landholders who
use their land for business activity. Whereas previously only the land service fee was
paid, in future payments would be made to the administration, the amount depending
on the size of land, the level of infrastructure provided, and the type of activity being
pursued. It is unclear, therefore, to what extent these regulations will affect urban
dairy producers, who, nearly without exception, carry out 'back-yard dairying.'
Further, unless urban dairying is regarded as 'industry', it is uncertain under what
category it will fall, since agricultural activities are not recognized by the regulations.
Finally, the public protests which arose at the issuance of these proclamations, under
which lease payments can be as high as EB 6000/sq meter/year, suggests that the
regulations may be reconsidered. At an exchange rate of EB 6/US$, this is $1,000/sq


rental transactions are rarely recorded or recognized officially. Official transactions

involving business or residential property occur mainly in urban areas.

Given the land resource base upon which dairy production must directly or

indirectly depend, changes in land rights are likely to shift incentives for patterns of

production. The potential savings available from more direct use of land resources

through shifts to rural production, however, will be offset by higher milk collection

costs. Management and technology considerations come into play as well. Given that

a rural resource base is supplying an urban output market, the choice of production

location will determine what type of material, whether input or output, will be

transported. While periurban production requires the transport of highly perishable milk

or milk products, urban production requires only the transport of durable feed materials.

Dairy Production in the Addis Ababa Milkshed

The following description of dairy production is based on the producer survey

carried out in 1992/93. As was described in the previous chapter, the survey was not

over a random sample, but instead a grouped-accidental sample accompanied by a

rigorous but subjective reliability deselection method. The data analysis was thus based

on responses from 77 small urban producers, 44 small periurban producers, 9 large

urban producers and 6 large periurban producers. While the large producer samples are

small, they represent a proportion of the total population as reported by the AADPA.

With only 131 large urban producers and 46 large periurban producers reported operating

in the Addis Ababa area, the samples represent 7% and 13%, respectively, of the total


populations. Further, given the reliability indicators used in the sample selection, the

researcher believes that the means calculated from the responses are representative of the

producer group populations.

Periurban Dairy Producers

In periurban regions, producers are generally rural farmers, whose primary

activities are agricultural. The use of animal traction, and accompanying traditions of

animal husbandry, allows dairy production to be incorporated easily into existing

agricultural practices. A minority of producers live in or near small towns and pursue

wage labor. No large periurban producers reported earning wage labor, while only 19%

of small periurban producers reported doing so. Table 3-2 shows relative household

sizes of the dairy producers surveyed.

Major crops in the area include wheat, barley, lentils, and oats. In some areas

of lower elevation, the traditional Ethiopian grain teff (Eragrostis tef) can be found.

Even those producers in small population centers generally have access to agricultural

land. Table 3-3 shows relative average land holdings by producer type. In addition to

land allocated to households, communal grazing land is available to producers.

Urban Producers

Urban producers in this study are considered to be those who live in the Addis

Ababa periurban area and have no access to agricultural land. They use zero grazing or

semi-zero grazing practices, depending on the level of street-side grazing

Table 3-2 Composition of Ethiopian Dairy-Producing
Households, by Producer Type

Average Adult Adult Children
Males Females < 16 yrs

Large Periurban 4.7 2.0 5.2

Small Periurban 2.7 2.4 3.1

Large Urban 3.0 2.4 3.7

Small Urban 2.5 2.5 3.0

Source: Producer Survey, 1992/93

Table 3-3 Average Land


Holdings in Hectares, Ethiopian Periurban Dairy



Food Crops


Large Periurban 8.0 0.4 2.9 11.3

Small Periurban 2.3 0.1 1.3 3.7

Source: Producer Survey, 1992/93


employed. They tend to rely more heavily on the income from their dairy activities. Of

those surveyed, 20% of large producers engaged in wage labor, while 30% of small

producers did so. For urban producers, many of whom are relatively recently

unemployed due to the considerable restructuring occurring in the Ethiopian economy,

dairy production is a means to generate income and stabilize cash flow.9

Urban Production Units

Production is carried out in 'backyard' operations, utilizing whatever space is

available in residential compounds. Compounds are not generally designed or enlarged

to accommodate dairy production. Milk sheds of larger urban producers average 275 m2

in area, while smaller producers' milk sheds were found to average 39 m2. Milks sheds

normally are simple structures of corrugated metal sheeting or mud and wattle and

floored with blocks of stone. Aside from milk sheds, fixed inputs are minimal, including

a few buckets, milk cans/jerry cans, a water barrel, and brooms. A few large urban

producers reported employing a water pump, and a sprayer for application of pesticides.

Disposal of animals wastes is carried out through open troughs which spill into

local runoff drainage systems. Producers have reported conflicts with neighborhood

authorities over disposal of wastes. Although smaller producers generally utilize or

market most of the manure, larger producers dispose of most of the waste produced.

Areas for exercise of animals are rarely provided, although some producers graze their

9 Stabilizing cash flow is likely to be an important objective for many urban
producers. Dairying contributes to this objective by providing cash flow throughout
the year.


animals for up to 8 hours a day on street-sides. Survey results show that 21 % of small

urban producers graze their animals in this manner, but no larger producers report doing

so. As it is widely known by producers that this practice exposes the animals to

additional health risks, such as increased contraction of liver flukes, it is possible that

such grazing is under-reported. Dairy animals are commonly seen grazing in Addis

Ababa neighborhoods.

Table 3-4 shows herd sizes and composition by producer type. Two types of

cattle are identified. Local cattle refers to indigenous breeds such as Zebu and Arsi,

which are resistant to disease but produce relatively little milk. Cross-bred cattle are

mixes of local with exotic grade dairy cattle. The primary exotic breed observed is

Friesian, although some Jersey mixes are evident. The degree of exotic blood in animals

is unknown to producers due to unsystematic breeding practices, and inadequate artificial

insemination (AI) services. These issues will be discussed in more detail.

No urban producers reported holding any cattle of local breeds. Because urban

producers carry out back-yard dairying for income generation, they choose to use the

higher-yielding cross-bred cattle exclusively. Periurban producers who carry out their

dairy activities within a traditional mixed farming system prefer to hold some cattle of

local breeds.

Within the categories defined for this survey, large urban producers have an

average of 14.3 cows each, most of which are milking at any given time. They also


Table 3-4 Ethiopian Dairy Herd Composition,

by Producer Type

Cattle/Household Large Small Large Small
Periurban Periurban Urban Urban

Cows (local) 2.1 0.5 0.0 0.0

Cows (x-bred) 15.4 2.5 14.3 2.3

Total Cows 17.3 3.0 14.3 2.3

Other Cattle (local) 2.5 0.6 0.0 0.0

Other Cattle (x-bred) 16.4 3.0 12.0 2.6

Total Other 18.9 3.6 12.0 2.6

Total local 4.5 1.1 0.0 0.0

Total x-bred 31.8 5.5 16.3 4.9

Total herd 36.3 6.6 16.3 4.9

Source: Producer Survey, 1992/93

have 12.0 other cross-bred animals, mostly heifer calves. These also include an average

of 1.8 cross-bred bulls. Small urban producers average slightly more than 2 cows each,

and 2.6 other cross-bred animals. Thirty five percent of small urban producers own only

a single cow. Approximately one in ten small urban producers owns a bull.

Periurban Production Units

Periurban production units resemble those in urban areas, although adapted to

rural conditions and practices. All use open grazing practices. While shed sizes are the

same (266 m2 for larger producers and 39 m2 for smaller producers), animals are not


generally confined to milk sheds during the day, but graze instead. Construction

resembles those in urban areas. Periurban producers use the same fixed inputs, although

fewer, than urban producers. Animals wastes are more heavily utilized, particularly for


Periurban producers tend to have local as well as cross-bred animals, because of

the mixed farming context of their production. Local cows are needed to breed the oxen

required for plowing. Local animals also produce milk with higher fat content, more

suited for the home butter production in which many producers engage. Of an average

of 17.5 cows owned by large periurban producers, 2.1 are local breed on average.

Large producers hold an average of 1.3 cross-bred bulls. Small periurban producers

average 3 cows each, one of which is local in every other herd. Twenty seven percent

of small periurban producers have only one cow.

Labor Use

Although household labor is important in dairy production, a substantial

proportion of labor is hired, as shown in Table 3-5. This is particularly true among

large producers, both periurban and urban, who utilize more hired than household labor.

It should be noted, however, that there is a great deal of variation in these averages, and

many small producers use no hired labor at all. This is particularly true among small

urban producers, 70% of whom use no hired labor. Tasks for which hired labor is

generally used includes milking, shed cleaning, and feeding of animals. Marketing of

milk is generally carried out by members of the household.

Table 3-5 Labor Use

by Ethiopian


Dairy Producers, Hours

Hours per Month

Average Male Female Hired Hired HH Not
Hrs/Month/HH Family Family General Herder Using Hired
Labor Labor Labor Labor Labor (%)

Large Periurban 118 76 370 340 0%

Small Periurban 81 80 77 167 30%

Large Urban 212 88 616 0 22%

Small Urban 94 90 74 9 70%

Source: Producer Survey, 1992/93

Further, labor hired to herd cattle generally falls under a separate category based

on the type of contract formed. Under this system, children hired from poorer families

or relatives are paid minimal cash wages but are provided with food and lodging.10

Many hours of labor are thus supplied to the producer at very little direct cost.

Although dairy tasks are described as traditionally women's work, producers

reported approximately equal use of male and female household labor, except in the case

of larger producers. Much of the male labor reported, however, was carried out by

young men or boys.

10 A practice called g'it'ir described by Dessalegn Rahmato (1991) as child
contracting, occurs mainly during periods of severe social stress, such as famine.
The child-herder labor observed during the survey shares some but not all of the
characteristics of this practice.

per Month


An ILCA study of cross-bred cattle productivity in a region just south of Addis

Ababa reported Friesian/Zebu (50%) crosses to yield on average 1,913 kgs annually

(Kiwuwa et al. 1983). The survey results show that producers reported yields somewhat

less than these, except in the case of large urban producers. Table 3-6 presents average

calving intervals, milk yields and feed conversion rates." Small producers and large

periurban producers report yields of between 1,450 and 1,750 kgs per cow annually,

lower than the ILCA study. Since it is likely that the majority of dairy cattle in all

producer groups are more than 50% grade (the grade of cattle in the above ILCA study),

these yields appear particularly poor. Calving intervals reported were in all cases longer

than the optimal 12-13 months although they did not differ substantially from the 15

months (458 days) reported for Friesian/Zebu crosses in the ILCA productivity study

(Kiwuwa et al. 1983).

Large urban producers report higher yields of nearly 2,700 kgs per year. These

yields are much better than those of other groups, and better even than the ILCA study

During the survey milk yields were measured in two ways; by reported daily
milking quantities and by quantities sold and consumed. Large producers generally
reported higher figures for milking than sales, while small producers reported higher
sales than quantities milked. The figures used in the analysis were a simple mean of
these two quantities. Producers may have various reasons for over or under-reporting
yields and sales, depending on their perception of tax risks posed by interviewers and
their desire to exaggerate the productivity of their animals.

Table 3-6 Milk Yields, Lactation and Calving Intervals, and Feed Conversion Rates
Reported by Ethiopian Dairy Producers Cross-Bred Cattle Only

Large Small Large Small
Periurban Periurban Urban Urban

Lactation (mos) 12.0 11.4 12.3 9.8

Calving Interval (mos) 14.6 16.8 16.2 15.0

Milk Yield/Day (kgs/day/cow) 5.9 6.1 9.8 6.3

Milk Yield/Lactation 2,120 2,065 3,613 1,864

Milk Yield/Year' 1,737 1,476 2,679 1,488

Feed Conversion Rateb (MJ of 10.1 9.6 7.7 12.5

a Annual yield = (yield per lactation) x (12 mos/calving interval in mos)
b Megajoules of metabolizable energy required to produce 1 liter of milk. Accepted benchmark
figure is 5 MJ/lt)
Source: Producer Survey, 1992/93

cattle." It is possible that large urban producers have cattle with higher exotic content,

due to better access to artificial insemination, but the degree of crossing is not

measurable. To better understand the reasons for this difference in yield results, feed

conversion rates were calculated. These are measured by calculating the amount of

12 Although the large urban results are taken from a sample of only 9 producers,
these represent 7% of the total large urban producer group. Although further study
might show the yield differences to be smaller than indicated here, the scale of the
differences leads this researcher to believe that the results cannot be dismissed as a
statistical anomaly.


metabolizable energy needed to produce one liter of milk (described in more detail in

Appendix A). Shown in Table 3-6, conversion rates range from 7.7 MJ/L for large

urban producers to 12.5 MJ/L for small urban producers. These are substantially higher

(indicating less efficient use of energy) than the accepted benchmark figure of 5 MJ/L.'3

Further, the differences in feed conversion rates between large and small urban producers

parallel their yield differences.1 This indicates that high yields among large urban

producers are not the result of simply higher levels of feeding, but result instead from

either higher grade cattle or relatively lower levels of morbidity. Further, since in all

groups feed efficiency is relatively poor, and since their cattle can be assumed to be

higher than 50% grade, this low feed efficiency can be attributed to generally high levels

of morbidity in all groups. The impact of disease, perhaps made worse by poor

accessibility to veterinary services, may be causing to low productivity, higher rates of

feed use, and lower than optimal profitability.

These productivity results have two important implications. The first is that there

exists, among large urban producers, the demonstrated potential for higher yields using

the same feed materials and generally the same cattle, although of potentially higher

grade. Lacking additional data, no further conclusions can be drawn as to the reasons

for these higher yields, but they present a compelling point for further study. Secondly,

1 It is interesting to note that Purcell and Stotz reported in the 1970s that feed
conversion ratios were 50%-90% higher than normal. The current feed conversion
ratios are even higher.

14 While large urban producer yields are 80% higher than small urban producers,
feed conversion rates are 62% higher. Given the uncertainty of reported feed use and
milk yields, these differences must be considered comparable.


the generally poor feed conversion rates in all groups points to high levels of morbidity,

confirming the observed poor accessibility to veterinary services. Improved veterinary

services which lead to better feed conversion rates can be expected to not only improve

profitability, but by reducing derived demand for feed, to make better use of available

feed resources on the market. Since shortage of feed on the market was the constraint

most cited by producers, improved services may be one key to reducing a primary

production constraint. Feed markets are discussed below.

Feed Usage and Feed Markets

Producer Feed Usage

Dairy animals are fed a variety of fodder materials, the mix depending largely on

availability. Table 3-7 shows the average annual kilograms of each type of feed per TLU

for each producer type, percent of annual feed expenditure, as well as the percent of

metabolizable energy provided by roughage and concentrate. There is great variability

among these averages, as many producers reported feeding only one or two materials,

as a result of sporadic market availability or bureaucratic obstacles to feed purchases.

The most important feeds are natural hay, wheat bran and middling, and niger (Guizotia

abyssinica) and cotton seed cakes. Also observed being used are beer waste, both from

home beer-making and commercial breweries. Commercial mixed concentrates, made

up of mainly grain, bran and oilseed cake, are not an important source of feed except in

the case of small periurban producers. This contrasts sharply with dairy production in

Kenya, where commercial concentrates are in many cases the only source of concentrate

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs