Milk in the city: profiles and development paths for urban dairy holders in Ethiopia

Figures & data

Figure 1. Four leadership styles as set out by the Situational Leadership Theory (elaborated from Hersey & Blanchard, 1969).

Figure 2. Radar chart showing how the different identified farm clusters score relatively in terms of each other for the discriminatory variables and compared to the average dairy holder.

Note: For each discriminatory variable, the relative scores were calculated by dividing the average value of each farmer cluster by the maximum observed value amongst all farmer clusters.

Table 1. Average values (mean ± SE) of the discriminatory variables used in the cluster analysis for the identified farmer clusters.

	Cluster 1 (N = 63)	Cluster 2 (N = 73)	Cluster 3 (N = 38)	Cluster 4 (N = 93)	Cluster 5 (N = 36)
Produced milk:feed cost ratio (Birr/L)	0.2^b ± 0.0	0.2^b ± 0.0	0.2^b ± 0.0	0.2^b ± 0.0	0.6^a ± 0.0
Share of produced milk being processed during fasting (%)	96.2^a ± 0.9	51.4^b ± 3.1	0.5^d ± 0.5	24.6^c ± 2.0	44.2^b ± 5.0
Share of produced milk being sold during fasting (%)	1.6^c ± 0.8	44.0^b ± 2.7	0.5^c ± 0.5	68.9^a ± 2.0	47.0^b ± 4.7
Minimum milk producer price (Birr/L)	10.1^bc ± 0.2	7.0^d ± 0.1	9.6^c ± 0.3	11.4^a ± 0.2	10.3^b ± 0.2
Milk price fluctuation (Birr/L)	3.0^bc ± 0.2	6.0^a ± 0.1	3.5^b ± 0.2	2.0^d ± 0.1	2.7^c ± 0.2

^a–dMeans within a row with different superscripts differ significantly (P < .05).

Table 2. Socio-demographic characterisation of the sampled dairy farming households and their heads: mean ± SE for the continuous variables and frequencies for the categorical variables.

	Cluster 1 (N = 63)	Cluster 2 (N = 73)	Cluster 3 (N = 38)	Cluster 4 (N = 93)	Cluster 5 (N = 36)
Household size (number)	4.8^ab ± 0.3	4.4^b ± 0.3	4.1^b ± 0.4	5.4^a ± 0.2	5.2^ab ± 0.5
Household income (Birr/member/month)	2,846.7^b ± 295.5	2,513.4^c ± 422.9	5,429.2^a ± 666.0	1,899.3^c ± 162.8	1,434.0^d ± 292.5
Dependency ratio^1	0.6^a ± 0.0	0.5^a ± 0.0	0.5^a ± 0.05	0.5^a ± 0.0	0.5^a ± 0.1
Share dairy income in total household income (%)	67.0^b ± 3.2	87.4^a ± 2.7	64.7^b ± 3.75	65.1^b ± 2.0	55.2^b ± 4.2
Close network size of household^2 (%)
1–2 people	3.2^b	53.4^a	0^b	62.0^a	5.6^b
3–4 people	3.2^b	17.8^a	0^b	5.4^b	2.8^b
5 or more people	93.7^b	21.9^a	100^b	30.4^a	91.7^b
Household participation (%)
Neighbourhood committee	19.1^ab	8.2^bc	31.6^a	2.2^c	0^c
Religious group	7.9^a	5.5^a	2.6^a	11.8^a	13.9^a
Finance group	85.7^a	16.4^c	100^a	52.7^b	88.9^a
Age distribution head (%)
18–39	34.9^ab	30.1^bc	50.0^a	18.3^c	16.7^bc
40–59	54.0^a	45.2^a	44.7^a	60.2^a	47.2^a
60 and above	11.1^bc	24.7^a	5.3^cd	21.5^ab	36.1^a
Female head (%)	28.6^a	19.2^ab	10.5^b	15.1^b	5.6^b
Marital status head (%)
Married	62.9^bc	61.6^b	52.6^b	78.5^a	80.6^ac
Single	8.1^bc	20.6^a	21.1^ab	5.4^c	8.3^abc
Divorced/separated/widow	29.0^a	17.8^a	26.3^a	16.1^a	11.1^a
Educational background head (%)
No schooling	19.1^b	36.1^a	13.2^b	28.0^ab	13.9^b
Informal education	11.1^a	12.5^a	7.9^a	11.8^a	25.0^a
Junior primary	14.3^a	16.7^a	18.4^a	15.1^a	30.6^a
Senior primary	42.9^a	12.5^b	39.5^a	18.3^b	13.9^b
General secondary	4.8^a	9.7^a	18.4^a	18.3^a	11.1^a
Preparatory secondary	4.8^a	2.8^a	0^a	4.3^a	2.8^a
Higher education	3.2^a	9.7^a	2.6^a	4.3^a	2.8^a
Secondary/tertiary occupation head (%)	61.9^b	23.3^c	60.5^b	79.6^a	36.1^c

^a–dMeans and frequencies within a row with different superscripts differ significantly (P < .05).

¹Dependency ratio was calculated by dividing the number of non-working household members by the total household size.

²Close network size of a household was measured by asking how many people are willing to assist the household when facing difficult time periods (such as death of household head) (see Grootaert et al. 2004).

Table 3. Characterisation of the dairy farm systems: mean ± SE for the continuous variables and frequencies for the categorical variables.

	Cluster 1 (N = 63)	Cluster 2 (N = 73)	Cluster 3 (N = 38)	Cluster 4 (N = 93)	Cluster 5 (N = 36)
Engagement in dairy farming (years)	7.3^b ± 0.7	7.9^b ± 0.7	6.7^b ± 0.8	9.6^a ± 0.6	9.7^ab ± 1.2
Dairy herd size (number)	3.5^c ± 0.2	4.5^abc ± 0.4	3.9^bc ± 0.4	4.8^ab ± 0.4	5.4^a ± 0.6
Share of local breeds (%)	0^bc	0^c	0^abc	2.3^ab ± 1.1	2.5^a ± 1.4
Main motivation for dairy farm (%)
Cash income	67.2^b	86.3^a	97.4^a	60.2^b	36.1^c
Investment opportunity	18.0^b	6.9^c	0^c	23.7^b	41.7^a
Saving opportunity	4.9^a	0^a	2.6^a	3.2^a	11.1^a
Family tradition	9.8^a	6.9^a	0^a	11.8^a	11.1^a
Household heads being sole owner of farm (%)	38.1^a	38.4^a	47.4^a	21.5^b	16.7^b
Relatives in dairy farming (number)	1.2^a ± 0.2	2.1^a ± 0.4	1.0^a ± 0.2	1.3^a ± 0.2	1.5^a ± 0.2
Fellow dairy farmers known personally (number)	7.3^c ± 1.7	13.7^a ± 1.3	1.6^d ± 0.6	7.6^b ± 0.5	12.6^a ± 1.8
Feed sellers known personally (number)	7.4^a ± 0.5	4.1^b ± 0.3	8.3^a ± 0.4	3.6^b ± 0.2	3.4^b ± 0.3
Dairy buyers known personally (number)	11.3^a ± 1.1	5.1^b ± 0.8	12.8^a ± 1.0	3.2^c ± 0.2	2.4^c ± 0.4
Selling and purchasing dairy animals (%)	100^a	94.5^a	100^a	49.5^b	94.4^a
Artificial insemination (%)	85.7^bc	94.5^ab	94.7^ab	78.5^c	100^a
Livestock extension services (%)	81.0^c	97.3^a	81.6^c	94.6^ab	86.1^bc
Hired labour (%)	30.2^a	42.5^a	34.2^a	38.7^a	27.8^a
Dairy farm costs (Birr/animal/month)
Animal health	105.5^a ± 16.3	91.4^a ± 12.4	87.7^a ± 15.7	103.4^a ± 17.7	56.8^a ± 11.7
Hired labour^1	155.5^ab ± 16.2	121.5^bc ± 15.7	232.9^a ± 34.3	118.5^c ± 13.6	98.6^c ± 10.1
Feed	1396.0^a ± 147.9	923.4^b ± 44.7	1504.3^a ± 128.4	844.3^b ± 42.2	263.7^c ± 31.1

^a–dMeans and frequencies within a row with different superscripts differ significantly (P < .05).

¹This cost is only for those farms that use hired labour.

Figure 3. Average daily milk production and feed cost for the different identified farm clusters and the average dairy holder (indicated by the black line).

Table 4. Feeding practices within the identified farm clusters: mean ± SE for the continuous variables and frequencies for the categorical variables.

	Cluster 1 (N = 63)	Cluster 2 (N = 73)	Cluster 3 (N = 38)	Cluster 4 (N = 93)	Cluster 5 (N = 36)
Number of feed types used	3.5^cd ± 0.2	3.9^c ± 0.2	3.1^d ± 0.2	4.3^b ± 0.1	4.6^a ± 0.2
Farmers who report to feed (%)
Green grass	1.6^c	16.4^ab	5.3^bc	20.4^a	8.3^abc
Grass hay	36.5^d	54.8^c	18.4^d	69.9^b	94.4^a
Crop residues	82.5^a	80.8^a	89.5^a	80.7^a	36.1^b
Wheat bran	95.2^a	97.3^a	100^a	97.9^a	94.4^a
Atella	73.0^b	95.9^a	73.7^b	95.7^a	86.1^ab
Salt	33.3^b	30.1^b	18.4^b	31.2^b	77.8^a
Prepared feed mixture	1.6^b	2.7^b	0^b	12.9^a	19.4^a
Molasses	1.6^a	5.5^a	5.3^a	1.1^a	0^a
Fresh cultivated forage	1.6^a	2.7^a	5.3^a	1.1^a	0^a
Leftovers (food, vegetable)	17.5^b	4.1^c	0^c	14.0^b	44.4^a
Share of purchased feed in diet (%)	94.7^bc ± 1.3	97.0^b ± 0.9	100^a	93.5^c ± 1.1	91.2^c ± 2.0
Farmers reporting feed shortage problem (%)	82.5^bc	90.4^b	76.3^c	98.9^a	100^ab
Reported feed shortage coping strategy (%)^1
Destocking	55.8^b	24.2^c	31.0^c	71.7^ab	86.1^a
Storage of feed	26.9^b	33.3^b	55.2^a	20.7^b	5.6^c
No coping strategy	5.8^b	34.9^a	0^b	3.3^b	8.3^b

^a–dMeans and frequencies within a row with different superscripts differ significantly (P < .05).

¹Percentages reported are only applicable for those farms that indicated that they faced feed shortage problems.

Table 5. Correlation matrix of marketing variables and fasting coping strategies.

	Variable	1	2	3	4	5
1.	Produced milk being sold during fasting (%)	1
2.	Produced milk being processed during fasting (%)	−0.54**	1
3.	Size of milk price fluctuation (Birr)	−0.14*	0.14*	1
4.	Minimum fresh milk price (Birr/L)	0.16**	−0.13*	−0.90**	1
5.	Maximum fresh milk price (Birr/L)	0.07	−0.03	0.01	0.44*	1

*Correlation is significant at the 0.05 level (2-tailed).

**Correlation is significant at the 0.01 level (2-tailed).

Table 6. Milk use and fresh milk marketing strategies among the identified farm clusters: mean ± SE for the continuous variables and frequencies for the categorical variables.

	Cluster 1 (N = 63)	Cluster 2 (N = 73)	Cluster 3 (N = 38)	Cluster 4 (N = 93)	Cluster 5 (N = 36)
Produced milk being consumed during non-fasting (%)	5.1^a ± 0.8	2.4^bc ± 0.6	1.9^c ± 0.9	2.8^b ± 0.5	6.0^a ± 1.6
Produced milk being consumed during fasting (%)	0.5^c ± 0.4	2.7^b ± 1.0	3.4^c ± 2.7	7.0^a ± 1.0	4.9^b ± 1.7
Produced milk being processed during non-fasting (%)	3.3^a ± 1.7	0.4^a ± 0.3	0.1^a ± 0.1	0.3^a ± 0.3	1.1^a ± 1.1
Farmers giving away milk for free to family or friends (%)	3.2^c	5.6^bc	5.7^bc	32.3^a	17.1^ab
Farmers processing milk (%)	100^a	98.6^a	5.3^c	73.1^b	80.6^b
Farmers selling butter (%)	76.2^a	69.9^a	2.6^b	8.6^b	11.1^b
Fresh milk buyer (%)
Neighbours	39.7^b	34.3^b	23.7^b	62.4^a	44.4^ab
Cafés, restaurants or hotels	50.8^bc	63.0^ab	79.0^a	54.8^bc	41.7^c
Traders	7.9^bc	15.1^b	0^c	11.8^b	36.1^a
Farmers who have more than 1 type of milk buyer (%)	4.8^c	15.1^b	2.6^c	33.3^a	22.2^ab
Arrangement for fresh milk (%)
Written	3.2^b	2.7^b	23.7^a	3.2^b	0^b
About milk volume	12.7^c	58.9^b	2.6^c	74.2^a	5.6^c
About milk price	7.9^b	82.2^a	5.3^b	72.0^a	13.9^b
About milk quality	82.5^ab	50.7^c	94.7^a	21.5^d	77.8^b
Payment frequency for fresh milk (%)
Upfront	20.6^a	0^b	2.6^b	0^b	0^b
Daily	38.1^b	19.2^c	60.5^a	18.3^c	44.4^ab
Weekly	22.2^b	61.6^a	5.3^c	25.8^b	30.6^b
Less than weekly	20.6^c	23.3^c	34.2^bc	65.6^a	41.7^b

^a–dMeans and frequencies within a row with different superscripts differ significantly (P < .05).

Table 7. Short summary of the identified urban dairy systems.

Cluster	Cluster name	Input efficiency and market integration	Main characteristics
1	Processing female farms	Low milk:feed ratio Medium-strong market integration	Significant contribution of female labour force Processing specialisation in milk Attention for milk and feed quality Extended feed supplier network
2	Surviving farms	Low milk:feed ratio Weak market integration	Dairy is main or only income generating activity Illiterate or poorly educated household heads Adverse and volatile milk and feed prices
3	Educated male entrepreneurs	Low milk:feed ratio Medium market integration	Educated, young, male entrepreneurs Dairy as a lucrative business Attention for milk and feed quality Extended feed supplier network Main milk buyers are cafés, restaurants and hotels
4	Established output-efficient farms	Low milk:feed ratio Strong market integration	Long dairy experience Large herd size Dairy as a side business Extended milk buyer network
5	Established output-input-efficient farms	High milk:feed ratio Strong market integration	Long dairy experience Large herd size Attention for milk and feed quality More diversified dairy diets and hay as main basal feed Extended milk buyer network

Figure 4. Required intervention styles for the different identified farm clusters.

Hersey, P & Blanchard, KH, 1969. Life cycle theory of leadership. Training and Development Journal 23(5), 26–34.

 Google Scholar

Grootaert, C, Narayan, D, Jones, VN & Woolcock, M, 2004. Measuring social capital: An integrated questionnaire. World Bank working paper 18. The World Bank, Washington, DC.

 Google Scholar